Same commit also seems to cause T69087.
Same commit also broke T69083, not sure if these are the same.
Since this is an assert only, I cannot forsee the consequences it really has / how bad it really is (and if this should be High prio? (sounds bad to me, daring High prio, please change back if unappropriate...)
This particular file started crashing with rBd357e7b06535.
It is meant for faces as well as vertices [well only one at a time for either faces or vertices].
In the case of faces, it just copies the face normal.
But it should work for vertices as well, try this:
- open .blend
- (it already has averaged Custom Split Normals)
- Copy Vectors from the already selected vertex
- Paste Vectors to the other vertex (circled in blue)
- should come out like, no?
I don't see why you break the whole logic here, that new version of the check will also be slightly more expansive since you fetch the other poly without first checking the first poly...
Just change the stupid comparison mistake (return (mp_other->flag & ME_SMOOTH) != 0; to return (mp_other->flag & ME_SMOOTH) == 0;) and you are done here. :)
- Revert "Fix T68807: smoothing group issue"
- just correct comparison mistake
Glad to hear, closing then
gg is Edge Slide(Vert Slide) and that is how snapping works for EdgeSlide(Vert Slide) afairc.
This was explicitly disabled in rB159bf9d19c77.
In 2.80, I can see object jumping around, but for me, 2.81/master is stable (in that it ignores the sim completely)
-i did find a work around, object/apply/make instances real
is that the preferred method? is it that im not supposed to be able to bake geometry together with convert to mesh?
Thu, Aug 22
@Brecht Van Lommel (brecht): is this expected behavior?
Well, was just a try, according to https://code.blender.org/2019/04/supported-gpus-in-blender-2-80/, your GPU is not supported (Haswell and up only which the N2840 doesnt seem to be...)
Looks like you have an outdate Graphics driver.
Try upgrading these https://downloadcenter.intel.com/product/80939/Graphics-Drivers and let us know if this helped.
Looks like this is also caused by rB9c010c44f42.
Not sure if this is really a candidate, but since this can make the UV Editor basically unusable, will dare setting it to High...
No idea how the volume code ties into this, Alpha Blend usually suffers from the sorting problem
I will dare raising this to High.
Yes poly_edge_loop_islands_calc or something related seems broken in a way (will check further).
Could you post that as a simple .blend file (before converting to mesh)?
(Makes it easier and more effective if multiple devs are looking at the problem, so not everyone has to redo those steps...)
cannot reproduce here in a simple test scene:
Has been reported before, will merge into T69051
Nope, still happening here, try lowering the radius to 0.005, get a coffee :)
Regarding the Copy / Paste functionality:
- documentation is a little sparse here :)
- from reading code it looks like you can only copy from a single face or single vertex (so Copy Vectors / Paste Vectors should be renamed back to Copy / Paste [or at least Vector -- singular], was done in rBdf712d74a25b and also mentioned as being bad in T67370)
- in case of copying from a single selected vertex (thats what you do) there is also a check that all loop normals are the same (this is not the case in your video), otherwise the previous copy buffer is left untouched [or zero even]
- so in your case it would only work if all split normals are actually the same on the selected vertex when copying (e.g. by running Average > Custom Normal prior) -- it is kinda hard to tell in the video if that is the case..
- if these split normals are all the same when copying, then pasting works like a charm here...
@Pablo Dobarro (pablodp606) : will keep this open with low prio (as a documentation thingie...)
Same is true for weightpaint.
I dare setting this to High, will check what caused this...
Looks like T68954 is the same thing, need to investigate both...
Committed a fix for the crash, the Copy / Paste issue needs another look...
Since in the case of calling this from the menu, this is coming from RGN_TYPE_HEADER not RGN_TYPE_WINDOW, so doing RegionView3D *rv3d = ar->regiondata fails.
I think we had similar issues before, let me dig them up...
@Abid Maqbool (cto.abid) : if I understood correctly, and you've already assigned this to something other than ESC, and you want to assign to ESC again, then this isnt working in the upper box, it will cancel instead (leaving it at the former assignment...)
OK, getting it from the menu, too.
Works from Alt+N though...
I can reproduce, @Pablo Dobarro (pablodp606) : is this intentional?
I guess it is... I am getting:
BKE_mesh_nomain_to_mesh: YEEK! this should be recoded! Shape key loss!: ID 'MECube'
I can reproduce, @Pablo Dobarro (pablodp606) : is this intentional?
Please follow our submission template and guidelines and make a complete, valid bug report, with required info, precise description of the issue, precise steps to reproduce it, small and simple .blend and/or other files to do so if needed, etc.
A guideline for making a good bug report can be found here: https://wiki.blender.org/wiki/Process/Bug_Reports
Assume this is because in the upper box the prop is drawn with full_event / UI_ITEM_R_FULL_EVENT. allowing to cancel...
Would think this is done intentionally?
We have T58191, T59333, T60733 open already.
Without having your .blend file it is kinda hard to decide which report to merge into, my guess is we are dealing with ngons here? (then it would be one of the later...)
Does this happen with Factory Settings as well?
oh easy HDRI addon always there i try to uninstall it then load default then save startup to get ride of it but it always there
Is this happening on every restart?
poll of VIEW3D_PT_tools_meshedit_options_automerge seems to be correct [just checks active object], needs further investigation.
@SecuoyaEx (SecuoyaEx) : thx for checking on this.
Cannot reproduce here, just getting a pink background (as expected).
This is just a clear warning/error that that modifier has severe limitations in sculptmode.
Spitting this to the console was added in rB552b2287db86.
I personally dont mind having this in the console as well and would lean towards not calling this a bug...
This seems like a duplicate of T65397
I still have to read through all of the above mentioned tests/nodesetups, but do we agree here, that this is not really to be considered a bug?
This looks like a duplicate of T56534
Glad to hear, closing then...
Wed, Aug 21
Not sure if I understand correctly:
Not sure about the first one, tested a couple UIs here, some do in fact consider scrolling as a immediate change in cursor position (thus start dragging -- QTCreator tree view), some dont (nautilus tree view, GTK tree view).
However all of them allow to drag after scrolling stopped.