openLCA 2.0 Feedback

Question

Dear Andreas and openLCA team,

Thank you for creating such a great LCA software and making it available for everyone! Thank you for you commitment and hard work! 

Version 1.11 was already a perfect tool for me and fits perfectly for the usecase of my company. I was excited to see the release of version 2.0 and I checked it out. I think that the development is going in the right direction, but it needs a bit more polish. Here are some problems that I encountered:

1. Sankey inconsistency: I have an electricity mix which is used in 5 processes, but in the Sankey diagram it is only connected to one process. Electricity emissions and contribution is however good, but the link between other processes is just not shown. Cut off was turned off. It does not happen all the time, but it is mostly the case.
2. Model graph and Sankey export: If I zoom in too much and export the pictures, the remaining processes (not shown on my desktop) are going to be cut off.
3. Model graph export: When exporting model graph and using dark theme, flows are almost not visible.
4. Theme switch: It would be nice to choose Dark/Light theme independently of the system settings.
5. Calculation type: I'm not sure what is the difference between "Lazy/On-demand" and "Eger/All" calculation types. In both cases I see the same results. I guess I should wait for the new manual to be released.

Points 1, 2 and 3 were working perfectly in openLCA 1.11 version.

Best regards

Nemanja Stipic

Comments

any idea when the new 2.0 manual supposed to drop?

Answer 1

Hi Nemanja, thanks a lot! You are right with 4 and 5, 4 is somewhat tricky though; 3 did not exist in 1.11 since it did not have a dark mode, right? You can also define other schemes for the model graph and sankey diagram, we are going to explain that, and that can change the contrast; 1: Can you maybe post an example? That would be useful. Thank you again, best wishes, Andreas

On 5, in short: lazy calculates just the result and does the contribution results (needed for the tree view and sankey diagram) only on the fly; all calculates directly all the contributions, upfront, and keeps them in memory.

Comments

For the point on lazy vs. eager calculations, is it then safe to assume that a contribution tree calculated via lazy will have the same values as those calculated by eager?

I'm using the olca-ipc package via Python where there is no option in CalculationSetup for Eager or Lazy, but the results I'm seeing in Python vs. contribution tree via Eager vs. contribution tree via Lazy all seem indicate the identical results regardless.