Long processing in AnnualDaylight component when reading many files

Hello,

I am relatively new to using DIVA for analysis (or any analysis software for that matter) and I am having trouble loading many AnnualDaylight files for reviewing the Data in the Grid Viewer.

I ran the AnnualDaylight component in "Write Only" mode to create .bat files for a matrix of 243 design options.

I then batch ran all of the .bat files, which takes a while, but I expected that (20.5 hrs).

My problem occurs when I am then trying to read the data of those 243 iterations through the AnnualDaylight component.

I've created a separate grasshopper file that contains only the AnnualDaylight component and a list of the names of each of the above mentioned iterations. But when I unlock the solver my machine can spend close to an hour loading. If I split the list of names to grab only a few of the iterations, it will take a minute or two, but as I grab 30 or 60 or all 243, it takes a long time to finish reading the data.

My question is this: my ultimate output is to extract the raw numerical data into a .csv with columns for each of several variables (ASE, sDA, avg. UDI, etc.) for input into a parallel coordinates chart in Thornton Tomasetti's Design Explorer 2; so, is there a way to bypass the use of the grasshopper components and just get the data from the .dlt through some other means?

I don't need to view the 243 analysis grids simultaneously, so I was hoping a more direct route to .csv would exist. If not, Rhino is locked down for an hour while grasshopper processes the data.

Thank you for your help,

michael

Tags: .dlt, AnnualDaylight, read data

Views: 33

Reply to This

Replies to This Discussion

Hi Michael,

Indeed, the components deserialize the entire DLT file before extracting any nuggets of data.  A 200MB file may take 5-10 seconds to load, which, times 243, quickly gets us to half an hour.  Unfortunately, this behavior is somewhat intrinsic to the current format, so we are stuck with it until DIVA 5.

In the meantime, I just pushed an update (v4.0.2.51) that writes a summary CSV file to the project folder at the end of a run.  This includes sDA, ASE, meanUDI, and a bunch of other data relating to grids and windows.  It won't help you retroactively, but may save you some time on future runs.

An example summary file is attached.  Let me know of other stats would be useful -- should be easy to augment.

Jon

Attachments:

Jon,

Thank you for the follow-up. I had to step away from this project for a while, but I think the CSV format that you've shown will be very helpful in the future.

FYI: My solution in the meantime was to have grasshopper iterate through the different analysis DLT files and write them to a CSV one at a time; which, may sound inefficient, but ended up being much faster. In fact, trying to load all the DLT files at once was never successful (at least, I gave up after 12 hrs of waiting).

Reply to Discussion

RSS

© 2017   Created by jeff niemasz.   Powered by

Badges  |  Report an Issue  |  Terms of Service