SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Advanced Micro Devices - Moderated (AMD)
AMD 250.37-0.9%10:09 AM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: TGPTNDR who wrote (236968)7/24/2007 1:45:37 PM
From: wbmwRead Replies (1) of 275872
 
Re: Fusion means that the CPU and GPU have to come out at the same beat rate.

>> Why? Do not, or have not, CPUs had different beat rates for their various parts?


It may not be crucial in the Fusion time frame, but eventually, as things get integrated, it will have to come out at the same beat rate. I used the floating point unit as an example. Intel used to make 287 and 387 discrete FPUs, but eventually integrated them onto the CPU. After that, the FPU design became part of the usual CPU design. I see the same thing happening with GPUs.

The only difference is that discrete GPUs will still exist, long after they get integrated onto the CPU. It's simply because having an entire die all for the purpose of graphics will beat sharing a die with a CPU every time, no matter how many times you shrink it, or what design optimizations you make.

At some point, it might make sense to derive the discrete design from the integrated design, rather than the other way around, which is what they do today.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext