Hi friends,
I am doing some concept testing around generative design studies that read and write to external databases for supplemental computing - in this case the “database” is excel. In the following operation, I am writing a number to excel file 1, reading a resultant value from excel file 1, then writing this value and a second number to excel file 2 in order to generate a second resultant value.
Ultimately the math of the equation looks like this: 2^a = b, b + c = d. d is my expected outcome.
This works without any issue when run in Dynamo, however when exported to a study it inconsistently produces incorrect values for ‘d’.
Dynamo Test.REV3.dyn (45.9 KB) Addition.xlsx (9.3 KB) Power.xlsx (9.4 KB)
Generative design will spin up six concurrent Dynamo Sandbox instances, and you’re running 2x excel interactions per graph… I don’t know if the Microsoft Office API will support that.
Try using a Data.Remember node to serialize the first excel sheet into the dynamo graph, do the look up there via dictionary or other means, and then let the resulting number be recorded to the results in Generative Design. Once the graph is finished you can pull that data into excel in one move for additional processing as desired.
1 Like
@jacob.small thanks for explaining that - I was not aware of the parallel Dynamo instances.
My AU session from last year might help she’d some light on that and other stuff that happens when you run a study: Generative Design at Hogwarts: Using Tech Instead of Magic | Autodesk University
1 Like
I confess I’ve had your presentation bookmarked to watch for a while - I’m going to take this as a sign and watch it haha!
1 Like
It was a lot of fun - my co-presenter Alexandra Nelson deserves a ton of credit too. 
Hopefully you enjoy it as much as we enjoyed making it.
