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 217.59+1.1%Dec 3 3:59 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: andreas_wonisch who wrote (30085)3/2/2001 2:02:09 PM
From: Harvey AllenRead Replies (1) of 275872
 
Andreas- Yes I have the same board. Initially I benchmarked it with a Duron 700 OC'd to 933 which was the only way to test the PC2100 at 266FSB. I wasn't overwhelmed by the memory numbers either but they were way higher than the P-III's I had been using with VIA chipsets and Intel's terrible i820 with SDRAM MTH.

My memory numbers were also better than any of the comparative Athlon systems listed but of course far lower than the P4/Rambus memory numbers.

I'm content right now with a 800MHz TBird running @ 904 using only 226FSB (PC1600 range). The TBird's cache makes up for the reduced FSB and overall my system feels quicker in this configuration than with the higher speed Duron. I believe I have a great increase of performance ahead of me and I'm only waiting for the right prices to come along and of course availability.

I'm really kicking myself for not bidding higher on this Blue Core T-bird 750@1000:

cgi.ebay.com

That would have solved my 266FSB problems through the end of 2001.

Overall, I don't think Sandra does a very good job measuring latency or L1/L2 cache effectiveness and the best measure is the performance of your primary job. I do database development and my feeling is TBird/DDR is the right environment for me. SMP offers a still better future.

Harvey

P.S. I read recently that the jump from EDO to SDRAM was only on the order of 10-15% and FPM to EDO even less so if we can get 15-20% out of SDR to DDR it's a pretty big deal.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext