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

A64 folding

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

Red_Silver

Registered
Joined
Aug 10, 2004
ok guys, little question here. does anyone have any of the benchmark scores for an a64 system for folding? or better eyt, some real world emIII or fahlogstats points per day calculatiopns
cheers!
 
I should have an FX-53 (new gaming rig) up this weekend. I'll let you know how it does.
 
The a64 rig in my sig fold's about 18\7 only time down is testing and gaming.
 
It took me a while longer than anticipated ( son took monitor and keyboard I planned to use off to college) to get my FX-53 (socket 939) up and running but I now have some results. Folding p1301_1RYP_AAAA_UM the FX-53 rig will produce 265 points per day in stock condition. Sig rig #1 & #2 produce right at 300 ppd and sig rig #3 produces 260 ppd. In terms of folding bang for the buck, nothing touches the Mobile XP 2500+. When I'm done overclocking, I imagine I'll get about the same production out of the nearly $900 FX-53 as I get out of a $182 P4 c. There are certainly lesser A64 processors that have acceptable price performance ratios. In my experience, CPU MHz is far more important than any other factor.
 
ChasR said:
In my experience, CPU MHz is far more important than any other factor.

this is almost certainly true, but as we all seen posted earlier this week that 2 Identical chips will not fold the same..2 different boards 2 different types of RAM 2 identicale CPUs and the DDR board outperformed the SDRAm by a landslide. I don't know enough about how all that works, but it makes sense to me with my limited knowledge. I would of though however that the 64's would have been golossal @ folding given that the memory controller was in the cpu..or am i mistaken about that...anywho This reafirms it isn't right for me to go 64 just yet. Thanks ChasR for the input.
 
memory is indeed a factor, but i think the bottleneck on the 64 comes not from the memory, but from mghz
 
ok then try to explain to me so I can understand why the AMD VS Intel always seems to favor the AMD with the same MHZ rating? if I clock my AMD to 2.6 and run the same wu as i am running on my p4 2.6 the AMD just hands the p4 it's butt.
 
i think that has to do with the work the amd does per cycle, when i say mghz vs mghz i mean within the same brand. like for intel i would say mghz is better than say faster ram.
 
FizzledFiend said:
ok then try to explain to me so I can understand why the AMD VS Intel always seems to favor the AMD with the same MHZ rating? if I clock my AMD to 2.6 and run the same wu as i am running on my p4 2.6 the AMD just hands the p4 it's butt.

AMD pwns Intel in folding because of the number of calculations per clock cycle. AMD does 9 calcs per cycle comparred to the Intel's 6. This is why you get ratings of 2500+ when the chip is running at 2.1Ghz.

Intel makes up for it in other apps by using larger L1 and L2 caches, superior memory clock speed, and software which is optimized for the chipset.

However, score yourself a 3.0 (or higher) Intel with HT and you'll pwn in production.
 
Ignoring overclockability due to its varibility, if you compare processors at the same price point, say the A64 3000+ at $175 and the P4c 2.8 at $182, I think the P4c will win the folding contest easily. If I get the FX-53 to 2.65GHz (as did anandtech) I expect it will produce 290+ ppd. A P4 560 @ $565 will produce upwards of 340 ppd. The AMDs do more work per clock cycle but can't overcome the MHz gap at the high end. At the low end, a Mobile XP running at a liesurley 2.4 GHz will produce 245 ppd the FX-53 at 2.4 Ghz only produces 20 ppd more (probably due to its on chip memory controller).
 
Last edited:
Back