[silent_guy]

Originally Posted by Alexko

Yeah, that and nobody wants the power consumption of 7Gbps chips either. That would be my guess, anyway.

I don't think the difference in power consumption will be much more than a linear scale up. If so, it'd be a simple choice between power and performance. One that many would be willing to make.

A bigger argument against the 7Mbps is that the performance increase doesn't scale as expected. There is a threshold (5.5Mbps?) above which addition restrictions are imposed wrt bank access patterns, which can result in performance hit. It was mentioned in an Anandtech article, but I can't find it right now. The result would be that, e.g. 6Mbps will be a net negative and you have to go higher to recoup that loss.

[rpg.314]

So this is the end of the road for GDDR5 then.

[silent_guy]

Originally Posted by rpg.314

So this is the end of the road for GDDR5 then.

I thought PCIe 3 was end of the line.
They find ways to improve, don't worry.

[rpg.314]

I am sure they will find a way to improve, but frequency seems to have maxed out for the most part.

Charlie posted a pic with interposers tagged as HD8000, and claims credibly that Intel is looking at interposers hard for Haswell. 2013 can't come here soon enough.

[Ninjaprime]

I seem to recall some hynix pdf slides that gave two possible branches for the future, one refered to GDDR5+ scaling up 32gbps, which would be 8ghz. So I think the possiblity is there, but with no demand for even their 7ghz products yet, they have no real reasons to launch anything. Or perhaps they are taking the other branch which refered to some new memory system entirely.

[mczak]

Originally Posted by Ninjaprime

So I think the possiblity is there, but with no demand for even their 7ghz products yet, they have no real reasons to launch anything.

I just see no evidence the memory manufacturers can actually produce 7gbps chips, despite claiming to be able to 3 years ago.
You'd certainly think they'd have 6gbps chips which were using standard 1.5V voltage instead of the factory overvolted 1.6V parts if it's so easy. I bet there IS demand for such such chips (as well as higher speed grade 1.35V parts for mobile products).
Noone is interested in parts needing even higher voltages (and hence with power consumption going through the roof) but I think claming no demand for 7gbps parts is silly (well if it actually can be made to work at standard voltage that is).

[Dooby]

I know this info is to be taken with a truck load of salt, but the wiki page for Southern Islands has changed.

7970: 2048:128:48 where 48 = ROPs. This definitely said 64 last week. Is someone in the know already?

I don't know why, but my expectation for a 7970 just cratered...

[AlphaWolf]

Last edited by someone in sweden, /shrug anyone can change it

[ToTTenTranz]

So any news about this rumoured event that should happen today in London?
It won't happen? Will it happen later?

I'm eagerly awaiting news about this launch, not only because I'm curious but also because I have a graphics card purchase pending.

BTW, I saw this:




Also, two new laptops have in the price lists appeared with SI GPUs:
- ASUS X53TK-SX058V w/ A6 3420M + Radeon HD 7670M for ~620€ (wow @ price point)
- HP Envy 17-3030ew w/ i7-2670QM + HD7690M for ~1200€

However, I haven't seen any specific rumour about what will be the HD76xxM GPUs, so there's a big chance they're just renamed HD67xxM Whistlers (VLIW5, 480sp, 24TMUs, 8 ROPs).

[no-X]

It's the old XDR2 fake...

[Kaotik]

http://lenzfire.com/2011/11/amd-rade...n-march-65733/

http://lenzfire.com/2011/12/amd-plan...ies-gpu-40341/

[homerdog]

Where is all this fake XDR2 crap coming from? I look at a chart that looks completely believable and then I see that and it's like

[no-X]

Originally Posted by Kaotik

Interesting. But it's not in line with the driver info, which indicates the same architecture for Tahiti and Pitcairn:

AMD68FA.1 = ati2mtag_Evergreen = "AMD Radeon HD 7300 Series"
…
AMD675B.1 = ati2mtag_NI = "AMD Radeon HD 7600 Series"
AMD6751.1 = ati2mtag_NI = "AMD Radeon HD 7600A Series"

…
AMD6859.1 = ati2mtag_R475 = "AMD Radeon HD 7400 Series"

AMD6840.7 = ati2mtag_R476 = "AMD Radeon HD 7600M Series"
AMD6841.7 = ati2mtag_R476 = "AMD Radeon HD 7500M/7600M Series"

AMD6799.1 = ati2mtag_R575 = "SI NEW ZEALAND"
AMD679A.1 = ati2mtag_R575 = "SI TAHITI PRO"
AMD6798.1 = ati2mtag_R575 = "SI TAHITI XT"
AMD683B.1 = ati2mtag_R575A = CAPE VERDE PRO
AMD6819.1 = ati2mtag_R575 = "PITCAIRN PRO"

AMD682F.1 = ati2mtag_R576 = "CHELSEA PRO"
AMD6824.1 = ati2mtag_R576A = "CHELSEA XT"
…

Which one is wrong?

[Kaotik]

http://lenzfire.com/2011/12/amd-rade...e-specs-89652/

Starting to think lenzfire is just spreading crap

[no-X]

Especially this paragraph is quite funny...

Another advantage of the 7000 series is that they will use the PCIe 3.0 bus on motherboards, while the use of new memories XDR2 seems to be left for later. A Rambus memory design would offer a higher bandwidth (up to 8,000 MHz) but the cost may pile up a bit more and that’s why AMD could have been suggested to go with GDDR5 itself.

Are they implying, that AMD replaced XDR2 memory controller by GDDR5 MC just a few weeks before launch?

[neliz]

Originally Posted by homerdog

Where is all this fake XDR2 crap coming from? I look at a chart that looks completely believable and then I see that and it's like

One Chinese post back in June/July and everyone keeps regurgitating it. It has had more public exposure than Kim Kardashian's ***

btw, Lenzfire is closer to the truth, but still spewing a lot of crap.

[Kaotik]

Originally Posted by no-X

Especially this paragraph is quite funny...

Are they implying, that AMD replaced XDR2 memory controller by GDDR5 MC just a few weeks before launch?

Well the XDR2 was never anything but rumors

[mczak]

Originally Posted by Dooby

I know this info is to be taken with a truck load of salt, but the wiki page for Southern Islands has changed.

7970: 2048:128:48 where 48 = ROPs. This definitely said 64 last week. Is someone in the know already?

I don't know why, but my expectation for a 7970 just cratered...

If you're assuming 384bit gddr5 it most likely can't really be anything else than 48 ROPs and I would guess that's why someone changed the wiki.

Originally Posted by homerdog

Where is all this fake XDR2 crap coming from? I look at a chart that looks completely believable and then I see that and it's like

Does not look believable to me anyway. In particular for GCN disabling of "clusters" should come in groups of 4 (and btw the power draw would be fantastic news but I see no reason why both nvidia and amd wouldn't go to the same limits as before).

Originally Posted by Kaotik

Hmm funny that so wichita/krishna cpu is scrapped and the IGP of it still scheduled to launch? That doesn't add up .

[Kaotik]

Originally Posted by mczak

Hmm funny that so wichita/krishna cpu is scrapped and the IGP of it still scheduled to launch? That doesn't add up .

Wichita/krishna scrapped since when?
The news about "starting fresh" and use TSMC are bollocks, we already knew sometime in early(?) summer that they're using TSMC at least for some of Wichita/Krishna combo

[no-X]

Kaotik:
http://www.xbitlabs.com/news/mobile/...t_Delayed.html
http://vr-zone.com/articles/report-a...led/13975.html

[Kaotik]

Originally Posted by no-X

Kaotik:
http://www.xbitlabs.com/news/mobile/...t_Delayed.html
http://vr-zone.com/articles/report-a...led/13975.html

Yes, I'm aware of those, "according to sources" stuff with no official info related to such, not even to this date.
Xbit for example mentions that "Brazos 2.0 will boost performance slightly" but won't "revolutionary" like Deccan was supposed to be, yet we already knew during summer that Krishna is supposed to use just enhanced Bobcat cores.
Also none of the sites broadcasting news about that seem to know that TSMC was already meant to be responsible for at least part of Krishna/Wichita chips.

To me it sounds like someone misunderstood something along some point, which then got spread as news. "Deccan disappearing" could just aswell mean that "Deccan" will be known as Brazos 2.0 officially, if the news have any reality behind them.
What could be true, of course, is GF problems, but that wouldn't affect all of Krishna/Wichita combo.

[fellix]

[Alexko]

Originally Posted by fellix

So in a nutshell, model [BLANK] will have [BLANK], up to [BLANK], and even [BLANK] for a power consumption of just [BLANK]. Impressive.

[neliz]

It looks like our "onepage saleskit" but this screecap has errors in it.. sorry to disappoint you guys

[UniversalTruth]

Originally Posted by Alexko

So in a nutshell, model [BLANK] will have [BLANK], up to [BLANK], and even [BLANK] for a power consumption of just [BLANK]. Impressive.

I understand you.

But this one is much better and at least there is a source.



Source

Originally Posted by neliz

It looks like our "onepage saleskit" but this screecap has errors in it.. sorry to disappoint you guys

Any hints about the DX support?
The card dimensions too...