if gromacs has crashed on your overclocked system, post here

looktall · Aug 25, 2002

okie dokie.

i'm trying to figure out if there is a pattern to the overclocked settings that causes the gromacs to crash.

if you've had a gromacs on your overclocked system and it crashed, how was the overclock done? fsb? multiplier? both?

for me it crashed 5 times in a row at random on my celeron @145mhz fsb.

did you lower the overclock and did the gromacs then run ok?
what settings did you use?

if you've had a gromacs on your overclocked system and it didn't crash, what are the settings?

IceMan3928 · Aug 25, 2002

ive run several Gromacs on my XP 1700 @ 1617, default multiplier (11) and 147mhz FSB. None of them have crashed so far.

jbslow · Aug 25, 2002

I've been getting alot of these the last 2 days if my systems were going to crash they would have by now. So far my 1600+'s are fine at 1.67ghz , @1.75ghz and the lone xp @1.91ghz the only thing the higher overclocks did was fold a little faster

. All my system are at low fsb 12.5x133=1.67ghz , 12.5x140=1.75ghz , 12.5x153=1.91ghz

Kendan · Aug 25, 2002

Early on those crashed my linux box twice. I was using an old client and after the second time I updated to the 3.** client and have been fine ever since. That box is an epox 8KHA+ with an XP 1600 not overclocked and running mandrake. I know this probably wont help but you never know so I posted. Good luck and Fold on!!!

muddocktor · Aug 25, 2002

I have my P3 server proc box crash after I had bumped it up to 154 fsb from 149 fsb. We had turned the AC temp way up in the house and the computer room got pretty hot. I'm pretty sure it was heat related because it had just finished a Gromacs WU a few hour before and the case I have it in suxors as far as case ventilation goes. I've since bumped it back to 151 fsb, but it hasn't gotten any more Gromaces WU's since then. Also, I know the crash wasn't caused by the pci bus speed because it's a GA7-OXET board with the pll circuit kicked in to lower the pci and agp bus speeds.

gtsimmo · Aug 25, 2002

i had three repetitive attempts crashed which i think was due to my overclock. 143 x 12.5. It was causing problems with not only Gromacs, but tinkers as well (but it was less likely to happen). Gromacs self destructed, i mean shut down to instability within like 1 or 2 mins. i back the multi down to 12 and it seems to work fine now, but i havent got another De Novo let so i cant know for sure.

hellrazrblade · Aug 25, 2002

i just crashed : "fah core 78 has encountered " blah blah

im running stock mult, 147 fsb, 1.85vcore, 2.85 memcore, 2.5-5-2-2-2 mem timings

what projects are gromacs? i think i was working on p200 when it crashed, maybe not, i wasnt here.

OpenFriday · Aug 25, 2002

Core 78's were crashing @139/140FSB but its at default now and ive got 1 de novo in water done and working on a second so i'd say im looking better.

JetMech · Aug 25, 2002

I'm running a DE NOVO PROTEIN in water on my fastest folder. So far no problems but that's the rub. Everything will be going fine and then suddenly the core dumps the work. Right now I'm getting 6.2min/frame, 10.34hrs./WU. Abit KR7A with 256MB Corsair PC2400 DDR, XP1800+@1633. I must admit though that this is below the highest point (1670) that I have run a core65. I agree with Flounder that it is getting very time consuming keeping the work up. I recently lost my WAN capability on the network (LAN works ok) so I have to keep track of which box is finishing next and move the ISP connection around accordingly. I haven't been doing any heavy troubleshooting yet so be warned. I will be brain picking next weekend. Lost a lot of production last week behind this. Ok I'm through

hellrazrblade · Aug 25, 2002

why would the 78's be harder on a processor than a 65? i mean, cpuburn/prime95/toast dont even make this setup unstable...

overdoze · Aug 25, 2002

I lower 2 rigs down 30MHz and all is fine after a number of crashes. I have both multiplier and FSB OC. Right now my T-bird is at 1356MHz (use to be 1400) and my XP1600 is at 1828MHz (use to be 1850). I have finish 3 gromacs WU so far and no errors, no crashes.

hellrazrblade · Aug 25, 2002

i guess i'll tone this oc back some then... WAAAAA

looktall · Aug 25, 2002

hellrazrblade said:
i guess i'll tone this oc back some then... WAAAAA

if you drop the oc back down a notch and it works (or even if it doesn't work) please post the detail in here.

it seems to me so far that there is not really a pattern. it either happens or it doesn't.

keep them coming guys.

JetMech · Aug 26, 2002

hellrazrblade said:
why would the 78's be harder on a processor than a 65? i mean, cpuburn/prime95/toast dont even make this setup unstable...

We went through the same thing debugging core_65 last winter and now we're moving into the big leagues. This thing we do is not simple and not for the faint at heart. We're a warrior breed and no matter what Stanford throws at us we'll find a work around. :beer:

hellrazrblade · Aug 26, 2002

well, i toned back 1mhz on fsb so im at 146*11
its rock solid as far as i can tell (its been more than a day)

i know, we're hardasses and we can take it!

hellrazrblade · Aug 27, 2002

i think progress quest might be to blame... i have crashed way too many times recently (3x) at different settings.
i was level 38 too!!!!
oh well

if gromacs has crashed on your overclocked system, post here

looktall

Member

IceMan3928

Member

jbslow

Member

Kendan

Senior Punk

muddocktor

Retired

gtsimmo

Member

hellrazrblade

Member

OpenFriday

Senior Member

JetMech

Member

hellrazrblade

Member

overdoze

Member

hellrazrblade

Member

looktall

Member

JetMech

Member

hellrazrblade

Member

hellrazrblade

Member

Similar threads