Ok thanks for the help.  I had stopped the Tracker 6701 WU when it was showing 1% completion.  I then restarted my previous 6701 WU that was created by the FAH-630 client (32% complete).  Figured I'd better finish one at a time in order of when they were started. 
 
My original 6701 (via FAH-630) will finish in about 2hrs.  Is the 6701 that the Tracker started still out there and can just pick up where it left off, or did I kill it permanently when I stopped it via the Tracker?  If it is still out there waiting to be finished, I'll hop on it once the current 6701 finishes.
 
Thanks,
 
Rick

Ok thanks a bunch for the help.  I prefer to finish any Folding that I start, so I didn't want to leave anything incomplete.
 
Rick

shogon

I just thought of this but not sure if it works, could the GPU tracker show the WU project, % complete, and PPD on the g15/19 screen? or something of the sort. 

Just wondering.

 
I don't know how to interface with the G15/G19 screen at the moment, but I like the overall idea. Ill see what I can do with this.

3.33 Stable has been released.
 
Important changes:
1. Added -nocpudetect flag that unlocks the SMP and -bigadv options in the config if the CPU detection isn't working
2. The Games.txt list included with the Tracker has been updated with the latest Nvidia driver game list dump (260.99)
3. An updater has been added for the Games.txt list (manual only)
4. Added TPF to the XML file output for use with 3rd party software

Also chiming in to say thank you jedi95, for making the set up so easy for a noob.
My rig has been folding for the cause for about 26 hours now.

The sigsend.dll should be located in the main folder where you installed the Tracker program.  Have you tried to uninstall/reinstall the Tracker?

Can't hurt to try right?

Unless you set the -one unit flag,  it'll just keep going.

NP.  Nice PPD for a 450 for sure!

n635dc

I am unclear on something regarding the tracker software. I can see checkboxes for Advmethods and for bigadv but how can I flag for big work units as I did when I was in the standford standalone client?

 
Big WU size is the default option. If you want to change it the only way is to edit the client.cfg files manually.

jedi95,
Reference that issue I'd brought to your attention earlier regarding HeatControlSettings.txt being deleted from the GPU0 folder. I've been able to recreate this anomaly by checking OneUnitMode for GPU0 while processing a WU on that core. The file was deleted after the WU completed. (I discovered this when setting OneUnitMode for GPU0 and GPU1 so that a SMP bigadv could process faster after the GPU WUs that were being processed finished.) I've replicated this anomaly two more times this morning. I did not experience this behavior with GPU1.
Chris

Evil_Monkey_153

jedi95,
Reference that issue I'd brought to your attention earlier regarding HeatControlSettings.txt being deleted from the GPU0 folder. I've been able to recreate this anomaly by checking OneUnitMode for GPU0 while processing a WU on that core. The file was deleted after the WU completed. (I discovered this when setting OneUnitMode for GPU0 and GPU1 so that a SMP bigadv could process faster after the GPU WUs that were being processed finished.) I've replicated this anomaly two more times this morning. I did not experience this behavior with GPU1.
Chris

 
I was finally able to reproduce this bug after trying lots of different scenarios. It only happens if you have Heat Control enabled, set on GPU0, and then you apply config settings without selecting a GPU in the Heat Control settings tab. I have added a check when applying Heat Control settings so that it won't apply if no GPU is selected.
 
This fix will be released with 3.34 later today.

jedi95

rjbelans

I have a suggestion jedi. With the very demanding nature of some of the newest GPU3 units, I would like to be able to disable -advmethods on my GT430. I cannot seem to find it anywhere, would you be able to add this functionality?

Right now if you want to do -advmethods on some of your GPUs but not all of them, your only choice is to disable -advmethods in the Tracker's config. After that, you will need to edit the client.cfg files for the GPUs you want to run advmethods and add the flag in the extra_params= section.

Adding this to the Tracker would relatively straightforward, but also time consuming.

I thought that -advmethods in the tracker was only for CPU, I guess it's for both? That makes it really odd then, as I don't have advmethods selected on that computer. However, the GPU was running the new P6800 units (doing a P10958 right now).

Thank you so much Jedi, now I can have the smp/480 with adv.  flags while the 450 isn't :p

rjbelans

Thank you jedi, that was pretty quick.

 
By time consuming what I mean is that >90% of the time adding the feature is spent doing repetitive, simple tasks.