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.
Politics : Formerly About Advanced Micro Devices

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Ali Chen who wrote (26259)11/29/1997 11:40:00 PM
From: Elmer  Read Replies (3) of 1572630
 
<The main reason why the Slot-1 CPU runs faster is the internal design methodoligy that Intel stole from DEC Alpha design - a RISK-like "short-tick" approach, when the complex logic is broken into longer processor pipelines. >

Ali, this isn't what Dec claimed Intel copied. What you have described is basic to all RISC designs, nothing unique to Dec.
In addition, AMD used these same techniques in designing the K6. You can make all the excuses you want but it won't change the fact that Intel has a better .35u process than AMD, and they have had it much longer. Some of the process steps may have come from Sematech, but certainly not all of them. Most of it was developed at Intels development fab, just like AMD developed their own too.

<If you would ask why the Slot-1 design shows slightly better performance, the answer is: due to additional "back-side" bus into L2 cache. AMD is going to fix this in K6+3D chip with on-chip L2 cache, as per my understanding. >

The backside bus does help, no question there and AMD will close some of the gap in doing their own version, but Intel's split transaction bus archeticture also contributes. This is why the Intel design is so effective is SMP servers and why you don't see any socket7 SMP servers.

EP
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext