ipod touch 4G

[infinity4]

So, this device in comparison to iPhone 4, has no GPS, non IPS panel and an inferior camera - but most significantly, only has 256MB of RAM when its bigger brother has 512MB.

What is the strategy behind this? Cost reduction? To keep older gens longer?

Anyway, I am disappointed about this, since it kinda makes me feel that this won't last as long as other models.

[AlphaWolf]

The old ipod touch also lacked features of the original iphone. It's not like this is a new strategy for them to be selling something for half the price with less features.

[Arwin]

The RAM is surprising to be honest. So far, the iPod lost features versus the iPhone but often had better performance. Now it seems that they want the camera features in there that the iPod never had before, presumably because they also want those available in games and to help boost the usefulness of FaceTime. But apparently then 512MB got too expensive?

The strategy is clear though - it needs to be cheap. That's how I got my first iPod, and I only dared trigger the iPhone purchase for my wife once I new the iPod's interface gelled with her.

I have to say though - the iPod's tinny mono (at least there is zero stereo separation if it has stereo, but I doubt it) speaker sound has been a disappointment for me from day one, and I would gladly get a thicker iPod if it had the much beefier, stereo speakers from the iPhone.

[brain_stew]

What is the strategy behind this? Cost reduction?.

Yup. Well profit maximisation more accurately. The differences aren't going to cost many sales but they'll help Apple's bottomline.

Anand's review killed it for me and I was quite interested in picking one up.

[Mike11]

Is it certain that all new iPod touch versions have only 256MB RAM? The 8GB version was always a little different and I haven't seen a teardown or analysis of anything other than the 8GB version yet.

[Silent_Buddha]

Anand's review killed it for me and I was quite interested in picking one up.

Same here. Especially when he got to the part about FaceTime. Something I was considering picking this up for. But the built in speaker is inadequate in anything but a very quiet room, microphone is on the other side, and you have to shell out extra cash for a headphone with mic if you wanted to go that route.

Throw in the fact that the camera used is an absolute POS (the example photo of the ball coming out VERY green instead of Yellow), the CPU is underclocked, half the ram, cheap buttons, extremely subpar screen for this day and age, no flash memory slot, etc... And yet still charging an extremely high premium considering the cheap components used.

I think Apple just made my choice between an iPod Touch and a Zune (or possibly WP7 depending on price and contracts) a lot easier to make.

Regards,
SB

[Pressure]

With regards to the screen, it still looks a lot better than the majority of displays out there.

The thing is it is getting compared to the currently best display in the mobile segment and hardware-wise it is still faster than most smart phones.

I mean, the former iPod Touch didn't just all of a sudden become a shitty device.

[idsn6]

Same here. Especially when he got to the part about FaceTime. Something I was considering picking this up for. But the built in speaker is inadequate in anything but a very quiet room, microphone is on the other side, and you have to shell out extra cash for a headphone with mic if you wanted to go that route.

Throw in the fact that the camera used is an absolute POS (the example photo of the ball coming out VERY green instead of Yellow), the CPU is underclocked, half the ram, cheap buttons, extremely subpar screen for this day and age, no flash memory slot, etc... And yet still charging an extremely high premium considering the cheap components used.

I think Apple just made my choice between an iPod Touch and a Zune (or possibly WP7 depending on price and contracts) a lot easier to make.

Regards,
SB

Wait. The new iPod touch has a worse microphone, camera, RAM, and display compared to the flagship iPhone 4...and this has somehow driven you into the arms of the Zune, which has no microphone, no camera, a quarter the RAM of the iPhone, and a screen with about a fifth the resolution?

[Lazy8s]

Apple's missing an opportunity to push the platform by not reinforcing the iPhone 4 spec with a follow-up device that shares its hardware.

[Mike11]

Apple's missing an opportunity to push the platform by not reinforcing the iPhone 4 spec with a follow-up device that shares its hardware.
I'm not so sure that an iPod touch that's identical to the iPhone 4, minus cellular modem, would have been overall beneficial for Apple. Especially if the iPhone comes to all US carriers in the next 4-9 months and subsequently will also be available contract- and SIM-lock free, like it already is in most countries. Apple makes way more money on the iPhone and a too good "cheap" iPod touch would just cannibalize iPhone sales and lower Apple's margins and earnings.

[Silent_Buddha]

Wait. The new iPod touch has a worse microphone, camera, RAM, and display compared to the flagship iPhone 4...and this has somehow driven you into the arms of the Zune, which has no microphone, no camera, a quarter the RAM of the iPhone, and a screen with about a fifth the resolution?

Not the current Zune, but a supposed future Zune device that would share some of the features of WP7, such as Xbox Live gaming for example.

With regards to the screen, it still looks a lot better than the majority of displays out there.

The thing is it is getting compared to the currently best display in the mobile segment and hardware-wise it is still faster than most smart phones.

Well the Anandtech piece compared it to a variety of displays, and the iPod touch came in a very VERY distant second to last place. It was even remotely close. For the price they are asking 229-399 USD, they should have done a lot better than that crappy screen, IMO. Sure it may be better than the iPod 3GS display, and better than the majority of last years displays, but the fact is, it is releasing now when much better displays are commonly available.

Regards,
SB

[RecessionCone]

Sure it may be better than the iPod 3GS display, and better than the majority of last years displays, but the fact is, it is releasing now when much better displays are commonly available.


Better with regards to displays is very subjective. Many people seem to like AMOLED screens with the Pentile matrix (like the Droid Incredible, Nexus One, etc.)
I think they are hideous. They have only 2/3rds the resolution they claim to have (since each pixels has only 2 sub-pixels instead of 3), and they can't render text well at all, since sub-pixel antialiasing is completely defeated by their asymmetric sub-pixel layout.
They may look good for photos and videos, but for text, which is the main reason I use mobile devices, they are hideous. Everything looks like it's being viewed through a screen door.

I would choose the iPod touch's poor black level and narrow viewing angle, coupled with its extremely high pixel density, over the purportedly superior AMOLED display - any day.

[Arun]

I actually cancelled my iPod Touch 4G pre-order today given just how bad the screen seems (I knew it might be worse, but not by that much) and how disappointing the 256MB RAM is. The weak speaker doesn't help either. Of course, Apple benefits since that means I've finally decided to bite the bullet and buy an iPhone 4...

The lower RAM is nearly certainly pure greed, but the screen very likely isn't (although it might have helped). The quantities Apple needs for these products is phenomenal, and there were rumours that they had run into supply shortages for the iPhone 4 screen. I don't know if that's true, but there's no way they could have used the same screen for the iPod Touch 4G and not run into big supply problems. So they had to choose another screen from another supplier, and it had to be the same size and the same resolution. It's plausible that there were no better choices in the ones they could buy in that kind of volume.

[idsn6]

Not the current Zune, but a supposed future Zune device that would share some of the features of WP7, such as Xbox Live gaming for example.

Wait. You're so offended that the iPod touch doesn't have all the features of the most desired phone on the planet that you've instead chosen some nebulous future device that does not exist outside your imagination?

[Exophase]

Wait. You're so offended that the iPod touch doesn't have all the features of the most desired phone on the planet that you've instead chosen some nebulous future device that does not exist outside your imagination?

Or has chosen to wait and see anyway. Is that such a big deal? Apple is clearly releasing a product that's far below what it could have been, and that says something. Other companies might be more willing to release a higher grade design, for a higher price premium of course.

[idsn6]

Or has chosen to wait and see anyway. Is that such a big deal? Apple is clearly releasing a product that's far below what it could have been, and that says something. Other companies might be more willing to release a higher grade design, for a higher price premium of course.

I'm confused more than anything. A Macbook doesn't have all the bells and whistles of a Macbook Pro. Does that "say something"? How does it make sense to feel cheated out of feature for feature parity with a phone more than twice the price? It's just strange to see these complaints that the touch doesn't measure up to some fevered dream machine when, really, what else in its market category even comes close?

[Exophase]

I'm confused more than anything. A Macbook doesn't have all the bells and whistles of a Macbook Pro. Does that "say something"? How does it make sense to feel cheated out of feature for feature parity with a phone more than twice the price? It's just strange to see these complaints that the touch doesn't measure up to some fevered dream machine when, really, what else in its market category even comes close?

You're comparing two devices in the same class. iPhone and iPod Touch are not supposed to be. Apple is breaking their own precedent here, in making an iPod Touch that isn't an iPhone minus the phone features. I think they're deliberately trying to segment the market to not steal away iPhone sales. People want a high end iPod without a phone contract and in a few critical ways this isn't something Apple wants to deliver on.

The price isn't the issue. I think if Apple were willing to sell a range of models with the current Touch's specs and ones more in line with the iPhone 4's people would have less problem. But it's quite subtle, because the "on paper" specs of a lot of Touch features aren't worse, they're just using markedly inferior components. Can't hardly do a good job selling the iPod Touch model A with the "worse Retina Display, camera sensors, and speakers" vs the model B with the "good kinds." The only thing they could really do is sell a 512MB RAM version. I mean, they sell ones with larger flash, so why not? The difference in cost between the 256MB and 512MB of RAM for them is probably quite small and they could easily justify a cost of > $25 more for it. But they don't want people to have 512MB iPod Touches.

Of course, people are fully going to expect iPhone 4 level components and are probably going to feel ripped off at being given anything less..

[aaronspink]

The lower RAM is nearly certainly pure greed, but the screen very likely isn't (although it might have helped). The quantities Apple needs for these products is phenomenal, and there were rumours that they had run into supply shortages for the iPhone 4 screen. I don't know if that's true, but there's no way they could have used the same screen for the iPod Touch 4G and not run into big supply problems. So they had to choose another screen from another supplier, and it had to be the same size and the same resolution. It's plausible that there were no better choices in the ones they could buy in that kind of volume.

The display issue is a pretty general issue right now. All the suppliers have been caught off guard with the demand for the high res/high quality small displays. Though in certain ways, it is hard to blame them since the market for larger displays has primarily been a race to the bottom going from MVA/IPS/etc to completely poor but cheap TN displays. They've seen the market completely fall out first at 19", then 20/21/22", now 24/26/28" and soon likely 30". Samsung just recently announced an extremely large investment to bring forward production capacity at their new factory. LG which produces the iPhone 4 display is doing the same. There have been reports that LG has so far been having a lot of trouble even keeping up with the volume requirements for apple. They reports say that LG has volume capacity for ~4M per month but yields due to defect requirements have so far prevented them from hitting that level. It will be interesting to find out who the source is for the new touch display but I've wager it might be sony who apparently has excess capacity right now because everyone wants either the samsung AMOLED or the the LG displays.

There is no doubt that the IPS display in the iphone 4 is incredibly high quality. Having compared it with the latest AMOLED HTC based phones, I really think that in general use there is no question its the best display on the market. I really hope samsung can sort through the various issues and get out 3 color pixel AMOLED within a reasonable timeframe.

[aaronspink]

Or has chosen to wait and see anyway. Is that such a big deal? Apple is clearly releasing a product that's far below what it could have been, and that says something. Other companies might be more willing to release a higher grade design, for a higher price premium of course.

The simple issue is that pretty much no one can release a new high end handheld display atm. Next year when LG/Samsung get their new capacity online, yes, but right now its almost impossible to get either a high pixel density AMOLED or IPS display in the 3-5" form factor. The entire supply has been allocated by HTC/Samsung for AMOLED and by apple for IPS. You quite literally cannot buy the high grade high res displays atm and it isn't an issue of money, every single one coming off the lines are already under contract. Both Samsung and LG should be able to significantly increase supply starting sometime next year with a ramp to 10x current capacity as the medium term goal at both companies. In the meantime, we'll have to deal with people either shipping lower grade high res displays or high grade lower res displays.

[Xmas]

Well the Anandtech piece compared it to a variety of displays, and the iPod touch came in a very VERY distant second to last place. It was even remotely close.
Black levels and contrast ratio alone don't tell the whole story, though. Viewing angles, sunlight readability, colour reproduction, pixel density and subpixel pattern can play an equally important role. Had Anand chosen to put pixel density in a chart it would have been a clear winner along with the iPhone 4.

The best way to judge a display panel is to look at it for yourself.

Better with regards to displays is very subjective. Many people seem to like AMOLED screens with the Pentile matrix (like the Droid Incredible, Nexus One, etc.)
I think they are hideous. They have only 2/3rds the resolution they claim to have (since each pixels has only 2 sub-pixels instead of 3), and they can't render text well at all, since sub-pixel antialiasing is completely defeated by their asymmetric sub-pixel layout.
They may look good for photos and videos, but for text, which is the main reason I use mobile devices, they are hideous. Everything looks like it's being viewed through a screen door.
I completely agree. Interestingly, after seeing the excellent close-up photo in this review (http://www.swedroid.se/samsung-galaxy-s/) I think I now understand where the screen door effect comes from. It's not the Pentile arrangement per se, but the fact that there is a little gap between the red and blue subpixels of alternating lines. I really wonder what the reason for this gap is.

The display issue is a pretty general issue right now. All the suppliers have been caught off guard with the demand for the high res/high quality small displays. Though in certain ways, it is hard to blame them since the market for larger displays has primarily been a race to the bottom going from MVA/IPS/etc to completely poor but cheap TN displays. They've seen the market completely fall out first at 19", then 20/21/22", now 24/26/28" and soon likely 30".
I really don't think it came as a surprise. For monitors the primary cirterion for most buyers is size. With a limited budget for their computing needs they can't justify paying twice as much or more for higher quality.

For mobile devices we also see the size trend, but pocketability sets a much tighter limit here. The premium for a quality screen is much smaller, both in absolute and in relative terms (since you're buying much more than a screen). Thus it's much more likely that a high quality screen is within a person's budget.

I'm kind of hoping that we're close to reaching a limit to a monitor's useful size (desk space, human FOV), so progress will concentrate more on quality than on further size increases at the same price.

[Dr Evil]

I was quite disapointed to see the new Touch unveiled. I have a 2nd gen touch that is falling apart and I was considering upgrading to this new one, but with them cheap buttons that are already causing me problems and the absence of 16GB model (wtf...) I'm not going to bother... Iphone 4 feels great, but I'm getting fed up with Apple. I was anxiously waiting for the new Mac Mini aswell only to find out that it costs 50% more than the old model with similar outdated specs... So I bought the old model.

[ltcommander.data]

I'd be interested to see a screen analysis on view angles, brightness, contrast, etc. between the 3rd gen iPod Touch and the 4th gen iPod Touch. If they are the same or improved then combined with the new higher resolution, it may be fair to be disappointed that it's not as good as the iPhone 4, but it's hard to be greatly offended. The same with the 256MB of RAM. It's disappointing, but you are still getting the faster A4 and at least you're not worse off than the iPad. It may seem like rationalizing, but it's not like Apple owed potential iPod Touch customers something.

Beyond supply constraints on the screen, the other reason for the hardware choice difference between the iPhone 4 and 4th gen Touch is probably not just the cost of the BOM of the 4th gen Touch itself, but the profit margin mix between the iPhone 4 and 4th gen Touch. Last years 3rd gen models were no doubt comparatively high margin products since there was a new SoC, but the housing and other components were largely the same as the 2nd generation. This year's changes to the iPhone 4 were drastic and the new housing/frame is no doubt especially high cost with BusinessWeek's recent interview with Foxconn's CEO revealing that the iPhone 4 frame is being made with low-volume prototyping machines because Apple's spec was too tight to transition to actual production machines. Sadly, lower margins on the iPhone 4 likely means that margins had to be more generous on the 4th gen iPod Touch to compensate.

These things also seem to ebb and flow. When the 2nd gen iPod Touch shipped with a faster clocked SoC, iPhone 3G users were annoyed too. The 2nd gen Touch also ended up getting Bluetooth 2.1 vs 2.0 and now Game Center support over the iPhone of that generation. Of course, the 1st gen Touch was slower than the original iPhone. Hopefully, the advantage shifts to the iPod Touch for 5th gen devices, which doesn't help people hoping to buy now of course.

[Mike11]

Everybody is comparing the new iPod touch to the iPhone 4. But I just got the new entry level iPod touch 8 GB ($229) for my father to replace the previous entry level 8GB model that was offered until last week for $199 (went to my mother), and it's a lot better in every aspect for just $30 more. For people who just want to have easy and cheap access to the App Store it's a very compelling offer (Yes, I know it's technically comparing a 2nd gen device with a 4th gen, but in practice it's comparing this year's entry level model with last year's. I also know the 32GB version gives you more bang for the buck, but for my parents 8GB is more than enough).

[Exophase]

Okay, I can give on the display, there are probably factors outside of Apple's control here like everyone's saying.

But the RAM.. why Apple, why?

[darkblu]

Okay, I can give on the display, there are probably factors outside of Apple's control here like everyone's saying.

But the RAM.. why Apple, why?
Perhaps apple view the new ipod touch is an mini ipad rather than a phone-less iphone? Besides, RAM is a major factor in determining the capabilites of a device, and launching the current gen of ipods with specs closer to the phone than to the ipad would leave the ipad in the position of the ipod touch 1st gen - the slowest (ipad has the lowest relative GPU specs, due to the having the biggest screen to feed), most 'cut-down' specimen of the lineup. One would guess apple don't want to cannibalize the ipad sales. That's all speculation, of course.

[Exophase]

I think that if Apple is afraid of iPod Touch cannibalizing iPad sales then it speaks a lot to what they think iPad's useful role in the market is.

Not that I'm saying you're wrong, of course.

[Squilliam]

No risk of them eating their own kind here. I got an iPad a while back and I recently ordered an iTouch. So the former caused me to buy the latter.

Damn Apple stuff is addictive!!!

[Exophase]

Yeah, iPod Touch strikes me as the thing you use to do stuff on the go (music, videos, gaming, wifi hotspot internet stuff, etc), iPad the thing you use while lounging on the couch at home.

If anything I expect iPod Touch and iPhone to be competing with each other, especially if monthly plans start dipping closer in price to dumb phone plans. No one is going to buy both.. and Apple probably wants to make iPhone 4 the more compelling choice.

[wco81]

I actually cancelled my iPod Touch 4G pre-order today given just how bad the screen seems (I knew it might be worse, but not by that much) and how disappointing the 256MB RAM is. The weak speaker doesn't help either. Of course, Apple benefits since that means I've finally decided to bite the bullet and buy an iPhone 4...

Knowing what you know about SOC trends, are you concerned that next year will bring an A5 SOC using a Cortex A9 core? Not to mention probably a better GPU?

If not Apple, you figure one of the Android vendors will offer A9 cores with better GPU than what is currently available in the market. The pace of HW improvements seem at least as rapid as the CPU clock speed leap-frogging in the '90s.

Battery life may or may not take a step back ...

[wco81]

Yeah, iPod Touch strikes me as the thing you use to do stuff on the go (music, videos, gaming, wifi hotspot internet stuff, etc), iPad the thing you use while lounging on the couch at home.

If anything I expect iPod Touch and iPhone to be competing with each other, especially if monthly plans start dipping closer in price to dumb phone plans. No one is going to buy both.. and Apple probably wants to make iPhone 4 the more compelling choice.

Despite Android's success on Verizon out here, apparently there are a fair number of Verizon customers who buy iPod Touch for access to the App. Store content.

If as rumors suggest, iPhone comes to Verizon in the near future, then the raison d'être of the iPod Touch diminishes that much more.

[Rys]

If not Apple, you figure one of the Android vendors will offer A9 cores with better GPU than what is currently available in the market. The pace of HW improvements seem at least as rapid as the CPU clock speed leap-frogging in the '90s.
There aren't any announced consumer designs yet, but there's at least one SoC with those features already announced and available in devkit form with Android, right now. So you can expect it to show up in phones at some point next year :smile:

[Arun]

Knowing what you know about SOC trends, are you concerned that next year will bring an A5 SOC using a Cortex A9 core? Not to mention probably a better GPU?I honestly don't know. The question is whether Samsung 32nm will be ready in time for the iPhone 5 in June, and I expect it would be but only barely so. Knowing how aggressive Apple can be sometimes, it's not impossible, even with an Intrinsity-optimised core. It's also possible that they'll try to do that but it ultimately delays them to September.

I wouldn't expect the new iPad to have that new chip anyway. I think the rumours of a smaller refresh in late Q4 or Q1 (so still with the A4) make sense.

If not Apple, you figure one of the Android vendors will offer A9 cores with better GPU than what is currently available in the market. The pace of HW improvements seem at least as rapid as the CPU clock speed leap-frogging in the '90s.I don't think we're going to see anything better than OMAP4/Tegra2/Orion before 4Q11. Then we might see QSD8672 and Tegra3, which are both pretty awesome on paper.

Battery life may or may not take a step back ...I think TDPs are likely to increase a bit but battery life for real-world usage will be stable at the worst. We'll see.

If as rumors suggest, iPhone comes to Verizon in the near future, then the raison d'être of the iPod Touch diminishes that much more.Talking of CDMA iPhones: unless Apple surprises everyone and picks Icera (which had secret CDMA R&D going on at one point but they might have cancelled it), in which case I risk instant death from an orgasmic heart attack (whatever that means), the chips to watch for are the MDM6600 and the QTR8600. And yes, that would be multimode so Infineon would lose the socket completely.

One pet theory I have right now is that Apple is going to come up with a five-band WCDMA iPhone Nano using Infineon chips that goes to T-Mobile (not necessarily very different from the 3GS spec-wise and would use the weaker-than-A4 45nm SoC found in the iPod Touch 3G), and then next year's iPhone 5 would use either the MDM6600 or a LTE/CDMA/WCDMA chip (the latter is doubtful). Fun stuff.

[Arun]

There aren't any announced consumer designs yet, but there's at least one SoC with those features already announced and available in devkit form with Android, right now. So you can expect it to show up in phones at some point next year :smile:That kinda fits both OMAP4 and Tegra2, except that Tegra2 has officially been announced to be inside a LG Android phone in Q4, so I'm confused now :) Unless of course you're implying you wouldn't even consider Tegra2 superior to the GPU in the A4, hehe.

[Rys]

Oh, I was talking about next year, rather than the tail-end of this one, or now. You can buy Tegra 2 today, so I wasn't really thinking about it (rather kudos to NV for bringing dual-core ARM to market first, by some margin).

[Exophase]

I don't think we're going to see anything better than OMAP4/Tegra2/Orion before 4Q11. Then we might see QSD8672 and Tegra3, which are both pretty awesome on paper.

I think we could see QSD8672 some time sooner than that. We also may see something with MSM8260 Q1 2011 if not sooner, which at the very least should be added to your initial list as comparable. Phones designed around 2010 Qualcomm releases are already making the rounds and possibly just held back by WP7 being out officially.

[Arun]

I think we could see QSD8672 some time sooner than that.I don't have any significant disagreement with your other points, but this is definitely wrong. See this: http://www.engadget.com/2010/09/08/qualcomm-1-5ghz-dual-core-snapdragon-devices-to-arrive-at-end-o/
Oh, I was talking about next year, rather than the tail-end of this one, or now. You can buy Tegra 2 today, so I wasn't really thinking about it (rather kudos to NV for bringing dual-core ARM to market first, by some margin).Ah, right, I was thinking phones specifically here. Tegra2 did manage to beat everyone else both for devices in general and (assuming the LG phone delivers) even phones, although arguably it's not a massive advantage. I think in terms of sampling ST-Ericsson's U8500 was probably first (although the way they prototype their stuff and their definition of sampling make things rather vague) but they were overly dependent on Symbian^4 at Nokia so they didn't manage 2010 devices (they've got design wins for Android in 1H11 and MeeGo later though)

[Helmore]

Phones designed around 2010 Qualcomm releases are already making the rounds and possibly just held back by WP7 being out officially.
I was under the impression that the first WP7 phones are going to use Qualcomm's QSD8X50 chipset and that pretty much all WP7 devices being launched this year will make use of the same chipset. It's just that I never found any conclusive sources about what will really be used in the first WP7 phones though, do you have any?

And about that QSD8672, I believe Qualcomm recently announced a delay to Q4 of 2011 for products on the market using that chip. Source (http://www.engadget.com/2010/09/08/qualcomm-1-5ghz-dual-core-snapdragon-devices-to-arrive-at-end-o/)

[Exophase]

I see, nevermind then >_>

There are some HTC WP7 phones slated for MSM7630 and MSM8655. My reasoning is that if an early WP7 phone can ship with a Qualcomm SoC released in mid-2010 then it shouldn't take much longer to ship with one released a few months later. It looks like QSD8672 is slated for release a clear year later than MSM8660, but maybe I'm completely misunderstanding the devices the latter is supposed to be in.

[metafor]

I see, nevermind then >_>

There are some HTC WP7 phones slated for MSM7630 and MSM8655. My reasoning is that if an early WP7 phone can ship with a Qualcomm SoC released in mid-2010 then it shouldn't take much longer to ship with one released a few months later. It looks like QSD8672 is slated for release a clear year later than MSM8660, but maybe I'm completely misunderstanding the devices the latter is supposed to be in.

I don't believe the 8672 is targetted for phones. The primary difference is the larger resolution display support and dual-channel LD-DDR3 memory (and a bump in frequency). This all points to a tablet chip.

For phones, the 8x60 series is going to be the flagship device for quite some time before the 28nm shrink. It's a relief in that if you get a phone with it early next year, it won't be obsolete in 3 months. But it also means they could easily be surpassed by A9-based devices coming out next year.

[Exophase]

Okay, good, I'm on the same page then ^^

I think 8x60 should be on a pretty level playing field against OMAP4 and Tegra 2, even if it's in-order; the clock speed advantage probably helps negate that a little. Hopefully dual core is going to be the real selling point for these platforms.

[wco81]

It's a selling point for geeks.

But there really haven't been apps. which require that kind of performance, unless generally faster UI helps sell given platforms.

Responsive UI though may be more about software optimization than brute force HW resources.

Video and games could be differentiators for dual-core CPUs but I think most people would be content with Youtube-quality videos (in which case network performance would be more pertinent than CPU/GPU). As far as games, the most popular smart phone games are very simple, like Angry Birds.

No matter how good that Epic Citadel demo looked, I don't see quasi-console quality shooters taking off on smart phone platforms either, especially if they're sold at higher prices than the most popular iOS games, which go for a couple of bucks.

[Exophase]

I hope there are at least some apps that do benefit. I hope dual core helps push that sort of thing. I'm sure it'll still be a high-end feature only, but it'll still be there.

Video should be moot, I mean, the fixed function hardware should have that covered still.

Really, any argument against dual core should work as an argument against ever increasing single-threaded performance capabilities of smart phones, but that hasn't stopped them from increasing rapidly and it's still a selling point, even for the non-hardcore. If anything, dual core helps better distribute the power consumption at lower clocks for properly threaded software.

[metafor]

Okay, good, I'm on the same page then ^^

I think 8x60 should be on a pretty level playing field against OMAP4 and Tegra 2, even if it's in-order; the clock speed advantage probably helps negate that a little. Hopefully dual core is going to be the real selling point for these platforms.

Scorpion is actually partially OoOE. It's just very restrictive of what instruction combos can be dispatched OoO. Drystone puts it at around 2.1-2.2 DMIPS/MHz which, while not as high as the A9, is a bit above the A8.

[Exophase]

I've heard that claim, but I've also heard Intel claim that Atom was partially OoO because it could integer and floating point in parallel, like most other designs (including Cortex-A8 and the original Pentium). Without really knowing anything else about it one can't say for sure how genuine the claim is. To me the biggest strength of OoOE are being able to execute ahead of loads and their immediate dependencies (especially in the even of L1 misses) and speculative execution/register renaming.

I take Dhrystone figures from different vendors with a grain of salt too. Especially when they're this close. There are a lot of small improvements that could be made to Cortex-A8 without making it out of order in any sense.

Of course, I welcome any further information if you have it.

[metafor]

I believe Dhrystone is entire integer bound. That being said, I agree that it shouldn't be the final word. I'm not sure if anyone's done native code testing of a Scorpion device, should be relatively easy to see based on instruction ordering whether any operation reordering is occuring.

[Exophase]

Of course, "integer bound" really just means "no floating point", which is fine and all.. it may also mean that it runs entirely within L1 cache, which should be the case for Dhrystone. But it definitely still makes memory accesses, pretty extensively, so some simple parameters like load-use penalty, dual issue on loads/stores, address generation interlock.. things like that, can make a difference in performance. Also branch prediction.

A problem with Dhrystone is that it's a C program, so the performance is heavily dependent on the compiler and runtime used. Even if both CPUs use the same instruction set (as is the case with Cortex-A8 and Scorpion) they could still be using different compiler versions with different compiler options applied. Even if they're using the same version and options, the compiler can still be generating better code for one platform than the other. For instance, if GCC's scheduler for Scorpion is somehow superior to the Cortex-A8 scheduler. For (mostly?) in-order processors, compiler scheduling is very important.

[metafor]

Well yes, by "integer bound" I mean it doesn't use the FPU. The integer and load/store usually share a dispatch queue whereas FP (including FP load/store) is its own separate entity.

My point is that Dhrystone is supposed to (don't know how well it does) flood the processor with little memory interaction. Now, it may fit in L1 or in L2 but how well the CPU schedules from those and mask read latencies is part of the processor's charm if you will.

I'm going to go ahead and trust that ARM -- with their compiler -- publishes what they tweaked to be the best out of Dhrystone for their A8 publications. I don't think Qualcomm has their own compiler team so I assume they either used GCC (which, IIRC, does not generate very good ARM code) or more likely just compiled the program to target an A8; I don't recall seeing an option in the armasm for Scorpion.

Now, you're right in that many various things can account for a slightly higher DMIPS/MHz. However, it does mean the claims of a partial OoOE isn't out of the question.

[aaronspink]

Scorpion is actually partially OoOE. It's just very restrictive of what instruction combos can be dispatched OoO. Drystone puts it at around 2.1-2.2 DMIPS/MHz which, while not as high as the A9, is a bit above the A8.

The fact that anyone anywhere are comparing anything using drystone in this day and age is just shockingly sad. Drystone is effectively completely useless for comparing performance for any application written in the past 15 years. Its really quite bad and bad form to even use it anymore, its like comparing performance using bogomips or setting your locker combination to 1234, only an idiot would do it. <insert spaceballs joke here>

[Laurent06]

The fact that anyone anywhere are comparing anything using drystone in this day and age is just shockingly sad. Drystone is effectively completely useless for comparing performance for any application written in the past 15 years. Its really quite bad and bad form to even use it anymore, its like comparing performance using bogomips or setting your locker combination to 1234, only an idiot would do it. <insert spaceballs joke here>
I couldn't agree more. One huge dependency dhrystone has is the performance of the string functions in the C library. Also the rule that forbids function inlining is utterly stupid.

Even Coremark, though less dumb, doesn't seem that meaningful.

[Exophase]

I'm trying to google more on this.

The only really root source I could find was Anand's comment:

"Qualcomm claims the ability to do some things out of order, but by and large the pipeline is in order which ultimately keeps it out of the A9 classification."

This can, of course, mean just about anything.. but I think we should take that "by and large" comment as detracting from its pertinence.

[metafor]

I'm trying to google more on this.

The only really root source I could find was Anand's comment:

"Qualcomm claims the ability to do some things out of order, but by and large the pipeline is in order which ultimately keeps it out of the A9 classification."

This can, of course, mean just about anything.. but I think we should take that "by and large" comment as detracting from its pertinence.

OoOE is a varying scale. Even the A9 isn't OoOE in the sense used by desktop processors. There is no rename register (albeit there is a WB buffer), and stalls halt entire sub-pipes.

Scorpion is obviously a lot closer to strict in-order than out-of-order (at least, based on its relative performance to the A8). But we don't know how much of its performance is due to design similarity and how much is due to trade-offs; it has a longer pipeline, which should theoretically, even with a better branch predictor, lead to penalties in branchy code.

[Exophase]

OoOE is a varying scale. Even the A9 isn't OoOE in the sense used by desktop processors. There is no rename register (albeit there is a WB buffer), and stalls halt entire sub-pipes.

I'm not sure if you're referring to something specific to the load-store pipe by "a rename register", or if you merely mean register renaming which Cortex-A9 TRM indicates:

"The register renaming scheme facilitates out-of-order execution in Write-after-Write (WAW) and Write-after-Read (WAR) situations for the general purpose registers and the flag bits of the Current Program Status Register (CPSR).

The scheme maps the 32 ARM architectural registers to a pool of 56 physical 32-bit registers, and renames the flags (N, Z, C, V, Q, and GE) of the CPSR using a dedicated pool of eight physical 9-bit registers."

Writeback buffering has been a feature of ARM processors since ARM9, for what it's worth (many other pretty old small/embedded designs have it too)

I take it the long poll ramification if stalls blocking the sub-pipe would suggest in ability to reoder loads and provide for instance hit under miss. I figured Cortex-A9 did have this capability since ARM11 did, but I can't find any mention of it. The product brief does say this:

"Dependent load-store instructions can be forwarded for resolution within the memory system -
further reduces pipeline stalls and significantly accelerating the execution of high level code accessing complex data structures or invoking C++ functions."

But I have no idea if that has any actual relevance.

Scorpion is obviously a lot closer to strict in-order than out-of-order (at least, based on its relative performance to the A8). But we don't know how much of its performance is due to design similarity and how much is due to trade-offs; it has a longer pipeline, which should theoretically, even with a better branch predictor, lead to penalties in branchy code.

We don't even have good benchmarks, so how are we comparing them? Dhrystone probably predicts well, even if you ignore all the other flaws in using its results. We also don't really know what anything about its pipeline outside of being "longer" (and I've seen mention of differently sized pipelines, ala ARM11, and this is quite likely where the OoO aspects come from), that doesn't even really tell you the mispredict penalty.

[metafor]

I'm not sure if you're referring to something specific to the load-store pipe by "a rename register", or if you merely mean register renaming which Cortex-A9 TRM indicates:

"The register renaming scheme facilitates out-of-order execution in Write-after-Write (WAW) and Write-after-Read (WAR) situations for the general purpose registers and the flag bits of the Current Program Status Register (CPSR).

The scheme maps the 32 ARM architectural registers to a pool of 56 physical 32-bit registers, and renames the flags (N, Z, C, V, Q, and GE) of the CPSR using a dedicated pool of eight physical 9-bit registers."

Interesting. For some reason I was under the impression it handled WAW and WAR's with a large write-back buffer instead of renaming registers. My mistake.

Writeback buffering has been a feature of ARM processors since ARM9, for what it's worth (many other pretty old small/embedded designs have it too)

Having a write-back stage, yes. Using it as your OoOE data window, with early-out support of your pipeline, no.

I take it the long poll ramification if stalls blocking the sub-pipe would suggest in ability to reoder loads and provide for instance hit under miss.

Re-ordering can occur but depending on the conditions for dispatch into the execute pipe, a load can still stall the entire execute sub-pipe, for instance, if it does not return data in the time expected. Waiting for data availability before dispatching can avoid this but you lose performance in the common case.