mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   GPU to 72 (https://www.mersenneforum.org/forumdisplay.php?f=95)
-   -   GPU to 72 status... (https://www.mersenneforum.org/showthread.php?t=16263)

flashjh 2012-01-13 03:57

[QUOTE=flashjh;283531]I'll be using my [URL="http://www.mersenneforum.org/showpost.php?p=282106&postcount=252"]Bulldozer[/URL] system that can't run GPUs for P-1. It's been running ~12 P-1s a day, so I can pickup most of your slack. I have it on 2 LLs also, so if the P-1 cache starts to fill up, let me know and I'll transfer the LLs to another system or add a P-1 or two to my Phenom system.

Jerry[/QUOTE]

My P-1 system is down for bad memory (multiple system resets per day, sometimes every 45 minutes). With 16 DIMMs to test, it's going to take a day or two to figure out which one(s) when bad. Hopefully it's not the MB.

kladner 2012-01-13 04:10

[QUOTE=flashjh;286107]My P-1 system is down for bad memory (multiple system resets per day, sometimes every 45 minutes). With 16 DIMMs to test, it's going to take a day or two to figure out which one(s) when bad. Hopefully it's not the MB.[/QUOTE]

Ouch! That's very disappointing. I hope you find the culprit quickly (and can RMA it.)

flashjh 2012-01-13 05:10

[QUOTE=Dubslow;285881]And welcome to the team Flash![/QUOTE]

Thanks
[QUOTE=kladner;286112]Ouch! That's very disappointing. I hope you find the culprit quickly (and can RMA it.)[/QUOTE]

If it's the memory, I'm still ok to RMA. If it's the MB I have to see what ASUS will do. It has a warranty, but I've never dealt with ASUS before.

bcp19 2012-01-13 05:59

[QUOTE=flashjh;286123]Thanks


If it's the memory, I'm still ok to RMA. If it's the MB I have to see what ASUS will do. It has a warranty, but I've never dealt with ASUS before.[/QUOTE]

Which ASUS board do you have?

flashjh 2012-01-13 06:06

[QUOTE=bcp19;286129]Which ASUS board do you have?[/QUOTE]

It's the [URL="http://www.newegg.com/Product/Product.aspx?Item=N82E16813131643"]KGPE-D16[/URL]. Awesome board for dual G34, but no overclocking at all (which is expected, I guess)

LaurV 2012-01-13 07:00

1 Attachment(s)
Looking occasionally to the lifetime top500, I just found out that I am better even then the ghosts... :P

[ATTACH]7554[/ATTACH]

KyleAskine 2012-01-13 13:31

[QUOTE=bcp19;286078]My 332M TF to ^80 just completed, 368 GHzD credit. So I 'lost' around 650 GHz on the 3 levels I did, allowing you to catch up a bit.[/QUOTE]

It's on!

:boxer:

:fight:

LaurV 2012-01-13 16:05

[QUOTE=LaurV;286139][I]Last fiddled with by chalsall on 13 Jan 12 at 20:18 Reason: Removed inappropriate joke. [/I][/QUOTE]
So, you are policeman... or you either have no sense of humor...
After I spent half an hour to write it down... You could say, or send a PM. I don't believe I ever posted inappropriate things. I also don't believe you don't have jokes with policemen or you never heard one... You lost me here.

oswald 2012-01-13 17:43

If you TF to 80, can you split it up?

Say up to 77 on mfaktc copy 1, 78 on mfaktc copy 2, 79 on mfaktc copy 3 and 80 on mfaktc copy 4.

I was wanting to have four cores and four mfaktc threads running on one LLTF number.

Does this make sense?

James Heinrich 2012-01-13 17:52

[QUOTE=oswald;286183]If you TF to 80, can you split it up?
Say up to 77 on mfaktc copy 1, 78 on mfaktc copy 2, 79 on mfaktc copy 3 and 80 on mfaktc copy 4.[/QUOTE]You can, but remember that each bit level takes twice as long as the previous one, so if it takes 1h to do 76-77, it'll take 8 hours for 79-80. What could work is doing (for example) 76-79 on one instance and 79-80 on a second instance; both will take roughly the same amount of effort (for example [url=http://mersenne-aries.sili.net/credit.php?worktype=TF&exponent=3323732333&frombits=76&tobits=79]32.2GHz-days[/url] for 76-79 vs [url=http://mersenne-aries.sili.net/credit.php?worktype=TF&exponent=3323732333&frombits=79&tobits=80]36.8GHz-days[/url] for 79-80). But, of course, if you find a 77-bit factor you will (probably) have wasted (part of) your 79-80 GPU time.

oswald 2012-01-13 17:56

James,
Thanks. I had not thought of wasted time after a found factor. I do believe I will not split the number up between cores.
Thanks,
Roy


All times are UTC. The time now is 22:33.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.