Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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 (44063)	12/26/1998 8:44:00 PM
From: Ali Chen	Read Replies (2) \| Respond to of 1573044

<Funny..what you say are nothing but facts, instead of the dumb opinions that they truly are.> Model CPU Drive Disk Winmark ----------------------------------------------- Dell XPS R450 P-II IBM DTTA-371010 3570 GTW G6-450 P-II IBM DTTA-371440 2460 ----------------------------------------------- If you call 2460kb/s vs. 3570 kb/s for the same chipset, CPU, and disk drive as "dumb opinions that they truly are", all my name calling are justified and you need to go back to 2 or 3 grade and learn the distinction between "fact" and "opinion". <Funny how you have to keep reminding me how so mature you are by using the kind of remarks that I outgrew back in high school.> I never attended an American high school, and I am learning these remarks from other, much younger sources, just to keep up with your way of discussion. Sorry if some of my remarks are a bit off the line, I am trying to do my best :) <Meanwhile, you offer no logical proof that Xeon's full-speed L2 cache is no better than Pentium II's half-speed L2 cache> I do not need to proof anything logically. For the market we are concerned here, the common benchmarks are the proof. If you want some logical explanation, I could speculate that, for a modern superscalar superpipelined microprocessor, if the data delivery time (latency+bandwith) drops below certain limit (I would suspect the length of pipeline is the characteristic factor), furhther increase in L2 speed does not matter - there is enough stuff wating to be executed, and it shadows the processor's bubble due to L1 miss. Is this comprehendable for you? <..benchmarks which aren't really designed to do the things that a Xeon was meant to do.> I also could care less of what Xeon is meant to do on not to do. We are talking about the market segment of personal computing. The benchmarks are designed for applications, and not meant for a certain processor. For THIS SORT OF APPLICATIONS the full-speed Xeon cache is waste of money. The tres-biffoons (and you) are desperately trying to justify the "full-speed" cache. You are trying to compare 128k "full-speed" Celeron with 512k half-speed P-II and derive from this "advantages" of a full-speed L2. The Xeon argument has been thrown for the same reason - to talk down K6-3 and upcoming K7 L2 design. (Actually, by lowering the expectations you are doing a good job for AMD stock.) However, there is another explanation of why all your three processors behave so equally: faster or bigger L2 does not matter much with a non-blocking system bus! Apparently you still don't get it. <Then you even stick in a few outlandish statements like "K7 is coming with up to 8MB of VERY AFFORDABLE L2 off-the-shelf cache" and "No matter how much money Intel has, industry will resist Intel's strategy and favor AMD's" and try to pass them off as facts.> Again here you are having problems with distinction between facts and opinions. Your problem. Be happy that I warned you... <Please, Ali Babble, please show me those benchmarks showing that K7 servers are going to be faster than Xeon servers.> IMO, these servers may not be faster on clock-per-clock basis, but I have some gut feeling that K7 servers will be much cheaper and run at much higher frequencies. Bookmark this. <Guess Mary Clueny was right in calling you the "insufferable Ali Chen."> What does this mean? I would appreciate the reference. She is funny sometimes.