Getting a new laptop with a 730M. Anyone used one before? Wondering how long it will take to do a WU.

---

All tasks have 5 day deadline, with bonus for 24hr and 48hr completion.

GK208 card is similar to my (2)GK107/GDDR5/128bit memory bus that currently (Noelia) completes, depending on type Long tasks- in ~40hr, while some are 24-36hr. Current short Noelia tasks finish in 17hr.
Older Short Nathan's finished in 6hr.
Each Long and Short tasks have a different amount of steps. Time (ms) for each step varies depending upon work unit.

Perfect. Thanks for the info. I will post times once I get the laptop.

Well, it took 48 hours to run through one WU. The time doesn't bother me but the fan noise and heat on the laptop was too much. I let it finish but stopped crunching after one WU. Just not worth burning the laptop out.

Just not worth burning the laptop out.

Yes, that's how it ends with pretty much any laptop GPU. Or with a burnt laptop ;)

Kidding aside, some people let laptop GPUs crunch for longer times, but almost all of those devices were not made for sustained load and are IMO too expensive to risk like this.

MrS

---

Scanning for our furry friends since Jan 2002

Just not worth burning the laptop out.

Yes, that's how it ends with pretty much any laptop GPU. Or with a burnt laptop ;)

Kidding aside, some people let laptop GPUs crunch for longer times, but almost all of those devices were not made for sustained load and are IMO too expensive to risk like this.

MrS

Custom VBIOS/BIOS help keep temps in check by opening more options than a OEM BIOS/VBIOS setting(s).

No custom BIOS here. This is a work laptop. I have a 660ti that is crunching 24/7. That is good enough.

Crunching on GT740M since March. No problems yet, except some over heating. I hope there won't be any problems in the near future. :)
Or if happens, I hope it happens somewhere near Nvidia GPU and I can continue using the Intel GPU. :)
I crunch on the GPU but prefer crunching without no other CPU job. I changed my thermal paste and it runs about 5 - 10 degrees cooler. GPU runs at turbo clock 1058MHz until it reaches 95 degrees and starts to decrease its clock then. However, this is a temperature I can get only with Furmark/Gputest/MadShaders running. Gpugrid doesn't heat it more than 80 degrees, and that's only in hot days.
Other than GPU, GPU VRM chips and inductor coils get really hot, too. These can cause problems before the GPU if they don't have some cooling. Mine doesn't have a cooling but I plan something using a thermal adhesive.

I looks like most of your WUs are ending in error:

process exited with code 212 (0xd4, -44)

Anyone know what the above error means?

I had 1 machine start throwing this error a couple days ago after rebooting it:

(unknown error) - exit code -44 (0xffffffd4)

Reloaded/updated NV drivers, etc. Finally reset BIOS and noticed time was off. Reset time again and it started working, has been fine since then.

I needed to update the driver to run newer cuda6.5 tasks. Errors was caused by old cuda42 enabled 331.38 drivers, trying to crunch cuda 6.5 tasks. Now, I ran 343.22 and finished the first task too.