In NSC mode, when I use an imported CAD (*.stp for example) file and execute a ray trace command within a macro, it takes a very long time (during which the RAM usage slowly increases) before the ray tracing actually begins. I can track this behavior by observing the CPU and Memory Usage in Windows Task Manager. The memory "ramp up" may occur over a period of 10 minutes or more, and then at some point the ray trace occurs noted by the CPU jumping up to 100% for a short while. I have noticed that on a different computer (same Window, same ver of OS, same files involved), this RAM ramp up is significantly shorter but not insignificant.
The step file sizes involved are say 20 MB. The correspond ZOF file might be 80 MB.
Any quick suggestions as to what is going on here or hints that might help alleviate?
Thank you.
Page 1 / 1
Are you running a ray trace on multiple cores?
When using multiple cores, OpticStudio will load the CAD on each core and this can be quite long. One way to alleviate this is to lower the number of cores until you get an acceptable "loading time". It is hard to tell what setting is the best, but basically the idea is to find a compromise between the CAD loading time (less core is better) and multi-threading ray-tracing (more core is better).
I hope this helps.
When using multiple cores, OpticStudio will load the CAD on each core and this can be quite long. One way to alleviate this is to lower the number of cores until you get an acceptable "loading time". It is hard to tell what setting is the best, but basically the idea is to find a compromise between the CAD loading time (less core is better) and multi-threading ray-tracing (more core is better).
I hope this helps.
Yes, running multiple cores. I will experiment as you suggest and write back with what I find.
Thanks!
Thanks!
Reply
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.