I was playing around with the Excel connector and created a list in Grasshopper.
Receiving that in Excel takes about 4 long minutes for about 450 rows with 4 columns each.
Is that the expected performance or is there something wrong with my setup?
I created a test stream here: Speckle (for some reason I can´t access speckle.xyz with my speckle connector, but the data should be the same as in the example from yesterday)
A few things I noticed:
If I use ‘Receive last selection’ it only takes a few seconds
If I receive the root object it also only takes a few seconds
and select the child of the root object (the list with 455 elements, see above) to get it into excel properly, which takes now only 2 long minutes not being behind a vpn today…
I Just debugged the slowness issue, it seems to be because each element is detached and that’s causing an extra API call. As a matter of fact, 455 * 0.15-0.3 sec results in approx 2 min.
Can you disable detaching on your source elements and try again?
I’ll also ping @dimitrie to see if it’s normal that the object loader takes so long to retrieve such simple objects. Ofc, the calls could be parallelized as well…
PS
Load last received works much faster as it’s pulling some cached data
Ok, for everyone else eventually reading this thread: setting ‘range with headers’ in Excel did the trick (pretty obvious actually), now receiving in Grasshopper works like a charm
Can we make the connector remember the stream & branch of each individual sheet?
If you send to multiple branches in the same book, you most certainly end up sending to the wrong branches since it seems to pick the last used one (of a different sheet eventually).