Cerbera

You should still have the various options, Tiny-Huge, but they are under Scene (previously called Project) Settings, under Display section. Or you can set it to custom to set both limits. CBR

Oh nice 🙂 He's got much better eyes than he used to have ! Great boots too ! And lovely lighting and SSS too... CBR

When I open that file, I see the hair is not linked to any geometry, despite the hair appearing on it, which means it must have been there at some point ! i don't know what caused it to not be there now, but if you drag the mesh into the link field it fixes your problem. CBR

Difficult to do anything but guess without scene file, but have you groomed the hair at all, which would require resetting its dynamics ? CBR

Hi Brandon Yeah we can do that. If you pop up here... ... you can double click the Instance icon to select all the instances in the scene in one go... Then you can right-click one of them, choose Add to New Layer, and then hide that layer from the object manager, or anywhere else should we need to... CBR

I would guess that the stacked booles and the various modes chosen therein, in combination with an incredibly small voxel size in the the VB are what is breaking that. I remain confused about why there is any boolean action needed here at all when using the volume builder, which viably replaces all of its functionality ! Here is what happens if you simply put the Cube, and Right and Left 'Big Circles' directly under the VB with no booles required... now your rotation segments have an actual effect on the smoothness.... After that, you can start to subtract other operands, also children of the VB, which is how I removed the small circle block on the left there, and could go on if so minded to complete the whole model using this sort of approach, which is how the Volume Builder workflow is meant to work... To get the result above my stack is simply this... Now, as you have seen, booleans can be used inside VBs but I have not yet found any circumstance where doing that helps the workflow more than confounding / obstructing it, or making it needlessly complex and even more calculation-full ! We could also talk about whether VB is the best approach for this kind of object - and I would say it is, IF the primary focus for the project is massive speed at the expense of polygon efficiency ! If you had a bit more time to throw at it you could model this via regular Sub-D modelling (no booles) quite easily and you'll get a nicer overall result, with about a 1/100th to 1/1000th of the polygons required to do it the VB way, and it would also be eminently more suitable for easy UVs and any subsequent animation, were that to be required... just a thought... Hope that helps CBR

Nice to hear. I regularly pop to GitHub to see how RS is doing with various OSL shaders, and most of them work great, with a few notable exceptions. Pleased to hear that list is getting smaller and one of the most useful ones is working for us RS-heads... CBR

You know what I think, on the whole, but I will say AMD processor again, just to reiterate how much extra money you have to throw at graphics because you are not paying X00 extra to have Intel written on it... CBR

Yes, either with a correction deformer, should you need the SDS to remain parametric, which does limit component operations to mainly selections only, or, as mentioned in first post, you can APPLY the current subdivision by hitting C with the SDS object selected rather than the base model which is under it. Note the level of subdivision that gets applied is the RENDER SD value, not the viewport level. Making SDS editable (usually only one Level of it at a time ) is a valid and valuable technique for increasing the level of subdivision evenly all over a mesh, which is sometimes more helpful than having to add specific loops to the base mesh, which is preferable in most cases only if you need additional loop detail / density in one particular area and not spread across the whole mesh. Here is your fish with various levels of SD applied - at the top no SD, then L1, then L2. A valuable note on this technique is that your total level of applied subdivision shouldn't ever total more than 3, unless you are using SDS edge weighting in which case the sensible maximum is L4 or 5. So if you apply 1 level of subdivision to a base mesh, and then put that mesh inside a another live SDS, then the live SDS should only have a maximum of L2 because you have already applied one level, if that makes sense. CBR

Not sure I understand the question. You haven't pressed C on the SDS or it would be editable and isn't in your screenshots. That would be what you wanted to do next when your low poly base mesh doesn't permit any more refinement of the shape, and you didn't want to manually insert extra loops but merely wanted to apply the existing subdivision. Still not sure what you are asking though... Perhaps you are confused about isoline editing ? If that is turned on in view settings, and display mode is also set to isoparms then you should be able to directly edit points on the subdivided mesh result, and the outer low poly cage should be hidden. Is that what you want to happen ? Here is a similar fish under isoline editing in poly mode - is that what you want yours to look like ? CBR

I don't think we stand much chance of finding out why either until we get the scene file ! CBR

This is the sort of question where a scene file would massively help anyone trying to assist - pls can you upload it ? I mean we'll probably have to do something with f value, but I'm damned if I'm recreating that from scratch to find out ! 🙂 CBR

Yes and yes ! CBR

Feel free to post your file ! I'd be interested to see what formulas you had going on... CBR

No idea which formula we'd need, but I can get very much that sort of thing merely using honeycomb array mode of cloner and some R transform on the clone... Just a thought... and I still don't understand why I don't have to do 45 degrees of rotation on all 3 axes, but results are results ! CBR

Yes that can be most frustrating. It's compounded by having different behaviours for different tools. So, if you are in poly pen and try and create a cut I have noticed that this wants you to actually cut directly to the line of choice, whereas Line Cut has the most chance of success if you cut BEYOND the target edge so that you cross it instead of meeting it. If you can get those 2 operations right a lot more of your cuts will succeed, but the disparity in behaviour between the 2 tools is confusing for new users and generally quite frustrating for everyone... CBR

The trouble with producing 1 long correctly proportional strip is that it is a horribly inefficient use of UV space - to fit it in the square UV canvas proportionally it will be a tiny long thin strip somewhere at the top with about 80% of the canvas space unused. And that necessitates huge image maps to get enough detail in there. Sometimes that is what we need though, and it is possible to get, although not necessarily via entirely automatic means. If cylindrical projection and a subsequent relax and pack doesn't restore your perfect UV proportions there are fall-back methods that will, such as projecting each wall face frontally and then terracing the islands together into the strip you need, or, where islands do not have matching / continuous topology even just placing them correctly without stitching should work. CBR

I have a lot of thoughts about this, most of which I have already shared with Maxon. The few bugs that remain aside I am mostly happy with the current SDS modelling toolset, but there are lots of things they could add to it, and few I hope they will improve about it. 1. Smooth edges needs an 'equal spacing' option, so we can turn that bit off when we don't need it, as we often don't ! 2. The brush tools and their smoothing / relax modes need a flood button that applies a set level of brush function to just the selected polys, effectively replicating what HB modelling Bundles's HB_relax used to do. We need to remove the human, brush-based element from these operations and apply these things exactly evenly sometimes. It is noted that Iron tool does similar things, but lacks curvature awareness, so this is mainly needed for the relax mode of the brush. 3. We also need an equivalent of HB_Round edge that works in the way that did, which is not the same as the way smooth edges works. We need both, and both integrated ! 4. Last one from HBMB - the whole start retopo setup and project mesh buttons. With some limitations these tools do still work in latest Cinema, but they really should be native by now. 5. My oldest ongoing fix request - make UV mapping work with OpenSubDiv. It still doesn't since its initial integration back in R17 or so - we get weird SDS distortion that is unsolvable. 6. I want Mirror back in the r-click menu ! 7. Normal Move / Scale / Rotate work on all components now but only appear in menu for polys modelling ops. 8. Q needs to only operate SDS and not any other generator. 9. One sided Bridge patch tool needs improvement so it can patch huge gaps around corners without bulging out / ruining its usabiility. 10. As always, I'd like a boole tool that performs more reliably in more difficult (animated particularly) circumstances, and with a workflow that doesn't involve having to combine operands under nulls or connects at any point... 11. Select similar. 12. Straighten Edges also needs a make planar button set. (Flatten does this, but would be nice to have within SE tool too) 12b. I sometimes miss a dedicated 'Make Planar' button for all component selections, outside the Flatten tool. Whilst I like that tool, nothing quite beats Max's system where we make a selection and can hit a single button to instantly make it planar to any axis and in any coordinate system. We can do it with zero scaling of course, so it's more a small QoL request. 13. Axis in Normal Mode needs to orient more helpfully to single edge selections and retain changes in Axis controls needs to be more predictable and robust. 14. Slide tool needs to not fight polygon snap, as it can do sometimes. 15. Line Cut needs to work more reliably, not insert the extra points it sometimes does, and actually make cuts on any visible line it touches. Currently it misses far too many that have to be repeated. 16. Although its algorithm (particularly in advanced mode) works very well, and much better than anything we had before) I remain unconvinced that Thicken works in the best way it could. With the current system of a mere percentage between inner and outer thickness it is impossible to type specific values in each, or to get specified differing thicknesses inner and outer, which is sometimes required when working with arch-vis blueprints for example... 17. Highlight a)ngons and b) tris buttons in the Mesh Checker so we don't have to piss about with Structure manager to find them ! 18. Slide tool needs negative extension beyond borders. 19. Terminate loop - function that takes a poly selection and terminates a loop within it. CBR

Like the new look and color scheme - all very readable and comfortable from over here. I'm still regularly hanging around after all this time ! Good to hear more members are joining ! CBR

If there is an option for that I can't find it either ! Yes, I concur that does happen. I'm afraid I don't yet know why, or if that is the intended result. Tell me - do you have any luck running untriangulate on the reimported mesh (Cinema) ? CBR

I don't think we can win by having a jiggle in a group with the resulting VB mesh - I think it has to be a child of a moving object. So, if we move the jiggle directly under the animated sphere then it continues to work in the VB. The problem then is that there is no movement (or jiggle) on the larger sphere, and so no jiggle appears in it, even though the other sphere continues to jiggle correctly inside it. Of course none of these things involve any sort of simulation, so we can't expect the larger sphere to physically react to the impact / absorption of the other one unless it ALSO has its own jiggle, and you animate it to move slightly as the other one joins it. And that works quite convincingly... I was also thinking we could try baking down the VB result to alembic, and them try jiggling that, but that doesn't work in an impressively explode-y type way, so I guess that's a no... CBR

Wires, or AI did it 😉 Lols... Nice to see a Transformer before they went mental complicated in the films ! Happy New Year dude CBR

I see - I guess that would explain it !

Ah yes, if we 'cross the streams' in a single move that does indeed report the created error state polys as non-planar (and produces visible polygon display errors in most cases). Even if that is terrible technique, and can't create a usable surface out of such cuts it is still a bug and confusing to users that it reports the error as non-planars when they are not. CBR

Why is non-planarity 'serious' ? Surely it is expected to some degree in the vast majority of modelling projects, and is often necessary to describe complex surfaces with low poly base geometry. Nor do I understand why that (specifically) leads to light leaks, which are caused by non-contiguousness, and as long as that is airtight, shouldn't be affected by the planarity or not of that geo, unless I am radically misunderstanding what you mean... could you upload a pic or video that shows this being problematic ? If I Line cut even some relatively complex shape into a flat plane I am unable to make it go / show as non-planar until points are actually moved, so that's me confused ! 🙂 CBR