I am trying to model a calibration target (similar to these larger and higher dot density). I'm not allowed to share the .prt file but here's the process:
Extruded a sketch for the overall shape.
Extruded a rectangular sketch to cut the first slot, linear pattern to the edge of the part
Created a sketch for a dot on the "high" rows, did a extrude cut and, then a fill pattern using the outline of the part
Created a sketch for a dot on the "low" row, did a extrude cut and, then a fill pattern using the outline of the part.
All in all there are around 6000 dots. Regeneration time often exceeds 30 minutes on a relatively fast Xeon machine. Is there any suggestions to more efficiently model this. I would like to be able to see the dots because I use the ray trace plug-in to visualize how they look behind various shaped glass but they don't have to be extruded.
Take a look at the 'Identical' pattern type:
Also watch the video on this page about 'turbo patterns'. The video is old but the concept still applies today. Using patterning quilts and then solidifying the quilts is very efficient.
Take a look at Geometry patterns and for a really large number like what you are talking about possibly a pattern of a pattern using geometry patterns. I'm not sure about what exactly you are looking for, I couldn't tell from your link if it was meant to be a random pattern or not but take a look. I'll attach 2 presentations I did from a number of years ago on Geometry Patterns and Pattern of a Pattern. It was on Wildfire 5 so the menus/icons look very different.
I can only concur on the geometry patterns. We have some parts that have thousands of rectangular cutouts and before learning about this option it took a couple of hours to regenerate a model. Now it takes minutes. Be aware that this only works with identical geometry (no partial cutouts).
Thank you everyone for taking the time to share suggestions! I've tried them all so far.
As a bonus, I did discover I could launch multiple instances of Creo at the same time and try alternate strategies in parallel since each instance only consumes ~9% CPU and under 1 GB ram while trying to regenerate.