OK - I think my appconfig.xml file screwed over my initial work on the new Prometheus, Computer ID: 7440554. I initially received 8 to 10 WUs and ALL immediately received Compute Errors.

First thing I did before releasing BOINC to crunch, (BOINC 7.6.22), was to Install Lunatics 45 Beta 6. CPU was set for AVX, NVIDIA GPU was set to SOG. I then set BOINC to Allow New Tasks and performed an Update. I received the 8 to 10 Tasks, and with NO other changes to BOINC, I released BOINC to crunch. ALL WUs failed with Compute Error.

I Suspended BOINC. Then Exited BOINC and removed the appconfig.xml File. For good measure, I Upgraded BOINC to 7.12.1 64 Bit, (as the system IS 64 Bit), and Relaunched BOINC. I set to Allow New Tasks again, and NOW am crunching ONE Unit at a time on the GTX-1050.

For some UNKNOWN reason, BOINC is IGNORING the Primary GPU - a GTX-760. Both cards are EVGA and both cards have 2GB GDDR5 VRAM. I would like to be able to crunch on both cards, if possible.

Here is the Original appconfig.xml that was in Prometheus when it had ONE EVGA GTX-750TI SC.

[appconfig.xml]

<app_config>
<app>
<name>astropulse_v7</name>
<max_concurrent>2</max_concurrent>
<gpu_versions>
<gpu_usage>.5</gpu_usage>
<cpu_usage>.5</cpu_usage>
</gpu_versions>
</app>
<app>
<name>setiathome_v8</name>
<max_concurrent>2</max_concurrent>
<gpu_versions>
<gpu_usage>0.50</gpu_usage>
<cpu_usage>0.04</cpu_usage>
</gpu_versions>
</app>
</app_config>


------------------------------------------------------

What should a Proper appconfig.xml have set for this new hardware?

CPU: Intel i7 7700K at 4.2GHz

32GB DDR4 Corsair Vengeance RAM

Gigabyte GA-Z270-HD3 MOBO

Corsair HX7501 Platinum 750 Watt PSU

One Internal Western Digital Black 1TB Hard Drive.

NVIDIA Driver 388.13 - Which, when I checked, is supposed to work for both the 760 and the 1050... (Performed Clean Install of Driver. Old Driver was 353.30.)

With BOINC 7.12.1 now installed and running, and the appconfig.xml pulled, Lunatics 45 Beta 6 is NO LONGER crashing Units. However; I'd really like an appropriate appconfig.xml File to make use of both GPUs.

Any and all help greatly appreciated.


TL

---

TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join Calm Chaos

Need an cc_config.xml for the use all gpus command

<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
</options>
</cc_config>

Will have to look at your errors to see if they show anything

Edit..

Looks like you had 1 work unit trying to build the kernal then immediately 18 erred out and now you have 4 after those that are pending. Check your antivirus and see how the rest of the work units fair.

Just re-read your comments. I'll give you my app_config and you can play with it

<app_config>
  <app_version>
    <app_name>setiathome_v8</app_name>
    <plan_class>opencl_nvidia_SoG</plan_class>
    <avg_ncpus>1</avg_ncpus>
    <ngpus>0.5</ngpus>
    <cmdline></cmdline>
       <max_concurrent>2</max_concurrent>
  </app_version>
   <app_version>
    <app_name>astropulse_v7</app_name>
    <plan_class>opencl_nvidia_100</plan_class>
    <avg_ncpus>1</avg_ncpus>
    <ngpus>0.5</ngpus>
    <cmdline></cmdline>
      <max_concurrent>2</max_concurrent>
  </app_version>
</app_config>

---

Need an cc_config.xml for the use all gpus command

<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
</options>
</cc_config>

Will have to look at your errors to see if they show anything

Edit..

Looks like you had 1 work unit trying to build the kernal then immediately 18 erred out and now you have 4 after those that are pending. Check your antivirus and see how the rest of the work units fair.

Just re-read your comments. I'll give you my app_config and you can play with it

<app_config>
  <app_version>
    <app_name>setiathome_v8</app_name>
    <plan_class>opencl_nvidia_SoG</plan_class>
    <avg_ncpus>1</avg_ncpus>
    <ngpus>0.5</ngpus>
    <cmdline></cmdline>
       <max_concurrent>2</max_concurrent>
  </app_version>
   <app_version>
    <app_name>astropulse_v7</app_name>
    <plan_class>opencl_nvidia_100</plan_class>
    <avg_ncpus>1</avg_ncpus>
    <ngpus>0.5</ngpus>
    <cmdline></cmdline>
      <max_concurrent>2</max_concurrent>
  </app_version>
</app_config>

Thanks, Zalster. :-)

I've copied and created the cc_config.xml and saved to the Desktop. I've also copied your app_config.xml and saved that over my original to replace the contents I had therein. I've got Prometheus set to NNT so that I don't accidentally create ghosts should something go wrong. When the queue is finished, where do I move the cc_config.xml file to??? I've never used one before. (I know where the app_config.xml file goes...)

As to AV, I got rid of PCMatic and am just using Windows Defender and Malwarebytes. If things go well, I may save up and install ESET on Prometheus, it does well for Macs, so I anticipate it should also do well for PCs.


TL

---

TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join Calm Chaos

cc_config.xml goes in the BOINC folder 2 levels up from the Seti@home folder.

computer-->local disc--->ProgramData--->BOINC

---

cc_config.xml goes in the BOINC folder 2 levels up from the Seti@home folder.

computer-->local disc--->ProgramData--->BOINC

Uh, on a hackintosh shouldn't that be: Macintosh HD/Library/Application Support/BOINC Data
It also will have to have owner/group name and permissions set.

Pretty sure he is talking about his dad's window computer and not the Mac Pro computer

---

1.) Hackintosh-Andromeda officially died 7-15-2018... Hackintosh-Andromeda WAS a Dual Boot System WITH Original Prometheus, Win7 Pro x64.

2.) New Andromeda is the reassigned Daedalus, the MacPro 5,1.

3.) New Prometheus is NOW an i7 7700K System with new MOBO, new RAM, and new PSU.

4.) Farragut STILL is my dad's i7 7700 System, (Note: NO "K" on CPU), and ONLY crunches occasionally.

------------------------------------------------------------------------------------------

[Update:]

Prometheus has now been Updated... I successfully removed the first EVGA GTX-1050 from New Andromeda and replaced that with the Hackintosh-Andromeda's ONE good EVGA GTX-750TI SC. New Andromeda accepted the changeover and is crunching away on the 750 card. I then removed the EVGA GTX-760 from New Prometheus and replaced that with the first EVGA GTX-1050. New Prometheus, (for some reason), required a Reinstall of NVIDIA Driver 388.13. I performed another Clean Install of said Driver.

Upon reboot from Reinstalling Driver 388.13 onto New Prometheus, I Updated BOINC and NOW both 1050 GPUs are recognized WITHOUT cc_config.xml, or app_config.xml files. BOINC immediately pulled down new Work Unit queue and began crunching upon setting to Resume.

[Spare GPUs now on hand:]

One EVGA GTX-760 with Extended Warranty through 2024

One EVGA GTX-750TI SC with bad Video Pipeline, BUT still can crunch


TL

---

TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join Calm Chaos

[Active Crunching Systems:]

New Andromeda, MacPro 5,1 - ID: 7952666

New Prometheus, i7 7700K - ID: 7440554

Farragut, i7 7700 - ID: 7066606


[Retired IDs:]

Daedalus, MacPro 5,1 - ID: 8508902

Exeter, AMD A6-6400K - ID: 7301566

---------------------------------------------------------------

I hope this helps to straighten out where my systems are now at.

Thanks again, everyone, for all the help. Especially TBar for the MacPro side, and Zalster for the new Win7 Pro x64 side.


TL

---

TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join Calm Chaos

[Update:]

In the last half hour, I changed Location from Home to Work to pick up CPU Units on Prometheus. Now have four CPU Units crunching away along with two GPU Units - one GPU Unit on each GTX-1050. :-)


TL

---

TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join Calm Chaos

The i7 7700K crunching CPU Units seems to be doing VERY well. It's completing CPU Units in 36 to 37.5 Min. Per WU. :-)

The 1050s are crunching Units between 10 and 10.5 Min. Per WU.

Lunatics 45 Beta 6 seems to be working as it should. :-) Can't wait to get a few APs on CPU and GPU.


TL

---

TimeLord04
Have TARDIS, will travel...
Come along K-9!
Join Calm Chaos