- User Since
- Nov 28 2009, 10:22 PM (391 w, 5 h)
Don't think this is the way to go. This is a more general problem in Blender (as in, exists for lots of various operations) and if any of API is needed to solve such problems those must be more generic and non-conflicting.
lease submot that as a bug report with file demonstrating the issue.
Fri, May 26
This issue was already solved by rB34b6898.
in bli_winstuff.h we replace the stock fseek function with a 64 bit version
Thu, May 25
Wed, May 24
The reason why saving multi-layer layer with F3 gives different name is because Blender uses passes names for EXR passes. But you can call passes arbitrary in EXR, could also be "mammonth's banana". So not sure how this indicates a problem?
Think it is more important to prevent loosing your data here. Having some editor which visualized what IDs are used for would help here, but it's out of the scope of bug tracker. So it is indeed a TODO.
Tue, May 23
This issue was fixed in rB34b6898.
Mon, May 22
Please don't mix multiple reports into a single one, reports unrelated issues as separate reports. Otherwise things runs out of control quite quick.
Ok, did some more testing.
@Lukas Stockner (lukasstockner97), mind having a look here? :)
@Pascal Schön (VanCantus), mind having a look? Thanks!
There might be some change indeed which caused issue with your particular mouse, but lot's of other mouses are working just fine here in the studio. This means it's something in particular with your hardware, which we don't have access to and can not troubleshoot.
Seems the current spline IK is not really suitable for such cases.
@Sybren A. Stüvel (sybren), render pipeline will call the frame change post/pre callbacks (this is likely happening via the depsgraph update). So seems reasonable thing to do for alembic export as well?
As @ronan ducluzeau (zeauro) explained here (thanks for that!), this isn't really intuitive thing to support. Missing features are not considered a bug.
Hair system is a subject for complete reconsideration in 2.8. This isn't a regression and has a work-around mentioned by @ronan ducluzeau (zeauro) above.
The issue seems to be related on OpenGL's lists, which don't have good performance here. This code is being reworked in 2.8 branch, so trying to fix it in master is not really best investment of time.
Surely it will be nice to support, but it is more a feature request than a bug, which we don't accept in the bug tracker.
@Bart Crouch (crouch), is that one for you? :)
This is yet another bug caused by no-functional-changes commit :(
@Sybren A. Stüvel (sybren), mind having a look here? Thanks! :)
It is unclear for me how to reproduce the crash. opening the attached .blend file. It renders all fine.
Wasn't able to reproduce the crash on Iris Pro 580 (nuc6i7kyk machine with driver version 22.214.171.12490). There is a a newer version (126.96.36.19964) of driver available at .
Sat, May 20
Interesting. @Mai Lavelle (maiself) might also be interested to look here then.
Attached file here so it is never lost:
This is a floating point precision issue happening in ray offset (required to avoid self-intersections) and triangle intersection. This is an error i've tried to address in D1212, but there is some bug somewhere which breaks shading.
Fri, May 19
You're right, no idea why my GCC does not complain. Committed fix.
This isn't really a bug. It is caused by some numerical changes and extra clamping applied after 2.76. Those fixes were needed to fix other artifacts in compositor.
The idea was to move to core profile ASAP without loosing ability to see modifiers in BI for until we've got proper depsgraph system in place.
Generally i'm fine, but worried a bit about such a generic name, which also looks like a variable name. Random ideas:
Please attach .blend file which demonstrates the issue with steps about how to reproduce the issue with that file.
Brought behavior back to how it was before. But i find this all weak, and some cleaner design (which follows artists' expectations instead of dictating them how to use something) is really needed in compositor.
@Aaron Carlisle (Blendify), so "squished" part i think we can investigate (as in "resolution" if you wish).