Additional to all...

Because of VGA card driver uninstall and install... : Message 1733115.

The newest DDU you could get here (at the developer website): http://www.wagnardmobile.com/forums

---

It was a long way to find the best way... ;-)

- - - - - - - - - -
BTW.
I tested Crimson 16.3.2 for around 3 1/2 days, the result:
SETI results are OK.
AstroPulse results go invalid.

I'm back to Crimson 15.12.
- - - - - - - - - -

Like Mike mentioned already, SETI r3330 ATI app don't work well with default cpu_lock (at least at my system). All 4 GPU apps were fixed at Core#0.
(This could be the reason that your Fury Nano's have ~ 55 % GPU Load each.)
I need to use -no_cpu_lock in cmdline.txt file (more slightly further down).

You use the stock SETI@home apps.
You should/could go with opti apps, at least higher RAC for the CPUs (and just the correct/best/fastest GPU apps installed/used).
Example here (in the middle of the site). If you decided to go with opti apps, we will help you.


If you would like to go still with stock apps... :

(I tell you what I would do (if I would go with stock apps))

You should make an app_config.xml (on the bottom of the site) file with:

<app_config>
  <app>
    <name>setiathome_v7</name>
  <gpu_versions>
    <gpu_usage>1</gpu_usage>
    <cpu_usage>1</cpu_usage>
  </gpu_versions>
  </app>
  <app>
    <name>setiathome_v8</name>
  <gpu_versions>
    <gpu_usage>1</gpu_usage>
    <cpu_usage>1</cpu_usage>
  </gpu_versions>
  </app>
  <app>
    <name>astropulse_v7</name>
  <gpu_versions>
    <gpu_usage>1</gpu_usage>
    <cpu_usage>1</cpu_usage>
  </gpu_versions>
  </app>
</app_config>

(I don't think there are SETIv7 tasks still around, but it's still possible to check it in the project prefs, so I included it.)

This make: 1 CPU-thread reserved for 1 GPU app (on this CPU-thread no CPU app/task (theoretically)).

My ap_cmdline_win_x86_SSE2_OpenCL_ATI.txt file in my setiathome.berkeley.edu project folder have inside:
-cpu_lock -instances_per_device 1 -unroll 18 -ffa_block 2830 -ffa_block_fetch 2830 -tune 1 64 4 1 -tune 2 64 4 1 -oclFFT_plan 256 16 256 -hp

My mb_cmdline_win_x86_SSE2_OpenCL_ATi_HD5.txt file in my setiathome.berkeley.edu project folder have inside:
-instances_per_device 1 -no_cpu_lock -sbs 512 -period_iterations_num 20 -spike_fft_thresh 4096 -tune 1 64 1 4 -oclfft_tune_gr 256 -oclfft_tune_lr 16 -oclfft_tune_wg 256 -oclfft_tune_ls 512 -oclfft_tune_bn 64 -oclfft_tune_cw 64 -hp

(With stock apps, the .TXT files could have the same name.
But you could have more (a few) .TXT files for SETI (MB) apps.)

This cmdline settings should go with your FuryX, could be tested on your Fury Nano's.

That worked! Thanks Dirk and Mike for getting me on the right track. I had the app_config part correct, but that left Core0 fully loaded and looks like it was a bottleneck for the GPUs. I tried the command line modification you suggested and it works perfectly for the Dual Fury plus 1 R7870 system. Now I can see the system is thermally throttled, so I won't get full performance until I get waterblocks on these cards. I hesitated placing the order, since they were definitely not needed if the cards were not fully loaded. I posted a video of the results on my channel.

The optimized apps sound interesting, but I will hold off until I finish the system. In addition to waterblocks on the Nanos, I want to implement a by core OC profile. It will take a few weeks for me to get caught up.

---

GitHub: Ricks-Lab
Instagram: ricks_labs

You use the stock SETI@home apps.
You should/could go with opti apps, at least higher RAC for the CPUs (and just the correct/best/fastest GPU apps installed/used).
Example here (in the middle of the site). If you decided to go with opti apps, we will help you.

I just loaded Lunatics apps and I am having some problems. First is that the estimated time for completion of all tasks increased dramatically. From 15min to over an hour for GPU tasks and from 3 hours to 13 hours for CPU tasks. Also, the GPU underloading is back with core 0 being the bottleneck. Is the fix different than with standard apps?

---

GitHub: Ricks-Lab
Instagram: ricks_labs

I just loaded Lunatics apps and I am having some problems. First is that the estimated time for completion of all tasks increased dramatically. From 15min to over an hour for GPU tasks and from 3 hours to 13 hours for CPU tasks. Also, the GPU underloading is back with core 0 being the bottleneck. Is the fix different than with standard apps?

I used the same command line modifications recommended for standard apps, and the GPU loading issue is solved. Seems like GPU loading is higher than with standard apps. I reduced number of CPUs per task from 2 to 1.5 and loading GPU loading is still good.

---

GitHub: Ricks-Lab
Instagram: ricks_labs

I used the same command line modifications recommended for standard apps, and the GPU loading issue is solved. Seems like GPU loading is higher than with standard apps. I reduced number of CPUs per task from 2 to 1.5 and loading GPU loading is still good.

It looks like this system is not getting new tasks. I just did an Update project and only 1 astropulse task downloaded.

---

GitHub: Ricks-Lab
Instagram: ricks_labs

It looks like this system is not getting new tasks. I just did an Update project and only 1 astropulse task downloaded.

Went home for lunch and found the system nearly out of work and no new tasks downloaded. I downloaded a new installation package from Mike's site and this time installed it while BOINC was running (first time I had exited BOINC) and new work started downloading even before the installation was complete.

---

GitHub: Ricks-Lab
Instagram: ricks_labs

Ap's are not working right. I have 16.4.2 installed. I aborted this wonky ap.

---

Ap's are not working right. I have 16.4.2 installed. I aborted this wonky ap.

With 16.3.2 being rubbish on my system. I'll wait for the next released driver before giving another 16.x driver a chance.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

Ap's are not working right. I have 16.4.2 installed. I aborted this wonky ap.

That one has nothing to do with GPU drivers - it was a bog-standard stock CPU run

http://setiathome.berkeley.edu/result.php?resultid=4910350852

Looks like something damaged the checkpoint file and it restarted from the beginning - we saw that in the early days when Josh was still writing the app, but I haven't seen one like it for several years. Probably made sense to abort it and move on.

Just installed the latest from AMD - Crimson 16.5.2

The issues seen on Fury seems to be substantially improved! I saw an immediate 10% increase in loading on my Fury X. Loading on my Nano cards also increased. Since I have them in the same cooling loop as the CPU, the CPU started to get too hot! I am reworking the CPU OC profile now in order to reduce the thermal load.

I am not sure if the issue of running 2 tasks at a time on the Fury is fixed. I tried it with no errors, but tasks were taking much more than twice as long. Maybe this was just the luck of suddenly getting more challenging tasks.

---

GitHub: Ricks-Lab
Instagram: ricks_labs

Ap's are not working right. I have 16.4.2 installed. I aborted this wonky ap.

That one has nothing to do with GPU drivers - it was a bog-standard stock CPU run

http://setiathome.berkeley.edu/result.php?resultid=4910350852

Looks like something damaged the checkpoint file and it restarted from the beginning - we saw that in the early days when Josh was still writing the app, but I haven't seen one like it for several years. Probably made sense to abort it and move on.

Upgraded to 16.5.2. My desktop is acting friendly again. Had some issues with another app. Now seems to be working properly. Hope it stays that way.

---