• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Superpi prescott patch

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Scph9002

Member
Joined
Dec 6, 2004
It´s kinda boost of the score.... lowered my score from 35sek to 32 sek...
What happens if U try it on a northwood or amd??? I think just probably nothing... but do u pressie-guys out there use the prescottpatch for superpi?
 
Scph9002 said:
What happens if U try it on a northwood or amd??? I think just probably nothing.
Funny you should ask. AMD guys get all bent out of shape and call it a cheat. They say the Prescott sucks because it is too hot and slow per clock, but the second you try to run a program that takes advantage of its advanced instructions or cache, it's a "cheat."

You can't win with these AMD zealots (not all AMD owners are this way, but the ones who are are very vocal). They only want you to run the version which favors their processor, and of course Prescott people would like to actually use SSE3 rather than have it sit dormant.

Anyway, benchmarks are meaningless unless - unless - you only look at them as task-based. In other words, which platform can compute pi or primes or whatever fastest. But if you expect different hardware to run the same application, someone's gonna feel cheated.

I say, compile separate, patched versions designed to milk the most out of each processor, whatever it takes to get the fastest pi, or whatever.

But using a program written before a new processor even came out which completely ignores its new features doesn't seem so fair, IMO.
 
uclajd said:
Funny you should ask. AMD guys get all bent out of shape and call it a cheat. They say the Prescott sucks because it is too hot and slow per clock, but the second you try to run a program that takes advantage of its advanced instructions or cache, it's a "cheat."

You can't win with these AMD zealots (not all AMD owners are this way, but the ones who are are very vocal). They only want you to run the version which favors their processor, and of course Prescott people would like to actually use SSE3 rather than have it sit dormant.

Anyway, benchmarks are meaningless unless - unless - you only look at them as task-based. In other words, which platform can compute pi or primes or whatever fastest. But if you expect different hardware to run the same application, someone's gonna feel cheated.

I say, compile separate, patched versions designed to milk the most out of each processor, whatever it takes to get the fastest pi, or whatever.

But using a program written before a new processor even came out which completely ignores its new features doesn't seem so fair, IMO.

So true..

I think the reason they get upset us because the patch wasn't released by the coder of Super-Pi. Which would bassically make it a hack, since it wasn't meant to run with the original code. It shaves 2 seconds off my Intel, which is cool, but i still run the non-patched one for comparing to others. I think the maker of the patch should have made it so the screen shots would represent the patch. But it really doesn't matter, it's easy to spot a patched score.
 
cpus hardware is there to be used and software should be compiled to take the best advantage it can. Perhaps all software should come with 2 instals the installer decides which will be the fastest for your hardware.
Is the built in memory controler cheating on the A64? Nope: I am looking forword to owning a numa system one day.
 
aftermath said:
cpus hardware is there to be used and software should be compiled to take the best advantage it can. Perhaps all software should come with 2 instals the installer decides which will be the fastest for your hardware.
Is the built in memory controler cheating on the A64? Nope: I am looking forword to owning a numa system one day.

Best argue I´ve ever heard of this matter!
 
Im sure the amd users in the future will do this when the A64 gets sse3, does pi not use the clawhammers 1mb cache fully?
 
Sucka said:
I think the reason they get upset us because the patch wasn't released by the coder of Super-Pi. Which would basically make it a hack, since it wasn't meant to run with the original code.

I'm sort of inclined to think this way too. If it was an "official" patch, I'd jump on it in a heartbeat. But. since it's a hack, then I can see the side of the argument that is calling it a cheat. As much as I'd like to gain an advantage, I'll just use the official version for now. Besides, my score of 29 seconds is hard to beat (no patch).
 
Back