Manual feedback

[vladimi...@gmail.com]

Hi!

While using Biocellion v1.2, I've had a couple of times when manual wasn't clear and it took a bit to figure things out. I'm listing the issues I have had here in case it might help update the manual:

- The manual doesn't state that the order of cells in the output files is stochastic. I may be missing a more obvious solution, but I had to assign each cell an ID and reoder their data in the output.

- "Info" namespace is missing in the manual, but the identifiers from section "3.1 Simulation Instance Information" are declared within it.

- As mentioned in the previous post I made here, the "updateIfSubgridKappa" function doesn't provide neighboring voxel agent data (which is needed to calculate kappa) and it's not obvious how to solve this. While I was given the solution in the other thread, it might be helpful to users to have this in the manual.

- In "updateFileOutputInfo" function, if "fileOutputInfo.particleOutput" is set to true, radius, color, and agent location are automatically printed and can't be disabled. I only needed to print extraScalars so these extra variables were using IO unnecessarily. Additionally, it might be helpful to provide datatypes other than float64 for output.

- For "updateIfGridDirichletBCVal" and "updateIfGridNeumannBCVal" functions, it's not clear what the unit for the "pos" variable is, it only says it's a VReal. I had to test to make sure I had it right (the unit is the same as the agent offset and can span multiple voxels), so it might be helpful to say it in the manual.

Cheers,

Vladimir

[Chris K]

Hi Vladimir,

Thanks for the feedback. I'm sure others can offer better feedback (on your feedback!) than I can, but to address a few of your points:

- The manual doesn't state that the order of cells in the output files is stochastic. I may be missing a more obvious solution, but I had to assign each cell an ID and reoder their data in the output.

Could you elaborate on the order of cells, and what you mean by stochastic?

- "Info" namespace is missing in the manual, but the identifiers from section "3.1 Simulation Instance Information" are declared within it.

Good catch. I can try to update the manual.

- As mentioned in the previous post I made here, the "updateIfSubgridKappa" function doesn't provide neighboring voxel agent data (which is needed to calculate kappa) and it's not obvious how to solve this. While I was given the solution in the other thread, it might be helpful to users to have this in the manual.

[I'll read up on this more...]

- In "updateFileOutputInfo" function, if "fileOutputInfo.particleOutput" is set to true, radius, color, and agent location are automatically printed and can't be disabled. I only needed to print extraScalars so these extra variables were using IO unnecessarily. Additionally, it might be helpful to provide datatypes other than float64 for output.

These outputs (radius, color, and agent location) were setup so you could output them as .vtp files and read them as spheres in Paraview. So unless you want Biocellion to custom output in any format (which I think would be great), I think we're stuck here...

- For "updateIfGridDirichletBCVal" and "updateIfGridNeumannBCVal" functions, it's not clear what the unit for the "pos" variable is, it only says it's a VReal. I had to test to make sure I had it right (the unit is the same as the agent offset and can span multiple voxels), so it might be helpful to say it in the manual.

I think 'pos' is always assumed to be the unit offset from the center. I can try to update the manual.

[vladimi...@gmail.com]

- Could you elaborate on the order of cells, and what you mean by stochastic?

Sorry, I guess I wasn't very clear there. What I meant is that (presumably because of the multithreaded environment), the order in which data about each agent/cell appears in the output isn't deterministic. E.g. if you add a file output scalar to agents, if the order of agent data in simulation step 1 file output (.vtp) is 1, 2, 3 (the numbers referring to agents), for simulation step 2, it might be 3 1 2, or whatever permutation. This is an issue if you care about knowing which agent/cell a piece of output data is coming from.

- These outputs (radius, color, and agent location) were setup so you could output them as .vtp files and read them as spheres in Paraview. So unless you want Biocellion to custom output in any format (which I think would be great), I think we're stuck here...

Yeah, I guess I made less of a manual comment there, as much as a general comment on how Biocellion works.

Cheers,

Vlad