So because they build CPU they cant build GPU's. OK I see,
Well according to Intel they a concentrating their efforts on the video card market.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
And the Cypress shader count is...
- Thread starter Tom|Nbk
- Start date
So because they build CPU they cant build GPU's. OK I see,
Well according to Intel they a concentrating their efforts on the video card market.
Associate
- Joined
- 30 Sep 2005
- Posts
- 144
Well, they designed the dispatcher to be scalable, as the number of SIMDs are increased, so I don't see why it would be a bottleneck. Besides which, if they could make the required alterations for RV770, they certainly can do so for RV870.
"R3D: How did you change the Ultra Threaded Dispatch Processor for the RV770? I presume it's been widened due to the 6 additional SIMDs, but did you retain the dual Arbiter/Sequencer arrangement that the R6xx had? Do vertex and texture fetches still have one arbiter-sequencer pair dedicated to each, thus making it possible to schedule them independently from math done in the SIMDs? Were the command queues increased?
Eric: The dispatcher was changed through nearly 1.5 years of work from a design team standpoint. It was made to be much more scalable in design, allowing for the additional SIMDS, while also offering new features and better performance. It's an evolution of the previous version and inherited the best parts, like as the ability to issue to multiple blocks in parallel such as texture and ALU and others. As for the command queues, some tweaking and optimizing of sizes was done to achieve better balanced in the new design."
http://www.rage3d.com/interviews/atichats/undertheihs/
Whats more interesting is the info on the sideport from that link - somthing thats been somewhat shrouded in mis-information and mystery...
Associate
- Joined
- 30 Sep 2005
- Posts
- 144
It was never all that mythical; it just provided a little bit more bandwidth for inter-chip communication, bandwidth that turned out in retrospect to be unneeded.
"This is simply a case of our software capabilities catching up to our hardware capabilities. When the initial design of the RV770 was taking place and concepts such as Sideport were being kicked around our ATI CrossFireX™ software wasn't in the place it is right now, so there was a much higher reliance on inter-chip communication.
While having lots of bandwidth is rarely a bad thing, the ATI CrossFireX communication bandwidth between two discrete cards is less than local bandwidth - even though Sideport doubles the inter-GPU communication bandwidth on an X2 type solution it's still not significant enough to really change the disparity in local frame buffer and inter-GPU bandwidths.
The software work that occurred in the space of time between the RV770 design and product saw significant improvements in inter-GPU communication. Internal to the driver we now have a number of "alternate frame rendering" (AFR) profiles, with many parameters that can be tweaked in order to control how the rendering behaves over multiple GPU's and reduce the inter-GPU communication as much as possible. By the time we put two RV770's on a board and started testing Sideport, the current ATI CrossFireX software capabilities delivered more than enough bandwidth, obviating the need for Sideport."
http://www.guru3d.com/article/interview-with-ati-dave-baumann/2
"This is simply a case of our software capabilities catching up to our hardware capabilities. When the initial design of the RV770 was taking place and concepts such as Sideport were being kicked around our ATI CrossFireX™ software wasn't in the place it is right now, so there was a much higher reliance on inter-chip communication.
While having lots of bandwidth is rarely a bad thing, the ATI CrossFireX communication bandwidth between two discrete cards is less than local bandwidth - even though Sideport doubles the inter-GPU communication bandwidth on an X2 type solution it's still not significant enough to really change the disparity in local frame buffer and inter-GPU bandwidths.
The software work that occurred in the space of time between the RV770 design and product saw significant improvements in inter-GPU communication. Internal to the driver we now have a number of "alternate frame rendering" (AFR) profiles, with many parameters that can be tweaked in order to control how the rendering behaves over multiple GPU's and reduce the inter-GPU communication as much as possible. By the time we put two RV770's on a board and started testing Sideport, the current ATI CrossFireX software capabilities delivered more than enough bandwidth, obviating the need for Sideport."
http://www.guru3d.com/article/interview-with-ati-dave-baumann/2
Associate
- Joined
- 30 Sep 2005
- Posts
- 144
Sure, before ATI clarified why it wasn't used, shortly after the 4870 X2 launched.
They got through the fx5xxx series so I don't see this as a major problem for them in the long run.
If it's true (or even close) I think the 5850 will be my next card. Roll on 10/09.
5900XT was only good because with decent cooling it could overclock enough to beat the 9800 performance... but it still looked like ass doing any newer shaders. I had one of the gainward GS engineering samples that could clock more than 50% on core and memory and that was the only way I stuck with nVidia during that period.
Gainward used to make good cards 
I had a 512Mb DDR3 6800 off them clocked to ultra extreme performance that would show a 7800GT a clean pair of heels... all these rebadged reference cards is depressing... the few innovative cards that do come out are too late and too expensive.
Still have the screenshot yay (ignore the 7950GX2 stuff):
http://aten-hosted.com/images/GPUZ.jpg
Reached 105C while gaming... but it was fast! for an AGP card.
I had a 512Mb DDR3 6800 off them clocked to ultra extreme performance that would show a 7800GT a clean pair of heels... all these rebadged reference cards is depressing... the few innovative cards that do come out are too late and too expensive.Still have the screenshot yay (ignore the 7950GX2 stuff):
http://aten-hosted.com/images/GPUZ.jpg
Reached 105C while gaming... but it was fast! for an AGP card.
Last edited:
Yeah I was a big Gainward fan myself. If fact I still have a 4200 and 4800 Ti GS ES.
The 5900 series put that Gainward out of business though.
The NV30 - 35 was a great big pile of fail. Do you remember the amount of dirty tricks Nvidia used to sell the cards.
The 5900 series put that Gainward out of business though.
The NV30 - 35 was a great big pile of fail. Do you remember the amount of dirty tricks Nvidia used to sell the cards.
Soldato
- Joined
- 28 May 2007
- Posts
- 10,222
What do you mean about the way they handle driver releases? A lot of people seem to like the fact that they know they will get at least 1 driver release every month of the year. I hope these cards perform as well as the specs suggest it should as i think a 5870 will be my next card.
I know I like it. And their drivers have been problem free for me, whereas I've read lots of issues relating to the way PhysX has been included with nVidia drivers. I don't think either company has got it right yet but I wouldn't cite monthly driver releases as an issue.
Caporegime
- Joined
- 8 Jul 2003
- Posts
- 30,063
- Location
- In a house
Yeah shocking support from ATi that, im going Nvidia next time
Don't think the problems are due to the way physx is in the drivers - there was only one beta release where there was a problem with the physx install launching and thats it...
Most of the problems with 185.xx and 190.xx are due to nVidia using a sloppy driver team...
Most of the problems with 185.xx and 190.xx are due to nVidia using a sloppy driver team...