[QUOTE=Madpoo;419332]Methinks Manfred4 might want to have a go at these triple-check lists I put out. That way if he's running tests on a machine that's a little iffy, these lists have two mismatched tests already and the odds are strong that one of them is already correct, so he'll have something to match.[/QUOTE]

I would do that, if that machine still was running prime. It is not though. I am only participating with my one computer, since I get free electricity here.
And i know, that this computer here ist rock solid, there was no mismatch at all from this computer and i am running many DCs as well (and that one doublechecked 100M exponent)

[QUOTE=chalsall;419343]or use the manual assignment page on Primenet.[/QUOTE]

That doesn't work for cat 1/2/3 exponents.

[QUOTE=Mark Rose;419350]That doesn't work for cat 1/2/3 exponents.[/QUOTE]

Oh... Thanks. I didn't think of that.

So using a slaved Prime95/mprime instance (probably already "trusted") is the way to do this.

[QUOTE=chalsall;419354]Oh... Thanks. I didn't think of that.

So using a slaved Prime95/mprime instance (probably already "trusted") is the way to do this.[/QUOTE]

Yeah, quite a few of the ones I list are in the lower ranges (DC's below 39M) so they're not available for manual reservation. If it's higher though, feel free to try and manually reserve specific things.

I think this page gives you the current list of cat 1/2/3 ranges:
[URL="http://www.mersenne.org/thresholds/"]http://www.mersenne.org/thresholds/[/URL]

That certainly explains the current high level of "churn" for exponents > 39M since they get picked up and abandoned by anonymous/new users.

[QUOTE=Mark Rose;419350]That doesn't work for cat 1/2/3 exponents.[/QUOTE]

Well, cat 3 assignments [b]can[/b] be obtained via the manual assignments page as long as you've selected to get low assignments.

[QUOTE=kladner;419124]I will take [CODE]DoubleCheck=36840007,71,1[/CODE]Registered.[/QUOTE]
Matched the only other completion. :smile: This reassures me about my current settings for the MSI GTX 580 Lightning.

One major test is running an entire DC with the VRAM at BIOS default speed of 2100 MHz. It was also run at the MSI stock OC of 833 MHz. It runs way cooler than when it's slamming out LLTF at over 900 MHz. EDIT: And this with the GPU core jacked up to 1.075 v, which can run TF at 900 MHz or more. This is almost certainly overkill, though it won't do CuLu at stock voltage. I just hate to poke at things when they are working with no warning signs.

I actually think that it would run LL successfully at some of the higher frequency steps. Though there are steps between, I mostly jump to 861, 872, 882, 891, 900, 911(!), 918, 924, selecting on the basis of temperature while running TF. I have gone higher, but voltage requirements mount up, as does heat. The above-mentioned 580, which has outstanding air cooling, is running 918 MHz, 1.075 v, 77C. This is my usual max.

At 900 MHz, the card puts out ~498 GHz-d/d. I can't track power changes. I decided to stop using the Killawatt when it started looking brown, and a bit melted around the outlet slots:max:

This would have been up to 850 w 24/7 consumption with a 125w CPU, two 580s in various states of OC. In addition, there are six fans- some pretty hungry, 5 spinners, 1 SSD.

I guess some of this is a bit OT, though maybe with some logical connections to the whole task of running D-T-Q-Q....checks

[QUOTE=kladner;419381]

At 900 MHz, the card puts out ~498 GHz-d/d. I can't track power changes. I decided to stop using the Killawatt when it started looking brown, and a bit melted around the outlet slots:max:
[/QUOTE]
So a kilowatt kills a killawatt?:buddy:

[QUOTE=VBCurtis;419391]So a kilowatt kills a killawatt?:buddy:[/QUOTE]
Looks that way. :smile:

[QUOTE=VBCurtis;419391]So a kilowatt kills a killawatt?:buddy:[/QUOTE]

I just checked their product page and it says it's rated to 15A @ 115V. 850W isn't even half of that... good grief, that shouldn't be a problem. Sigh...

[QUOTE=Madpoo;417532]Here's a batch...
[CODE]exponent Bad Good Unk Sus Solo Mis worktodo
54394937 2 1 67 2 66 3 DoubleCheck=54394937,73,1
54398093 2 1 67 2 66 3 DoubleCheck=54398093,73,1
[/CODE][/QUOTE]

I've taken these 2 from post #577.

I took these:
[CODE]44062201 DoubleCheck=44062201,72,1
44463481 DoubleCheck=44463481,72,1
45089507 DoubleCheck=45089507,72,1
45551903 DoubleCheck=45551903,72,1
45713539 DoubleCheck=45713539,72,1
46141541 DoubleCheck=46141541,72,1
46203917 DoubleCheck=46203917,73,1
47000207 DoubleCheck=47000207,72,1
[/CODE]