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 -- Ignore unavailable to you. Want to Upgrade?


To: Tenchusatsu who wrote (60355)6/3/1999 7:58:00 AM
From: Scumbria  Read Replies (4) | Respond to of 1571146
 
Ten,

regarding the benefits of moving to PC66 to PC100

The dependency of High End Winstone on memory bandwidth is interesting. Good catch!

I should point out that the numbers you reported were for a Celeron, which has a very small cache structure (32K L1, 128K L2.) Any other CPU will show an even smaller performance increase with bus speed. In addition, the performance/MHz curve tails off as you move higher up, so that will diminish the effect further.

Scumbria



To: Tenchusatsu who wrote (60355)6/3/1999 10:22:00 AM
From: Steve Porter  Read Replies (1) | Respond to of 1571146
 
Tench,

Yeah but pc-133 memeory will help my compiles go faster ;).. man those thing can be memory hogs.

Steve



To: Tenchusatsu who wrote (60355)6/3/1999 12:39:00 PM
From: Marco Polo  Read Replies (1) | Respond to of 1571146
 
Can you link to where he said the Celeron 400 wasn't multiplier locked? Because I was under the impression that all C400's are! Using PC-100 RAM with a Celeron doesn't mean the system bus is running at 100 MHz. I'm using PC-100 RAM right now with a Celeron running the bus at 75 MHz.

It seems everyone on this board doubts the importance of the bus speed. Put it this way, even if the Coppermine was as good as the K7 (it's not), the 200 MHz non-bus speed of the K7 would put it over the top in a lot of areas vs. the 133 MHz of the Coppermine, even if they were both using the same memory.



To: Tenchusatsu who wrote (60355)6/3/1999 11:16:00 PM
From: Ali Chen  Respond to of 1571146
 
<Pretty complicated picture, no?> Not at all.
All these things are totally transparent
for specialists. For one of the model,
see some previous posts from Petz if you
want to understand impact of caches and memory
performance.