If you would like to run multiple tasks per GPU, you can do this with an app_config.xml file in your Kessler project directory, the same as you do for any other project.

1. Create a file named "app_config.xml", the app name for this project is simply "kessler"

2. Example contents:

<app_config>
<app>
<name>kessler</name>
   <gpu_versions>
      <cpu_usage>1.0</cpu_usage>
      <gpu_usage>0.5</gpu_usage>
   </gpu_versions>
</app>
</app_config>

the above example is for two (2) tasks per GPU. the gpu_usage value is the inverse of the number of tasks you want to run. 0.5= 2x tasks, 0.33= 3x tasks, 0.25= 4x tasks and so on.

3. place the app_config.xml file in your Kessler@home project directory
- Linux/Ubuntu standard install dir /var/lib/boinc-client/projects/kesslerathome.org
- Linux/Mint standard install dir /var/lib/boinc/projects/kesslerathome.org
- Windows standard install dir C:\ProgramData\BOINC\projects\kesslerathome.org
- or wherever you custom installed BOINC, or maybe a different distro uses a different location.

4. Either re-read config files or restart BOINC client:
- GUI: BOINC Manager Options -> Read config files
- boinccmd CLI tool: boinccmd --read_cc_config (*reads all config files, not just cc_config)
- or restart BOINC

in addition to the app_config method above, I have added the ability to just select this from the website in your Project Preferences

in each venue, you can define how many tasks per GPU you want. up to 10.

the presence of an app_config.xml will override this. so if you have one, remove it, set No New Tasks, and reset the project once your cache is empty. then you can resume work fetch with the web setting if you prefer.

additional note for users running BOINC client versions 8.2.0 - 8.2.8; this function will not work without a project reset to take effect. so you will have to empty your cache and do the project reset to see a change.

for clients before and after this range, the setting should take effect for new tasks downloaded after making the change. existing tasks in your queue will remain at the old setting.