FVerbaas posted at 7:07PM Fri, 20 August 2021 - #4425715

The two on the left look all right. The two on the left have too sharp folds as they are, even for shapes held in a bra.

Completely agreed! Also too balloony, assuming that's a word. :-) I've been tweaking that morph a lot more in the meantime.

(I assume that as a westphalian you prefer your feedback straight? If not forgive my Dutch honesty)

Absolutely!

Of course one must take into account that I'm also Andalusian, whatever that implies. :-)

That on average you are a Gaul ;-)

:-D

New version of the breast size morph after a lot of tweaking. I think I'm ready to move on now. :-)

Better.

Seriously!

It's the crucial difference between "does not look like a bag of beans" and "does actually look decent and somewhat realistic." :-) Anyway, I'm still tweaking. Someone pull me away from the computer!

I like more of a natural look for my clothing racks. Less silicone. Also with rather low poly models like this one:

adp001 posted at 11:32PM Sat, 21 August 2021 - #4425823

I like more of a natural look for my clothing racks. Less silicone.

Yep, working on it. That's a cute model, but the way. Someone should make her a sandwich, though, those breasts won't carry themselves. :-)

My plan with the breast morphs was to do a basic one for size that looks reasonably organic - not screaming silicone, but not necessarily representative of the average large natural breast - and have separate ones for sag, areola size and so on. Also I've set myself the challenge to make a morph that works at a large range both negative and positive, which really helps me identify problems, but also limits the possible shapes a little bit. If I make them sag realistically at strength 3, I'd have to lift them up a bit more than I'd like at strength -1, a.k.a. flat chested. The ones posted above are still way too bulgy at the top and sides, which I've fixed in the meantime.

My opinion on this is that the typical concept doesn't really work. The problem: If you have a "basic shape" (size, position, mass distribution, etc) that you like, and it's not a mini size, then you better not move the model. Because no one can build such universal morphs that will preserve that particular shape, when the model bends forward or backward, moves its arms or even lies down (damn physics). What comes out of this can be admired on many sites (although I often ask myself whether those who create these images, which are contrary to any reality, have ever seen a living model).

My approach would be: morph sets that are based on sizes, shapes and masses and allow movement in all directions. Each specifically tailored to the "floor plan". Per set, not "one for all", because that only works in theory.

I'd have to agree with ADP here, although I think it's useful to find your own formulas for doing things. I have a bit of experience making breast morphs sets for V4, my Missus figure, La Femme (Eldritch Breast Morphs, which is in the process of getting an overhaul and update to be included with a larger scope project), and Sera.

Here's the morph set for Sera, included in my Sera's Secret project. I take an overkill approach to morphs, with breast morphs ranging from non existant, realistic ranges, to hyper ridiculous hentai style, but this set animates REALLY well if you take the time to do it.... Don't have the time or inclination to post any images, that can wait, but here's the dial arrangement. Might give you some ideas...

I was planning to also make a set for Antonia to go along with the other things, but that's taking a back seat to a more near completion La Femme project for now. It's interesting to watch your progress on this though.

I'm not sure what the typical concept is that doesn't work, but I'll take your point that animating large breasts realistically is difficult. Coming up with good morph sets is also extremely difficult, I find, but maybe that's just coming up with morph sets that I personally can use. When I see enormous lists like the above with numbered variations, I get lost and will simply give up. My brain just can't deal. And those are very typical in the Poser world, which might be the reason I have a tendency to just make my own ad hoc custom morphs for whatever I have in mind at the moment. There are enormous morph sets for Antonia that I've never even looked at because I get intimidated by the sheer numbers.

That said, apart from the numbers, primorge's list up there actually looks pretty well-structured and logical. I can see myself getting something done with that after a bit of practicing, unlike other morph collections I've seen.

You're right, there are a lot of comprehensive morph sets for Antonia. I haven't looked at them yet though. Just the ones that ship with the figure. Like I said earlier take forum advice with a grain of salt in terms of aesthetics stuff and follow your own vision. The technical things might be more prudent to heed, it's objective. As soon as you started posting breast morphs I really wanted to comment, mostly about how it was going with the topology. I haven't tried morphing anything other than Antonia's head and head parts as I'm still working on the textures but I did notice the non typical way the boobs were modeled and worried about it a bit in terms of morphability. Guess I'll cross that bridge when I come to it, as you are now. Looking pretty good so far, especially after the revisions. It would help if you can morph "Post Transform" for lack of a better word, that is morphing over top of a morph that's treated as a default shape. You can do this with ADP's script, PML, or GoZ... and I'm pretty certain with the morph brush. There's a way to do it with iterative subtraction and reapplying results via spawning props/export import also but that's a PITA. Anyway, interesting stuff...

As I said above, putting poles on the breasts and so close to the areolas proved to be an enormous mistake (no pun intended). It might not be quite as bad if one does the morphs on the standard resolution mesh, but going low-res as I prefer to for larger scale shaping is hellish. The few polygons I saved there are definitely not worth the price. Can't wait to see how you'll be able to deal with it.

Flatten brush in mudbox I imagine. It's really the only way to deal with high-valence vertexs, smooth makes them worse. Causes pinching artifacts. In the middle of refitting Ali's HR 224 to La Femme, snugging the Skull Cap...

Oh, the ones on Antonia's chest are valency three, not five. I don't recall having any really badly placed valency five vertices, although I may be wrong. The ones on the upper lips should probably be closer to the nose, but they're not completely terrible. I've avoided six or higher like the plague.

I'll have to take another look at her boobs, I don't recall it being horrible. I think my main concern was not as many continuous concentric edge loops to slide and make adjustments with, which I kind of rely on for fixing texture stretching adjustments and scaling of the areola and nipples. There's some road blocks there.

Just about finished the refit. Needs a couple simple morphs to bring down the hairline over the forehead and ears but otherwise a completely perfect refit. Painless to bake everything down into a single morph for an inject, it's a cute hair model. I like that long straight hair behind the ear look... time for dinner, some music and then bed.

Ah yes, I wasn't particular generous with the edge loops, that's also true. Not as crucial to the way I usually model, I guess. But again, one or two unbroken loops around the areolae would have been nice in hindsight.

Alright, I'm reasonably sure that this is my final word on this particular morph. I changed the strength scaling so that now -1 is completely flat. Values shown are -1, 0, 1, and 2. For my next trick I think I will - inspired by adp001's feedback - attempt to make some lie-down and bent-forward morphs.

@odf: Looks good! However, the number 4 does not look as realistic as the number 3.

The size of the areola is a problem with all Poser models when a certain breast volume is exceeded. The only thing that helps then is sensitive vertex pushing. And don't use the Smooth brush afterwards. Then all the work is gone again... I helped with the low poly model shown above with a modified UV map. That's the advantage of not having to build universal morphs :)

Some of the Antonia morph sets have areola size morphs, but of course as with any morph they are deltas relative to the zero-ed mesh and results may be off when posed or morphed. Magnets are another option but the magnet base and fied do not move with the vertices. There are so many factors now that influence the world xyz of vertices that is is difficult to achieve consistent results.

I am thinkng about ways to use direct references in geometry, like vertices by their sequece number. This is of course not direcly supported by Poser. One would need to keep a sort of registry and have a Python method that iterates over the refereces and make the updates, in this case that would be moving the magnet base and field to the average postion of a set of vertices. In theory tis could be a callback function, but that would be like calling for slowness and trouble. Better be it called by the user if he wants to synchronize things prior to render.

Callback functions work without problems at least in my Poser 11. I also had concerns. But based on experience with P9/10.

I use a script to feed a MQTT server. This monitors each dial with a Poser callback and passes changes to the server in quasi real time. And the other way around.

No matter what I do (delete morphs, remove actors): Nothing happens. Everything continues to run.

adp001 posted at 6:30PM Mon, 23 August 2021 - #4425875

@odf: Looks good! However, the number 4 does not look as realistic as the number 3.

Agreed! I can dial that morph up to strength 4 without obvious artifacts, but of course the breasts will look increasingly like floatation devices. Once I have a few shaping and gravity morphs, we'll have to check again how far we can go in the direction of an Angela White clone. :-D

Back in the day folks were discussing things like breast bones for getting around the limitations of linear morphs. Does anyone know if that has been done successfully for any figures? Just out of curiosity, mind you, I'm not working on a gargantuan list of future Poser projects here. :-)

@FVerbaas I have made and deleted two areola size morphs, so far. It's tricky. I really like to play with areola sizes, but in order for one morph to work on a large range of breast sizes and shapes, one has to be careful not to change the orientation of the areolas too much in all those other morphs. But of course the surrounding curvature has to change, so there's a limit. The best strategy I can think of without having tried anything in practice yet is to scale up the areola on the flat-chested Antonia and have an extra bulge morph that can be added in when the first one flattens out the area too much (and of course can also be used on its own).

But obviously I'm just tinkering and imagining here. Just an amateur having fun playing with polygons.

PS: No, actually scrap that. Scaling up the areola on the flat chest will make them protrude when there's curvature. Adding a bulge might even that out a bit, but in order to fix the protrusion one would have to pull the areola inwards. Might not actually be that big a problem in practice, so I'll just wait and see.

Import it into wings and make a series (not just one) of edge slide morphs of varying degrees, ranges and strength. Apply the lot and fiddle with mixtures and bake the results that work for what you are looking for. If you mess with scaling on the boobs quite a bit it's useful to have more than one version of an adjustment morph dial available to make finer adjustments when the limits of the initial dials usefulness reaches it's apex for a particular scale. This probably doesn't seem to make sense but it's just a practice I've adapted from my personal experiences with this. Sculpting tools are limited for such things, you want to slide loops if possible.

primorge posted at 7:07PM Mon, 23 August 2021 - #4425910

Import it into wings and make a series (not just one) of edge slide morphs of varying degrees, ranges and strength.

Excellent idea! Why haven't I thought of that myself? [bigthumbsup]

I'd probably use Blender, though, these days. My Winxpertice is completely gone.

That's what I meant in my first post. The traditional methode works up to a level depending on shape and volume. So better make different "bases" for some shapes along with fitting morphs to move that mass. Easier at the end. And you don't need tons of morphs (to fix problems caused by the last morph) and time for trial and error :).

If you need to rotate vertices, you will be better off with bones. An underestimated solution is to use deformers (magnets). Very useful, especially the weightmapped ones. At the Latest when your figure moves its arms or shoulders forward/backward, there is rotation. The more volume, the more visible rotation. You can catch this up to a certain degree with a linear morph - considering the effort, a bone is more effective.

So let's say you have a areola size morph that you created on the default scale. So you decide to inflate the boobs up to some large size. And you realize that you want to scale the areola down back to its default size while leaving the breast still large and not changing the shape overly much. So you spin the areola scale dial down and you realize it's not working quite the way you expected. The areola is still huge. If you have another adjustment dial it makes things alot easier. You can mix varying strengths, make finer adjustments, and sliding loops is the best way to do this and not distort the overall breast shape. It's one of the reasons you see numbered morphs in the set I showed.

It's extra dials, yes.

You can also fix stretching of the UVs in either direction with slide dials.

It's not a perfect solution; it's fiddly and might be confusing and cause some nasty results for a user that doesn't experiment or know what it's for. But for me it's the best solution I've found for the problem. Considering that your morph isn't ridiculous it should be a pretty easy process.

I think fiddling with large, convincing gravity undercuts on poser figures is far more challenging.

odf posted at 7:45PM Mon, 23 August 2021 - #4425913

primorge posted at 7:07PM Mon, 23 August 2021 - #4425910

Import it into wings and make a series (not just one) of edge slide morphs of varying degrees, ranges and strength.

Excellent idea! Why haven't I thought of that myself? [bigthumbsup]

I'd probably use Blender, though, these days. My Winxpertice is completely gone.

Either or. ?

@primorge Yep, that make sense and explains the numbered dials. I'd probably try to avoid the numbering and find some more descriptive names, but I know that can be very tricky.

A typical case of I have to run into the problem first and think it through before I can appreciate the solution. One of the reasons I had such a hard time studying maths.

Maybe include something like "tweak" or "adjust" in the morph names to make the purpose more obvious? Or introduce a "Breast adjustments" group?

This is going to sound a little weird but I'll give you a "for instance" example. Let's say I wanted to use morphs/ blend shapes to animate a convincing breast bounce jiggle. If you look at the breast up down dial 1 it's a range of influence encompassing the entire breast and a bit of the skin surrounding the breast mass. With breast up down 2 the falloff of influence is tightened to encompass more toward the center off the breast. With up down 3 it's tightened further to encompass just the nipple and surrounding flesh there. Now if you take that and a set of side to side or swing left right morphs with similar fall off ranges and you mix that all together meticulously in animation frames you can achieve a realistic undulating bounce with the effect of gravity pulling in varying degrees during motion. It's the reason for the numbering in the movement morphs, it indicates falloff or degree of force.

I have a similar set up for buttocks too lol. But that's another tale :D

@primorge Yep, that makes perfect sense. Thanks for the explanation.

Oh emm gee, what if all numbered morphs follow a logical schema like that and I've just never been smart enough to notice? :-O

I don't think so. Maybe? I kind of got my morph conventions from morphs ++ conventions... probably like most people have really. The numbering I just use because it seems the most immediate, takes up less space. I figure the user will figure it out or I'll make notations in the readme/documentation. The things I'm making for La Femme will need some explanation for sure. There's morphs that only work correctly with other particular morphs as a pre-requisite, that sort of thing. I've taken to using matching asterisk multiples. Morph parameter custom groups are helpful too.

For shaping stuff that's more self explanatory I use ADJ and Fix.

I meant maybe there's a pattern, not maybe you're not smart enough lol.

primorge posted at 2:12AM Tue, 24 August 2021 - #4425931

I meant maybe there's a pattern, not maybe you're not smart enough lol.

Yeah, I got that. No worries! :-)

Solution would be to morph or deform in uv space.

FVerbaas posted at 3:38AM Tue, 24 August 2021 - #4425939

Solution would be to morph or deform in uv space.

Do you mean the solution to texture stretching?

I thought it would be nice to have a weight-mapped low-res Antonia, so I wondered how easy it would be to transfer Diogenes' rig to the low-res mesh. The answer is dirt-easy. All hail the fitting room! The result looks almost perfect, as far as I've checked, with one or two "small" exceptions. :-) (I think the problem is that propagatingScale is translated into explicit dependencies, but both actors are at 100%.)

odf posted at 3:42AM Tue, 24 August 2021 - #4425942

... The result looks almost perfect, as far as I've checked, with one or two "small" exceptions. :-)

Yes there is a 'scale' parameter playing havock. Ususally appears also on hands.