wPrime - hwbot edition launched

[richba5tard]

wPrime - hwbot edition launched

http://www.hwbot.org/blog/wp-content/wprime127.PNG

We are happy to announce that wPrime, an new benchmark application that takes full benefit of having a multiplecore processor, has been added to the hwbot benchmark suite.

wPrime has the same ease of use as SuperPi, but has 4 key benefits:


multithreaded: take full use of your dual / quad core processor
simplified: click 32M for a quick test, 1024M for a stability test. No 10 different methods to choose from.
hwbot integration / one click comparison: no need to manually enter your result on hwbot, or post it in a bot monitored forum. Just click ’submit’ and you can compare your result to ten thousands of other overclockers. If you belong to a team which uses a bot to monitor your forum, the bot will will update your forum ranking too.
hardware recognition: in contrary to superpi, wPrime dedects your hardware automatically.


The current version is v1.28. Previously submitted results by beta testers with a client < 1.27 will be deleted.


original anouncement (http://www.hwbot.org/hwbot.post.do?postId=270)
download wprime v1.28: local (http://www.hwbot.org/blog/?attachment_id=269) - mirror (http://www.wprime.net/wPrime_gui128.zip)
wPrime info page (http://www.hwbot.org/benchmark.application.info.do?applicationId=14&name=wPrime%2032m)
wPrime website (http://www.wprime.net/cgi-bin/page.exe?q=../home.html)
32M: hall of fame (http://www.hwbot.org/hallOfFame.do?type=result&applicationId=14) - statistics (http://www.hwbot.org/benchmark.application.statistics.do?applicationId= 14)
1024M: hall of fame (http://www.hwbot.org/hallOfFame.do?type=result&applicationId=15) - statistics (http://www.hwbot.org/benchmark.application.statistics.do?applicationId= 15)
hwbot forum discussion (http://www.hwbot.org/forums/viewtopic.php?pid=3773#p3773)


Known issues:

checksum validation sometime fails on hwbot
if you submit another result without closing your browser window first, your browser may display cached data. Hit CTRL+F5 if you encounter this issue, or close your default browser if you submit a second result.

[speed bump]

Can we run this on one computer then save the result and transfer it to another computer? If not that might kill it popularity and this looks like it could be a pretty good benchmark.

Also your mirror does not seem to be working.

[richba5tard]

Yeah, saving scores to a file and opening it in another instance of wprime is not a problem.

mirror link is back up

[El<(')>Maxi]

Get on this guys, more points for the Team.

[Gautam]

Will do, sick of single stage though. :p

[sno.lcn]

Get on this guys, more points for the Team.
Got my scores in :D I'll push for more on some of my other rigs tho :beer:
Will I get better times if I use a slower rig that has four threads (2.4 xeons to be exact) than the one in my sig?

[El<(')>Maxi]

I don't know sno, thats a good question though. Best try it out ;)

Hey Gaut, I got a new chunk of copper on the way. I promised the rev 1 mousepot to gwild first but you're wlecome to it when he's finished.

[Gautam]

Thanks for the offer, but TBH while the ol Chilly1 container isn't the best, its still good enough for it to not be worth your trouble. I'd much rather have it to go to someone with no subzero anything on hand but wants to get in on it.

[Gautam]

Several problems...the ability to save results offline is a must. Also, it doesn't detect my speeds properly when I use ClockGen, furthermore, it seems to perform as if it were at the speed it detects. 23.922 at 4 GHz. :S

[El<(')>Maxi]

Sounds like it has some reason to be concerned. I'd post you're findings over at hwbot forums to be sure those guys see this, I havn't had a chance to try it yet.

[Gautam]

lol, looking at the forum it seems like Fredrik himself has this issue. :p

Benchmark seems to have a lot of potential, but its a bit rough around the edges ATM...

[richba5tard]

Several problems...the ability to save results offline is a must. Also, it doesn't detect my speeds properly when I use ClockGen, furthermore, it seems to perform as if it were at the speed it detects. 23.922 at 4 GHz. :S

The ability to save exists, maybe it's not super clear. It's the 'save' icon next to your benchmark score.

A more correct readout of the clockspeed is being worked on. :)

[Vento1]

High there gave this a go using latest version uploaded two 32m runs the checksum on there site says invalid so no points for the team but i can run seti at this speed for over 12hrs with no errors any sugestions on what i am doing wrong.

[Gautam]

Nice rig there. :D

This is a known issue apparently. I don't think it has anything to do with stability. Was the clock speed really 4.4GHz or 4.2GHz?

Try running it again, and if that doesn't work, hopefully Fredrik will come over soon to resolve this. Great score again, this should be taken care of soon!

Edit- actually the checksum doesn't seem to matter. Your points will probably be added within 24 hours.

[Vento1]

I had it at 4.4 but my abit board does not like the OCZ memory i have, I will run the 1024 test again tonight i did run it but did not upload the score when it showed invalidated it did the test in 10.5 mins

[El<(')>Maxi]

Is anyone gonna try this or should we delete the rankings for wPrime?

[sno.lcn]

I'll try and get some better runs in, but I think I've about maxed out the rig I've been benching with. I'll try it real quick on one of my dually xeon rigs and see if it does any better.

[El<(')>Maxi]

Dual Xeon, thats sounds good to me :)

[David_N]

I plan on doing some runs this weekend, I'm just trying to get some spare time.

I have a 1250 word economics essay due in an hour and a half, along with a 4000 word essay comparing blu-ray and HD-DVD due on Monday... That should be an interesting essay, provided I can find enough resources (which is actually scaring me to be quite honest... It might prove to be quite difficult).

Anyways, enough of my babbling... My point is, I'll do my best to get some results posted this weekend! Wish me luck.

[sno.lcn]

lol the dually xeons did over 56 seconds for the 32mb. I'm guessing it does a different one for each thread/core?

[Gautam]

Is anyone gonna try this or should we delete the rankings for wPrime?
Hey hey, practice what you preach. :p

I did a run at 4.1ghz a while ago, but I scored worse than most people at 3.5GHz....:( so obviously I didn't even bother with it since then.

Maybe its largesystemcache that kills this bench or something.

[El<(')>Maxi]

I was going for some clever motivation but you just blew my cover, thanks!!! ;) :D

[speed bump]

I ran this then cried at the results. However I think i'll give it another shot when I get around to ordering a Brisbane and 8600.

[El<(')>Maxi]

What is Brisbane? I'm soo out of AMD at the moment :)