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: Elmer who wrote (118938)	11/22/2000 5:27:07 PM
From: pgerassi	Read Replies (2) \| Respond to of 186894

Dear Elmer: Did you compare the P4 against the Athlon on hand tuned optimized assembly code. I have not seen those benchmarks. I have always said that the SPECmark as currently used is broken. And the P4 does not perform far better than a Athlon given equal optimization. In fact, it performs no better than an Athlon at 80% of its clock rate. The only time it goes faster is when it is executing code that is SSE or SSE2 optimized only but, not 3DNow or dual pipe FPU optimized. That violates your recompiled for argument. All currently available compilers (there does not exist one publically yet for SSE2 (IFC 5.0 is not yet available last I looked)) do not have SSE2 optimizations, yet you conveniently deny Athlon hand tuned assembly routines when P4 currently requires them as well. So all we have is current benchmarks and those hand coded for P4. The current champion CPU in Primordia, QMC, Moldyn, and Linpack is an Alpha. 2nd is an Athlon. P4 is below those and sometimes not even in third. These are real world benchmarks. And you forgot the list where P4 performs so bad, that it is near the bottom of currently available CPUs. Rc5 Expendable (640x480) Sysmark 2000 SPECviewperf 6.1.2 awadvs-04 3D Studio Max 2 Linux Kernel Compilation IEEE high quality iDCT MPEG-4 encoding And this list is growing faster than the other one. Pete

To: Elmer who wrote (118938)	11/22/2000 7:48:32 PM
From: muzosi	Read Replies (1) \| Respond to of 186894

You're just rehashing the old argument that it doesn't run old code* quite as well as new code.* So one can go to Fry's and buy this new code you're talking about ? I think you meant to say current code vs future code. I don't know anyone doing their work by running spec. I am running HSpice almost everyday and from what I can gather P4 would completely and utterly suck with HSpice compared to Athlon. When do you think Avanti will recompile Hspice with this unavailable Intel P4 compiler ? I also run Xilinx PAR tools, when do you think Xilinx will compile their tools with Intel compiler ? The fact is P4 does very poorly in GP applications and it will do very poorly for an unknown period of time. First Intel has to release their P4 compiler and get someone to use it. Who do you know uses the Intel Plugin ? Avanti, Altera, Xilinx, Adobe ? Any names with urls you can give ? Muzo