Issue Running Tunny in Grasshopper (v1.0.6-RC.2)

Hi @hiron,

I’m trying to use Tunny with my Grasshopper definition, but I’ve run into an issue and would appreciate your help.

I followed the steps outlined in the documentation:

  1. Downloaded the latest version (v1.0.6-RC.2)
  2. Dragged the appropriate version for Rhino 8 into Rhino
  3. Double-clicked the Tunny component to install the required libraries

I confirmed that the .tunny_env folder was created. However, when I press the Run button, nothing happens—even after waiting several minutes. When I press Stop, I receive an error (attached).

My Grasshopper script runs fine with Opossum, and I’d really like to get Tunny working.

I also experimented with different .NET settings in Rhino (NetCore and NetFramework, versions 7 and 8). With NetFramework, Tunny fails to load due to a licensing issue with the license.php module. So I’ve been using NetCore 7 and 8, but the issue persists.

Any guidance would be greatly appreciated.

Best,
Pooyan

Hi @pooyan_k,

Three possibilities can be considered:

  1. What is the status of the Python folder in .tunny_env? There may have been a failure in installing the Python files necessary for optimization.

  2. Is Wallacei installed? Depending on the environment, it seems that sometimes it doesn’t work properly when Wallacei is installed.

  3. Which sampler are you using? Since some samplers may fail to execute, could you please check if you get the same results with the TPE sampler?

1 Like

Hi @hiron,

Thank you very much for your prompt and detailed response.

I believe the first possibility you mentioned is more likely. My Grasshopper definition includes several GhPython components, and when I open the file with the Tunny component added, I encounter errors related to all GhPython components. However, if I remove Tunny and reopen the definition, these errors no longer appear.

Additionally, when using Tunny, the optimization process takes 5–10 minutes to start and runs very slowly (with TPE). In contrast, the same definition runs much faster with Opossum, and even with Tunny 0.10, the performance was significantly better. Moreover, I’m unable to stop the optimization without force-quitting Rhino via Task Manager.

I’m also wondering if Tunny might conflict with other plugins, such as Speckle. I’ve attached a screenshot showing an error from the Speckle Receive component, which might indicate a conflict between Tunny’s Python environment and other software—or possibly an issue with missing or improperly installed packages.

Here are a few specific points I’d appreciate your help with:

  1. There is a Python folder inside the Tunny_env directory. How can I check which files or packages might be missing?
  2. I don’t see a Wallacai folder inside the Tunny_env directory. Is this expected?
  3. When using TPE, the optimization does run, but it’s extremely slow.

Thanks again for your support!

Bests,
Pooyan

Hi @pooyan_k

  1. There is a Python folder inside the Tunny_env directory. How can I check which files or packages might be missing?

The optimization seems to have worked without problems, so it does not appear that the Python folder is causing the problem.

  1. I don’t see a Wallacai folder inside the Tunny_env directory. Is this expected?

Wallacei is a another optimization plugin from Tunny. I mention this because we have received reports that it sometimes conflicts with Tunny when it is installed.
But since you don’t seem to be using it, it is not a problem.

  1. When using TPE, the optimization does run, but it’s extremely slow.

TPE can be a slow optimization technique.
For example, is it similarly very slow when Random is selected?
If so, a possible cause is the location of the result file.

Tunny creates the result file on the desktop by default. If your desktop is set to automatically sync to the cloud, such as Onedrive in Windows, this may slow things down as the results file is synced every step.

About Images

It seems that the sqlite library used by Tunny is conflicting with other plugins. This problem also occurs with another plugin and will be fixed in the next version.

1 Like

Hi @hiron

Thanks a lot for the comprehensive explanation!

I just noticed that in the latest release the SQLite dependency has been removed—really appreciate the update. Looking forward to using it!

1 Like

Hi @pooyan_k

Thanks for checking new version. I hope the version help you to use Tunny!

1 Like