• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

WPrime 2.00 Benchmark Thread

Uriel

Uriel

Uriel

Soldato
Joined
15 Jan 2006
Posts
7,770
Location
Derbyshire
Just thought we could do with another benchmark thread. Some have said that SuperPi is optimised for Intel. Others have suggested that Fritz strongly favous hyperthreading.

So wPrime 2.00 - available to download here: http://www.wprime.net/Download/ Please calculate to 32m.

To make CPUs easy to compare I suggest we benchmark with threads set to cover the main combinations of cores and hyperthreading. I'd suggest setting to threads to 8, 6 (in preparation for the hexacores - quad cores with hyperthreading can get in some early results now) 4, 3, 2 and 1 if available. wPrime

Please include in the screenshot your user-name and how many threads you had running for each run, plus a CPUz shot. Please also indicate your CPU and clock speed.

If someone could try an i7 @ 4 threads with hyperthreading on and off to see if it affects the results that would be handy.

Edit : Leaderboard (work in progress)
4341124513_285341a948_o.jpg

Note: * indicates unlocked cpu. 0 reading indicates no result - not the ultimate ultimate uber CPU. I'll sort that out in an update at some point :rolleyes:

Edit: Results work in progress.
4341103069_77c9999fce_o.jpg


Here's mine to kick us off:
4337539436_988d0df3bc_o.jpg

Athlon X2 5000+ 45nm @ Phenom FX-5000 3.3GHz (NB/HT 2.1GHz)
 
Last edited:
I'd be interested to see a few hyperthreading vs native core benchmarks to get an idea what sort of effect it has. FWIR, hyperthreading gives a nice boost to wPrime so the i7s should beat higher clocked Qxxxx CPUs. Having said that, I would expect the results for Intel vs AMD to be much closer than in SuperPi.
 
Some benches and testing software won't run with a slightly dodgy overclock. S&M won't for example. Also try the previous version of wPrime to see if it will work as it's known to be more stable.
 
Topitoff - just a query - your benchmark results show 3900 mhz. CPUz shows 3600. Which is it? Is there a turbo mode or something running?

Still - impressive results from the i3. Getting the same 4 thread result as my Deneb @3.3GHz, a quad core (albeit unlocked from dual :)).

Edit: Topitoff's shot also indicates 4GHz in the notepad file :confused:
 
Last edited:
Just thinking about making this thread more about info, with a hint less epeen (if that's possible round these parts).

I give you the first results graph - a bit of info on scaling vs clockspeed, cores and architecture:
4339130422_79d30188aa_o.jpg


I decided not to use deneb x4s running 3 cores in the deneb x3 results for example. Any thoughts?

Edit: Already we have an anomaly @ Elfmeister something's up with your result. Try disabling CnQ.
 
Last edited:
It may have taken a little time for the CPU to get up to speed. All the other Deneb results seem to line up in an almost linear fashion with clockspeed. You end up getting about the same time at mine at 3.2GHz. There could be something else causing the issue but my money's on CnQ, unless you were running something else at the same time.

Edit: my hypothesis seems to fit with your times with less threads being closer to the other Denebs. It would make sense for the shortest test to be the most severely affected.
 
We'll have to see. On Deneb and Bloomfield it looks like a very close correlation between pure clockspeed and the results (as long as the number of cores / threads remains the same and CnQ etc is off).

I'm interested to see the effect of cache. If we get some Athlon II results that should make it clearer.

NB speed and memory timings don't seem to make a huge difference for Deneb at least. There's an effect but it's not huge.
 
Last edited:
Interesting on the v2.01. I got 7.359 vs an 11+ seconds run with version 2.00.

It's just a reminder how much these benchmarks can be CPU optimised. SuperPi anyone?

Anyway - I consider this thread rather redundant now. :(

If people that have already submitted scores could add some v2.01 results (in the other thread) I'll do a comparison chart. It will be interesting to see whether everything gets a boost or only selected CPUs.

After that, I'm afraid I don't see the point in keeping the charts and leaderboard up-to-date. Sorry everyone. Just bad timing I guess.

New thread in here people: http://forums.overclockers.co.uk/newreply.php?do=newreply&p=15929431
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom