By: Ben (benbowen.delete@this.zoho.com), August 27, 2014 10:56 pm
Room: Moderated Discussions
Gabriele Svelto (gabriele.svelto.delete@this.gmail.com) on May 8, 2014 9:24 am wrote:
> I found the article quite interesting and it got me thinking about the potential impact
> this might have on GPUs and made me remind about one datapoint we have on this topic.
>
> Back in the RV7xx days when AMD introduced the RV790 XT (aka Radeon HD 4890) they added a large ring of decoupling
> capacitors to the existing RV770 XT design to boost its maximum clock speed. The RV770 was originally released
> at 750MHz and the RV790 started at 850MHz but was widely available in overclocked configurations up to 1GHz.
> Interestingly the die-size increased significantly from 256 mm² to 282 mm², a >10% increase.
>
> While those chips are now relatively old and the techniques used in GPU design might
> have improved, they still give us an idea of how much benefit could be reaped by getting
> rid of decoupling capacitors. Current high-end GPUs sport both higher power consumption
> and larger dies than the RV7xx family so the benefits might be very tangible.
Interesting.
> I found the article quite interesting and it got me thinking about the potential impact
> this might have on GPUs and made me remind about one datapoint we have on this topic.
>
> Back in the RV7xx days when AMD introduced the RV790 XT (aka Radeon HD 4890) they added a large ring of decoupling
> capacitors to the existing RV770 XT design to boost its maximum clock speed. The RV770 was originally released
> at 750MHz and the RV790 started at 850MHz but was widely available in overclocked configurations up to 1GHz.
> Interestingly the die-size increased significantly from 256 mm² to 282 mm², a >10% increase.
>
> While those chips are now relatively old and the techniques used in GPU design might
> have improved, they still give us an idea of how much benefit could be reaped by getting
> rid of decoupling capacitors. Current high-end GPUs sport both higher power consumption
> and larger dies than the RV7xx family so the benefits might be very tangible.
Interesting.
| Topic | Posted By | Date |
|---|---|---|
| New article: Adaptive Clocking in AMD’s Steamroller | David Kanter | 2014/05/06 12:33 PM |
| New article: Adaptive Clocking in AMD’s Steamroller | itsmydamnation | 2014/05/06 10:12 PM |
| New article: Adaptive Clocking in AMD’s Steamroller | Xi Yang | 2014/05/06 11:08 PM |
| Decoupling capacitors on GPUs | Gabriele Svelto | 2014/05/08 08:24 AM |
| Decoupling capacitors on GPUs | Ben | 2014/08/27 10:56 PM |
| New article: Adaptive Clocking in AMD’s Steamroller | Poindexter | 2014/05/10 05:06 PM |
| New article: Adaptive Clocking in AMD’s Steamroller | Rob Thorpe | 2014/05/12 04:42 AM |
| New article: Adaptive Clocking in AMD’s Steamroller | David Hess | 2014/05/12 09:02 AM |
| New article: Adaptive Clocking in AMD’s Steamroller | Rob Thorpe | 2014/05/12 09:46 AM |
| New article: Adaptive Clocking in AMD’s Steamroller | Ricardo B | 2014/05/13 02:15 AM |
| New article: Adaptive Clocking in AMD’s Steamroller | Rob Thorpe | 2014/05/13 05:30 AM |
| Multiple PLLs wouldn't work, but DLL jitter is no problem | Robyn Henry | 2014/05/13 11:58 PM |