Folding Forum

Hello to everybody!
First I've to apologize for my bad English...
I've bought a new gpu few days ago , the R9 280x toxic and I've noticed that the gpu stops folding for 45-50 seconds after reached 1% of the wu , is that normal? I had an amd 6870 and it didn't stop at any time ( I guess )
The temperatures are good cpu @ 60-65 C' gpu @ 60

My pc components are these
I7 2700k @stock
Asus p8z68
4x2 gb 1333 corsair
R9 280x toxic
Samsung ssd 128gb OS drive

Welcome to the folding@home support forum eddy_spaghetto.

Unfortunately your thread title doesn't match the body of your post but I can tell it is not unusual for there to be a dip in GPU usage at regular intervals. During those times a checkpoint is being written to disc. The length of time between each checkpoint is set by the project leader for core_17 work. Previous cores used a client setting to determine when to write a checkpoint.

Try the 14.x cat driver if not using it already. It performs better. Be sure to let us know the driver version and the project number in your next post.

Thanks for the answers ( I've changed the thread title ) currently i'm using 13.251 , project 9401 fahcore 0x17.
I tried to change checkpoint frequency with 30 min but it didn't change anything.

The Core_17 checkpoint frequency is determined by the settings put into the WU you download, so changing it in FAHControl will not have any effect. The frequency setting in FAHControl will affect the checkpoints of the CPU cores A3, A4 and A5.

Thank for the info , I'm a real newbie... I've installed the newer beta 14.3 catalyst and there's the same dip...

That is as expected. The checkpoints and validation of the data are still done periodically, and that uses your CPU. So the GPU will be not be actively folding for a brief period during that period. The performance improvement with the later AMD drivers applies during the actual GPU calculations.

Thanks for helping me , right now I'm earning more than 200k ppd. The dip appears less often with project 13001 , about 18 min .

That's is ok, as mentioned above that is different by each project.
I do (for fun) long term monitoring of temperature of my GPU and can see on the graph what WU is running. Each checkpoint gives the GPU enough time to cool down a few degrees. Gives each project an own characteristics.

Oh, and welcome to the world of folding.