SETI.Germany invite Team Crunching @ EVGA to the SETI@Home Wow!-Event 2019. It takes place from 15th August, 16.00 UTC, until 29th August 2019, 16.00 UTC.

We cordially ask all team leaders to forward the invitation to their team. Information about the Wow!-Race 2019 and a registration form can be found here:  https://www.seti-germany.de/Wow/register/en .

SETI.Germany wishes you and your team members much fun with the Wow!-Event 2019.

Best regards
Terminator
SETI.Germany

Done here as wellLeo

Hi,thank you for register and welcome to the SETI@Home Wow!-Event 2019.
 
SETI.Germany invite Team Crunching @ EVGA to the SETI@Home Wow!-Event 2019.Everyone is welcome to join in the Event.Register here:
 
https://www.seti-germany.de/Wow/register
 
Best regards
Terminator
SETI.Germany

 

Thanks for Posting the Invite

ipkha
Is there a trick to get GPU tasks? I'm not getting any. Just CPU tasks so far.

If using BAM! make sure you are allowing GPU.

On your Project Preference and set to only use GPU.
 
But the site is Down ATM
Notice: Trying to get property of non-object in /disks/carolyn/b/home/boincadm/projects/sah/html/inc/user.inc on line 36
Notice: Trying to get property of non-object in /disks/carolyn/b/home/boincadm/projects/sah/html/inc/user.inc on line 36
 
Standing Outage Notice: Every Tuesday morning (Pacific Time)
we havea regular 3-4 hour outage for database backup (and other system maintenance).
Most servers and web site functions will be off during these outages.        
[As of 13 Aug 2019, 16:00:05 UTC]

Project down for maintenance

SETI@home is temporarily shut down for maintenance.  Please try again later.
 
app_config.xml
<app_config>
   <app>
     <name>setiathome_v8</name>
     <gpu_versions>
         <gpu_usage>1.0</gpu_usage>
         <cpu_usage>0.1</cpu_usage>
     </gpu_versions>
   </app>
   <app>
     <name>astropulse_v7</name>
     <gpu_versions>
         <gpu_usage>1.0</gpu_usage>
         <cpu_usage>0.1</cpu_usage>
     </gpu_versions>
   </app>
</app_config>
 
Or 2 Tasks Per GPU
 
<app_config>
   <app>
     <name>setiathome_v8</name>
     <gpu_versions>
         <gpu_usage>0.5</gpu_usage>
         <cpu_usage>0.1</cpu_usage>
     </gpu_versions>
   </app>
   <app>
     <name>astropulse_v7</name>
     <gpu_versions>
         <gpu_usage>0.5</gpu_usage>
         <cpu_usage>0.1</cpu_usage>
     </gpu_versions>
   </app>
</app_config>

You have to run one task and write the time down then run 2 Tasks and compare them.
The 2080 runs them Fast less than 3 Minutes running 1 Tasks Per GPU.
Running 2 Tasks Per GPU can take from 6 to 8 minutes depending on the Tasks.
But it is best to test this on your host to see what will give you the Higher Return.

The home page is now a forced video

bill1024
It started, good luck everyone. Lets find some Martians!!!
No more video.
Everyone running the correct optimized apps for your GPU type and OS?  

Yeah there server was down this AM, but it has been back up for a while now, since noon or so.
I have this in my CC_config file so it reports tasks right away. They have it set to limit your reports to 5 min apart
This way you do not have a days worth stuck on your box. Works on my Lenux boxes too.
<cc_config>
<options>
<report_results_immediately>1</report_results_immediately>
<use_all_gpus>1</use_all_gpus>
<http_1_0>1</http_1_0>
</options>
</cc_config
 
EDIT go in the BOINC mgr and tell it to update Seti and see what happens.
May hold you off for five min or so.

I heard if you restart a task it has a good chance of being inconclusive, invalid or giving bad data.
Also said "to set the request checkpoint to longer then the task takes in the options section."
Default is 60 seconds, so I set mine to 777 just to be safe.
If you restart the client it will start back at 0% or go corrupt right away.
That's ok by me, not running CPU and tasks do not take long and I don't have to restart BOINC that often.

 
Add  <max_concurrent>3</max_concurrent> or if running 2 Tasks Per GPU  <max_concurrent>6</max_concurrent>
What is the % you have the CPU set to, I am not running CPU and have mine set to 90% 
 
app_config.xml
<app_config>
   <app>
     <name>setiathome_v8</name>
   <max_concurrent>3</max_concurrent>
     <gpu_versions>
         <gpu_usage>1.0</gpu_usage>
         <cpu_usage>0.1</cpu_usage>
     </gpu_versions>
   </app>
   <app>
     <name>astropulse_v7</name>
     <gpu_versions>
         <gpu_usage>1.0</gpu_usage>
         <cpu_usage>0.1</cpu_usage>
     </gpu_versions>
   </app>
</app_config>
 
Some set another file to use all GPU's but do not recall the name. Looking it up now.
cc_config.xml
Still Looking....
<use_all_gpus>0|1</use_all_gpus>
 
If 1, use all GPUs (otherwise only the most capable ones are used). Requires a client restart.
 
I have never had to use this before even with 4 GPUs in one Host

This is my cc_config.xml
<cc_config>
<options>
<report_results_immediately>1</report_results_immediately>
<use_all_gpus>1</use_all_gpus>
<http_1_0>1</http_1_0>
</options>
</cc_config
 
 
it tells BIONC to use all the GPUs, Also for when using two different types of GPUs   660Ti & 780Ti...
 
It also has a line to have BOINC report the tasks as soon as they are done.
That helps with PrimeGrid so finished tasks do not sit there waiting to be returned. First one in is the prime finder
Seti right now after uploading will set it to 5min delay everytime it uploads so not to overload their server.
 
It also tells BOINC to report using http, otherwise it uses some other protocol 
I had an issue with a couple projects where my tasks would not upload to the server.
Was odd cause not every system on the same router had the issue. But I use this cc_config on all my sys. anyway.
 
 

Sweet

I switched to linux where it picks up all 3 gpus.