Change Log: cntlc4d 0.584 alpha (2019 December 31)
[core] Fixed z-fighting render artifacts for coplanar polygons in the scene, precision is improved.
[core] Fixed few rare crash-bugs.
[cntlc4d] Cinema R21 support.
CentiLeo for Cinema 4D 0.585 available - speedup 2x to up to 10x and R21 support
Some render tests :), Render tests done on a already made geometry of a rim
Administrator Posts:
Hi Juan! Almost done, it seems the new solution is working, and a couple of more issues were fixed. The reason of delay is that I didn''t want to sacrifice some memory/speed and there were some conflicting pros and cons here and there (as always). Need to test and release, hopefully soon! In days
![]()
CentiLeo Chat:
|
Some render tests :), Render tests done on a already made geometry of a rim
Administrator Posts:
Still fixing the problem with artifact, not so easy, but hope will progress asap.
CentiLeo Chat:
|
Some render tests :), Render tests done on a already made geometry of a rim
Administrator Posts:
CentiLeo Chat:
|
Some render tests :), Render tests done on a already made geometry of a rim
Administrator Posts:
Hi Juan, I can't yet reproduce the crash, but haven't looked very deep there because I looked on Max2019-Win7 and Max2020-Win10. Will look deeper today, probably wil rework some open/close render staff.
Did you open IPR and couldn't manually save? Or you closed the IPR and it didn't want to save anyway? As for artifact - there is a reason: coplanar triangles in artifact area. For the whole artifact area there are 2 triangles and when secondary rays launch from one triangle they meet the z-fighting problem of the duplicate triangle. Removing unneed triangles (and leaving their duplicates) by user clears the problem. But I prefer to fix it on our part. In current version we workaround this problem analyzing triangle IDs and without ray offset. But sometimes there are scenes with coplanar triangles (like this one) and so we will introduce a ray offset render setting. At first time it will be user defined, later we will do things automatically to reduce this not very nice setting.
![]()
CentiLeo Chat:
|
Some render tests :), Render tests done on a already made geometry of a rim
Administrator Posts:
Hi Juan, great thanks for file! Nice model btw! I reproduce the problem, there is something wrong either with normals or geometry translation. Investigating it further
CentiLeo Chat:
|
first render test
Administrator Posts:
Sure, the nature never has perfect and same colors
CentiLeo Chat:
|
first render test
Administrator Posts:
Hi remi! Very good! I would recommend a good CG trick: add green (yellow, brown) variations to the tree leaves. In CentiLeo it's quite easy to do:
1) make cntlRandom shader (with randomize by object parts checkbox). This means: any locally connected triangular object groups will get different color/ID 2) make cntlRamp shader and setup several knots with green, yellow, brown and other colors you like 3) connect cntlRandom to the input_map of cntlRamp (it can be done inside cntlRamp). 4) connect cntlRamp to diffuse slot of CentiLeo Material. The other way is to do it even without Ramp, only with Random shader. But in this case in Random shader you should switch off black and white output and tune the HUE variation range (e.g. from 0 deg to 90 hue value). That way the Random shader will generate Random colors specified by you. I know with node editor GUI it would be much easier to observe the shaders. I am workin on it now. The speed is ok for 980. And let's hope we can improve it further. As for Volumetrics - not soon yet, too many urgent tasks.
CentiLeo Chat:
|
first render test
Administrator Posts:
Hi remi, how would you use these renders in a game? For adverticement of game? I think some day we will have to make lighting baking to texture
![]() I think some reflections + bump are missing on top of sss
CentiLeo Chat:
|
Some render tests :), Render tests done on a already made geometry of a rim
Administrator Posts:
Juan, hi! Let's fix the artifact! E.g. you may look how the UV mapping works if you plug the cntlUVProjection node to diffuse slot of CentiLeo material. In cntlUVProjection you should select the same mapping type as the in the Bitmap which causes the problem. And there we will see what's going on.
CentiLeo Chat:
|
CentiLeo for 3ds Max 0.586 available - speedup 2x to up to 10x, added many new shaders and emission
Administrator Posts:
![]() Will soon start publishing the documentation online with few articles, will begin from materials and shaders.
CentiLeo Chat:
|
Some render tests :), Render tests done on a already made geometry of a rim
Administrator Posts:
Hi Juan, there seems to be a problem (probably with UV or XYZ mapping). Does it happen when you apply a bump map? Can you please send me a file for test and debug to
![]()
CentiLeo Chat:
|
CentiLeo for 3ds Max 0.586 available - speedup 2x to up to 10x, added many new shaders and emission
Administrator Posts:
CentiLeo Chat:
|
first render test
Administrator Posts:
Hi remi bouffort, to make a normal map you can make CentiLeo bitmap shader for the Bump slot of CentiLeo material. Then in CentiLeo Bitmap you select the normal map file location and enable the "Is Normal Map" checkbox. That's all. There are also other flags that flip the normal map r/g/b to adapt for various normal map formats.
Also this way of doing a normal map allows additional combinations with greyscale bump maps.
CentiLeo Chat:
|
CentiLeo for 3ds Max 0.586 available - speedup 2x to up to 10x, added many new shaders and emission
Administrator Posts:
Change Log: cntlmax 0.583 alpha (2019 December 6)
[cntlmax] Fixed broken GUI layout for increased windows font size and high resolution monitors.
CentiLeo Chat:
|
Roadmap for CentiLeo 0.55-0.61
Administrator Posts:
Hi remi, sure, do it!
![]() ![]()
CentiLeo Chat:
|
first render test
Administrator Posts:
Hi remi! Good start! Btw, you can make colored flakes if you connect flake shader (with greyscale ids) to the Ramp (to input map) or Random shader (to seed map).
CentiLeo Chat:
|
Roadmap for CentiLeo 0.55-0.61
Administrator Posts:
Hi remi bouffort, and welcome to our forum! Thanks! I try to test as much as can remember but there are still several known bugs there. As for the nodal GUI - it's in development currently
CentiLeo Chat:
|
CentiLeo for 3ds Max 0.586 available - speedup 2x to up to 10x, added many new shaders and emission
Administrator Posts:
Hi a1cat cato, a lot nicer!
![]() This is diffusion based SSS (but ray traced). It's pretty fast, but not precise on the small details because may them. As for usability - it's the same as in Arnold or Reshift (tested a year ago). Their documentation may apply for us too. 3 layers which are mixed together. For example to make white you may create one red, another green and another blue. And they will be combined to white. Another way is to make all layers white and it will be the same. To make some color effect from under surface you can make some red channel with larger radius. There is more precise method (but a bit slower/noiser) - Random Walk SSS, we haven't yet implemented it, but it's in the TODO list. Will describe such things in manual.
CentiLeo Chat:
|
CentiLeo for 3ds Max 0.586 available - speedup 2x to up to 10x, added many new shaders and emission
Administrator Posts:
Btw I tried to fix the bug with IPR - final render conflict. But hasn't reproduced it quickly and released with other fixes. For me the pauses, launches of final render haven't produced the crashes. Note that final render will not start until IPR is closed because we want full resources in either case.
But I will test more on Win10 (because I usually do tests in Win7) and hope to cactch the problem. Good image! I like them with DoF. Next step is to add SSS for candles. And small fire ![]()
CentiLeo Chat:
|