Open gumyr opened 3 years ago
lorenzncode
commented
3 years ago Hi, I tested this script. I see that CQ-Editor memory usage continues to increase until process exit.
CQ-Editor was able to display the part after increasing the Deviation value in 3D Viewer settings: Deviation: 0.01 Angular deviation: 0.1
The very large memory usage case for me was with default Deviation 1e-5.
gumyr
commented
3 years ago Thanks! When I change Deviation to 0.01 my "real" threads are visible in cq-editor again. I'll put a note in the documentation for my thread library that this value should be changed. Is the only downside a slight drop in viewing fidelity?
Cheers, Roger
On Thu, Aug 26, 2021 at 7:43 AM Lorenz @.***> wrote:
Hi, I tested this script. I see that CQ-Editor memory usage continues to increase until process exit.
CQ-Editor was able to display the part after increasing the Deviation value in 3D Viewer settings: Deviation: 0.01 Angular deviation: 0.1
The very large memory usage case for me was with default Deviation 1e-5.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/CadQuery/CQ-editor/issues/286#issuecomment-906334693, or unsubscribe https://github.com/notifications/unsubscribe-auth/AUNL25SJJQH7JWZ2IA2LAADT6YSHDANCNFSM5CZN723A . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email .
lorenzncode
commented
3 years ago Yes, I think that's right - it should be a render issue. The trace, while this is running, is in Tessellate, and @adam-urbanczyk mentioned increasing Deviation when a similar render problem came up in discord.
A few test results:
| Deviation | Time | RSS peak |
|---|---|---|
| 1.5e-5 | ~1m45s | ~8.3GB |
| 2e-5 | ~45s | ~5GB |
| 1e-4 | ~10s | ~1.4GB |
| 1e-3 | ~5s | ~1GB |
gumyr
commented
3 years ago Wow, that;s a big difference. I've used ray tracers that render pixelated images to start and then recursively refine the image to eventually get to the final result. If the same was done here those first few loops would take relatively no time at all.
Thanks for the help, Roger
On Thu, Aug 26, 2021 at 5:28 PM Lorenz @.***> wrote:
Yes, I think that's right - it should be a render issue. The trace, while this is running, is in Tessellate, and @adam-urbanczyk https://github.com/adam-urbanczyk mentioned increasing Deviation when a similar render problem came up in discord.
A few test results: Deviation Time RSS peak 1.5e-5 ~1m45s ~8.3GB 2e-5 ~45s ~5GB 1e-4 ~10s ~1.4GB 1e-3 ~5s ~1GB
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/CadQuery/CQ-editor/issues/286#issuecomment-906755575, or unsubscribe https://github.com/notifications/unsubscribe-auth/AUNL25WLZU3M6F2NQH4QFZLT62WX5ANCNFSM5CZN723A . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
To Reproduce
While working on the creation of a thread library I discovered that cq-editor crashes on extrusions with rotation and was able to reproduce the problem with the following (note that FreeCad is able to open the generated step file which looks surprisingly good for such a rough approximation) :
Environment
OS: Ubuntu 21.04
Output of
conda listfrom your active Conda environment: