Beyond3D Forum

rpg.314 · 25-Jul-2010, 18:50

Quote:

Originally Posted by NathansFortune

It wouldn't make sense for Nvidia to stick A2/3 and get a neutered chip when a silicon respin would fix a lot of their problems and increase yields...

A silicon respin could certainly improve upon many performance and efficiency metrics. However, a bigger question, at least for me, is whether there is a reasonable probability of this happening, assuming Fermi 2/Fermi's shrink is due this winter.

Alexko · 25-Jul-2010, 19:05

There can't be any shrink this winter, since TSMC's 40nm process is the smallest available.

NathansFortune · 25-Jul-2010, 19:26

Quote:

Originally Posted by rpg.314

A silicon respin could certainly improve upon many performance and efficiency metrics. However, a bigger question, at least for me, is whether there is a reasonable probability of this happening, assuming Fermi 2/Fermi's shrink is due this winter.

TSMC and Global Foundries have stated 28nm isn't going to be ready until H2 2011. That means Nvidia need to get more mileage out of their current designs.

DegustatoR · 25-Jul-2010, 19:44

Quote:

Originally Posted by Silent_Buddha

Unless they were ditching GF100 entirely as ATI did with R520/R600. And instead there's a new modified chip coming out for the fall.

GF100B (or whatever they'll call it in the end) is what's coming in the Fall. There may be a more or less new (still Fermi-based at it's core) 40G GF100B replacement down the road but its fate will depend on a lot of factors, and I won't be surprised if they'll wait for 28HP for their next top-end GPU.

rpg.314 · 25-Jul-2010, 19:48

Quote:

Originally Posted by NathansFortune

TSMC and Global Foundries have stated 28nm isn't going to be ready until H2 2011. That means Nvidia need to get more mileage out of their current designs.

Must have missed it. Do you have a recent link?

Also, it means at best, we can expect a hybrid part this year from AMD.

NathansFortune · 25-Jul-2010, 20:00

Quote:

Originally Posted by rpg.314

Must have missed it. Do you have a recent link?

Also, it means at best, we can expect a hybrid part this year from AMD.

I can't find the link, but it was from the TSMC Fab 15 article somewhere. The CEO said 40nm was their concern right now and 28nm is delayed until later in 2011, probably H2. Global Foundries have a similar outlook as 32nm for AMD and ARM is their primary concern.

Blazkowicz · 25-Jul-2010, 20:48

How interesting, as People's Republic of China domestic CPU industry (loongson processors) stated they aim for 32nm at end of 2011

.

Alexko · 25-Jul-2010, 21:11

Quote:

Originally Posted by NathansFortune

I can't find the link, but it was from the TSMC Fab 15 article somewhere. The CEO said 40nm was their concern right now and 28nm is delayed until later in 2011, probably H2. Global Foundries have a similar outlook as 32nm for AMD and ARM is their primary concern.

As far as I'm aware, this is GlobalFoundries' latest public roadmap:

And I haven't heard of any changes to the 28nm schedule since then.

aaronspink · 26-Jul-2010, 01:16

Quote:

Originally Posted by Blazkowicz

How interesting, as People's Republic of China domestic CPU industry (loongson processors) stated they aim for 32nm at end of 2011

.

They've stated a lot of things over time and they contract out the fabrication to non Chinese companies.

Man from Atlantis · 31-Jul-2010, 17:21

GF104, GF100 Core Architecture Comparison
http://news.mydrivers.com/Img/20100730/02501268.jpg
GF104 SM architecture (part of the speculation)
http://news.mydrivers.com/Img/20100730/02503995.jpg
GF100 SM architecture
http://news.mydrivers.com/Img/20100730/02504021.jpg
NVIDIA graphics core in recent years, the evolution diagram
http://news.mydrivers.com/Img/20100730/02521912.jpg
G80, GT200, GF100, GF104 contrast the core memory and multithreading
http://news.mydrivers.com/Img/20100730/02521937.jpg

trinibwoy · 02-Aug-2010, 04:08

Did Nvidia beef up GF104's texture units? Was just browsing Damien's english review and it seems FP16 and RGB9E5 are now full speed as opposed to half speed on GF100.

Alexko · 02-Aug-2010, 09:39

Damien's reviews usually deserve a bit more than a quick browsing…

Quote:

Moreover, the texturing units have been improved to filter FP16 textures (as well as FP11, FP10 and RGB9E5) at full speed.

http://www.behardware.com/articles/7...e-gtx-460.html

trinibwoy · 03-Aug-2010, 05:05

Of course, thanks. Saw it on my second read through

Wonder why they bothered.

Chalnoth · 03-Aug-2010, 07:23

Quote:

Originally Posted by trinibwoy

Of course, thanks. Saw it on my second read through

Wonder why they bothered.

My first guess would be that it was something that was intended for the GF100 all along, but there was a bug in the hardware implementation that forced them to implement these modes with reduced performance.

As for why they would have wanted to go this route in the first place, well, that would make sense if they feel that these modes will become more and more common as time goes forward, and if the added hardware cost was minimal.

mczak · 03-Aug-2010, 13:54

Maybe the full-speed fp16 was just a later addition which didn't make it for GF100.
That said, it would imho make more sense for GF100 than GF104, since GF100 has lower tex:alu ratio (and also higher memory bandwidth / tex). Unless you think it doesn't matter for GF100 since it looks more useful for non-gaming usages anyway..

ShaidarHaran · 03-Aug-2010, 15:00

Interesting that the fp formats have seen performance increases from GF100->GF104, but the int formats have seen performance decreases. Also, there appears to be a hard cap @ 33.3 GTexels/s for 3 of the formats. Any thoughts as to what might be causing this? Is it a lack of cache or cache bandwidth? Some other architectural limitation? I don't think it's a lack of VRAM or VRAM bandwidth since GF104 out-performs GT200b in 2 of the 3 formats.

TKK · 03-Aug-2010, 16:38

Quote:

Originally Posted by ShaidarHaran

I don't think it's a lack of VRAM or VRAM bandwidth since GF104 out-performs GT200b in 2 of the 3 formats.

Also, if it was the case there should be a difference between the two GTX 460 variants, which isn't the case.

Gipsel · 03-Aug-2010, 17:57

Quote:

Originally Posted by ShaidarHaran

Also, there appears to be a hard cap @ 33.3 GTexels/s for 3 of the formats. Any thoughts as to what might be causing this? Is it a lack of cache or cache bandwidth? Some other architectural limitation? I don't think it's a lack of VRAM or VRAM bandwidth since GF104 out-performs GT200b in 2 of the 3 formats.

It's the theoretical max throughput of the 56 TMUs * 0.675 GHz = 37.8 GTexel/s. Obviously the efficiency (88%) is slightly lower than on AMD GPUs (~98% or so) for this simple tasks.

mczak · 03-Aug-2010, 19:34

Quote:

Originally Posted by Gipsel

It's the theoretical max throughput of the 56 TMUs * 0.675 GHz = 37.8 GTexel/s. Obviously the efficiency (88%) is slightly lower than on AMD GPUs (~98% or so) for this simple tasks.

I think the more interesting comparison is GTX470/480 - 60 TMUs *0.7 GHz = 42 GTexels/s and it is achieving 41.4 GTexels/s (for int8 only though) - 99%. So for some odd reason GF104 can achieve less of the peak potential of the tmus.

CarstenS · 03-Aug-2010, 20:20

I'm showing (almost) the same here. 33.8 GTex is the maximum i can get out of a stock GF104 with bilinear filtering. With trilinear it's a more expected 18.9 GTex/s. Together with the point sampling result of - again - 33.8 GTex/s I'm guessing, it's maybe interpolation or adress bound.

An HD5830 is literally miles away at 43.6 and 22.4 GTex/s.

02-Aug-2010, 04:08	#4011
trinibwoy Meh Join Date: Mar 2004 Location: New York Posts: 9,809	Did Nvidia beef up GF104's texture units? Was just browsing Damien's english review and it seems FP16 and RGB9E5 are now full speed as opposed to half speed on GF100. __________________ What the deuce!?

03-Aug-2010, 05:05	#4013
trinibwoy Meh Join Date: Mar 2004 Location: New York Posts: 9,809	Of course, thanks. Saw it on my second read through Wonder why they bothered. __________________ What the deuce!?

03-Aug-2010, 20:20	#4020
CarstenS Senior Member Join Date: May 2002 Location: Germany Posts: 2,842	I'm showing (almost) the same here. 33.8 GTex is the maximum i can get out of a stock GF104 with bilinear filtering. With trilinear it's a more expected 18.9 GTex/s. Together with the point sampling result of - again - 33.8 GTex/s I'm guessing, it's maybe interpolation or adress bound. An HD5830 is literally miles away at 43.6 and 22.4 GTex/s. __________________ English is not my native tongue. Before flaming please consider the possiblity that I did not mean to say what you might have read from my posts. Work\| Recreation Warning! This posting may contain unhealthy doses of gross humor, sarcastic remarks and exaggeration!

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

25-Jul-2010, 19:05	#4002
Alexko Senior Member Join Date: Aug 2009 Posts: 2,023	There can't be any shrink this winter, since TSMC's 40nm process is the smallest available.

25-Jul-2010, 20:48	#4007
Blazkowicz Senior Member Join Date: Dec 2004 Location: Toulouse Posts: 4,142	How interesting, as People's Republic of China domestic CPU industry (loongson processors) stated they aim for 32nm at end of 2011 .

31-Jul-2010, 17:21	#4010
Man from Atlantis Member Join Date: Jul 2010 Location: Istanbul Posts: 727	GF104, GF100 Core Architecture Comparison http://news.mydrivers.com/Img/20100730/02501268.jpg GF104 SM architecture (part of the speculation) http://news.mydrivers.com/Img/20100730/02503995.jpg GF100 SM architecture http://news.mydrivers.com/Img/20100730/02504021.jpg NVIDIA graphics core in recent years, the evolution diagram http://news.mydrivers.com/Img/20100730/02521912.jpg G80, GT200, GF100, GF104 contrast the core memory and multithreading http://news.mydrivers.com/Img/20100730/02521937.jpg

03-Aug-2010, 13:54	#4015
mczak Senior Member Join Date: Oct 2002 Posts: 2,437	Maybe the full-speed fp16 was just a later addition which didn't make it for GF100. That said, it would imho make more sense for GF100 than GF104, since GF100 has lower tex:alu ratio (and also higher memory bandwidth / tex). Unless you think it doesn't matter for GF100 since it looks more useful for non-gaming usages anyway..

03-Aug-2010, 15:00	#4016
ShaidarHaran hardware monkey Join Date: Mar 2007 Posts: 3,905	Interesting that the fp formats have seen performance increases from GF100->GF104, but the int formats have seen performance decreases. Also, there appears to be a hard cap @ 33.3 GTexels/s for 3 of the formats. Any thoughts as to what might be causing this? Is it a lack of cache or cache bandwidth? Some other architectural limitation? I don't think it's a lack of VRAM or VRAM bandwidth since GF104 out-performs GT200b in 2 of the 3 formats.

Beyond3D Forum

Welcome, Unregistered.