Canadian Overclock Database

[Babrbarossa]

Great idea. I think for it to work (ie. for lots of people to want to contribute) it would have to be easy to post your results without too many screen shots and proofs-

I also agree with el donko in that it is more useful to post benches as performance is the ultimate goal of overclocking and the successful completion of the bench test can be the proof of stability (albeit marginal)- but as he says, it's well covered by hwbot.

Another key ingredient is a self-managing database which allowed members could just click a link to open an entry form which then ranked them and presented the results in an organised way.

I wonder if the forum admins have the resources to set up a database with an easy form-based entry system.

[Patriote]

Originally Posted by Eldonko

I would rather see a canadian benching database myself, PI times, 3d, bandwidth, max RAM clocks, etc, but I guess hwbot runs that already..

I would contribute either way though, although I hate running stability tests for that long. Why not 30 min-1 hour OCCT, stress CPU and RAM? It detects errors faster than prime or orthos usually. Tests other that that could be bonus, as long as there is a minimum requirement set.

Damn, i would of expected that a good 8hours at least was what you would of recommended Eldonko lol

Oh well, whatever ppls likes, i'll post my results anyway!

[Eldonko]

Im sure I stability tested my 24/7 oc for 8 hours orthos at some point, I just dont bother doing SS of stuff like that. If I can SMP fold and run every program needed for 3 months I think it's good. ;)

[Patriote]

Originally Posted by Eldonko

Im sure I stability tested my 24/7 oc for 8 hours orthos at some point, I just dont bother doing SS of stuff like that. If I can SMP fold and run every program needed for 3 months I think it's good. ;)

Yeah, thats true. But i still like to Orthos stress for 12 hours just to make sure it's stable instead of having the comp crash during folding

[3.0charlie]

[off-topic]
you guys are not folding for 54196.
[/off-topic]

I think that with the amount of databases around the 'net, we need to keep ours as simple as possible, both in the testing (I totally forgot about OCCT...) and in the result posting. Speaking of which, could we use Goggle Docs and create a Spreadsheet accessible to members?

[Supergrover]

Originally Posted by Eldonko

Im sure I stability tested my 24/7 oc for 8 hours orthos at some point, I just dont bother doing SS of stuff like that. If I can SMP fold and run every program needed for 3 months I think it's good. ;)

If it folds without throwing units then it's rock solid stable IMO, if you are right at the edge of your overclock and pass prime/orthos 12 hours I have still errored a WU.

Normally I would say 12 hours as I seam to always get a error @ 10-11 hours
Put something together

CPU-Z/GPU-Z validation would be a must, it takes 2 seconds .

I say prime/orthos stable 12 hours, might be better to get the average of everyone's answer

SS of course

If you are going to include the videocard then what are you going to bench the GPU with?

Specs, in the initial rules clearly state what is expected.

Non stable should have it's own section for those do or die runs

[BALISTX]

What about benching the video cards with 3DMark06 graphics tests.

[3.0charlie]

Don't forget there was 1 3Dmark that heavily used the CPU - but I don't remember which one.

[Eldonko]

Originally Posted by Supergrover

If it folds without throwing units then it's rock solid stable IMO, if you are right at the edge of your overclock and pass prime/orthos 12 hours I have still errored a WU.

I agree, SMP is best test of all.

Originally Posted by 3.0charlie

[off-topic]
you guys are not folding for 54196.
[/off-topic]

I'm team captain for 50141 so am unable to contribute to 54196, but its all for the same cause. :)

[BALISTX]

[off-topic]
you guys are not folding for 54196.
[/off-topic]

Since the computer will be my only PC, I will be able to fold when I'm at work.