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: Elmer who wrote (101268)	3/31/2000 10:09:00 AM
From: Scumbria	Read Replies (1) \| Respond to of 1571808

Elmer, I'm really looking forward to some T-Bird benchmarks on a neutral compiler, which blow away Intel, and end this absurd discussion forever! Intel is selling a few overclocked chips labelled as 1GHz, but their design and manufacturing have fallen behind AMD. Scumbria

To: Elmer who wrote (101268)	3/31/2000 12:59:00 PM
From: tejek	Read Replies (1) \| Respond to of 1571808

The argument can be made that the webpage is simply old and out of date and this is probably the case, but people here don't seem to believe that about Dell's website. If AMD can have an old webpage why can't Dell? EP, First, I can not believe you are still sawing over this issue on your little violin (that you bought as a set with your tin drum)! Secondly, it is one thing to leave an outdated page re bench marks on your website and another thing to have an outdated one regarding pc availability. Why? Most people don't give a rats ss about bench marks and so AMD's business will hardly suffer over its bench mark page. People do care when they are to get the product they have ordered. Therefore Dell's leaving an outdated web page showing ship time longer than what it really is is not only poor business practice but plain out stupid. Now the question is: Is Dell a lousy seller of PC's or are you a horse's ss? ted

To: Elmer who wrote (101268)	3/31/2000 4:41:00 PM
From: Joe NYC	Respond to of 1571808

Elmer, I was just looking up stuff, and I came across this page on IBM website: commerce.www.ibm.com As of 3/31 it says about availability of most of the coppermines: within 4 weeks as of 03/31/2000 As far as Pentium III 500, it says: In-Stock as of 03/31/2000 Joe

To: Elmer who wrote (101268)	3/31/2000 9:25:00 PM
From: hmaly	Respond to of 1571808

Elmer Re..<<<<<<Everyone wants to point out that the new compiler won't help Athlon but the point is that the old compiler hurts CuMine and AMD falsely claims it used the newest one available, a claim that is clearly false, while producing unexplanably low FP benchmarks which no one else in the world can reproduce. <<<<<<< Elmer, maybe that is the latest compiler which runs on Athlon. Maybe AMD wrote their own compilers which show up athlon and hurt cumine just like Intels new compilers do to athlon. Maybe AMD is taking their sweet time publishing it just like Intel did. Secondly, how do you know if no one else in the world can reproduce the results if you don't know which compiler is used? Thirdly you claim this compiler hurts cumine. Duh Isn't that exactly what Intel did to AMD with their new compiler. Finally just maybe AMD'S webpage is up to date if they were AMD'S compilers. According to someone else who posted on this thread. AMD is not required by law to help the competition.

To: Elmer who wrote (101268)	3/31/2000 9:54:00 PM
From: pgerassi	Read Replies (1) \| Respond to of 1571808

Dear Elmer: I have written compilers and work in highly optimized assembly language code for embedded systems and scientific calculations. I could write a compiler that recognizes SPEC code and substitutes highly optimized assembly code tuned for ONE processor (i.e. a 0.18 micron 512K athlon at 1/3 cache) for the normal code generated. This would only occur during the optimising phase. Now if I did that, a Williamette would fail no matter what it was clocked at, but an Athlon at 700 would whip any 1.5Ghz Williamette scores out on any compiler. But take any decent third party compiler (gcc is in public domain), and this would not be the case. This type of speciality compiler would fit your rules, but you would cry FOUL! The best would take a public domain compiler (i.e. gcc) and run both processors on that code. Call that the BASE SPEC. Then Intel or AMD could write optimzed versions of gcc (and show the source used to generate those optimizations to protect against unethical practices) and display those as OPTIMIZED SPEC. This would allow anyone to verify the scores given the hardware and protect against cheating. There are many claims that Intel's 4.5 compiler can not even generate code that runs on an Athlon with the optimsed switches on. This would be easy to modify to do the reverse (after some reverse engineering) to a Coppermine. Thus, the Great 1.0 Compiler generates a SPEC 2000 score of 80 for a lowly Athlon and 0 for any Pentium III, Coppermine, or Williamette. Consider my diatribe on your emphasis on the Intel 4.5 SPEC compiler ended. Pete