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.

Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Tenchusatsu who wrote (137210)	6/12/2001 4:25:25 PM
From: Jim McMannis	Read Replies (2) \| Respond to of 186894

RE:"And if the Pentium 4 does well on that, you'll complain the creator of that benchmark is actually under Intel's payroll, yada yada" Does everyone play Quake? Are there other games available? <G> Tom used to love monster truck madness...what will he come up with next? P4s best benchmark is always something called memory bandwidth. I don't complain about that. I'm still trying to find the "memory bandwidth" program so I can try it myself. Is this like MS Office or AutoCad? Is it a spreadsheet? LOL Jim

To: Tenchusatsu who wrote (137210)	6/13/2001 8:17:41 AM
From: fyodor_	Read Replies (1) \| Respond to of 186894

Tench: By the way, if you haven't noticed, BAPCo's 3DMark and ZDNet's 3D Winstone were attempts to create the "all-encompassing 3D graphics benchmark." Given the controversial nature of benchmarking in general, what do you think it would take to come up with the benchmark to end all benchmarks? Everyone agrees that benchmarking is a sensitive issue. Personally, I rather see benchmarking of REAL programs than synthetic benchmarks any day (e.g. Quake3 over 3DMark GameX). I especially don't like the "complex" synthetic benchmarks. At least with the "simple" ones, the measured quantity can be used to estimate how the system will perform in certain real programs (that may not be easily benchmarked). Another major problem with "complex" synthetic benchmarks (besides yielding a result which is extremely hard to extract useful information from) is that they can be easily manipulated by processor-specific optimizations. In benchmarking of REAL programs, these optimizations can also be present, but they are REAL programs, so you can actually USE the program on a processor with those functions and benefit from the optimizations. No one benefits (directly, anyway) from processor-specific optimizations of synthetic benchmarks. As for the impressive Q3 scores of the P4 processor, the best evidence to date suggests that it is due to the extremely skillful programming, rather than any processor-specific enhancements. Quake3 is quite possibly the best-programmed (complex) game out there today. The tight inner loops reduce the performance loss due to small L1 cache of the P4, while enhancing the gains from the low latency of the L1 (and L2) cache. While this may not be indicative of 3D performance in other games (and isn't, if you judge by the other 3D game benchmarks available today), it is a REAL program and the performance showed by the benchmark is REAL. If you play the game (which I don't, but apparently many do), this performance can be achieved. -fyo