wooooow. at 4ghz my score changed from 32s to 29s. please god let it be legit :)
as i understand it it's no cheat as the program does still the same calculation but optimized for p4e?
Printable View
wooooow. at 4ghz my score changed from 32s to 29s. please god let it be legit :)
as i understand it it's no cheat as the program does still the same calculation but optimized for p4e?
impressiveQuote:
Originally Posted by xsky
Makes it pretty much impossible to compare with original superpi results - which is not a good thing IMO. Where's the A64 patch?? :DQuote:
Dunno how legit this patch is that updates FISTTP instructions for prescott:
Yeah, but with this out there its impossible to tell which version its runnning so just run the fastest version? :p:
20-21s, the new world record??
Damn where is the A64 patch :(
After the patch its good for comparing different Prescotts but thats about it. Silly prescott-only bench.
I mean superpi has been the same for several years now and become a popular 'standard bench'. It was fairly easy to see the performance level of any given hardware (AMD or intel) just by running the SPi 1M test. Now someone (working for intel??) got a great idea and edited the code and made it 'prescott-optimized'. He should have edited the name of the benchmark too.
and I think that's not just optimized for the 1M, 8M and 32m too?
I test
It's not funny if there is nothing for the A64
yeh it would be more fun if there was an a64 patch but somehow it's ok as it is a prescott feature. maybe simular things are possible in real useful tasks and prescott becomes superior to the rest.
i don't know enough about these things but maybe someone can tell me...
Its not funny that it wont run on any other CPU than Prescott (just tested W/ my P4EE and it crashes instantly).Quote:
It's not funny if there is nothing for the A64
I dont see how this edited-superpi has anything to do with any kind of realworld performance now.
:banana::banana::banana::banana:ty thing that in weekend gonna cool with co2 and ln2 myh precious FX-53 and with out that patch. :( It ain't so funny anymore
5m16s at 4097MHz!!!!!!!!!!!!!!!!!!
bad trip
hehe =)Quote:
5m16s at 4097MHz!!!!!!!!!!!!!!!!!!
bad trip
I wonder if Aguri has been useing this patch for his latest DDR2 Superpi benchmarks? If not then it seems that he is the only person on the planet getting decent performance out of DDR2 mobos :D
edit: or maybe there is a 'DDR2 patch' available too??
yes he has because he gets 31s at 4148MHz like me but with 230/306Quote:
Originally Posted by macci
http://members.jcom.home.ne.jp/aguri...50-E0_test.htm
Yep, its definitely way off in terms of Real world performance
Have no idea at all who wrote the patch
Quote:
Originally Posted by LardArse
Hmm checked ABOUT in the patch
-={ CodeFusion Runtime }=-
Generator: CodeFusion Wizard
Super PI was reverse engineered by
Igor Levicki
Here's an improved one of mine, nothing special, but i was happy to get it.
http://www.sevenlabs.net/sucka/pictures/superpi5.jpg
Sorry, just keep trying for more...
http://sevenlabs.net/sucka/pictures/superpi6.jpg
Here's mine AKA Vflux/Henesseynalert
Sucka, whats up!!bro!!! :)
Air Cooling 3.2E 5:4 Cas2.5-3-3-7 BL8 PAM Auto/Standard
http://images.dr3vil.com/uploads/Prescottsuperpi2.JPG
LS cooled 3700+ @ 1.66v:
http://server5.uploadit.org/files/nexus2525-1M30s.JPG
Messy WinXP, stuff run in background. Only had this 3700+ for less than a day. I think it can do better.
it's realy "ScrenShot" :DQuote:
Originally Posted by Tek
hmm CO2 didn't improve my 1M time at all, but 8M 3 second faster than with prometia machII r404. Now meh best time in 8M is 4min 49s
SPi 8M (non-patched) w/ 2.8A (D0):
http://www.akiba-pc.com/2.8A/424.gif
Here's the 8M Ranking at the moment and as far as I can tell none of the top 7 results is done w/ the patch.
http://www.akiba-pc.com/2.8A/spi8mrank.gif
Neither is mine at place #10 ;)
5468Mhz and 4:36
Congrats on the 4:24 time!
:banana: