iPhone/Zune/iPod & More Prediction Thread [Archive]

Arun

27-Apr-2009, 15:54

Go wild. FWIW, here are my predictions:

iPhone 2Q09 @ ~$99
Same app processor as iPod Touch 2G
3.2MP with ISP - see product brief (http://www.ovt.com/uploads/parts/OV3642_PB_(1.1)_web.pdf)
Qualcomm MSM6290 (http://www.eetasia.com/ART_8800489081_499488_NP_950125aa.HTM) Baseband
802.11g/BT2, unknown chip(s)

Zune HD 3Q09
Tegra (http://www.nvidia.com/page/handheld.html), who knows which SKU(s)
SiPort SP1010 (http://www.siport.com/products/sp1010.html) for HD Radio
Atheros AR6002 (http://www.en-genius.net/site/zones/networkZONE/product_reviews/netp_111207) 802.11g
AMOLED

iPhone 4Q09 @ ~$299
App processor with SGX & VXD for 1080p decode
5MP with ISP & BSI - see product brief (http://www.ovt.com/uploads/parts/OV5642_PB(1.0)_web.pdf)
Qualcomm MSM6290 (http://www.eetasia.com/ART_8800489081_499488_NP_950125aa.HTM) Baseband
802.11n/BT2/etc. BCM4329 (http://www.broadcom.com/products/Bluetooth/Bluetooth-RF-Silicon-and-Software-Solutions/BCM4329)
AMOLED

P.S.: In the unlikely event this gets picked up by a third party and someone goes nuts over my prediction of Infineon losing the baseband socket to Qualcomm, please note that this is mostly speculation. However, Infineon will nearly certainly not have its HSUPA solution ready for products released in June, and their performance/spectrum efficiency remains much lower than Qualcomm's which is a problem for carriers with the device generating the most data traffic in the world today. Finally, the main reason I'm betting on Qualcomm is that an analyst who uniquely leaked the iPod Shuffle refresh back in December also indicated (http://www.appleinsider.com/articles/08/12/31/new_shuffle_cheaper_iphone_and_macbooks_seen_in_20 09.html) that Qualcomm had won the iPhone socket.

As for the Verizon iPhone rumors in 2010, I believe those would likely be based on Qualcomm's multimode HSPA+/EV-DO/LTE chipset sampling in mid-2009. It should be usable for a phone launching in 4Q10 (I believe that's what Nokia will use too), and has the advantage of being a viable solution for every single non-WiMax network in the world. No exception, not even China Mobile. It's probably not a cheap chip, but at least there is no redundant application processor on it, and remember it wouldn't be a cheap phone either... Anyway this is the only way it would make sense, but that doesn't mean anything will happen either.

EDIT: It looks like Infineon is being very bullish on next quarter's wireless revenue - so it looks like I was probably wrong at least about the 2Q09 iPhone having a Qualcomm baseband... We'll see.

Ike Turner

27-Apr-2009, 17:27

And what about other phones spéculations ? ;)

Here's mine

HTC Star 4Q09 :
Qualcomm SnapDragon QSD2850 @ 600Mhz
5Mpix cam with 720p encode
Qualcomm MSM6290 Baseband
WM 6.5

HTC SuperStar 4Q09/1Q10:
Tegra AP16 @ 600Mhz
Qualcomm MSM6290 Baseband or Ericsson EMP
5Mpix with 720p encode
HDMI out with 720P/1080P decode
Capacitive screen
WM 6.5

Arun

27-Apr-2009, 17:37

That seems reasonable, and mentioning both MSM6290 and ST-Ericsson as possibilities for SuperStar is probably a very good idea. I'd personally bet on the latter FWIW. Just two corrections:
- Snapdragon includes the baseband (not the WiFi/Bluetooth/etc. though), and it also only supports 720p decode, not encode which is still VGA.
- The phone SKU of the AP16 only supports 720p decode, not 1080p. Unlike AP15, it supports VC-1 Advanced Profile though, which is nice.

And obviously feel free to include any other flagship device you can think of... :) A couple of things that seem worth to point out there:
- LG very likely to use Tegra for WM6.5 phones, IMO. Would be interesting to see if they use it on lower-end models too.
- Samsung should probably have more OMAP3 models but seems to be the most diverse in terms of both platforms and OSes.
- Nokia likely to have a slim modem-based part of its portfolio with discrete app processors in addition to announced platforms.
- Motorola's custom OMAP3 platform with integrated baseband and Infineon RF still slated for 2H09 for first products. Two interesting posts on their Android progress... (1 (http://forum2.mobile-review.com/showpost.php?p=788204&postcount=2092) & 2 (http://forum2.mobile-review.com/showpost.php?p=789242&postcount=2105))
- SE seems to be in the toughest position of all with the least optimism from insiders, not clear what Idou is based on, probably 65nm U380 (OMAP3+EMP baseband) which still existed in January but not mentioned by ST-Ericsson at all in favour of 45nm U8500 Nomadik+Nokia baseband which comes later.

Ike Turner

27-Apr-2009, 18:01

oops forgot that QSD2850 and Tegra AP16 has only 720p Decode.

LG is realy quite lately though (nothing since the LG GM730 (MSM7200a) announcement at MWC).
Isn't Nokia going to start using Qualcomm MSM 7XXX/8XXX chipsets starting 2010 ?
see here http://www.forbes.com/2009/04/21/nokia-qualcomm-mobile-technology-wireless-nokia.html
http://www.slashgear.com/qualcomm-and-nokia-announce-us-mobile-device-collaboration-1734603/

Seems like Qualcomm is the big winner latetly (besides the 800m settelement with broacom today...).

Arun

27-Apr-2009, 18:32

oops forgot that QSD2850 and Tegra AP16 has only 720p Decode.AP16 has 720p Encode, it just doesn't have 1080p Encode. I guess that's what you meant though... :)

LG is realy quite lately though (nothing since the LG GM730 (MSM7200a) announcement at MWC).Yeah, they have been gaining quite a bit of market share though! Certainly in relative terms they are gaining the fastest right now, not sure in absolute terms though. FWIW, I think I mentioned this before but Arena uses an Infineon baseband and an AMD A250.

Isn't Nokia going to start using Qualcomm MSM 7XXX/8XXX chipsets starting 2010?For the US market, yes - and obviously not just for LTE despite that being the only thing I mentioned.

Seems like Qualcomm is the big winner latetly (besides the 800m settelement with broacom today...).Yes, and they'll have probably the strongest early positioning for LTE. However, from a fundamental point of view, I am extremely pessimistic about them in the mid/long-term. Basebands will soon become even more worthless (ASPs for deals signed in 2H08 already plunged pretty badly, and a certain UK company is partly to blame for that) and the real cost benefit of integrating them with application processors/connectivity/etc. is actually going down AFAICT. The licensing environment should also be less attractive for them in LTE.

I suspect people are going to look at Qualcomm's integrated HSPA+ solutions (including app processor, bluetooth, gps, fm, etc.) and realize discrete solutions from competitors can actually be cheaper and better for every subcomponent. We'll see, maybe I'm wrong, it's obviously hard to guess at this point!

Wishmaster

27-Apr-2009, 18:38

Arun: Edited so that this doesn't take a trillion different posts for something that isn't the original topic (devices!) and can be discussed practically anywhere else... ;)
Arun you're sure that snapdragon has only VGA encode? I thought that it was 720p encode and decode. Besides when you check qualcomms website it clearly states its 720p encode, although you have to search to find it.
This is taken out straigh from their website:Other features of the Snapdragon platform include integrated cellular broadband as well as support for Wi-Fi, Bluetooth and mobile TV; high definition video recording and playback; up to WXGA resolution (1280 x 768); sophisticated positioning with Assisted-GPS; hardware-accelerated 3D graphics; and industry-leading integration to enable smaller, sleeker devices.
It's possible that it was upgraded to 720p Encode. After all theirs is a partially software solution - however if so I fear the bitrate offered might not be very impressive. We'll see. It's also worth mentioning Toshiba never mentioned HD encode for the TG01 AFAIK...
They never mentioned 720p decode either AFAIK ;)
They did, but really not much indeed. I seem to remember someone from Toshiba mentioning it wasn't yet exposed in the prototypes at the launch event but that it eventually will be. Either way this is nothing but a small detail, why exactly are we discussing it again? Damn Ike for unwillingly turning this into yet another Snapdragon vs World discussion ;)
If so it means that Coreplayer will get full support of QTv before TG01 launches (cause they use CP as their default player for audio/video) and that is something I hope will happen sooner than later :)
Wherever you go you will always find qualcomm vs the world topic :cool: is it bad that I think that snapdragon will be great :lol:

tangey

28-Apr-2009, 11:16

iPhone 2Q09
Customised Samsung S5PC100 processor
with SGX graphics.

Zune HD 3Q09
Tegra

4Q09
iphone "pro"
which could be an iphone with better camera abilities
or could be an "upsized" iphone/itouch with larger screen
5MP camera.
Same processor and graphics as above.

2Q '10
iphone, in-house designed Soc from PA semi +
SGX graphics (MP ?)

3Q '10
Sony PSP2 Cortex A9 with multi-core SGX graphics.
Although, I'm starting to see some signs that sony may
bring out a PSP2 4Q '09 due to seeing how well the iphone
platform is doing. It is unclear how Sony will deal with
software compatiblity between current and next gen platforms.

SE seems to be in the toughest position of all with the least optimism from insiders, not clear what Idou is based on, probably 65nm U380 (OMAP3+EMP baseband) which still existed in January but not mentioned by ST-Ericsson at all in favour of 45nm U8500 Nomadik+Nokia baseband which comes later.

Idou is the Omap3 platform, IMG have been referring to the Idou....

"Imagination’s technology could be seen
in many new platforms at the show this
year. As well as appearing in the new
Samsung Omnia HD, Sony Ericcson Idou
and Palm Pre there were several......"

wco81

28-Apr-2009, 16:04

What no prognostications on the rumored tablet/netbook with multitouch?

tangey

28-Apr-2009, 16:14

What no prognostications on the rumored tablet/netbook with multitouch?

That *could* turn out to be what I conjectured for Q4 '09 (upsized iphone/itouch)

Ike Turner

28-Apr-2009, 22:53

http://www.engadget.com/2009/04/28/microsofts-pink-smartphone-could-rival-iphone-on-verizon/
http://online.wsj.com/article/SB124093915558664239.html (can't read the full article)

Who's willing to bet with me that it's just the HTC SuperStar (MS' reference design for Pink smartphone devices) ?
My humble opinion is that Pink is MS' new media platform : Zune+WM7+Marketplace+Xbox360 = Pink

darkblu

28-Apr-2009, 23:07

iTablet 3Q (read: 4Q) 09 @ ~$400
an iphone os device,
multitouch
2x Cortex A9 MP
SGX5xx,
same camera as the iphone of the day
802.11g/BT2

Mike11

29-Apr-2009, 01:40

iPhone 2Q09 @ ~$99
Same app processor as iPod Touch 2G
3.2MP with ISP - see product brief (http://www.ovt.com/uploads/parts/OV3642_PB_(1.1)_web.pdf)
Qualcomm MSM6290 (http://www.eetasia.com/ART_8800489081_499488_NP_950125aa.HTM) Baseband
802.11g/BT2, unknown chip(s)
I tend to agree. But I think we'll see a SGX530 in that model (new GPU baseline).

iPhone 4Q09 @ ~$299
App processor with SGX & VXD for 1080p decode
5MP with ISP & BSI - see product brief (http://www.ovt.com/uploads/parts/OV5642_PB(1.0)_web.pdf)
Qualcomm MSM6290 (http://www.eetasia.com/ART_8800489081_499488_NP_950125aa.HTM) Baseband
802.11n/BT2/etc. BCM4329 (http://www.broadcom.com/products/Bluetooth/Bluetooth-RF-Silicon-and-Software-Solutions/BCM4329)
AMOLED
Same app processor as above? Hm, I think just a higher clocked ARM11 and SGX540 plus VXD. And a little more RAM as in the iPhone3G, because of the exclusive video and high resolution picture features in this model, plus up to 64GB storage (2x32GB flash chips in this high-end iPhone, like the touch).

After SGX in 2009, Apple will directly switch to Cortex-A9MP and a new display resolution/size in 2010 for the high-end iPhone. So no Cortex for Apple in 2009.

wco81

29-Apr-2009, 02:29

iTablet 3Q (read: 4Q) 09 @ ~$400
an iphone os device,
multitouch
2x Cortex A9 MP
SGX5xx,
same camera as the iphone of the day
802.11g/BT2

That price seems uncharacteristically low from Apple. Especially if unsubsidized

Mike11

29-Apr-2009, 03:45

That price seems uncharacteristically low from Apple. Especially if unsubsidized
Well, the iPod touch starts at $229, so if the only hardware difference between this years next gen iPod touch and the iTablet is a bigger display (plus bigger battery and maybe more RAM because of the higher resolution), then $400 could be possible. But $499 seems more reasonable for me. If the iTablet is not just a bigger iPod touch and has a different SoC etc. then even $499 is too low.

Laurent06

29-Apr-2009, 08:22

3Q '10
Sony PSP2 Cortex A9 with multi-core SGX graphics.
Although, I'm starting to see some signs that sony may
bring out a PSP2 4Q '09 due to seeing how well the iphone
platform is doing. It is unclear how Sony will deal with
software compatiblity between current and next gen platforms.
I'd be interested to track where this rumor comes from :)

Arun

29-Apr-2009, 10:01

Wow, it's nice seeing all of you people with your hopes not yet crushed... :)
WRT the iPhone/iPod: the current version is 1,2 for the iPhone 3G and 2,1 for the iPod Touch 2G. How do you explain the existence of the following versions in various hacked files: iPhone 2,1; iPod 2,2; iPhone 3,1; iPod 3,1? If you can't explain that, it's difficult for me to understand how it could all make sense, heh.

I don't see how you could justify two new SoCs for 2009 (unless one was iTablet-only) or the next SKU in June being SGX-based unless it was announced at the same time as a new non-SGX one. And then you're incompatible with the Digitimes "3.2MP first, 5MP later" leak, which could be wrong I guess. As for the PSP2, 2H10 seems plausible (unlike 4Q09), but even that seems a tad optimistic. I definitely hope I'm wrong there and it thankfully seems likely I am.

Finally, wrt Verizon, I'd ignore everything you see on that front. The "cheap iPhone for Verizon" thing is nearly certainly bullshit. At best, it's random made up stuff. At worst, it's aimed at stock manipulation. So the MS stuff doesn't seem much more credible - and at best, what it would mean is that MS is trying to convince Verizon to have a Windows Mobile 7 ODM phone available at launch (April 2010 according to the latest leaks...) - which I think is kinda a given anyway, and they'd be talking with everyone about that and not just Verizon.

tangey

29-Apr-2009, 10:22

I'd be interested to track where this rumor comes from :)

http://www.engadget.com/2009/04/28/sony-unveiling-umd-less-psp-with-slide-out-buttons-at-e3/
http://www.pocketgamer.co.uk/r/PSP/Playstation+Portable+2/news.asp?c=12612

Additionally an acquiantence of mine has heard that a new PSP is being readied for launch. I do not know how accurate, if at all, that info is, and if true it is most likely a cosmetic refresh.

Personally, I don't think a major PSP hardware change will happen this year, unless they rush something due to a perceived threat from iphone (the itouch alone sold 3M units last quarter, iphone+itouch was 7M). IMO, it is reasonably certain that Sony has taken an SGX licence from IMG. But they havn't had enough time to use that licence.

Sorry for being OT.

Mike11

29-Apr-2009, 14:27

Wow, it's nice seeing all of you people with your hopes not yet crushed... :)
WRT the iPhone/iPod: the current version is 1,2 for the iPhone 3G and 2,1 for the iPod Touch 2G. How do you explain the existence of the following versions in various hacked files: iPhone 2,1; iPod 2,2; iPhone 3,1; iPod 3,1? If you can't explain that, it's difficult for me to understand how it could all make sense, heh.

I don't see how you could justify two new SoCs for 2009 (unless one was iTablet-only) or the next SKU in June being SGX-based unless it was announced at the same time as a new non-SGX one. And then you're incompatible with the Digitimes "3.2MP first, 5MP later" leak, which could be wrong I guess.
Hm, why can't the June iPhone (3.2MP) be SGX based?

Two SoCs for 2009:
One SGX-upgraded 2008 iPod Touch SoC for the $99 3.2MP June iPhone (iPhone 2,1) and the 2009 iPod touch (iPod 2,2). And the other SoC for the high-end (5MP) Q4 iPhone (iPhone 3,1) and the Q4 iTablet/iPod touch HD (iPod 3,1).

Arun

29-Apr-2009, 14:35

tangey

29-Apr-2009, 15:08

I don't see how you could justify two new SoCs for 2009

My thoughts are that there will only be one new Soc in '09 used for multiple products, including the much speculated itablet ithingy

Okay, so, err - why was the iPhone 3G 1,2 and the iPod Touch 2G 2,1? The two main differences, AFAICT, are the application processor and the WiFi/Bluetooth chip.

Is it not just the case that the latest itouch is clocking the processor quicker as opposed to being a different app processor, allowed by the fact that the device doesn't have any telephony stuff to power. ?

Arun

29-Apr-2009, 15:21

Is it not just the case that the latest itouch is clocking the processor quicker as opposed to being a different app processor, allowed by the fact that the device doesn't have any telephony stuff to power. ?Not at all. The iPhone 2G, iPod Touch, and iPhone 3G use the S5L8900 (http://theiphonewiki.com/wiki/index.php?title=S5L8900). Meanwhile, the iPod Touch 2G uses the S5L8720 (http://theiphonewiki.com/wiki/index.php?title=S5L8720_(Hardware)). As I said, practically nobody in the press seems to have caught on to most of this, so I would be extremely surprised if Apple used a confusing versioning system on purpose...

Mike11

29-Apr-2009, 15:42

Okay, so, err - why was the iPhone 3G 1,2 and the iPod Touch 2G 2,1? The two main differences, AFAICT, are the application processor and the WiFi/Bluetooth chip. Surely the latter can't be such a big deal? Unless of course you are assuming Apple is using a senseless and confusing versioning system on purpose to further confuse everyone?

Frankly, I have yet to see someone in the press having a clue about this kind of thing, so I doubt Apple would bother doing that. But who knows...
Good point, but the iPod 2,1 number is most likely over a year old, a lot has changed since Apple decided to name the 2008 iPod touch 2,1. Maybe it's just that the development of new SoCs got accelerated because of the acquisition of PA Semi. Who knows, anything is possible.

So let's just ignore the June iPhone and concentrate on the speculated Q4 iPhone (3,1) with 5MP and a brand new SoC :smile: . What app processor and SGX version do you think we'll see? ARM11, Cortex-A8 or even Cortex-A9? The most logical thing would be a Cortex-A8 (after two years of ARM11) and then in 2011 Cortex-A9 (after two years of A8). And there's a small chance for a custom ARMv7 design (like Snapdragon). But personally I'd like to see them go directly from ARM11 to Cortex-A9MP, so my prediction for the Q4 SoC is a upgraded and higher clocked ARM11 (an then beat TI's OMAP4 with a A9 in 2010 :wink: ).

wco81

29-Apr-2009, 16:12

http://www.engadget.com/2009/04/28/sony-unveiling-umd-less-psp-with-slide-out-buttons-at-e3/
http://www.pocketgamer.co.uk/r/PSP/Playstation+Portable+2/news.asp?c=12612

Additionally an acquiantence of mine has heard that a new PSP is being readied for launch. I do not know how accurate, if at all, that info is, and if true it is most likely a cosmetic refresh.

Personally, I don't think a major PSP hardware change will happen this year, unless they rush something due to a perceived threat from iphone (the itouch alone sold 3M units last quarter, iphone+itouch was 7M). IMO, it is reasonably certain that Sony has taken an SGX licence from IMG. But they havn't had enough time to use that licence.

Sorry for being OT.

Is there really that much overlap between PSP and iPhone/iPod Touch customers?

The average selling price of a PSP game has to be at least 4 times that of the ASP of the iPhone game.

tangey

29-Apr-2009, 16:46

Is there really that much overlap between PSP and iPhone/iPod Touch customers?

The average selling price of a PSP game has to be at least 4 times that of the ASP of the iPhone game.

And yet the game playing capability of the two devices is not so dissimilar, which is where the problem arises for Sony. They have to make their portable gaming playing device much more appealing than the iphone otherwise why will people pay x4 the cost for the games. It wasn't an issue until the iphone came along...it now is. And lets not forget the itouch, which is a non-contract device, that is even more a competitive threat to the traditional portable game playing devices.

wco81

29-Apr-2009, 18:19

But most of the iPhone games which are doing well aren't even 3D or don't come close in terms of production values.

So far the only $10 games have been mostly ports of games from other systems, so they didn't have to build up a shrink-wrap quality game from scratch.

The cheaper games are about the equivalent of XBL Arcade and PSN games, no?

One thing though, PSP had good sales of hardware in recent years but not so much on the software side, other than a few blockbusters like God of War here and there.

On local Craiglist, there are a lot of ads for hacked PSPs along with DVDs full of ripped games. Could that partly account for the disparity in PSP hardware vs. software sales?

tangey

29-Apr-2009, 22:35

Again, the problem as I see it (for Sony and others ) is that there is literally thousands of games for the itouch/iphone, some free, some a few dollars some $10. Its an entire platform of software/hardware that they can not compete with, which has come from nowhere.

The App store is an extremely potent weapon....distribute 1 billion apps in 9 months from a standing start ?....totally unheard of at any level of computing.

on the app store, out of the top 20 "paid for" apps on the all-time list, 14 of them are games...do they necessitate 3D hardware?, I don't know (I don't have an iphone)...does it point the way forward?...it definitely does.

The method of getting software to end users, including giving just about anyone access to your entire userbase for 30%, is a threat to all traditional mobile gaming providers.

wco81

30-Apr-2009, 00:18

See I've downloaded mostly free games including Monkeyball Lite. But the game I've played the most is this little ski jump thing which is 2D.

A lot of these games look more closer to flash games or stuff you'd see on Yahoo Games, so it would appeal to a different audience.

No evidence that iPhone games are taking sales away from PSP or DS. PSP might be stalling now because it's old and there aren't any big blockbusters out for it now.

Arun

30-Apr-2009, 09:16

I don't think the question is so much whether the iPhone is taking sales from the PSP/DS (although the latter is IMO more vulnerable here) but rather whether the next-gen one on SGX will. Apple would be crazy not to more aggressively aim it at the same market...

BTW, Infineon released its quarterly results now and it looks like they're very bullish on wireless for the next quarter, so I guess it's safe to say the June iPhone will still be based on an Infineon chipset - my mistake.
EDIT: I'm listening to the call now and, when asked, they reluctantly implicitly said that it was existing products - not new ones - that will improve wireless in Q2. They also repeated HSDPA when the analyst said that. This would imply Apple is sticking to their current solution, and the rumours indicating HSUPA are wrong. Then the question is who wins the next socket in 2H09...

CEO

30-Apr-2009, 11:35

Cant wait for the new Zune.

wco81

30-Apr-2009, 14:31

Some stories in the business media about speculated Apple products. Business week says Apple has talked to Verizon about a lower-cost iPhone and a media pad. The BW rendering of the latter shows no buttons so not particularly useful for gaming.

A WSJ story talks about two former execs. of ATI/AMD joining Apple to build up in-house chip design capabilities. It makes it sound like eventually, Apple will do everything in-house to control confidentiality about future products.

Mike11

04-May-2009, 01:32

iPhone 4Q09 @ ~$299
App processor with SGX & VXD for 1080p decode
5MP with ISP & BSI - see product brief (http://www.ovt.com/uploads/parts/OV5642_PB(1.0)_web.pdf)
Qualcomm MSM6290 (http://www.eetasia.com/ART_8800489081_499488_NP_950125aa.HTM) Baseband
802.11n/BT2/etc. BCM4329 (http://www.broadcom.com/products/Bluetooth/Bluetooth-RF-Silicon-and-Software-Solutions/BCM4329)
AMOLED

I don't think AMOLED Displays are ready for prime time in mobile phones. One of the advantages of the current iPhone display is its great readability in sunlight, at least compared to other touchscreen phones. But right now it seems even the best and brightest AMOLED suck in sunlight:

Samsung OMNIA HD i8910 Review:
"Just as expected, the screen delivers beautiful, saturated colors albeit in artificial lighting only. Unfortunately, you can hardly see any details on the screen if you take the phone outdoors in bright sunlight, especially if you happen to forget to wipe the fingerprint-smudged surface clean."
http://www.phonearena.com/htmls/Samsung-OMNIA-HD-i8910-Review-review-r_2148.html

And keep in mind that after MWC the Omnia HD was praised all around the net for its "fantastic" and "best-ever" display.

But on the plus side, the 4.1 hours of HD video capture for the Omnia HD exceed my expectations! I'd love to see something like that in the next iPhone!

Ike Turner

04-May-2009, 12:21

But on the plus side, the 4.1 hours of HD video capture for the Omnia HD exceed my expectations! I'd love to see something like that in the next iPhone!

Err that's if you change your battery every 30 minutes ;). 4.1 hours is the memry limitation AFAIK.
As For the outdoor visibilty it has more to to with the glass (and digitizer on resitive touchscreens) use on top of the screen than the screen itself (see the variation between many of the WM phones out there.

Some more phone Specs:

Samsung GT-i8000 aka CUBIC37:
CPU/GPU unknown (PXA312@624mhz? PXA320@806Mhz? MSM7200a528Mhz?)
Baseband unknown (Qualcomm?)
Screen 800x480 (resistive or capacitive like the i7500?)
8MPix camera
8Gb/16Gb of mem plus MicroSD
FM radio
Windows 6.1/6.5 with Samsung's new Cube 3D interface (Sammy and LG sure do like to copy each other ALOT!)

Samsung Louvre (Omnia Pro)
CPU/GPU unknown (PXA312@624mhz?)
Baseband same as current Omnia
Screen 800x480 resistive
5Mpix camera (or maybe 8mp)
8Gb/16Gb of mem plus MicroSD
FM radio
Full QWERTY sliding keyboard
WM6.1/6.5

Mike11

04-May-2009, 15:47

As For the outdoor visibilty it has more to to with the glass (and digitizer on resitive touchscreens) use on top of the screen than the screen itself (see the variation between many of the WM phones out there.
I think AMOLEDs are just not bright enough for bright sunlight yet. Since they have no reflective part like transflective LCD Displays, they have to "shine" all on their own ;)

But maybe in Q4 they will be good enough, 6 months is a lot of time.

Lazy8s

09-May-2009, 18:49

I predict Sony will launch a PSPhone and a PSP2, which will be just a PMP/MID version, in 2010 sharing the same platform:

SGX543MP2 @ 250MHz + MVED (VXD/VXE) + an ARMv7 or MIPS generational equivalent

OpenGL ES 2 w/extensions and OpenCL

wco81

09-May-2009, 19:06

So with Kutaragi gone, Sony isn't going to roll their own?

Lazy8s

09-May-2009, 19:14

After using around 20mm^2 of silicon just to get the performance they did with PSP, I don't think they could justify an in-house GPU even if Kutaragi were still there.

Ike Turner

12-May-2009, 18:06

Some More WinMo 7 stuff...
http://blogs.zdnet.com/microsoft/?p=2752

WM7 Chassis 1 Specification

Core requirements:
Processor: ARM v6+, L2 Cache, VFP, Open GL ES 2.0 graphics HW (QCOM 8k, Nvidia AP15/16* and TI 3430 all meet spec)
Memory: 256MB+ DRAM, 1G+ Flash (at least 512MB fast flash – 5MB/s unbuffered read @4K block size)
Display: WVGA (800×480) or FWVGA (854×480) 3.5” or greater diagonal
Touch: Multi-touch required
Battery: Sufficient to meet Days of Use LTK requirements.
Controls: Start, Back, Send and End are required (soft controls allowed as long as they are always present).

Peripherals:
Camera: 3MP+, flash optional, 2nd camera optional (VGA resolution sufficient)
GPS: aGPS required
Sensors required: Light Sensor, Compass (3 axis, 5 degrees, 100 Hz sample rate), Accelerometer (3 axis, 2mg resolution, 100 Hz sample rate)
USB: High speed required, 20 MB/s transfer rate.
BlueTooth: BT2.1 required, must run MSFT BT stack, CSR BlueCore6 or later recommended.
Wi-Fi: 802.11B/G required, must run MSFT Native Wi-Fi stack, Atheros 6002 or Broadcomm 4325 recommended.
Connectors: Micro USB and 3.5mm Audio required.

Options:
FM tuner: If tuner HW is present it will be detected and supported by the Media application.
Haptics
SD Card (Micro SD recommended)
DPAD, qwerty or 12/20 key keyboards all optional

w00t !

Arun

12-May-2009, 19:14

Wow, that's a nice leak :) I'm very happy about WM7 being OGL ES 2.0-only too. That's a MASSIVE step in the right direction, and also one hell of a confirmation that even WM7's *default* GUI is 3D. Those chips have quite different performance levels though, so it'll be interesting to see how game developers handle it. Also this could be why Qualcomm revised the QSC7230 (now known as the MSM7230) so significantly after its pre-announcement: it wouldn't have met the WM7 specs! (see: http://www.linleygroup.com/Newsletters/LinleyMobile/lm090129.html)

I'm intrigued by the multi-touch requirement though; one day Ballmer says Capacitive is too expensive, the next they outlaw anything else?! Unless they're considering promoting Stantum's multi-touch technology as an alternative, of course.... (just Google that if you don't know what I mean). One tidbit about that, BTW: much of the cost of the Stantum approach seems to be because they're using an off-the-shelf microcontroller and mini-FPGA for control logic. If this was integrated into a single chip, or even in the application processor, the main extra cost versus vanilla resistive would be the greater number of wires going from the touch screen to the rest of the system - not free, but not so far from it either compared to alternatives I think.

wco81

12-May-2009, 20:08

MS is supporting OGL over whatever they call their graphics API these days?

Arun

12-May-2009, 21:21

MS is supporting OGL over whatever they call their graphics API these days?They basically gave up on their own internal API, it's still supported on quite a few chips but it will never get beyond DX7-level.

wco81

12-May-2009, 22:47

That's huge!

That means the next Xbox will use OGL?

Ike Turner

12-May-2009, 23:29

That's huge!

That means the next Xbox will use OGL?

Nope. They are just dropping Direct3D support from Windows Mobile, nothing else.

Laurent06

13-May-2009, 09:23

Core requirements:
Processor: ARM v6+, L2 Cache, VFP, Open GL ES 2.0 graphics HW (QCOM 8k, Nvidia AP15/16* and TI 3430 all meet spec)
I would put my bets on Tegra because:
1. since Tegra has been announced, it's been said to support WM (Android came much later)
2. a few days ago, nVidia said they had high hopes for Tegra ("ref" (http://www.theinquirer.net/inquirer/news/1137243/nvidia-posts-loss-pins-hopes-tegra))
3. OMAP3 is ARMv7; QCOM 8k also is IIRC.

Ike Turner

13-May-2009, 09:29

I would put my bets on Tegra because:
1. since Tegra has been announced, it's been said to support WM (Android came much later)
2. a few days ago, nVidia said they had high hopes for Tegra ("ref" (http://www.theinquirer.net/inquirer/news/1137243/nvidia-posts-loss-pins-hopes-tegra))
3. OMAP3 is ARMv7; QCOM 8k also is IIRC.

No need to bet on anything in this case. These are just the some of the minimum specs to run WM7. Tegra,SnapDragon,OMAP3, MS doesn't care cause all three are good enaugh. And we are going to see many more SnapDragon devices/Phones than Tegra anyway.

Arun

13-May-2009, 10:40

Just tweeted something that seems very relevant here: http://twitter.com/ArunDemeure/status/1782918819 - I don't think anybody else caught this quote ("Apple is also talking about other devices"), which is nice because it's a very reliable confirmation, although it shouldn't exactly be a surprise either. Also the fact they're implying Infineon is in a good position to win those sockets kinda makes me think my MSM6290 prediction was too hasty :)

Mike11

13-May-2009, 15:14

Why do you just assume that ALL WM7 devices will follow these specifications? It says "WM7 Chassis 1 Specification". Can't there be 2 or 3 more, maybe with lower specs? Maybe these Chassis 1 specs are just for the high-end smartphones.

Mike11

15-May-2009, 11:45

IF there will be an iPhone 2 in June AND an iPhone PRO in Q4, do you think Apple will announce the PRO at WWDC? Because if Apple doesn't even mention the iPhone PRO then a lot of enthusiasts/early adopters of the iPhone 2 will be very pissed when Apple releases a better iPhone less than 6 months later.

I think Apple should at least mention that a more expensive PRO version is coming in Q4 and highlight some of the biggest differences. For example "PRO in Q4, starts at $299, hardware keyboard, up to 64GB storage, 5MP and 720p HD video recording."

wco81

15-May-2009, 14:49

AzBat

16-May-2009, 02:33

OK, TeamXbox posted a pretty extensive rumor filled feature today about a "Microsoft Digital Entertainment Handheld". Here's the jist of it...

It's not a phone! It won't compete with Windows Mobile partners.
xYz(the letter 'Y' was highlighted by the original source) device sits BETWEEN Xbox & Zune; The device has a mix of each platform making it unique.
“think of a mashup of the Sony Mylo, the PSP, and the iPhone… errr, the iPod Touch; [the MS handheld] doesn’t need access to a phone network.”
Large WVGA touchscreen & “hardware features not found on any handheld on the market.”
“This is a Live Anywhere device”...“There will be a single online marketplace; the lines between the Zune, Xbox Live and Sky marketplaces will blur when the handheld launches.”
the graphical interface found in the New Xbox Experience will make its way onto the Microsoft handheld. The NXE user interface will be even easier to use on the handheld than on the Xbox 360, the source claimed.
May support WiMAX
Might be based on NVIDIA Tegra platform

http://news.teamxbox.com/xbox/19639/Microsoft-Working-on-Digital-Entertainment-Handheld/

Interesting. Would love to finally have a Live Anywhere device.

Tommy McClain

Mike11

16-May-2009, 07:51

They rarely announce a new product in advance, especially when they're hawking new products in the market.

That would be undercutting anything they may intro in the summer.

The original iPhone intro in January and then shipping at the end of June was an exception but then they didn't have an existing product to worry about.
True, but they normally also don't release a totally new high-end product just a few month after a product in the same category was released. Just imagine there's no MacBook Pro and never has been. Then in June they release the new MacBook and then suddenly out of the blue in October/November a MacBook Pro that was never even mentioned before. I think that would create bigger problems than announcing a product (with a significantly higher price point) a few months in advance.

Arun

16-May-2009, 14:41

It's a bizarre leak, doesn't really tell you much that's new or unpredictable if you understand the other ones, see for example LiveSide.net - except for the WiMax part, which is kinda insane and full of misconceptions sadly.

Here's my point of view: the logical "solution" would be to have a 3G SKU that's data-only; i.e. no voice support. However operators are dead scared of VoIP and will not allow this possibility in this timeframe unless there is essentially no way to install any VoIP application on the phone. Therefore, this is only believable if there was no generic application ecosystem for the Zune HD - and as I indicated earlier, I suspect that's a likely possibility and that the only SDK would be for gaming along with a controller specialized App Store. Other services, such as weather, productivity, and social networks, might be provided by Microsoft who'd probably invest quite a bit of money in these apps - with the goal of perhaps giving them away (with source code) to partners for WM7...

I would expect, however, that the first Zune HD announcement will focus on an WiFi-only SKU.

As for pre-announcing the iPhone HD, I don't buy it. However I wouldn't be surprised if they hinted at there being an OpenGL ES 2.0 device with more focus on gaming sooner rather than later, and this would be enough to make sure the most knowledgeable (aka bitchy) enthusiasts decide to wait if they want to.

tuna

17-May-2009, 14:28

May support WiMAX
Tommy McClain

Isn't wimax dead?

wco81

17-May-2009, 15:00

AzBat

17-May-2009, 17:24

Isn't wimax dead?

I don't think so. According to the article...

In the United States, Comcast, Intel, Time Warner Cable and Google are investing $3.2 billion to combine Sprint Nextel and Clearwire into a single company that will build the first nationwide WiMax network. The service has already launched in Chicago, Baltimore, and Washington D.C. as initial test markets, with Boston, Philadelphia, and Dallas to follow before a nationwide rollout.

Tommy McClain

Entropy

17-May-2009, 21:20

I still think the most popular games on the iPhone are simple things like Tetris or Bejeweled.

Even people who play games a lot are looking for at most a 5-10 minute session of gaming on their iPhones.

For the life of me, I can't understand what you're trying to say. Is this good or bad? If it means that games have to be made so that you can save anywhere rather than at specific save-points, isn't that a good thing? (Forcing the player to trudge through long parts of a game because he/she had to stop playing at an inconvenient time easily ranks one of the top ten most annoying games design choices.)

And from the view of a potential games designer, isn't it good that a good arcade style gaming idea can reach a large audience and reap nice revenue? And for that matter that there is a large still untapped market for strategy games (YES to Civ on the iPhone!), adventure games, role-playing games, networked racers, MMOs of various flavours, .....

That people can only devote rather short stretches of continous time to gaming is the norm. Most people have lives that don't revolve around gaming.
This is quite healthy.

wco81

18-May-2009, 00:53

I'm saying that the platform doesn't lend itself to the longer-form games with a lot of production values like those on dedicated handheld systems.

Apple could put in the 3D hardware and maybe even offer a usable control scheme.

But it appears the center of gravity in the iPhone games market is around casual, short games which are at best a few bucks.

EA would probably prefer to charge $30 for their iPhone Tiger Woods game rather than $10, so that the pricing is more in-line with versions on other systems. But even at $10, is it selling?

Or are iPhone gamers more than satisfied with Flight Control for 99 cents or whatever they're charging?

tangey

18-May-2009, 10:03

EA would probably prefer to charge $30 for their iPhone Tiger Woods game rather than $10, so that the pricing is more in-line with versions on other systems. But even at $10, is it selling?

There is the question. However 2 points.

I think game distributors are quickly coming to the conclusion that where Apple is at the moment is only the starting point, and that there are going to be various Apple products that will all be compatible with the App store, perhaps tablets, perhaps more dedicated gaming devices. nobody knows what exactly, but most generally beleive that the product base will grow AND diversify. The potential is there for all to see, 1 Billion downloads from the current user base, what will it be once there is an entire family of app-store compatible products ? Appleinsider is conjecturing that Apple made $45M to date from the app store (with a reasonable assumption of free->paidfor ratio), which translates into $150M in total sales from a brand new distribution system, impressive.

The wholely digital distribution system, effecting cutting out the traditional distributor and retailer, and resulting in zero media and packaging costs is extremely appealing. Zero wasted stock. I also would not be surprised if the "majors" negotiate better splits with Apple.

Ike Turner

27-May-2009, 07:15

www.zune.net/zunehd

wco81

27-May-2009, 07:18

www.zune.net/zunehd

Is that the official site?

Price and specs?

What's with the dock thing for outputting to HDTV?

Ike Turner

27-May-2009, 07:40

Is that the official site?

Price and specs?

What's with the dock thing for outputting to HDTV?

Yes it's the official site.
The dock isn't anything new (look @ the 50€ iPhone/iPod tv-out cable..)

Arun

27-May-2009, 14:11

The dock does make good sense in my mind, but it doesn't make sense not to *also* have a HDMI port on the device. Oh well, I really should stop expecting too much from MS. I bet this is the same kind of retarded reasoning that led to the Palm Pre not having microSDHC (zomg it'll ruin the exterior!)

Also, they are in the weird position where they are both cheaper and more expensive in different ways than an iPod Touch. This is not an ideal competitive positioning. They really should have made two SKUs like Apple will. Oh well, live and learn - at least it should be a very solid foundation unlike the original Zune. And the two biggest selling points IMO (battery life & gaming) haven't even been unveiled yet... :) I might be tempted to buy one if it's cheap enough, although I love the iPhone/iPod Touch apps too much to consider it as a potential replacement.

wco81

27-May-2009, 14:53

Probably licensing costs for HDMI ports?

Also, Apple has these specialized video out cables which are like $50 because they put in a DRM chip of some kind.

Maybe a similar situation with the Zune dock.

Arun

27-May-2009, 15:05

HDMI royalties are $0.05-$0.15 depending on whether you show the HDMI logo (see: http://www.hdmi.org/manufacturer/terms.aspx) - so if that's the reason, that's even dumber. BTW, another big deception I have is the lack of video streaming from PC to Zune (to TV) via WiFi - unless they just didn't talk about it.

BTW, I wonder if it's the same 3.3" 480x272 AMOLED as these two PMPs: http://www.tigerdirect.com/applications/searchtools/item-details.asp?EdpNo=4490709&SRCCODE=GOOGLEBASE&cm_mmc_o=VRqCjC7BBTkwCjCECjCE and http://www.amazon.com/gp/product/B001NXDC4O?ie=UTF8&tag=displ-20&linkCode=as2&camp=1789&creative=9325&creativeASIN=B001NXDC4O - and I wonder how much it actually costs in high volumes! Certainly these price points would tend to support a 16GB version at $199, which might indeed be a successful product if it's polished enough and the battery life numbers are what I expect them to be (hint: one more digit in every category than the iPod)

wco81

27-May-2009, 15:41

$199 would be terrific price. That really underprices the iPod Touch.

But I think they're holding off on the price to see what cards Apple plays.

They have to ship in volume or else this is little more than a tech demo.

For instance, the Pre may not be able to meet demand and they may not be able to commit to production of higher volumes until they see what the demand is or they get some cash flow coming in.

Mike11

27-May-2009, 19:10

Now that even the Zune HD comes with an OLED multi-touch screen, do you think this makes it more likely that the iPhone will have one as well in June?

Regarding the CPU of the next iPhone, there seems to be a consensus out there among analysts that it will have 600 MHz, the question seems to be if it's ARM11 or Cortex. Lately Cortex ("double the horsepower") seems to be the safer bet according to some sources.
http://arstechnica.com/apple/news/2009/05/iphone-hardware-details-and-predictions-continue-to-multiply.ars

Arun

27-May-2009, 19:13

wco81

27-May-2009, 19:19

If not this year, then hopefully next year.

Although, I'm not sure how heavily video is used on these things. I've viewed some podcasts here and there and I even loaded some movies, including the digital copies of The Dark Knight and Wall E from the Blu-Rays I bought.

But I've never sat down and bothered to watch them, not even on long flights. I would watch the Blu-Rays at home instead.

roninja

27-May-2009, 23:08

question is will iPhone 2,1 feature SGX520 and 3,1 SGX530?

wco81

27-May-2009, 23:23

Apple isn't known for pushing specs.

Question is how much silicon they need at a given time to support the software and the features they want.

I hope to be pleasantly surprised about components like OLED but not counting on it.

Mike11

28-May-2009, 05:17

The June/July iPhone is 2,1. The current iPod Touch is also 2,1. The 3,1 comes later, probably Q4. I'm really not sure how this isn't completely obvious but heh :) I wouldn't bother listening to analysts who are clearly speculating.
IF there are two iPhone versions that will be released this year, do we have any other rumors besides the camera one (3.2MP summer, 5MP later this year) that they will come out 3-6 months apart? Why can't they release them together this summer? A 8GB entry model (iPhone 2,1 based on the iPod touch hardware, so very little work needed) for $99 and the 16GB ($199) and 32GB ($299) "new" iPhone (3,1). Apple did something similar with the MacBook (white one went down to $999 and got upgraded hardware when the unibody came out).

question is will iPhone 2,1 feature SGX520 and 3,1 SGX530?
Well, if Arun's right then the iPhone (WWDC) will essentially just be last years iPod Touch with 3G, GPS, a 3.2MP camera and maybe a slightly higher clocked CPU (10%-20%). That would mean no SGX until iPhone 3,1 later this year.

$199 would be terrific price. That really underprices the iPod Touch.

But I think they're holding off on the price to see what cards Apple plays.
I don't think Microsoft can hold off on the price until September.

RudeCurve

28-May-2009, 06:51

I think PSP2 will have the same features as Zune HD in addition to full game support, tilt, mic and maybe even camera.

Arun

28-May-2009, 11:10

I think PSP2 will have the same features as Zune HD in addition to full game support [...]Uhm... if you think MS isn't going to announce an aggressive gaming strategy in the near future, you're nuts :)

RudeCurve

28-May-2009, 19:23

Uhm... if you think MS isn't going to announce an aggressive gaming strategy in the near future, you're nuts :)

I really hope they do because the Zune HD looks real sexy. Just wish they added tilt and a mic which costs what $5?

AzBat

28-May-2009, 20:01

I really hope they do because the Zune HD looks real sexy. Just wish they added tilt and a mic which costs what $5?

Although no mention of a mic, there was mention of an accelerometer. I'm sure details will be announced next week. Gotta be patient. ;)

Tommy McClain

RudeCurve

28-May-2009, 20:26

Sounds good guys, looking forward to the announcemnt. If priced right I will have to get a Zune HD.

AzBat

28-May-2009, 23:45

Gimodo hands on...

Fk79SByyNIY

Still don't show the Games menu!! :mad:

Photos here...

http://gizmodo.com/5272228/zune-hd-hands-on-photos-and-video-tour

Tommy McClain

AzBat

29-May-2009, 19:25

Paul Thurrott got a few more details...

http://community.winsupersite.com/blogs/paul/archive/2009/05/28/more-details-about-the-zune-hd-and-zune-xbox-live-integration.aspx

a new "Zune" link will replace the Video Marketplace link in the Xbox New Experience (NXE) UI that you see on the console today.
...
"Zune is growing," Seitz said. "It's a complete software + services layer. Zune is about music and videos. On the Xbox next week, we'll talk about Zune in the context of the Xbox Video Marketplace. We're making huge improvements to the navigation and the look and feel. We will also discuss other cool features and capabilities that the service will have."
...
"There are a total of three buttons on the thing," he said. "On the front, near the bottom, is a silver button that takes you back to the home menu. There's a rocker on the upper left that you can use for volume up and down and to trigger onscreen transport [Play, Pause, Next, etc.] controls. And there's a power button on the top. That's it."
...
The Zune HD web browser is based on a "flavor" of Internet Explorer 6, Seitz said, "but you won't recognize that. We did heavy, heavy customization work on the UI, and it's optimized for multi-touch, gestures, and finger-based navigation instead of a stylus."

Tommy McClain

idsn6

08-Jun-2009, 19:51

New iPhone 3Gs supports OpenGL ES 2.0.

Scott_Arm

08-Jun-2009, 19:53

New iPhone 3Gs supports OpenGL ES 2.0.

New still/video camera looks very nice. Oh, and HSDPA 7.2 MBit/s

Mike11

08-Jun-2009, 20:14

iPhone 2Q09 @ ~$99
Same app processor as iPod Touch 2G
3.2MP with ISP - see product brief (http://www.ovt.com/uploads/parts/OV3642_PB_(1.1)_web.pdf)
Qualcomm MSM6290 (http://www.eetasia.com/ART_8800489081_499488_NP_950125aa.HTM) Baseband
802.11g/BT2, unknown chip(s)

iPhone 4Q09 @ ~$299
App processor with SGX & VXD for 1080p decode
5MP with ISP & BSI - see product brief (http://www.ovt.com/uploads/parts/OV5642_PB(1.0)_web.pdf)
Qualcomm MSM6290 (http://www.eetasia.com/ART_8800489081_499488_NP_950125aa.HTM) Baseband
802.11n/BT2/etc. BCM4329 (http://www.broadcom.com/products/Bluetooth/Bluetooth-RF-Silicon-and-Software-Solutions/BCM4329)
AMOLED

Hm, the iPhone 3GS ($199/$299) looks to be too powerful (2x-4x) to have the same app processor as the iPod touch (even at ~600MHz). And since the iPhone 3G will still be sold at $99, I really don't see the possibility for an iPhone Pro/HD (as a 3rd version in Q4/09 starting at $349 or more) until next years WWDC.

Very strange, now we're back to having no idea what the iPhone 3,1 could be (unless it's just the "normal" 2010 iPhone)...

http://www.engadget.com/2009/06/08/iphone-3gs-announced/

roninja

08-Jun-2009, 20:44

Guessing its an Samsung Cortex A8 / SGX 520 combo...

Ike Turner

08-Jun-2009, 21:39

The Samsung GT-i8000 specs have been leaked http://www.mobili.lt/lt/telefonai/samsung/i8000.html?show=all . The samsung S3C6410 is on board (pdf: http://www.samsung.com/global/business/semiconductor/support/brochures/downloads/systemlsi/s3c6410_datasheet_200804.pdf ) and Qualcomm MSM6290 Baseband.

And I'm 90% surethat the iPhone 3Gs has the same CPU as the current just clock higher and maybe a die shrink.

darkblu

08-Jun-2009, 23:29

And I'm 90% surethat the iPhone 3Gs has the same CPU as the current just clock higher and maybe a die shrink.
what do you mean by CPU? if it's the entire application processor die, then no - the GPU is a large enough change that you would say it's not the same AP anymore. if it's the CPU part alone then it could be a higher clocked arm11 just as well, but i don't see how one could be so sure of that, unless you know something the rest of us don't.

tangey

09-Jun-2009, 00:22

wco81

09-Jun-2009, 01:34

Hmm, so could games support specific OGL 2.0 features and if the game detects 3G-S, then it would load with a different profile and offer extra graphics features?

If not this year, by next year, there would be a wide enough HW gap that developers would be faced with the dilemma of only supporting the latest models or supporting extra features on the latest models.

Then again, it seems the most popular iPhone games aren't the most graphically-intensive ones.

warmi

09-Jun-2009, 02:39

Given that they have stated its openGL es2.0, then its definitely NOT the same AP just clocked faster, as that AP has MBX which is only ES1.1 compatible.

I'm more confident that it is the S5PC100 Cortex A8, given that apple are saying:-
1) its about twice as fast
2) its GL Es2.0
3) it has better battery life.

I "outed" that processor back in mid-april as a likely candidate for the ihpone on this post:-
http://forum.beyond3d.com/showpost.php?p=1283065&postcount=106

And that post was thereafter referenced by many other online news sites.

a terrible thought is that it just *could* be the S3C6410 with samsungs own in-house Open GL2.0 (then do have one), but thats too horrible to think about given that both samsung AND apple are IMG licencees, and apple is an IMG shareholder.

It is SGX for sure.
They have some sort of middleware code emulating OpenGL ES 1.x with custom shaders etc ...

Quoting from the latest OpenGL ES doc on the OpenGL ES developer page
--------
The PowerVR SGX is the graphics processor in Apple’s newest iPhones and is
designed for OpenGL ES 2.0. The graphics driver for the PowerVR SGX also
implements OpenGL ES 1.1 by efficiently implementing the fixed-function
pipeline using shaders. More information about PowerVR technologies can be
found in the PowerVR Technology Overview. Detailed information about the
PowerVR SGX can be found in the POWERVR SGX OpenGL ES 2.0 Application
Development Recommendations.
------------

tangey

09-Jun-2009, 03:12

warmi...thanks for the confirmation about SGX. It doesn't happen to say whether its SGX520 or 530, 535 anywhere ?

warmi

09-Jun-2009, 04:19

warmi...thanks for the confirmation about SGX. It doesn't happen to say whether its SGX520 or 530, 535 anywhere ?

No.. nothing specific really ... just SGX and OpenGL ES 2.0.

Wishmaster

09-Jun-2009, 05:23

The problem is that openGL ES 2.0 isn't backwards compatible with 1.1...
Does that mean that all games will have to be rewritten for the 3GS?
If so "boo hoo" apple...

Grall

09-Jun-2009, 05:28

Pretty much everything about the new iPhone impresses me except for two things:

* Wut? No front-facing camera STILL? iPhone's pretty much the only 3G handset not capable of videocalls. Certainly the only one in its market segment.

* The pricetag. $700 for 32GB model, +$100 for 16GB flashram, someone please tell apple it's not 2006 anymore....

I'm very tempted by this phone, but the the idiot decision not to include front camera and the stupid pricing that is so clearly good ol' apple-tax ripoff of its customers makes me want to say no on general principle.

idsn6

09-Jun-2009, 05:44

Hmm, so could games support specific OGL 2.0 features and if the game detects 3G-S, then it would load with a different profile and offer extra graphics features?

Yes, you just request a 2.0 GL context for more advanced graphics. All existing games will continue to work with the 1.1 context.

If not this year, by next year, there would be a wide enough HW gap that developers would be faced with the dilemma of only supporting the latest models or supporting extra features on the latest models.

This is already a reality across the iPhone/iPod touch line. The differences in speed and built-in hardware mean that some apps have only been released for later models.

tangey

09-Jun-2009, 10:00

The problem is that openGL ES 2.0 isn't backwards compatible with 1.1...
Does that mean that all games will have to be rewritten for the 3GS?
If so "boo hoo" apple...

Existing games will run fine, new games running 2.0 won't run on old hardware. The inclusion of 2.0, and the certain appearance of decent "2.0 only" 3D games running on a quicker processor, might very well be another leverage that Apple wanted to get their current user base to upgrade to an "S" iphone.

The exclusion of the front facing camera makes perfect sense when viewed from a Steve Jobs perspective. The decision was made simply because its extra hardware that actually NO ONE uses, so instead fit hardware that people will use, such as the compass.

Grall

09-Jun-2009, 10:38

Where's the source for this supposition of yours?

Back in the real world, it's not hardware that NO ONE uses. Deaf people regularly make use of video calls, for reasons I'm sure you can figure out yourself :) I'm sure others find it useful as well.

I'd venture a guess and say video call capability would see far more use than the newfangled compass will. When in everyday life do you ever find a need for one? I can't think of a single goddamn time I've personally needed one in the last 20ish years...

tangey

09-Jun-2009, 10:43

Supposition ?, I'm in the real world, I don't know anyone that uses video calling other than as a gimmick.

I do not however have hearing-impaired friends.

Its a choice between including hardware that a very small percentage of the public will genuinely find useful, and including hardware that everyone will find useful. I'm glad apple are big enough not to follow a dead-end trend

Simon F

09-Jun-2009, 10:43

It is SGX for sure.
They have some sort of middleware code emulating OpenGL ES 1.x with custom shaders etc ...

Quoting from the latest OpenGL ES doc on the OpenGL ES developer page
--------
The PowerVR SGX is the graphics processor in Apple’s newest iPhones and is
designed for OpenGL ES 2.0. The graphics driver for the PowerVR SGX also
:?: Do you have a link?

darkblu

09-Jun-2009, 12:47

:?: Do you have a link?

it's in the platform notes from the latest edition of the gl es programming guite for the iphone. link (http://developer.apple.com/iphone/prerelease/library/documentation/3DDrawing/Conceptual/OpenGLES_ProgrammingGuide/OpenGLESPlatforms/OpenGLESPlatforms.html#//apple_ref/doc/uid/TP40008793-CH106-SW1)

oh, btw
OES_standard_derivatives
yay!

Mike11

09-Jun-2009, 14:19

Does anyone know why there are no iPhone 3G S hands-on articles? Gizmodo had one for a few minutes (after the keynote) but then it disappeared...

wco81

09-Jun-2009, 14:36

Does AT&T even support video calls?

In Europe, there's a premium on video call charges IIRC.

Ike Turner

09-Jun-2009, 14:54

Does AT&T even support video calls?

In Europe, there's a premium on video call charges IIRC.

Video calls (for the last 2 years) are part of all the unlimited data plans in Europe (in France for sure) .So if you have a 2h voice plan/month it means that you can also make 2 hours of video calls. No extra fee. A call is a call (with video or without).

Simon F

09-Jun-2009, 14:57

it's in the platform notes from the latest edition of the gl es programming guite for the iphone. link (http://developer.apple.com/iphone/prerelease/library/documentation/3DDrawing/Conceptual/OpenGLES_ProgrammingGuide/OpenGLESPlatforms/OpenGLESPlatforms.html#//apple_ref/doc/uid/TP40008793-CH106-SW1)
Ah.. I assumed the OP meant it was on the Khronos site. I'm not an Apple developer so I'll take your word for it :-)

warmi

09-Jun-2009, 15:48

No official info about the CPU yet .. but it looks like it will be Cortex.

TimothyFarrar

09-Jun-2009, 23:23

"The PowerVR SGX is the graphics processor in Apple’s newest iPhones and is designed for OpenGL ES 2.0. The graphics driver for the PowerVR SGX also"

SGX in the new iPhone, cool!

I'm going to also toss in a wild guess that if Apple doesn't fully support the unified shaders of the SGX (what if vertex texture fetch is missing) via GLES2, that they might be planning on moving on to Tegra in a future product line...

Arwin

09-Jun-2009, 23:42

Definitely going to hold out for the new model - I noticed that the old one has some among other speed related issues that even if I may be able to get used to them, my wife definitely isn't (and I'm expecting her to use it more)

tangey

10-Jun-2009, 02:34

SGX in the new iPhone, cool!

I'm going to also toss in a wild guess that if Apple doesn't fully support the unified shaders of the SGX (what if vertex texture fetch is missing) via GLES2, that they might be planning on moving on to Tegra in a future product line...

Than isn't going to happen. Apple took a multi-year, mult-product, multi-use license from IMG, and have also become a shareholder. Apple have also bought their own Soc design house (PA semi). I'd say there is zero chance of Apple turning to an off the shelf App Pro like Tegra.

Mike11

10-Jun-2009, 03:39

silent_guy

10-Jun-2009, 04:44

How long do you think it will take until we find out what CPU/GPU/RAM Apple used in the iPhone 3G S?
Friday next week, you'll see thousands of developers writing running their own benchmarks. It won't take more than a few hours before you'll find tons of blog posts about it.

warmi

10-Jun-2009, 04:51

"The PowerVR SGX is the graphics processor in Apple’s newest iPhones and is designed for OpenGL ES 2.0. The graphics driver for the PowerVR SGX also"

(what if vertex texture fetch is missing) ...

Indeed, it is missing :-(

tangey

10-Jun-2009, 10:12

How long do you think it will take until we find out what CPU/GPU/RAM Apple used in the iPhone 3G S?

I HOPE they used ARMv7@600MHz/SGX530/256MB but I guess something like ARM11/SGX520/128MB could be possible too.
I also hope they tweaked the display a little bit like last time (e.g. brightness, viewing angle etc.) and made the loudspeaker louder.

Anandtech have put up an article today that agrees with my view from April that its a customised samsung S5PC100 (they suggest running at 600Mhz) + SGX (they suggest SGX520),

http://www.anandtech.com/gadgets/showdoc.aspx?i=3579

Arun

10-Jun-2009, 10:36

Sorry for not having commented much on this yet, but I'm willing to bet it's at least a SGX530 given PowerVR's claimed tape-out timeframes for various SGX family members. And of course, I'm very impressed by Apple managing to bring this product to market in this timeframe - I guess I got a tad too pessimistic because of constantly being proven wrong when too optimistic... :) I'm a bit surprised my info was wrong, but heh. BTW, I don't think it's fair to call it a custom S5PC100, since so much of the front-end design work should already have happened inside Apple.

Regarding the missing texture fetch, I'm sure they had to do plenty of custom work to port SGX to MacOS X so I wouldn't be surprised if this was just a development focus issue. There is zero (read that again: zero) chance Apple goes for Tegra anytime soon, but it's true that they like keeping long-term flexibility when it comes to suppliers (witness the SGX license agreement mentioning other semi partners/foundries), presumably including IP, even if they often stick for a long time to the same one. But in that timeframe I can't imagine competitors not supporting vertex fetch anyway.

Mike11

10-Jun-2009, 11:01

Friday next week, you'll see thousands of developers writing running their own benchmarks. It won't take more than a few hours before you'll find tons of blog posts about it.
Hm, I think we're gonna have to wait a bit longer for all details like if the chip really is a derivative of Samsung’s S5PC100, what's the CPU cache size, or even the GPU clock speed etc.

NocturnDragon

10-Jun-2009, 11:49

it's in the platform notes from the latest edition of the gl es programming guite for the iphone.

Yay for respecting NDAs!

Ailuros

10-Jun-2009, 12:15

Sorry for not having commented much on this yet, but I'm willing to bet it's at least a SGX530 given PowerVR's claimed tape-out timeframes for various SGX family members. And of course, I'm very impressed by Apple managing to bring this product to market in this timeframe - I guess I got a tad too pessimistic because of constantly being proven wrong when too optimistic... :) I'm a bit surprised my info was wrong, but heh. BTW, I don't think it's fair to call it a custom S5PC100, since so much of the front-end design work should already have happened inside Apple.

The article contains a few tidbits about SGX I'm not so sure he could figure out from public documents (like Vec2 for USSE and Vec4 for USSE2). I don't know what the 3GS actually contains but it's likely that he has contacted IMG for some details on their articles to write his article.

One minor hair splitting would be that MBX's VGP isn't a fixed function unit. It's a VS1.1 geometry processor afaik.

darkblu

10-Jun-2009, 13:13

Indeed, it is missing :-(
how could you tell that?

warmi

10-Jun-2009, 15:55

how could you tell that?

Read the new OpenGL docs on their dev site ... it is there.

silent_guy

10-Jun-2009, 17:06

Hm, I think we're gonna have to wait a bit longer for all details like if the chip really is a derivative of Samsung’s S5PC100, what's the CPU cache size, or even the GPU clock speed etc.
Ok, but whether or not it's a derivative of something else is more of an academic question, right?
But things like L1 and L2 cache are trivial to determine. CPU clock speed likewise if you don't mind writing a couple of lines of ARM asm code. Absolute GPU speed is difficult, but what matters is the relative clock speed. Those numbers are going to be posted very quickly.

TimothyFarrar

10-Jun-2009, 17:25

Regarding the missing texture fetch, I'm sure they had to do plenty of custom work to port SGX to MacOS X so I wouldn't be surprised if this was just a development focus issue. There is zero (read that again: zero) chance Apple goes for Tegra anytime soon, but it's true that they like keeping long-term flexibility when it comes to suppliers (witness the SGX license agreement mentioning other semi partners/foundries), presumably including IP, even if they often stick for a long time to the same one. But in that timeframe I can't imagine competitors not supporting vertex fetch anyway.

My guess that Apple might not have full SGX hardware supported features stems from Apple's past history (IMO) of only supporting the lowest common denominator of GPU hardware via API support. Likely so they could swap in/out hardware and have differences between low and hi-end machines without ever thinking about compatibility. Since I don't know about the licensing end, I just have to take your word that this would NOT be a case of Apple planning ahead to optionally replace SGX with something else in later iPhone models.

It would be hard for me to believe that if Apple decided not to include API support for various unique SGX hardware features, that they would do so out of a time issue. For example, with unified shaders, something like VTF would have to be intentionally disabled because you'd be using the same JIT/compile system for both VS and PS. Also Apple would be getting reference driver source from the IHVs, if so wouldn't they have to remove SGX feature support from the reference drivers when porting them over?

Besides if Apple was tied down with a long term license, wouldn't it make more sense to make it a priority to fully support the unique advantages of the licensed hardware to both enable developers to better take advantage of the hardware (better iPhone apps compared to competition), and to possible make it harder to port the applications off the iPhone to competitive Windows/Linux devices and dilute the market.

tangey

10-Jun-2009, 21:04

T-mobile in holland has confirmed a 600Mhz CPU and 256M of Ram.

http://www.t-mobile.nl/iphone/specificaties.html?WT.ac=sc_iphone2_specs

tangey

10-Jun-2009, 21:08

Ike Turner

11-Jun-2009, 14:46

Beyond3D Exclusive!

http://img81.imageshack.us/img81/4564/img0141water.th.jpg (http://img81.imageshack.us/i/img0141water.jpg/)
The Samsung Louvre (B7610)

- 3.5 AMOLED 16Mcolors resistive screen WVGA (800X480)
- 800Mhz CPU (The samsung reps didn't even know wich one lol and the samsung UI skinning is so deep I couldn't even find the properties page in WM to see the CPU reported. But I'm guessing S3C6410 + MSM6250 baseband)
- 5.1Mpx AutoFocus Cam with LED flash
- Slide Out Qwerty
- 2 UIs (Pro & Media, there's a dedicated button to switch between them)
- TV-OUT
- 3.5" Headphone jack
- MicroUSB connetor (no more samsung connector to charge and headphone)
- Divx/WMV/H.264 accellerations
- GPS/WIFI/HSDPA/HSUPA/BT
- FM Radio
- DNLA support
- 1GB internal memory + SDHC up to 32Gb
- 1500mAh

Release date August 2009 (maybe later cause the current rom is slow as hell. The Omnia 2 (GT-i8000) has been delayed t'il novembre for the same reasons IIRC) and free upgrade to WM6.5 in Sep/Oct.

Sorry for the lame watermark on the pic ;)

Mak

AzBat

11-Jun-2009, 14:52

Where's the pics? :(

Ike Turner

11-Jun-2009, 14:53

Where's the pics? :(

Refresh the page it should be back.

I.S.T.

12-Jun-2009, 02:48

I got a porn pop-up from there. =\ Sex webcams. No nudity, but who knows if other pop ups might have some.

Tread with caution.

wco81

15-Jun-2009, 04:38

I didn't see it linked here but Anandtech speculates on the silicon in the 3G S.

http://www.anandtech.com/gadgets/showdoc.aspx?i=3579

Mike11

15-Jun-2009, 09:58

I didn't see it linked here but Anandtech speculates on the silicon in the 3G S.

http://www.anandtech.com/gadgets/showdoc.aspx?i=3579
Well, the only thing that isn't "clear" yet is the GPU. Anand speculates in the article that it is a SGX 520 (the smallest SGX) and writes in the comments when asked how sure he is about that: "The graphics side I'm also very sure about, although I can't post why :)".

Some people hope for SGX 530 or even more. SGX 530 is very possible, 540 a long shot. I think Apple plans to reuse this year's SoC in the 2010 model (@45nm and with some tweaks), so a more powerful SGX than the 520 would imho make more sense.

tangey

15-Jun-2009, 13:36

Well, the only thing that isn't "clear" yet is the GPU. Anand speculates in the article that it is a SGX 520 (the smallest SGX) and writes in the comments when asked how sure he is about that: "The graphics side I'm also very sure about, although I can't post why :)".

Some people hope for SGX 530 or even more. SGX 530 is very possible, 540 a long shot. I think Apple plans to reuse this year's SoC in the 2010 model (@45nm and with some tweaks), so a more powerful SGX than the 520 would imho make more sense.

Its is nearly guaranteed NOT to be 520. In Oct '08, IMG stated:-

"SGX540 is licensed to more than five partners, and has taped out this quarter in multiple SoCs, while SGX531 and SGX520 are in design with several licensees"
http://www.imgtec.com/News/Release/index.asp?NewsID=400

So 520 wasn't in Tape out at that time. Its pretty much impossible to go from "in design" to in a chip in production quantities in 9 months.

It would appear that 520 was present in some physical form in Nov '08, as it passed opengles2.0 conformance testing at that time:-
http://www.imgtec.com/News/Release/index.asp?NewsID=411

I think its highly likely to be 530, with a very outside chance of being 540.

darkblu

15-Jun-2009, 15:36

Read the new OpenGL docs on their dev site ... it is there.
i must be reading the wrong docs then, as i still can't see anything like that. technically, i see no reason for it, but then again, apple have been acting rather minimalistic re their iph/ipt GPU featureset exposure (VGP vertex shaders, hello?), in stark contrast to their desktop GPUs, where they'd go extra miles to offer as full feature support as possible, so go figure.

warmi

16-Jun-2009, 22:53

i must be reading the wrong docs then, as i still can't see anything like that. technically, i see no reason for it, but then again, apple have been acting rather minimalistic re their iph/ipt GPU featureset exposure (VGP vertex shaders, hello?), in stark contrast to their desktop GPUs, where they'd go extra miles to offer as full feature support as possible, so go figure.

Check out OpenGL ES Programming Guide for iPhone|Platform Notes|OpenGL ES 2.0 on the PowerVR SGX|Limits - look at the second bullet point.

manux

19-Jun-2009, 09:12

Teardown of iphone 3gs hardware

http://www.engadget.com/2009/06/19/iphone-3g-s-gets-the-quick-and-dirty-tear-apart-treatment-alrea/

They've identified the chipset as the SoC S5PC100 from Samsung and thus confirm an ARM Cortex A8 running at 600 MHz (operates at up to 833MHz though) and the ability to record 720p video should Apple choose to go in that direction (iPhone 3G S records VGA only). PowerVR SGX graphics and 256MB of RAM too as expected.

Mike11

19-Jun-2009, 09:39

Teardown of iphone 3gs hardware

http://www.engadget.com/2009/06/19/iphone-3g-s-gets-the-quick-and-dirty-tear-apart-treatment-alrea/

I wonder why the 3G S has a (quite big) additional chip on the system board compared to the 3G.
http://www.rapidrepair.com/guides/iphone-3g-s-repair/iphone-3g-s-dissasembly-repair-guide.html

rendezvous

19-Jun-2009, 10:19

I wonder why the 3G S has a (quite big) additional chip on the system board compared to the 3G.
http://www.rapidrepair.com/guides/iphone-3g-s-repair/iphone-3g-s-dissasembly-repair-guide.html

The flash memory is not visible on the iPhone 3G picture becouse it is mounted on the back side.

Follow link for pictures of iPhone 3G and iPhone 2G boards:
http://techon.nikkeibp.co.jp/english/NEWS_EN/20090114/164030/?SS=imgview_e&FD=-825822605

Mike11

19-Jun-2009, 11:34

The flash memory is not visible on the iPhone 3G picture becouse it is mounted on the back side.

Follow link for pictures of iPhone 3G and iPhone 2G boards:
http://techon.nikkeibp.co.jp/english/NEWS_EN/20090114/164030/?SS=imgview_e&FD=-825822605
Thanks for the info! I guess we need to see a photo of the back side to know if it's 2x16GB (front and back) or 1x32GB.

Anyone wanna bet how long it takes until the 3G S is jailbroken? I hope it's no longer than 3 months. I just got mine for a few hours and I already miss it (mostly Backgrounder, Intelliscreen and SBSettings) ;)

AzBat

20-Jun-2009, 01:17

Microsoft confirms Nvidia Tegra for Zune HD from Zune Insider Podcast...

“We actually wanted to confirm something; there’s been a lot of chat going on right now about us possibly putting the NVIDIA Tegra chip in the Zune HD. Guess what we did? Ok, right now, we’re going ahead and confirm that: Yes, the Zune HD does have the Tegra chip in it, which is so sick, so much better battery life, graphics acceleration, I mean, this thing is like a mini laptop in your hands.”

http://zuneinsider.com/archive/2009/06/19/zune-insider-podcast-24-is-available-now.aspx

Tommy McClain

Mike11

23-Jun-2009, 11:56

I was a little pessimistic about OLED touchscreens in sunlight, but it seems like they are getting much brighter very fast, they should be iPhone ready next year.
Samsung S8000 Jet with 3.1" 800x480 (WVGA) in sunlight:

http://www.mobile-review.com/review/samsung-s8000-en.shtml#3
http://www.mobile-review.com/review/image/samsung/s8000/add/pic01.jpg
http://www.mobile-review.com/review/image/samsung/s8000/add/pic02.jpg

And by the way, it's 4 days now since the iPhone 3G S was released and we still don't know what SGX it uses etc. Seems like the "we'll know everything on the first day" predictions were a bit overly optimistic.

Ailuros

23-Jun-2009, 12:47

wco81

23-Jun-2009, 15:06

But regardless of which SGX, there's a wide performance-gap between iPhone 3GS and 3G/2G.

So how is the iPhone gaming market going to develop from this new hardware?

Will developers support the new graphics capabilities right away? Can they when 99 cents games sell so much better than even $5 games?

warmi

23-Jun-2009, 16:38

tangey

23-Jun-2009, 17:07

One way to do it would be to create fat binaries .... ie one for ArmV6 and another for ArmV7 - this way you can still sell just one app while supporting both devices.

To make things even more complicated iPhone 3GS support for OpenGL 1.x is just an emulation layer running with shaders etc .... which means potentially different performance characteristics.
Interestingly, there are actually 3 different profiles now:

1. 3G with OpenGL 1.x - 2 texture units etc ...
2. 3GS with OpenGL 1.x - 8 texture units, stencil buffers , larger texture sizes etc ..
3. 3GS with OpenGL 2.x - same as 3GS 1.x but based on shaders.

Well you could do, but seeing how iphone apps are distributed purely via electronic download, its totally feasible for the app store to pick-up which device you are downloading to, and send the correct version transparently. Would keep the download size to a minimum which is important given many downloads will be via wireless/2G/3G as opposed to via PC/Mac.

Makes it more complicated for the app store to handle mind you.

I suggest that is part of the reason that Sony has gone for electronic distribution with the PSP GO. It paves the way for a complete hardware platform change (to SGX of course circa Q4 '10). The online store just has to pick up the hardware and offer/download the appropriate version of the game.

Ailuros

23-Jun-2009, 21:04

But regardless of which SGX, there's a wide performance-gap between iPhone 3GS and 3G/2G.

The performance gap of MBX for the first generation OpenGL_ES1.x cores compared to competing solutions was even wider than that at times.

So how is the iPhone gaming market going to develop from this new hardware?

I doubt any developer that would code a mobile game from scratch for the iPhone wouldn't use the iPhone3G as one of the lowest common denominator. The 3GS in that regard will probably be quite a bit faster with any possible result.

Will developers support the new graphics capabilities right away? Can they when 99 cents games sell so much better than even $5 games?

My own personal objections regarding ergonomics and game controls aside on mobile phones, you can play something like Quake3 mobile on a SoC with MBX just fine. If you now go to anything in the SGX league you could eventually enable antialiasing which is available in the mobile version. Just an example.

Simon F

24-Jun-2009, 09:13

Well for IMG the most interesting part about iPhone 3GS should be the claimed 1M sold units already; I can picture someone in the upper management drooling over a calculator right now while **zipping on his tea :lol:
I'd better go put the kettle on then - just got in a fresh box of "Kenya" tea bags.

**Tea is better uncompressed IMHO

Ailuros

24-Jun-2009, 20:21

**Tea is better uncompressed IMHO

No pun intended hm? ;)

roninja

26-Jun-2009, 11:52

MacRumours site suggesting the IPhone 3GS features an SGX535. Does that stack up?

Wishmaster

26-Jun-2009, 14:00

MacRumours site suggesting the IPhone 3GS features an SGX535. Does that stack up?

Yes it is true - proof (http://www.glbenchmark.com/phonedetails.jsp?D=Apple%20iPhone%203G%20S&benchmark=glpro)

Ailuros

26-Jun-2009, 14:56

Yes it is true - proof (http://www.glbenchmark.com/phonedetails.jsp?D=Apple%20iPhone%203G%20S&benchmark=glpro)

What am I missing in that link that doesn't tell me that it's a 535 at all?

Go ahead and run a comparison between a iPhone3GS/iPhone3G/Nokia N95 under low level 3D. Unless there's something seriously wrong with the driver the MBX with barely 1 TMU cannot under any circumstance yield higher fillrates than a SGX535 which has 2 TMUs (for which both the SGX in the iPhone3GS might even run at a higher frequency than the N95 MBX).

If you look at the fillrates between iPhone3G (MBX Lite) vs. N95 (MBX) the differences are easy to explain since the MBX has a slightly higher frequency and can execute one pixel/clock, whereby for MBX Lite it should be rather 1 pixel/2clocks.

It may very well not be SGX520, but then if it's a SGX53x variant there's a problem with the drivers. Those results point clearly in the 520 direction so far.

Xmas

26-Jun-2009, 15:02

What am I missing in that link that doesn't tell me that it's a 535 at all?
There's a GL Environment tab...

Please note: Almost all of these results are vsync limited. They are NOT representative of actual peak performance.

Simon F

26-Jun-2009, 15:03

There's a GL Environment tab...

Please note: Almost all of these results are vsync limited. They are NOT representative of actual peak performance.
Out of curiosity, I presume the benchmark is a Java application?

Xmas

26-Jun-2009, 15:14

Out of curiosity, I presume the benchmark is a Java application?
No, that's JBenchmark. GLBenchmark is a native OpenGL ES benchmark.

Ailuros

26-Jun-2009, 15:27

There's a GL Environment tab...

Please note: Almost all of these results are vsync limited. They are NOT representative of actual peak performance.

Argghh.....*runs against the next wall* I still don't get it why a 535 would give (even under a vsync limited scenario) lower fillrates than a MBX; in theory if both would have the same frequency the 535 would still have twice the fillrate (w or w/o overdraw).

(Oh and yes I'm awefully tired right now...)

roninja

26-Jun-2009, 15:29

GL_VENDOR Imagination Technologies

GL_VERSION OpenGL ES-CM 1.1 IMGSGX535-31.4

GL_RENDERER PowerVR SGX 535

whats the benefit of sticking an 535 inside the iPhone versus say the 530 in OMAP3?

roninja

26-Jun-2009, 15:40

re GLbenchmark - how low is the Samsung i8910 in the rankings. Nice to see PowerVR continuing to dominate though ;-)

warmi

26-Jun-2009, 15:56

This benchmark is still testing 1.x which means is running 1.x emulator within the OpenGL driver .

tangey

26-Jun-2009, 15:57

re GLbenchmark - how low is the Samsung i8910 in the rankings. Nice to see PowerVR continuing to dominate though ;-)

Something badly BADLY wrong with that samsung i8910, its overall performance is slower that the ARM11/MBX 8510.
Some of its triangle & fill rates are 1/3 of the 8510, and 1/4 of the iphone

Look how much stronger the samsung CPU Float/Int figures are compared to the 3GS

Appl Samsung
Int 2378 2680
Float 8920 17182

The samsung has TWICE the Float performance.

Xmas

26-Jun-2009, 16:40

I still don't get it why a 535 would give (even under a vsync limited scenario) lower fillrates than a MBX; in theory if both would have the same frequency the 535 would still have twice the fillrate (w or w/o overdraw).
Let me explain a few limits (for a vsync limited scenario):

GLBenchmark HD:
60 (fps) * 30 (seconds runtime) = 1800 frames

Swapbuffers:
60 (fps) * 10 (seconds runtime) = 600 frames

Geometry:
60 (fps) * 10216 (triangles per frame) = 612960 Tri/s

Fillrate:
60 (fps) * 320 * 480 (resolution) * 4 (layers) = 36864 kTex/s
40 (fps) * 320 * 480 (resolution) * 8 (layers) = 49152 kTex/s
40 fps is a 2:1 cadence thus one step below 60 fps. All this shows is that the fillrate in this particular test is below 73728 kTex/s. And it's entirely bandwidth limited.

frogblast

26-Jun-2009, 16:52

Argghh.....*runs against the next wall* I still don't get it why a 535 would give (even under a vsync limited scenario) lower fillrates than a MBX; in theory if both would have the same frequency the 535 would still have twice the fillrate (w or w/o overdraw).

(Oh and yes I'm awefully tired right now...)

Only if the test draws twice as much per frame, which it may not be doing. They could easily get around the vsync, simply by flushing rather than swapping, if they cared to do so. Or, they could render to an offscreen framebuffer.

Ailuros

27-Jun-2009, 06:48

Let me explain a few limits (for a vsync limited scenario):

GLBenchmark HD:
60 (fps) * 30 (seconds runtime) = 1800 frames

Swapbuffers:
60 (fps) * 10 (seconds runtime) = 600 frames

Geometry:
60 (fps) * 10216 (triangles per frame) = 612960 Tri/s

Fillrate:
60 (fps) * 320 * 480 (resolution) * 4 (layers) = 36864 kTex/s
40 (fps) * 320 * 480 (resolution) * 8 (layers) = 49152 kTex/s
40 fps is a 2:1 cadence thus one step below 60 fps. All this shows is that the fillrate in this particular test is below 73728 kTex/s. And it's entirely bandwidth limited.

Thanks Xmas now of course it's a LOT clearer. Why aren't you guys creating a techdemo along the Fablemark philosophy (witty and cute(tm)) that would also measure some performance aspects?

Ailuros

27-Jun-2009, 06:50

GL_VENDOR Imagination Technologies

GL_VERSION OpenGL ES-CM 1.1 IMGSGX535-31.4

GL_RENDERER PowerVR SGX 535

whats the benefit of sticking an 535 inside the iPhone versus say the 530 in OMAP3?

Twice the fillrate.

warmi

27-Jun-2009, 08:02

MacRumours site suggesting the IPhone 3GS features an SGX535. Does that stack up?

MacRumors:
iPhone developers, however, have discovered that the iPhone 3GS has extension files named “IMGSGX535GLDriver”

I wonder if they were referring to this post :
http://forum.beyond3d.com/showpost.php?p=1305812&postcount=16

It is like a hall of mirrors... .. hehehe

Ailuros

27-Jun-2009, 08:57

Yes you were on the right track. In retrospect it means that the 3GS graphics part by itself can turn out even above 4x times more efficient compared to the 3G.

warmi

27-Jun-2009, 09:41

Yes you were on the right track. In retrospect it means that the 3GS graphics part by itself can turn out even above 4x times more efficient compared to the 3G.

I am going to do a couple of fillrate tests this weekend just with 2d quads ... so far my game (3d) is maxing out at 60 fps (as compared to 35-40 on the old device) which doesn't tell me much.

Anyway, the real fun will start with ES 2.0 renderer specific code...

Ailuros

27-Jun-2009, 14:01

I am going to do a couple of fillrate tests this weekend just with 2d quads ... so far my game (3d) is maxing out at 60 fps (as compared to 35-40 on the old device) which doesn't tell me much.

Anyway, the real fun will start with ES 2.0 renderer specific code...

What kind of filtering are using for that game? Would it too bold to ask if something like anisotropic is possible at reasonable performance on the GS?

Lazy8s

28-Jun-2009, 03:52

Mobile SoC designers had been more likely to select an SGX variant with increased pipelines before TMUs, seeming to indicate their belief of a better balance of performance versus power and/or silicon consumption with SGX5x0 cores.

Someone suggested that Apple chose the 535 for increased video functionality... I wonder if the 535 actually is more capable in that way and whether that would matter to a VXD3x0 licensee like Apple. Maybe for video encoding, but wouldn't Apple be using some other dedicated core/DSP for that?

With their selection of the 535, I wonder if they would've picked the 540 had it been available sooner.

If Apple has specified their VXD as well into this new SoC, they'd have a single mobile solution with the best CPU, GPU, and video core.

warmi

28-Jun-2009, 04:02

OK ... I ran some basic 2d fill rate test displaying simple alpha blended 32x32 quads (single batch + single texture atlas)
Compared to the old iPhone, I was able to display about 5 times more sprites.

This test is part of my small 2d demo and with the old one I was able to display 350 32x32 blended sprites ( in addition to some non-blended and UV animated background images) while maintaining around 30 fps.
With the new one I was able to push the number of quads to 1750.

warmi

28-Jun-2009, 04:14

What kind of filtering are using for that game? Would it too bold to ask if something like anisotropic is possible at reasonable performance on the GS?

For my game I was just using GL_LINEAR/GL_LINEAR_MIPMAP_NEAREST.
This is the kind of game that wouldn't benefit much from having GL_MAX_TEXTURE_MAX_ANISOTROPY_EXT enabled.
I know it works but I haven't had a chance to see what kind of performance hit it generates.

Ailuros

28-Jun-2009, 12:31

For my game I was just using GL_LINEAR/GL_LINEAR_MIPMAP_NEAREST.
This is the kind of game that wouldn't benefit much from having GL_MAX_TEXTURE_MAX_ANISOTROPY_EXT enabled.
I know it works but I haven't had a chance to see what kind of performance hit it generates.

Thanks for the results above in your former post.

If you have time and can look into the AF performance it would be cool to know.

tangey

01-Jul-2009, 14:24

"The PowerVR SGX is the graphics processor in Apple’s newest iPhones and is designed for OpenGL ES 2.0. The graphics driver for the PowerVR SGX also"

SGX in the new iPhone, cool!

I'm going to also toss in a wild guess that if Apple doesn't fully support the unified shaders of the SGX (what if vertex texture fetch is missing) ..

Indeed, it is missing :-(

I see OS3.1 (beta) adds in additional OpenGl es2.0 extensions, so I wonder has this been addressed :-

http://www.appleinsider.com/articles/09/06/30/apple_posts_iphone_os_3_1_beta_sdk_with_new_video_ extensions.html

warmi

01-Jul-2009, 17:51

I see OS3.1 (beta) adds in additional OpenGl es2.0 extensions, so I wonder has this been addressed :-

http://www.appleinsider.com/articles/09/06/30/apple_posts_iphone_os_3_1_beta_sdk_with_new_video_ extensions.html

Uhhh ... can't really talk but :sad:

Tchock

07-Jul-2009, 07:17

AzBat

07-Jul-2009, 14:10

Anybody have any nice comparisons between Zii & Tegra?

Tommy McClain

Arwin

07-Jul-2009, 15:56

Dammit. One of the biggest reasons I want an iPhone is to develop my own software for it. But of course, Apple has to make that hard by only making the software available on OS/X ... sigh. Definitely not going to be buying a Mac anytime soon. :( And will have to get a PC of my own before I'll even attempt to install Leopard on that.

Captain Chickenpants

07-Jul-2009, 16:43

Scott_Arm

07-Jul-2009, 18:34

I have the equipment to develop apps, but I have yet to come up with a worthwhile idea!

CC

I was going to try and make some productivity apps for myself at work, but then I realized you had to pay $100 to become a developer if you actually wanted to test your app on the ipod/iphone. You can use the emulator free though. I'm pretty sure that's how it works.

AzBat

09-Jul-2009, 05:26

I was going to try and make some productivity apps for myself at work, but then I realized you had to pay $100 to become a developer if you actually wanted to test your app on the ipod/iphone. You can use the emulator free though. I'm pretty sure that's how it works.

Similar to how you write for XNA apps on Xbox, Zune & PC. All the developer tools are free. You can test or publish to Zune & PC for free, but to publish to the CommunityIndies Games on Xbox Live requires paying a $99/year or $49/4 months Premium Creator's Club subscription.

Tommy McClain

tangey

09-Jul-2009, 11:58

I think you'll find you have to pay $99 for the 3.0 SDK, the 2.1 SDK was free.

idsn6

09-Jul-2009, 12:41

That's not true.

Tchock

28-Jul-2009, 15:32

http://www.engadget.com/2009/07/28/creative-debuts-android-powered-zii-egg-for-developers-and-oems/

Cat out of bag, and I would actually get the EGG as is now. Quite impressive for Creative.

wco81

28-Jul-2009, 15:48

iTablet 3Q (read: 4Q) 09 @ ~$400
an iphone os device,
multitouch
2x Cortex A9 MP
SGX5xx,
same camera as the iphone of the day
802.11g/BT2

Rumors have picked up about this in recent weeks.

But the speculated prices is as much as $800!

That would be unsubsidized.

Maybe at that kind of price, the specs would be raised?

tangey

28-Jul-2009, 17:40

That's not true.

Great news, can you supply a link where I can download 3.0SDK for free ?

idsn6

28-Jul-2009, 23:06

Same place as always. http://developer.apple.com/iphone/

To access iPhone SDK 3.0 and additional technical resources and information, log in with your Registered iPhone Developer Apple ID and password, or sign up as a free Registered iPhone Developer today.

Arwin

29-Jul-2009, 13:02

I have the equipment to develop apps, but I have yet to come up with a worthwhile idea!

CC

I'm an idea fountain. Just name a category you're interested in and I'll hit you up with tonnes of ideas! :) With a 14 month old kid (and I guess too much time wasted on playing games on the PS3) I just a) don't push myself to realise them and b) being on a somewhat tighter budget financially as well, I become a lot more critical about which idea I'm actually going to invest in (in this case, buying a freaking mac ;) AND getting an iPhone 3GS).

I was looking hopefully towards the HTC Hero, but it's hardware seems to be too slow. Android seems to be open enough for me though, I think that it might be my platform of choice eventually. Windows Mobile wasn't too bad either though with the Compact framework, in that I could the dynamic UI I made for our business to it in a day.

wco81

29-Jul-2009, 15:11

Beyond3D.app?

Actually, I heard that someone is working on "augmented reality apps" which will overlay map data over a live video feed using GPS and compass data. Cool idea.

Arwin

29-Jul-2009, 15:41

Beyond3D.app?

Actually, I heard that someone is working on "augmented reality apps" which will overlay map data over a live video feed using GPS and compass data. Cool idea.

That was one of the ideas that I actually discussed on this forum earlier. I was thinking of making a 3d forum posts tree that sorts and filters new posts according to your interests. The forum RSS 2.0 feeds are actually already available here http://www.beyond3d.com/rss/feedslist and the iPhone wouldn't be a bad idea to try it on, considering its 3D capabilities and then using the touch screen for zooming and selecting branches of the tree for instance could be pretty cool.

That thing you mention by the way is something being developed by a Dutch company I think. I saw it somewhere a while ago. You can already try it out on Android I believe?

darkblu

29-Jul-2009, 16:04

We will Zii Egg you. I wonder if it's the Zii or still fixed function hardware handling "HD Camera" video capture (hopefully they meant HD video, because HD photo capability is obvious laughingstock)

http://www.engadget.com/2009/07/07/creative-zii-and-zii-egg-touchscreen-handhelds-served-up-by-fcc/
i didn't see that coming. and that ziilabs' zms05 is quite an interesting AP. it's the first piece of hw i've seen to conbine ARMv5TE-class (dual) cpu with a shader-based GPU. it's like as if nintendo ds was running with an gl es2 hw. or my cell phone, for that matter.

all of a sudden i'm very curious of that AP. fillrate is in the pre-3GS iphone range, but as we know that sufficed for quite a while with apple's devices.

any idea how one can get a ziiegg?

silent_guy

29-Jul-2009, 16:20

any idea how one can get a ziiegg?
You can pre-order it on their website for $400.

Two ARM 926 CPU's may be too slow. They also don't indicate clock speeds and nobody knows what this StemCell processors really do.

darkblu

29-Jul-2009, 17:11

You can pre-order it on their website for $400.

Two ARM 926 CPU's may be too slow. They also don't indicate clock speeds and nobody knows what this StemCell processors really do.

thanks for the heads-up.

from this datasheet here
http://www.ziilabs.com/products/ziilabs_zms05_datasheet_28JAN09.pdf

it appears the dual ARMv5 can go as high as 666MHz (that's the max rate of the AP's PLL) and the media engine is a flexible SIMD array with single and half fp support, multi-threaded and cache -backed. one can think of it as a unified shader engine, i guess.

o Media Processing Array - Architecture
. High compute density SIMD architecture
. 24 Processing Elements (PE) in 3 clusters
. Each cluster runs the same or independent code
. Multiple High bandwidth memory paths
. Advanced hierarchical cache structure
. Random access to memory per PE
. Shared access to ARM memory
. Independent DMA controller per cluster
. Integer, IEEE 32-bit and 16-bit floating point
. Pre-emptive task switching
. Independent frequency scaling
. Independent voltage domain

silent_guy

29-Jul-2009, 21:33

it appears the dual ARMv5 can go as high as 666MHz (that's the max rate of the AP's PLL) ...
Ah, I never stumbled into that data sheet.

They're pretty specific about the clocks of a number of blocks, but don't really say anything for the CPU's. According to ARM, a 926 goes up to 470MHz in 90nm, but they don't say anything about smaller processes.

The two 666MHz PLL's could also be the ones for MC or the processing array. Two is pretty low for such a device, I assume there's a couple more PLLs for specific interfaces.

roninja

29-Jul-2009, 23:14

Where's the gpu coming from. An inhouse creative design?

Tchock

30-Jul-2009, 09:03

Built upon 3DLabs' DMS chips. Counted as inhouse Creative.

darkblu

30-Jul-2009, 15:05

Built upon 3DLabs' DMS chips. Counted as inhouse Creative.
if somebody told me 4 years ago we'd be getting a 3dlabs-tech-powered ipod i'd have looked at them with, well, some disbelief ; ) but look where we are now.

it seems we're finally finding the ballance spot in the cpu/gpu battles. autonomous 'shader' engines that can work on a bunch of tasks, morphing to adapt various API forms, have been obsoleting the classic gpu, and now even in the most watt-sensitive segment. that indicates the concept is working, and is fairly mature.

i think i might order a ziiegg just on principle ; )

tangey

31-Jul-2009, 11:49

Same place as always. http://developer.apple.com/iphone/

I'm signed up to the 2.x SDK. However last time I tried to get the 3.x SDK, the website told me I have to sign up to the $99 program. That was a few months ago...has things changed ?

JohnH

03-Aug-2009, 12:44

...

it seems we're finally finding the ballance spot in the cpu/gpu battles. autonomous 'shader' engines that can work on a bunch of tasks, morphing to adapt various API forms, have been obsoleting the classic gpu, and now even in the most watt-sensitive segment. that indicates the concept is working, and is fairly mature.

I'd say you're jumping the gun somewhat here as there is insufficient information on performance and power consumption to really draw any concluesions. I would say that it remains a reasonable assumption that this type of architecture will not be as power efficient as dedicated HW when it comes to things like encode/decode, and, as far as 3D is concerned the quoted fill rate should be enough to raise a flag on how well it handles that type of workload.

John.

darkblu

03-Aug-2009, 14:27

I'd say you're jumping the gun somewhat here as there is insufficient information on performance and power consumption to really draw any conclusions. I would say that it remains a reasonable assumption that this type of architecture will not be as power efficient as dedicated HW when it comes to things like encode/decode, and, as far as 3D is concerned the quoted fill rate should be enough to raise a flag on how well it handles that type of workload.
well, i'm clealy being enthusiastic here. and i too think power efficiency has been the main question mark in designs like this one (i.e. larabee's and such). but at the same time, i don't think we need to think in absolute terms - a 'GP' shader architecture needs be in the same ballpark wth the GPUs in terms of power efficiency to already present a viable opportunity. i'm willing to trade in ultimate power efficiency for some versatiliy and programmability. after all, we've been paying the price of programmability in CPUs for generations now - how are the 'media' processing (i.e. vector stream crunching) tasks different in this regard?

as re the fillrate - the first gen iphone/ipod were doing fine with very similar fillrates (given, with HSR in addition) on the same output res. i don't see why the ziiegg should do much worse.

theBishop

03-Aug-2009, 17:48

I just found out my wife ordered the Zii Egg SDK for me as an anniversary present. I'm personally more interested in general purpose app/Android development than the raw 3D power, but the whole package is very exciting to me. They're supposedly shipping by the end of August.

JohnH

04-Aug-2009, 14:53

well, i'm clealy being enthusiastic here. and i too think power efficiency has been the main question mark in designs like this one (i.e. larabee's and such). but at the same time, i don't think we need to think in absolute terms - a 'GP' shader architecture needs be in the same ballpark wth the GPUs in terms of power efficiency to already present a viable opportunity.

My view on this is that you need to hit a base level of peformance before you can go fully programmable, that level of performance is probably two to three orders of magnitude higher than where we are in the handheld space right now. Incedentally I'm not convinced larabee has proven anything here yet either.

i'm willing to trade in ultimate power efficiency for some versatiliy and programmability.

But how much battery life are you willing to trade? You happy with only being able to watch 60 minutes of video vs 10 hours plus? Are happy having to recharge your phone every few hours vs every day or two?

after all, we've been paying the price of programmability in CPUs for generations now - how are the 'media' processing (i.e. vector stream crunching) tasks different in this regard?

It depends on the specific task you're looking at, if you look at video encode/decode you typically see up to an order of magnitude difference in the power envelope for fixed function vs programmable. This is primrily due to the standardisation of algorithms used which makes programmabiliy of these functions largely superfluous.

as re the fillrate - the first gen iphone/ipod were doing fine with very similar fillrates (given, with HSR in addition) on the same output res. i don't see why the ziiegg should do much worse.
Apple did a great job with the original iPhone, however the 3GS is a much more snappy user experience, given the choice I doubt people would accept the lower performance version. Basically why accept a step backwards?

To be honest, I think the bigger question is why do you think you need this level of programmability i.e. what accessible benifits do you think you're gaining from it?

Cheers,
John.

darkblu

04-Aug-2009, 17:07

To be honest, I think the bigger question is why do you think you need this level of programmability i.e. what accessible benifits do you think you're gaining from it?
my rationale is simple, John. i want 'decent' programmability from my pocketables for the same reason the industry seeks OpenCL from the desktops.

i don't view a pocketable as a 'video player', a 'music player' or any other type of appliance. i think of them as mobile computers. as such, i expect them to be useful (not necesserily stellar) in the broad range of tasks (or a sufficient subset of) i generally expect a personal computer to handle. so i'd rather have a pocketable device that is usable in a variety of situations rather than one which is the ultimate video player.

see, you can think of this as a sort of transistor metrics: usefulness of the transistors i have to carry in my pocket. by that, the best video-codec mobile ASIC in the world still does not fare as well as something less good at video codecs, but more widely used in my daily routine.

as regarding power efficiency - the moment a pocketable can last me a day on a recharge, that moment it has met my a-ok. yes, ideally i'd like my devices to be insanely power-efficient, but at the same time i am realistic enough to work with what's available. for the record, my home desktop machines are all selected on the principle 'just enough power to meet my routine' - my linuxbox is an efika machine, and my 'fancy desktop' box is a ppc mac mini. so as you see i'm not some powermonger (neither do i play games on my desktops - this is what game consoles are for).

wco81

04-Aug-2009, 17:42

Industry? Isn't OpenCL mostly an Apple thing?

Remains to be seen if the implementation lives up to the hype.

darkblu

05-Aug-2009, 01:32

Industry? Isn't OpenCL mostly an Apple thing?
how is it an apple thing? it's an (open) standartization of something that has existed before in one form or another, and whose mainstream adoption (or need thereof) has led to said standard. the fact that some os vendors would rather adopt an open standard than try to impose a proprietary one does not make it 'an apple thing'.

Remains to be seen if the implementation lives up to the hype.
it's a tool of the trade. those who need it and know how to use it may find it useful, or even indispensable. for the end user it has no significance whatsoever.

and whether particular implementations will be successful or not does not void the need for the tool. unless you want to argue that CUDA, dx compute, etc. are all artificial, unnecessary developments.

Ailuros

05-Aug-2009, 08:22

Don't mind the typical marketing parrots; what we don't have or aren't able to deliver naturally stinks. What else is new?

***edit:

http://www.xbitlabs.com/news/video/display/20090804231719_DirectX_Compute_Shaders_Less_Advanc ed_than_OpenCL_Head_of_Khronos_Group.html

http://www.techreport.com/articles.x/17321

JohnH

05-Aug-2009, 15:56

my rationale is simple, John. i want 'decent' programmability from my pocketables for the same reason the industry seeks OpenCL from the desktops.

i don't view a pocketable as a 'video player', a 'music player' or any other type of appliance. i think of them as mobile computers. as such, i expect them to be useful (not necesserily stellar) in the broad range of tasks (or a sufficient subset of) i generally expect a personal computer to handle. so i'd rather have a pocketable device that is usable in a variety of situations rather than one which is the ultimate video player.

see, you can think of this as a sort of transistor metrics: usefulness of the transistors i have to carry in my pocket. by that, the best video-codec mobile ASIC in the world still does not fare as well as something less good at video codecs, but more widely used in my daily routine.

as regarding power efficiency - the moment a pocketable can last me a day on a recharge, that moment it has met my a-ok. yes, ideally i'd like my devices to be insanely power-efficient, but at the same time i am realistic enough to work with what's available

I think you're missing my point here. If the power consumption is so high that when you are watching a video the battery runs down after an hour of use the device then becomes useless for the other task you want to use it for.

OpenCL is an interresting development, however it is also designed to run on GPUs so there is no inherent advantage to having a "sea of CPUs" vs a GPU with a sea of shaders, in fact the GPU has the advantage that it will also run graphics work loads (a lot) faster with less power consumption.

Incedentally there is no proof that this partuicular device will enable you to do any of the tasks you think you want to do with it any better than any other device (which is why I think you're jumping the gun!).

. for the record, my home desktop machines are all selected on the principle 'just enough power to meet my routine' - my linuxbox is an efika machine, and my 'fancy desktop' box is a ppc mac mini. so as you see i'm not some powermonger (neither do i play games on my desktops - this is what game consoles are for).

The correct question here is at what point do you think these machines met the bar for being "just enough for your routine" and how do you think that compares to the power of something like an iPhone or iPhone GS?

Alternatively why should we settle for things being merely adequate?

John.

darkblu

05-Aug-2009, 17:37

I think you're missing my point here. If the power consumption is so high that when you are watching a video the battery runs down after an hour of use the device then becomes useless for the other task you want to use it for.
if that's indeed the case then the device will most likely not meet its 'a-ok' status with me (even though i seldom watch videos on my handhelds for hours on). as of now we don't know how it fares there, though.

OpenCL is an interresting development, however it is also designed to run on GPUs so there is no inherent advantage to having a "sea of CPUs" vs a GPU with a sea of shaders, in fact the GPU has the advantage that it will also run graphics work loads (a lot) faster with less power consumption.
well, i never said the classic GPU does not hold the speed and/or efficiency crown at graphics. all i said was that if it only does that then it'd have less value to me than something else that does graphics not so well, but which is also usable in a bunch of other task domains. as re OpenCL, while it has clear provisions for GPUs, it does not limit itself to GPUs only.

regardless, i'd be very interested in a handheld that features OpenCL. i'm not seeing one coming, though. for instance, apple, the big adopter of the API, are quite stingent when it comes to fully exposing the programmable features of their beautiful fenced garden (vertex shaders on the VGP, anybody?). i'd love to be proven wrong, though.

Incedentally there is no proof that this partuicular device will enable you to do any of the tasks you think you want to do with it any better than any other device (which is why I think you're jumping the gun!).
it may just as well seem from aside that i'm jumping the gun, but from my perspective i'm just hopeful that they (creative) may be actually taking steps in the general direction i think these devices will move on in the future. IOW, i'm giving them the benefit of the doubt. i'll also be checking that first-hand as i've pre-ordered their SDK kit.

The correct question here is at what point do you think these machines met the bar for being "just enough for your routine" and how do you think that compares to the power of something like an iPhone or iPhone GS?
i'm taking you're asking me how i see the power of devices like the iphone in the context of their daily use?

i find them fairly adequate power-wise (the GS clearly more so. when it comes to graphics, at least). it's what tasks i can (respectively, cannot) use them for that bothers me. that latter is usually a resuilt of the platform's hardware *and* software features and policies, of course. or from a dev's perspective, it's the hardware and its 'de-facto' programmable features.

as it stands, the iphone GS' fairly-power-adequate GPU currently gives me a fancy GUI (which the the old MBX does almost as well), and the odd es2-tailored game (not yet, but any time now). it also allows me to do es2.x R&D, but for that i have to be sitting at my desk, with the device hooked to my desktop - nothing mobile there - for the purpose i could just as well use an emulator. and when you think of it, its transistor budget is comparable to the device's CPU. so why am i carrying all those transitors around, when i could be carrying an equal, but more useful bunch of them?

Alternatively why should we settle for things being merely adequate?
why shouldn't we? i find the adequacy criterion quite practical. we're not talking art or sports here, we're talking use and function.

Ailuros

06-Aug-2009, 09:59

Unless I'm reading something wrong out of that list that multimedia SIMD thing is actually SIMD per cluster with a MIMD connection between clusters.

If manufacturers need only a GP processor they could always alternatively license just IMG META where they'd also have pure MIMD units from the get go or go for a full GPU under SGX which has MIMD units at its heart already.

Each manufacturer should know why they're integrating in device X hardware Y. As for GPGPU would something like image processing do the trick for some of you?

Arun

06-Aug-2009, 12:31

darkblu, with all due respect to the ex-3DLabs/Creative guys, the Zii architecture isn't worth a billionth of the hype it has received. It's just not that great in practice. It's a relatively big chip with power consumption that's nothing out of the ordinary for what it does, and certainly much worse than good fixed-function-centric solutions. As for 3D performance - I don't have the exact data at my disposal, but I wouldn't be surprised one iota if the iPhone 3GS was a full order of magnitude faster. That's not what I call "good enough", especially given the higher die size and power consumption.

The ZMS-05 architecture makes me think of Broadcom's VideoCore. I actually think Broadcom's die size and power consumption are better though. Interestingly, it is my understanding that both architectures were designed in the UK. It's quite amazing how nearly every innovative/exotic parallel processor is engineered in the UK (I'm including ImgTech stuff in there John, don't worry). Good ole Inmos/Transputer legacy!

Certainly for unorthodox workloads, it's nice to have an innovative parallel processor at your disposal and it's good that they're making a SDK available. I'm sure it must be interesting to program for; if I had the time (which I obviously don't, see my lack posting in recent times) I might even be tempted to try. But that doesn't mean future architectures that do move towards much greater flexibility will look like that; it's not quite the holy grail Creative would like it to be.

Ailuros

06-Aug-2009, 12:35

There's a healthy portion of Transputer legacy also in IMG, so careful what you want to say you rotten all time lurker LOL ;)

Anyway something relatively OT but interesting nonetheless:

http://www.xbitlabs.com/news/other/display/20090805120152_Khronos_Group_to_Enable_Hardware_Ac celerated_3D_Graphics_on_the_Web.html

Ailuros

06-Aug-2009, 12:52

Geez I swear I hadn't read that link before I made my post on the former page LOL:

http://www.eetimes.eu/showArticle.jhtml?articleID=219100143&cid=NL_eeteu

Graphics core licensor Imagination Technologies Group plc (Kings Langley, England) is making a push to persuade chip companies to license its Meta multithreading processor as a housekeeping and operating system host in system-chips for multimedia devices.

JohnH

06-Aug-2009, 14:27

...
well, i never said the classic GPU does not hold the speed and/or efficiency crown at graphics. all i said was that if it only does that then it'd have less value to me than something else that does graphics not so well, but which is also usable in a bunch of other task domains. as re OpenCL, while it has clear provisions for GPUs, it does not limit itself to GPUs only.

GPUs are not only good at graphics, they're also good at OpenCL i.e. they are usable for a bunch of "other" things without compromising graphics performance.

regardless, i'd be very interested in a handheld that features OpenCL. i'm not seeing one coming, though. for instance, apple, the big adopter of the API, are quite stingent when it comes to fully exposing the programmable features of their beautiful fenced garden (vertex shaders on the VGP, anybody?). i'd love to be proven wrong, though.

I can't comment on Apples intentions but you need to bear in mind that VGP on MBX is a non standard HW specific API extension, OpenCL is a ratified standard API & Language.

it may just as well seem from aside that i'm jumping the gun, but from my perspective i'm just hopeful that they (creative) may be actually taking steps in the general direction i think these devices will move on in the future. IOW, i'm giving them the benefit of the doubt. i'll also be checking that first-hand as i've pre-ordered their SDK kit.

Benifit of the doubt is fine but from the limited inmformation available to date I think its hard not to be pessimistic in my opinion. Quite happy to be proven wrong.

i'm taking you're asking me how i see the power of devices like the iphone in the context of their daily use?

i find them fairly adequate power-wise (the GS clearly more so. when it comes to graphics, at least). it's what tasks i can (respectively, cannot) use them for that bothers me. that latter is usually a resuilt of the platform's hardware *and* software features and policies, of course. or from a dev's perspective, it's the hardware and its 'de-facto' programmable features.

as it stands, the iphone GS' fairly-power-adequate GPU currently gives me a fancy GUI (which the the old MBX does almost as well), and the odd es2-tailored game (not yet, but any time now). it also allows me to do es2.x R&D, but for that i have to be sitting at my desk, with the device hooked to my desktop - nothing mobile there - for the purpose i could just as well use an emulator. and when you think of it, its transistor budget is comparable to the device's CPU. so why am i carrying all those transitors around, when i could be carrying an equal, but more useful bunch of them?

Simple, because most people are not happy with merely adequate. Seriously, use an iPhone 3GS and then try and move back to a 3G, you'll find your expectations as to what is adequate will have moved up a significant notch.

As I've said before, you need to reach a base level of performance before can start compromising certain areas for the sake of additional (unnecessary) flexibility. Handheld is currently some way of hitting that base performance, in fact due to power and size constraints its likely to remain short of that performance point for some time.

why shouldn't we? i find the adequacy criterion quite practical. we're not talking art or sports here, we're talking use and function.

Sorry, but I think you need a reality check, "adequate" does not sell chips or gain you market share, pushing the boundaries is what drives markets forwards, if you don't push someone else will.

John.

wco81

06-Aug-2009, 15:33

Among the previews for the Zune HD in the last couple of days were claims -- not official ones yet -- that Tegra can deliver MP3 playback which last days, like even over a week?

Hype or possible?

darkblu

06-Aug-2009, 15:53

ok, i don't want to further hijack this thread, so i'll just make one comment on a subject that concerns me.

I can't comment on Apples intentions but you need to bear in mind that VGP on MBX is a non standard HW specific API extension, OpenCL is a ratified standard API & Language.
yes. at the same time OpenCL would be a major endeavor to bring to a hanheld, in comparison to which said IMG extension would have been a non-effort. incidentally, that extension also happens to make the MBX a good deal more useful than it is at its current exposure level - something i'm sure apple have known from the get go. so tell me again why i should be expecting apple to give us openCL on the GS and why i should not pay attention to other offerings like creative's product, that make bold, perhaps naive, strides in that direction.

see, i'm not attacking the GPU, and even less so the SGX in particular - far from that - i've ordered a pandora just on the premise of getting intimate with and making full use of that particular GPU. i cannot accept, though, that apple's application of this tech is the most the hardware can do for me (us, the industry), and in this line of thoughts, that devices like the iphone are what i could be best hoping for. sorry, John, i'm just not buying it. yes, maybe from aside it seems i'm getting ahead of your ideal 'baseline performance' curve, but i think i seek nothing out of our current reach.

ps: i'll surely report back on the zii's power efficiency once i have it.

Arun

06-Aug-2009, 16:29

Among the previews for the Zune HD in the last couple of days were claims -- not official ones yet -- that Tegra can deliver MP3 playback which last days, like even over a week? Hype or possible?Definitely possible, although the precise numbers depend on the battery size and the DAC used. NV claims 100-130 hours for smartphones and 600 hours for netbooks; the difference between the two is exclusively the size of the battery since both assume the screen is off etc...

Wolfson has revealed that the Zune HD uses their power management silicon, whereas MS has confirmed it also uses a Wolfson DAC. Therefore, it's most likely based on either the WM8351 or the WM8352, which also support the APX 2600 according to the list on Wolfson's own site: http://www.wolfsonmicro.com/productListings/powermanagement - alternatively, it could be a combo of the recently announced WM8310 and another DAC.

For optimal power consumption, you want a higher-end DAC such as Wolfson's WM8903 or Cirrus' CS42L55. My impression is that it's not very likely the Zune HD uses that, so I certainly wouldn't expect 130 hours - and given that PMP batteries are usually smaller than smartphone batteries, I probably wouldn't even expect 100 hours. But we'll see, and it should still be a very high number compared to existing products.

On that note, better poof again!

Lazy8s

07-Aug-2009, 08:23

I've been wondering if the processing demands of future mobiles wouldn't be better served by a META architecture processor instead of an ARM.

I expect Pure will expand META's deployment as they start expanding their own product line, and recent PowerVR video cores already integrate an MTX microcontroller.

Ailuros

07-Aug-2009, 08:51

Ailuros

07-Aug-2009, 09:06

see, i'm not attacking the GPU, and even less so the SGX in particular - far from that - i've ordered a pandora just on the premise of getting intimate with and making full use of that particular GPU. i cannot accept, though, that apple's application of this tech is the most the hardware can do for me (us, the industry), and in this line of thoughts, that devices like the iphone are what i could be best hoping for. sorry, John, i'm just not buying it. yes, maybe from aside it seems i'm getting ahead of your ideal 'baseline performance' curve, but i think i seek nothing out of our current reach.

ps: i'll surely report back on the zii's power efficiency once i have it.

Well as I said if you'd want a multimedia only solution for a mobile phone the better place to look for within IMG's IP portofolio would be a META GPP. Coincidentally it seems that it's the actual direction IMG is planning to drive META into for the foreseeable future.

For a multimedia only device a general purpose processor is better suited and I'm still wondering why you two haven't reached yet common ground. Since both of you seem to repeatedly throw the iPhone 3GS into the mix, it's by far not intended as a multimedia only device; IMHO Apple knows exactly what its doing and they are in the longrun building themselves up in the handheld gaming market. And no for that a pathetic multimedia only thingy isn't by far good enough.

AzBat

10-Aug-2009, 23:54

http://www.blogcdn.com/www.engadget.com/media/2009/08/zunehd-in-fcc-picture-17-rm-eng.jpg
http://www.blogcdn.com/www.engadget.com/media/2009/08/zunehd-in-fcc-picture-12-rm-eng.jpg
http://www.blogcdn.com/www.engadget.com/media/2009/08/zunehd-in-fcc-picture-13-rm-eng.jpg

Yes, that's the Zune HD & yes, that looks like Windows Mobile running on it. :)

http://www.engadget.com/2009/08/10/zune-hd-hits-fcc-in-prolific-photo-shoot-16gb-and-32gb-capaciti/

Tommy McClain

Ike Turner

11-Aug-2009, 01:49

Yes, that's the Zune HD & yes, that looks like Windows Mobile running on it. :)

Tommy McClain

That's Windows CE not Mobile. The Zune UI runs "on top" of CE. And Windows Mobile is a a subset of Windows CE. WM6.X = CE 5.2 kernel, WM7 = CE 7 "chelan" kernel. IIRC the Zunes are based on CE 6.

AzBat

11-Aug-2009, 02:06

That's Windows CE not Mobile. The Zune UI runs "on top" of CE. And Windows Mobile is a a subset of Windows CE. WM6.X = CE 5.2 kernel, WM7 = CE 7 "chelan" kernel. IIRC the Zunes are based on CE 6.

:p technically you're correct, but to most people WindowsCE is no longer in the public eye. :)

BTW...

http://www.blogcdn.com/www.engadget.com/media/2009/08/zune-hd-keyboard-rm-eng.jpg

http://www.blogcdn.com/www.engadget.com/media/2009/08/zune-hd-rm-eng-browser.jpg

http://www.engadget.com/2009/08/10/zune-hd-on-screen-keyboard-and-browser-caught-on-camera/

Neato, frito!

Tommy McClain

Arun

11-Aug-2009, 09:32

Looks like I was right, it's indeed the WM8352 PMU/Audio codec. Not the lowest-power audio codec to drive Tegra, but it's cost efficient and probably good enough. I can't make out what the chip on the right is; might be the AR6002 in BGA form (weird, I'd expect WLCSP) or, well, practically anything else.

Richard

11-Aug-2009, 19:10

AzBat

12-Aug-2009, 03:56

Loving the bulge on the key row when you press it. Btw, Neowin puts (http://www.neowin.net/news/main/09/08/11/zune-hd-prices-leak-for-16gb-and-32gb-versions) this at $220 for the 16gb and $290 for the 32gb version.

Great price. Here's confirmation at release date (Sept 15) & packaging...

http://cache.gawker.com/assets/images/4/2009/08/504x_IMG00013-20090811-1742.jpg

http://cache.gawker.com/assets/images/4/2009/08/504x_IMG00012-20090811-1429.jpg

http://gizmodo.com/5335353

Any guesses at how well their launch will go provided the pricing is right?

I think it will do a lot better than previous models, but somehow I think the lack of a phone will keep people jumping from the Apple ship. I just wished they had demoed the games & marketplace features. What's taking them so long?

Tommy McClain

wco81

12-Aug-2009, 04:39

So they seem to be underlining the HD aspect. HD Radio doesn't seem to be going anywhere. So HD video, how much is that dock and how much is HD content?

How easy will it be to load it with your own HD content, ripped Blu-Rays especially? Lets face it, PMPs would never be as popular as they are now if people couldn't at least easily rip their disc collections.

Finally they have to have a games and app. story, if not on day 1, then in the foreseeable future.

Are they going to roll this out internationally? If they announce plans to release in Europe say in a month or two after US launch, then it'll be a good indication of how confident and ambitious their strategy is.

AzBat

12-Aug-2009, 05:10

I wouldn't plan on an International release unless the demand explodes over here in the US. If MS can't do well here, there's no way they'll do well outside the country.

BTW, Engadget has some new pics & a glorious video hands-on...

http://www.engadget.com/2009/08/11/zune-hd-video-hands-on-chock-full-of-media-edition/

Tommy McClain

AzBat

12-Aug-2009, 05:19

So they seem to be underlining the HD aspect. HD Radio doesn't seem to be going anywhere. So HD video, how much is that dock and how much is HD content?

BTW, on the Zune.net site you can buy the Zune accessories @ 50% off. Right now they have the Zune A/V Pack with Zune Dock, component video cable, wireless remote & charger for $30US. Says its compatible with all Zunes, but the pics of the ZuneHD Dock look specifically molded for that device. Plus, not sure if the connector is the same. I would hope so.

As for the content I'm sure the content will be the same as the HD content on Xbox Video Marketplace. In fact, they will rebrand the Xbox Video Marketplace as the Zune Marketplace later this year. It's been speculated that will allow you to purchase content that will be compatible on either device.

Tommy McClain

Richard

12-Aug-2009, 05:55

Great price.

I know the price is lower than the iTouch but I can shake the feeling the profit margins must be huge. For that price you can buy an XBOX 360 and a new game!

Btw, it has been confirmed they're going internationally (http://www.neowin.net/index.php?act=view&id=54088) though no exact mention of dates or country list.

I'm still on the fence on whether to get one, partially because my music collection alone is a little over 10gb, nevermind my videos + pics which would make even the 32gb version inadequate. If I can't have my whole media collection on the portable device and if I have to compromise and stick with a fraction, why go for a high-end PMP at all then, might as well just buy a 30 euro cheapo 4gb MP3 player.

Tchock

13-Aug-2009, 14:03

33 hours MP3 (wireless off), 8.5 for video.

I know it's better than the Touch, but tiny batt? =/

wco81

13-Aug-2009, 14:29

Remind me again, is the Tegra a more expensive core than the SGX whatever core that people think the iPhone 3GS has (and presumes the new Touch will get)?

Probably depends on the volume deals they negotiate.

AzBat

13-Aug-2009, 15:05

The Zune HD Leaked prices & date have been confirmed by MS. You can pre-order from their web site...

http://www.zune.net/en-us/mp3players/zunehd/default.htm

Only Walmart, Amazon, Microsoft Store & Best Buy.

Also, got this from my email subscription...

Starting September 15th you can customize a Zune HD with 10 new exclusive designs on five different colors.

http://f.email.zune.net/i/38/1633616494/ZUNE081009_originals.jpg

They're still doing the Zune Originals program. Pretty cool.

Tommy McClain

Lazy8s

13-Aug-2009, 20:53

I've yet to learn of an IMR with a smaller die size than a comparably performing PowerVR TBDR, so I assume the 3G S's SGX535 -- probably less than 6 mm^2 -- is less costly than Tegra's graphics core.

Just as the chip technology development from Britain shows extensive roots from Inmos personnel and designs -- http://business.timesonline.co.uk/tol/business/article6788569.ece -- technology development from Colorado Springs is also largely the legacy of Inmos. I live less than a mile from their old plant here coincidentally (which next became a home to the Cray supercomputer company), and the number of local technology companies which owe to, as well as the degree to which they owe to, Inmos is surprisingly great.

http://www.inmos.com/

Reunion dinners are even held here downtown periodically by former personnel, apparently.

JohnH

14-Aug-2009, 09:52

Not sure if Tegra comes with a big chunk of embedded memory or not, seems likely to me, which would mean its area would be significantly higher than an equivalent performing SGX.

John.

Arun

14-Aug-2009, 10:13

Not sure if Tegra comes with a big chunk of embedded memory or not, seems likely to me, which would mean its area would be significantly higher than an equivalent performing SGX.It doesn't, unless you consider ~192KiB big nowadays... :) Certainly it's negligible to what the GoForce 4500 had back in the days. And no, it's not a tiler. Presumably this is used for HierZ and/or the upper layers of compressed framebuffer data.

Even if SGX 535 was bigger (and I have no idea about that), I'm not sure it's a fair comparison since it's not strictly a handheld GPU, and so has a richer feature set than Tegra. At the same time, given the 535's low ALU:TMU ratio and Tegra's less flexible and presumably less efficient *per unit* performance on the other side, I don't even want to start thinking about complex shader performance!

Regarding battery life - 8.5 hours video is about what I'd have expected, but 33 hours audio is surprisingly slow. Apparently they mentioned that was conservative and they'd probably revise it upwards - still, it does make me wonder how big the battery is and why they were too cheap to use a WM8903+PMU instead of a WM8352. Oh well, I'll probably be sticking to be 3G S anyway :p

JohnH

14-Aug-2009, 12:56

It doesn't, unless you consider ~192KiB big nowadays... :) Certainly it's negligible to what the GoForce 4500 had back in the days. And no, it's not a tiler. Presumably this is used for HierZ and/or the upper layers of compressed framebuffer data.

Hmm, not convinced 192KB buys you anything significant wrt IMR graphics, obviously you can include part of the compressed data structure, but ultimately the bulk of the data will sit in external memory. Do we know how wide the mem I/F is and what rate its running at?

John.

Arun

14-Aug-2009, 13:07

Hmm, not convinced 192KB buys you anything significant wrt IMR graphics, obviously you can include part of the compressed data structure, but ultimately the bulk of the data will sit in external memory. Do we know how wide the mem I/F is and what rate its running at?Sure, according to NV's public spec sheet on their site it's 32-bit LP-DDR1 @ 166MHz (333MHz effective) for every SKU except the Tegra 650 that's specced for 200MHz. Obviously this is shared with other subsystems.

That's 4 bits per pixel at 800x480, or 128 bits/16 bytes per 16x16 block (I'm taking 16x16 since that's how G80 works). Seems to me that could save quite a bit if combined with fairly expensive compression/etc. silicon... Remember even in the corner case where you really need 64 bits per pixel, you'll still save 6.25%. I know, I know, how useful! (my point simply being there might be plenty of tiles where you still get 25-50% of the benefit)

Of course, none of this helps at all for alpha blending.

wco81

14-Aug-2009, 15:45

This specs. page is listing 24 hours audio and 4 hours video for battery life:

http://store.microsoft.com/microsoft/Zune-HD-32GB/product/41941DC9#ctl19_tcla_a

Arun

14-Aug-2009, 16:00

That spec page is wrong, MS has said so themselves. As for why they haven't updated it yet, uhhh :|

wco81

14-Aug-2009, 19:06

What's odd is the footnote says it's based on testing in August 2009, using preproduction, you may not get the same results, blah, blah, blah and other disclaimers.

Ike Turner

20-Aug-2009, 11:50

Back to the Original topic (future phone specs)

HTC Leo: what to you tink ?
http://www.mobiletechworld.com/2009/08/14/htc-leo-renders-supposedly-leaked-again/

It's a shame WM7 is so late. If the specs are true it should be able to support it (WM7) without a problem.

Mariner

20-Aug-2009, 13:17

Regarding the Leo, it doesn't appear as though anyone really has much of a clue what it is going to be yet as various 'leaked' images seem to contradict each other directly. The videos showing the operation of the new TouchFlo3D Manila 2.5 and 2.6 are encouraging though. A nice-looking and reasonably complete interface. Because it is apparently so 'finger-friendly', many are assuming Leo will have a capacitive screen. I can't say I ever really need to use the stylus on my Touch HD running 2.1 though so who knows if this is actually the case?

Can't wait until someone has cooked one of these new Manilas into a ROM for my Touch HD. :cool:

Ike Turner

20-Aug-2009, 13:55

I wouldn't be surprised if the Leo and Firestone have capcitive screens. A Samsung rep told me a few months ago that HTC was working on WM 6.5 devices with capacitive panel for the holydays season but that they (samsung) are probably going to wait for WM7 and the sucessors of the Omnia II to have capcitive panels on their WM products.

theBishop

26-Aug-2009, 12:49

Just a heads up on the Zii, they shipped late last week. Fed-Ex has mine in Anchorage Alaska (i'm in Philadelphia), so we should start seeing personal pictures pop up in the next couple days.

The initial devkit release is available on their website, and it's almost entirely 32-bit .deb packages. There's only one windows exe and it's just for exploring the filesystem. Also more detailed SDK is now available.

frogblast

27-Aug-2009, 01:25

warmi

27-Aug-2009, 14:46

darkblu

27-Aug-2009, 15:05

So this thing is running Android but with ability to run native C++ code , correct ?
it can run either android or its native 'plaszma os' - another proprietary-enhanced linux distro.

theBishop

27-Aug-2009, 21:06

it can run either android or its native 'plaszma os' - another proprietary-enhanced linux distro.

Yep, and while Android support is drawing most of the headlines, Plazma is likely to make better use of the Zii's features. At least in the near term.

theBishop

27-Aug-2009, 21:12

So, what you're saying is that you'll have a set of performance numbers ready for us to tear apart by next week? :wink:

I'm less interested in the raw performance personally, but if there are some common things you'd like to see, I'd be happy to work on it. It may take me a little while to figure out the "array processor" elements. Creative's technical documents sound very similar in philosophy to Sony's Cell processor. Though obviously on a smaller scale.

darkblu

28-Aug-2009, 16:38

my zii egg arrived yesterday and here's my first recon of the device. due to limited chances to dig deeper* it will be focused entirely on

power efficiency

the most power-demanding task i could run on the device since its first full battery charge last night, was video playback. for the purpose i used one of the two demo flicks, 'elephant's dream' (a really nice open project short-film production) of resolution 1280x720p, and duration 10:53. i played that back 6 times, to a battery meter drop of 50% (2/4 bars). i belive it's safe to extrapolate from that that the device can playback 720p video for at least 2h straight. in the course of the playback, the device's back got slightly warm, but nowhere near what an ipod touch 2g gets under similar tasks.

so these are the dry facts. the more i thought of those, though, the more my respect for the 3dlabs guys grew. bear with me:

the device's video playback is achieved not through some specialized codec silicon, but through GP stream processing. that is, in the same way my desktops decode video. the funny thing, though, is that zii is actually better at that task than one of my desktops - a 1.25GHz altivec-wielding ppc, which reaches its performance limit at 720p; under VLC (the fastes player i've come across on that desktop) it already tends to drop frames at that resolution and 30Hz framerate. not the zii, though - its 720p is butter-smooth. with superb down-scaling to its native 480x320.

the implications?

i have a device fitting in my pocket, which device outperforms my everyday desktop (15-30W) in one particular GP streaming class of tasks, without the use of dedicated silicon. and said device can do that for more than 2h on a single battery charge. and by virtue of GP, i, too, can make use of that compuational power for my own purposes.

wow.

i don't know about you, but that makes me glee like a schoolgirl. if i were at nasa i'd be stock-piling zms chips by racks as we speak.

seriously, though, as you can see i'm deeply impressed so far. i hope to be able to give you my impressions from the sdk next.

* my zii sdk installation was bogged down by a persistent problem with ubuntu hardy's networking under parallels on my mac mini. it's a known issue involving wifi, parallels and ubuntu. i plant to revert to wire tonight and continue with the sdk installation.

JohnH

29-Aug-2009, 13:27

the most power-demanding task i could run on the device since its first full battery charge last night, was video playback. for the purpose i used one of the two demo flicks, 'elephant's dream' (a really nice open project short-film production) of resolution 1280x720p, and duration 10:53. i played that back 6 times, to a battery meter drop of 50% (2/4 bars). i belive it's safe to extrapolate from that that the device can playback 720p video for at least 2h straight. in the course of the playback, the device's back got slightly warm, but nowhere near what an ipod touch 2g gets under similar tasks.

so these are the dry facts.

Bitrate @ video codec used? Show me 2h+ 40mbs,1080p, h264 HP, then I might be impressed ;)

John.

frogblast

29-Aug-2009, 19:00

so these are the dry facts. the more i thought of those, though, the more my respect for the 3dlabs guys grew.

That isn't too impressive. The 2nd Gen iPod Touch turns in much better battery life than that (most reviewers claim of 5-6 hours w/ 50% screen brightness), and with a much smaller battery (750 mAh vs. 1200 mAh). Now, it does depend on what video you're decoding, but that is a fairly massive gap.

As far as performance benchmarking goes, I am more interested in the mundane, conventional 3D benchmarks (vertex processing rate, blended fill, texture sample rate, etc).

For example, the wikipedia page stats: 42M textured pixels/sec, 21M vertices/sec. Vertex rate is pretty good for a device in this category, but efficient software vertex processing is fairly easy. You would expect it to scale fairly well with the GFLOP rate, which it appears to do.

However, the texture sampling is one place where a pure-software solution has never been shown to perform well, and their advertised texture rate very sub-par. Does anyone know if they have dedicated texture samplers like Larabee, or if they're trying to do it all in software?

Lazy8s

30-Aug-2009, 03:29

Two hours plus of decent HD is just good enough to be useful, so achieving usefulness on a fairly demanding, highly specialized task with general purpose hardware is impressive in its own right.

I suspect its relative performance, however, compared with some alternate architectural approaches to the same GP streaming end would be underwhelming.

darkblu

01-Sep-2009, 03:28

Bitrate @ video codec used? Show me 2h+ 40mbs,1080p, h264 HP, then I might be impressed ;)

John.
you know of a < 1W stream processor/dsp that does that? ; )

to answer your question, though: AVC1, VBR (peak 4.2Mb/s, avg across the stream 2.66Mb/s), 1280x720p @24fps.

Arun

01-Sep-2009, 10:23

The question is not one of programmability, but rather of specialization. Would you consider a Tensilica Xtensa-based DSP with plenty of specialized video instructions to be a "DSP"? If so, there's one in Qualcomm's Snapdragon that does 720p at well under 1W (moving to 1080p in Snapdragon2), even much less than the Zii. And of course, at a bit more than 1W because it's optimized more for area than power, it's also what you have in either NVIDIA or AMD graphics cards.

There's also Broadcom's VideoCore as I said earlier, which was also designed by an UK company (Alphamosaic) with a similar design philosophy as the Zii; they do 720p H.264 High Profile decode & encode at well under 1W. Like the Zii, it's not very specialized and can even do OpenGL ES 2.0. They're moving to 1080p now and the original version of this very same core aimed at VGA video decode shipped in millions of devices: specifically, the original iPod with Video! Once again, zero hardware acceleration. If they go on, I'm sure they could easily do <1W for 1080p 40Mbps H.264 HP on 28nm (or on 40nm with more cores at a much lower voltage/frequency maybe).

There are plenty of weird little startups with unclear architecture I won't get into here, but perhaps most interesting are the wireless baseband startups (I know, I always get back to them!) that prove it's all about specialization, not programmability. Icera, with a fully programmable but very exotic approach (the PHY is a single instruction stream although there are multiple 'programmable' units that each have multiple execution units) that is the unambiguous HSPA+ leader, and Altair for WiMax/LTE with small super-specialized processors working on small parts of the overall problem one step after the other, each seemingly more efficient in both area and power than any commercially available fixed-function alternative implementing all the possibilities.

In the end, specialized solutions that are also programmable or at least highly configurable (i.e. PowerVR VXD) continue to have a massive advantage over the general-purpose approaches of architectures like the Zii or even GPGPU. The question is always whether the target market (and the silicon's average utilization when it is available) is large enough to warrant dedicated hardware; for things like video encode/decode, it seems very clear that is the case and the Zii's approach feels like a historical oddity.

For other problems, exotic architectures --including GPGPU to a certain extent-- holds a lot of promise although it is certain to be too complex or sometimes limited in some ways for certain consumer apps. The serial-centric CPU is also a historical oddity, and as you imply it's unclear how long it will prosper in the most performance or power-centric applications where special-purpose hardware either makes no sense or cannot be afforded. Intel's best shot at fighting this, besides playing the same game to a certain extent with Larrabee, is their 'many mini-cores (plus a few big OoOE ones)' strategy they've talked about in the past; it's still unclear to me when and even whether that will ever come to fruition.

3DLabs should be applauded for actually getting a seemingly very solid solution to market (even if I don't really approve of the marketing tactics) and it's good to see them attract attention to exotic architectures; however, there is nothing fundamentally unique about their design philosophy.

JohnH

01-Sep-2009, 15:31

you know of a < 1W stream processor/dsp that does that? ; )

Thats exactly my point, fixed function video decoders do exactly that for a lot less power.

to answer your question, though: AVC1, VBR (peak 4.2Mb/s, avg across the stream 2.66Mb/s), 1280x720p @24fps.

Doesn't sound like a particularly demanding clip to decode, which profile?

Captain Chickenpants

02-Sep-2009, 16:53

Also worth remembering that just because a stream is marked as a particular profile does not mean it is necessarily using all the features of that profile, the encoder could be using just one fairly simple feature from that profile in order to qualify as a high (or main) profile encoder.
It is the decoder that is more tightly specified, a decoder that supports a particular profile must support all the features of that profile.
It is what makes developing encoders/decoders interesting. Decoders are more complicated in that they have to support everything, whereas encoders can give you freedom to be innovative.

CC

darkblu

02-Sep-2009, 17:14

Thats exactly my point, fixed function video decoders do exactly that for a lot less power.
whereas mine was that i have little use of video decoder silicon per se ; )

Doesn't sound like a particularly demanding clip to decode, which profile?
i dunno, i'm not a video codec guy - couldn't tell mjpeg from mng. i thought the fourcc was sufficient to tell the profile.

ps: Arun, your post was informative, and has been appreciated (the alphamosaic guys blipped once on my radar not long ago, and now they're dully tracked), i just don't feel like delving into industry-vs-market-vs-efficiency-vs-utilization discussions atm; i'm yet to carry out my gl tests on the zii, and i hope to be able to come up with more informative numbers next time i post. so far, though, the battery drainage of the device has been nothing out of the ordinary for the class, considering i've been doing little but demoing it around.