Member Possibly-mef Posted March 18 Member Report Share Posted March 18 I have noticed that Autotopo via decimation seems to be broken since 2022.49 I just tried 2023.07 and I it still seem broken. I have tried it on Windows as will a Mac. I am attaching an image that demonstrate the results I am now getting. I am taking a 4M face mesh down to 400k. It has worked flawlessly until any of the releases since 2022.49 (mac). Quote Link to comment Share on other sites More sharing options...
Advanced Member Elemeno Posted March 18 Advanced Member Report Share Posted March 18 got a file? by looking at it ... just seems its gone too low Quote Link to comment Share on other sites More sharing options...
Advanced Member Elemeno Posted March 18 Advanced Member Report Share Posted March 18 mines working fine... Quote Link to comment Share on other sites More sharing options...
Member Possibly-mef Posted March 19 Author Member Report Share Posted March 19 Here is the same file but 2022.49. Works like a charm Quote Link to comment Share on other sites More sharing options...
Member Possibly-mef Posted March 19 Author Member Report Share Posted March 19 Thanks for the quick reply. Here is the same file after Autotopo w/ decimation from 2022.49. Any version later than this produces the result I previously posted. Quote Link to comment Share on other sites More sharing options...
Carlosan Posted March 19 Report Share Posted March 19 File was not attached. Quote Link to comment Share on other sites More sharing options...
Member Possibly-mef Posted March 19 Author Member Report Share Posted March 19 sorry. Here is a screenshot of result from 2022.49 and a link to the .3b file https://www.dropbox.com/s/4y7c83v1ab0w821/3DC-Angela-Take01-M4.3b.zip?dl=0 1 Quote Link to comment Share on other sites More sharing options...
Advanced Member Elemeno Posted March 19 Advanced Member Report Share Posted March 19 mines ok Quote Link to comment Share on other sites More sharing options...
Oleg_Shapo Posted March 19 Report Share Posted March 19 The workaround is voxels Quote Link to comment Share on other sites More sharing options...
Oleg_Shapo Posted March 19 Report Share Posted March 19 Try this before decimation. Can be several times. 1 Quote Link to comment Share on other sites More sharing options...
Carlosan Posted March 19 Report Share Posted March 19 Hi, I forwarded your issue to mac support team. You can also send this mac support-related question to Sergii Kryzhanovskyi at support.linux@pilgway.com Thanks Quote Link to comment Share on other sites More sharing options...
Member Possibly-mef Posted March 19 Author Member Report Share Posted March 19 Carlosan, Thank you. I have encountered the same problem with the Windows version as well. I will follow up with Sergil. 1 Quote Link to comment Share on other sites More sharing options...
Member Possibly-mef Posted March 19 Author Member Report Share Posted March 19 Elemno. What value did you use for the decimation? -- based on the density of the mesh in the image you sent, it seem quite a bit high. I will try a higher density and see what happens. Quote Link to comment Share on other sites More sharing options...
Member Possibly-mef Posted March 19 Author Member Report Share Posted March 19 Ok so the problem gets a little weirder - I did as Elemeno suggested: converted to voxels (went from from 4 011 720 tris to 5 208 516 tris) Decrease Object 2X (Increase Density) Autotopo via decimation (500 000) polys While it did not exhibit the same damaged mesh as in the prior example the resolution is too low for good renders in C4D (or other apps)The result is a 40 000 poly group no matter what value it put in to the dialog. I tried a few values between 300k and 400k). Always resulted in a 40k poly group BTW I ran this test on a Windows machine using version 2022.58 starting from scratch in a new 3DC document. Quote Link to comment Share on other sites More sharing options...
SERGYI Posted March 24 Report Share Posted March 24 I have thoroughly tested "Sculpt Tree > Retopo via decimation" in "3DCoat 2023.08" for macOS using standard models (Reduction Percent 99%) and the attached scene "Angela" (90%). The results are on the attached screenshots. It seems like the decimation itself is working correctly. Try to rename the user data folder "Documents > 3DCoat" into something like "---3DCoat". Run "3DCoat 2023" and try decimation again. Some data in the user data folder or some setting inside "3DCoat 2023" forces the decimation to work incorrectly. Quote Link to comment Share on other sites More sharing options...
Member Possibly-mef Posted March 24 Author Member Report Share Posted March 24 Hello Sergyi, Thanks so much for following up. I tried a couple of tests with both 2022.60 and with 2023.08. I deleted the Documents->3DCoat folder between each test. I am running this test on at 2017 iMac Pro (8-core Xenon W - 32Gb) running Ventura 13.2.1. I seem to be getting the same sort of results. See attached screen shots. The results are slightly different but still not right. Although likely not significant I noticed that after I deleted the 3DCoat file from Documents, each time I started the app a 3DCoatV4 folder was create as well at 3DCoat. And also as a side note it seems the the imported object in 22.60 and 23.08 is now centered vertically in the grid. Is that an expected change?? Quote Link to comment Share on other sites More sharing options...
Member Possibly-mef Posted March 24 Author Member Report Share Posted March 24 I ran the same test on a Windows machine and on an M1 MacBook Pro using different initial data (i imported a different scanned mesh with 4M triangles.) Same results. One notable thing I discovered was no matter value I put in for the Polygon Reduction (200k, 400k, 1M) it alway resulted in a 40k retopo PolyGroup m Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.