Predictor@Home race started!

[NeoGen]

I rounded up the base values for everyone. Start dumping those workunits.


Race Stats Links:
Pharmers League
Regulars League

[Nflight]

Dumping, ! I feel like a little kid with all this excitement in the air! :laughing8:

[gamer007]

Whoa? Feels early to start. :? I thought it was Mondays @ 4pm PST for me.

Ah well... DUMP!

EDIT: Oh crap.. maybe that's why I had a little disadvantage.. I thought it started a day later. :mad:

[vaughan]

Mondays at 9:00 am here.

Its goin' to take some time to switch the pharm off Sztaki and on to Predictor but the process has started :!:

[Strongbow]

I've dumped all mine (not many) and I seem to have finally aquired a load of WU's to run on each system - why didn't I get all these before the race started? :roll:

Oh well...

Now can anyone tell me why this is happening; does Predictor have different job applications which take a different amount of time to complete and is it based on processor type?...

* My Dual Core 2.8 Intel is taking 1.18 to complete a WU :roll:

* My 2way Opteron is under 40mins to complete a WU

* My Mac is taking over 3 hours to complete a WU :shock:

* My old boy Intel P4 2GHz is taking 2.40 to complete a WU :?

* My other old boy Intel PentM 1.6 is taking 1.01 to complete a WU

All of them have the latest BOINC and the x86/x64 systems all have Crunch3r's client. The DC Intel shouldn't take that long surely and it certainly should be quicker than a PentM 1.6??? ...and the Mac G5 might have to retire from the race unless it gets it's finger out and starts crunching!

Strange how the benchmarks do not reflect the time taken to complete!

Any ideas? ...I know if they were all AMD I probably wouldn't have this problem but I am using all these disparate systems for a comparison and yet Predictor has thrown me completely off target when compared with other projects!

[drezha]

Yes they do have different applications...
http://predictor.scripps.edu/apps.php

All are avaliable for all the OS's you mention.
BUT

I cant say whether they favour a chip make or not. Coudln't really help there because I only have AMD chips on it...and they all have mbfold

[Strongbow]

They're all crunching "mfoldB125 4.28"

EDIT:

Just found out that my Mac is running "mfoldB125 4.29"

[Lagu]

Hi

My AMD Athlon 2300+ crunch both mfold8125 4.28 bprion 10 and 11 between 42.50 to 44.01 minutes.
My AMD Athlon 1001 crunch 4.28 bprion 10 and 11 between 2.17.11 to 2.18.53 hours.

Predictor is a little harder to crunch than SZTAKI.

Lagu

[Strongbow]

Hi Lagu,

Your benchmarks reflect the difference in compute time required, mine however look like this (in order of completion time)...

* <=41mins - Opteron248
Measured floating point speed 3748.04 million ops/sec
Measured integer speed 12229.73 million ops/sec

* 1.01mins - Intel PentM 1.6
Measured floating point speed 2205.72 million ops/sec
Measured integer speed 8466.15 million ops/sec

* 1.22mins - Intel Dual Core 2.8
Measured floating point speed 3109.09 million ops/sec
Measured integer speed 9982.19 million ops/sec

* 2.40mins - Intel P4 2GHz
Measured floating point speed 1724.88 million ops/sec
Measured integer speed 6668.02 million ops/sec

* 3.02mins - iMac G5
Measured floating point speed 965.23 million ops/sec
Measured integer speed 2986.3 million ops/sec

It's very strange :cry: :?

I must admit though if Predictor really does favour the PentiumM then I have 3 more 1.6's I could power up (wish it was 3 more Opty's mind you).

[Nflight]

Patience is something I just seem to not have much of lately. Especially when it comes to seeing results for work done, and a lot of work done, but no results are showing, What gives?
I dumped a full two days of work Units and the result is exactly what Mitros machine did for me and nothing of my own. Maybe I am giving someone elses machine points and not mine!

Signed Frustrated