New Figures for Poser any update?



  • @erogenesis said in New Figures for Poser any update?:

    PE has handles to control the breasts. Although I'm not a big fan of cutting the mesh up into even more parts, if the majority say its better to do so, let me know.

    But soft body dynamics? That's a complex simulation and setup. Wouldn't it be faster just to zbrush it?

    Well, for me it would have to be Blender rather than ZBrush, but you are probably right :)

    Amanda



  • @morkonan said in New Figures for Poser any update?:

    @erogenesis said in New Figures for Poser any update?:

    But soft body dynamics? That's a complex simulation and setup. Wouldn't it be faster just to zbrush it?

    Without an engine dedicated to it, then absolutely! If one wanted to simulate such a thing for... chests, one could try regrouping and dynamics, one supposes. But, it probably wouldn't be worth it and that's got too many issues for the model-maker to support in this sort of figure, made for general use. (IOW - Users so inclined can, given willingness to work, do some of those things themselves, so that the general users don't have to jump through unaccustomed hoops.)

    I rarely do in-scene simulations mainly due to the extra complexity involved, and how one is stuck to a fps that would make a 400 page comic a massive pain in the butt to repeatedly simulate (partly to check if it works), especially if I'd want to edit something in the middle of it (or even get 'right' in a scene). Its just way easier to zBrush things in that respect. In the case of a squashed bosom, I could imagine that it would look cool in an animation, but even that can be zBrushed easily since its a rather simple morph to do. However, do people even make actual animations with Poser these days? I only know of three folks that bother to make them over at Rotica, and even those animations are like 3 seconds long usually. Hardly worth the effort, I'd almost say.

    But anyway, if people want all this convex hull stuff, sure, but I'd have to be told how and what. I hope its ok for me to be critical about these things because I don't wanna give myself too much extra work for those few that want to do simulations like this. But yeah, let me hear the arguments.

    Question on the face topo - The nose. Right where the nostril joins the tip, do you think that may be an issue when creating facial morphs? Hard to see, since my eyes suck these days, but it looks like it could "pucker" easily. The same goes for the nostril, itself, getting mushed/blended-in around during deformations.

    you'd have to give me an example of what you mean. In my experience that mesh there would be ok. But let me know what you mean.

    Also, what's the geometry like for around the eyes? The "detail" bits, in other words. And, of course, is the texture map for the lacrimals an appropriate 4kx4k? !! :)

    The eyes are very 'true to reality'. Here's an excerpt from my blog:

    "e.mesh5's eyes have been modeled true to nature. There is an outer skin that flows over into the outer surface of the cornea. The tissue and cornea have own mat zones, but share the same map. Therefore you can use a transmap to define the transition between sclera and cornea.

    There is also an inner skin that starts as the inner surface of the cornea, folds back into the shape of the iris, folds back again to the back of the iris (forming a gap that is the pupil) and then ends up as the inner surface of the eye, or just the retina. Yes the pupil is now a 3D hole. In theory you could try to get the red-eye effect by making the retina reflective and shining a spotlight into her eyes. I haven't tried that yet. There is no lens behind the iris however."

    alt text

    A question I should have asked, if I already didn't, a long time ago - So, about eyebrows... Yes, here it comes: What's your though on separate eyebrow groups vs including them in the face map? Are you looking for cross-compatibility for texture-map conversion for the most common maps (without separate mat group for eyebrows) or have you other thoughts on that? (I don't care, either way, and am fine with no separate group, since it can make certain things easier. But, others may disagree or have firmly held preferences.)

    This is more personal preference than anything objectively throught about but I prefer as little mat zones as possible; so yeah no lip zones, they're bloody annoying. However I know there are a bunch of folks out there that love splitting even the slightest thing up into mat zones so I'm fine with providing a totally split up version for them if need be. That requires very little work.

    PS - And, it should be understood - I am very willing for you to take my money as soon as possible!

    And I am very appreciative of that idea!!!! :D



  • PS: the red eye trick worked haha:

    alt text

    PSS; one more thing. the bulging cornea actually deforms the eyelids with a magnet. I don't have a render of that yet but I'll make one shortly.



  • @erogenesis said in New Figures for Poser any update?:

    But soft body dynamics? That's a complex simulation and setup. Wouldn't it be faster just to zbrush it?

    What would be helpful is morphs of a few typical 'corsetted' body and breast shapes. Add a few also for typical 'bulge' legs over stockings and gloves and you have a morph package that may sell well in all markets.
    Not all morph artists are clothing makers and not all clothing makers are good at making morphs. If the body shape is there with a suitable 'squeeze' building a realistic corset or bodice to fit it relatively easy.



  • @F_Verbaas that's actually a cool idea. I'll make a note of it!



  • @erogenesis said in New Figures for Poser any update?:

    you'd have to give me an example of what you mean. In my experience that mesh there would be ok. But let me know what you mean.

    It's difficult to tell if there's a star there or diamond that might pucker since the geometry there flows, or should, at a bit of an angle to it. (Can't see the pic atm) For the sides, IIRC, it may be difficult to keep them deformed properly during expressions without edge-loops/harder edges so they don't "crawl" around the face or spread. Just my two coppers, will try to find an example when able.

    The eyes are very 'true to reality'. Here's an excerpt from my blog:

    Are there separate lacrimals, "tears" (like in V4) or basically "soft-tissue" bits that reflect the soft tissue around the eyes that isn't properly "skin."

    This is more personal preference than anything objectively throught about but I prefer as little mat zones as possible; so yeah no lip zones, they're bloody annoying.

    It'll be very annoying for people who like to use separate shaders for the lips, like coloring or, especially, increased speculars for "lipstick." Without a separate lip mat, they'll have create masks if they want it and that may effect the rest of the mat group having access to similar effects. IOW- I get ya, but it's a bad idea not to have a separate lip mat group there, in my opinion. (This could also apply to the soft-tissue around the eyes for those who like ultra-closeups. IMO, that's why we see them in popular Poser figures in the first place.)

    And I am very appreciative of that idea!!!! :D

    Credit-card is on "Ready 5" status... :)



  • @erogenesis said in New Figures for Poser any update?:

    you'd have to give me an example of what you mean. In my experience that mesh there would be ok. But let me know what you mean.

    Sorry for the delay! Incoming example pics: (Quick and dirty links)

    1. http://imgur.com/y9muCPX
    2. http://imgur.com/NMeV5UC
    3. http://imgur.com/sEdDsPS
    4. http://imgur.com/tMe4QaD

    Each is a progressively older Poser-compatible mesh. V4/V3/V2 base?/NEA (I have Posette, somewhere, but can't remember where.)

    See how the meshes each have an edge-loop on the outside of the nostril, as you do, but that is further augmented, using various methods, so that there is more emphasis placed on the deformation between where the nostril flesh joins the tip of the nose and/or the ridge? Yours has an edge loop that crosses that, which is fine, as exampled in V4. But, to help shape that area where the nostril/ridge meet, each face topo has some sort of control built-in which will influence how the mesh deforms during things like expression morphs and even during editing for constructing new "character" morphs. For instance, if someone smoothed that area in the other meshes, that region would react differently than its neighbors, simply due to mesh density/topology.

    Without that, it's possible that definition will be lost and, since it's a fairly "rigid" part of the face and, even in extreme expressions, retains relative definition, it's usually protected.

    Some photo-ref expression examples:

    http://masochisticcannibal.deviantart.com/art/LAUGH-Reference-Front-192887792?q=gallery:masochisticcannibal/45347816&qo=8

    http://experimental-magic.deviantart.com/art/My-Facial-Expressions-309394007

    http://zobodahobo.deviantart.com/art/Expression-meme-329239157

    See how that region is preserved during expressions, even sometimes extreme ones? In certain instances, it is partially lost, but there's still a good landmark deformation just above the nostril, where the top of the nostril towards the midline of the face where it meets the ridge is usually emphasized. (Muscle/facial fat/cartilage/structural stuffs) In the last photo ref, we have someone who doesn't naturally have a pronounced "alae" and/or pronounced landmarks between upper nasal carriage and lower. (Ref pic, and not a bad general topo sort of guide thingie, either, which we can see reflected in the topos for the example meshes, for the most part - https://i.stack.imgur.com/SSuyk.jpg ) However, notice how that specific area gets pronounced during facial expressions, even though in her "default face pose" she doesn't have a well defined joining of the carriages.

    *Edit-Add: Descriptions of various protection/definition-preserving methods in above topo examples:

    1. Edge loop terminus and secondary edge loop at a diamond-ish junction.
    2. Edge loops, mesh density, tris (Tri-splits to increase edge-loop density)
    3. Edge loops, mesh density, diamonds as well as tris (A lot of tris)
    4. Edge loop density, tris and a big star as the result

    Edit:Add - Expression "morphs", of course, would be created manually, so they may not be directly, themselves, subjected to this. Combinations may and, certainly, standard control handles would unless their zones are weight-mapped, which is, itself, a different form of customized protection.

    You've got face control handles built it, so that may help. OR, it may hurt, depending upon what vertices are effected. My only suggestion would be to take a look at that area during deformation with morphs and handles to see if it will retain that human-specific facial distinction. Take it and hit it with a smooth brush in 3ds and, even better, check to see how well the nostril retains its shape or whether it simply "blends" in with the tip/ridge of the nose, losing a great deal of definition during native Sub-D in Poser, itself.

    PS - I KNOW retopo isn't a much beloved experience by anyone who's been refining their figure over a long period of time. :) Just take my comments as "just another noise to check on" in the vast engine of your ongoing creative efforts. If you examine it and think more attention might be warranted, that's fine. If not, that's cool too - It's your mesh and I don't consider myself offering any sort of professional, credible, commentary. Just observation. (And, again, with much positive enthusiasm! ;) )



  • @morkonan hey. ok so I read your posts a few times to make sure I understood what you were trying to say, I hope I understood it.

    Honestly, don't think its even remotely an issue. One thing I've noticed in Poserworld is that expressions are the most overrated most misunderstood aspects of poser figures. And this results in artists either overdoing their expressions (almost like Hitchcock-type 50's over-acting), or artists fearing to touch them resulting in totally lifeless characters. While the secret is that: people just don't use their face as much as you think.

    alt text

    Its just a few dials, mostly. Its probably because we communicate with our face that we think that the movements and positions of the features need to do something dramatic, when in fact you're just raising a lip or an eyebrow. A woman's private parts has more 'expressions' than a face, I'm not even kidding. I have 300% more dials for her gens than her face, and that's not because I am a sex maniac.

    alt text

    So when it comes to the nostrils, hardly anything really happens there. Nostrils can widen, they can get pulled up with a sneer, and then a smile can stretch out the nose... but that's really it. So when it comes to geometry, its just a matter of making sure everything gets pulled along either through WM or morph deltas.

    alt text

    Which brings me to the subject of geometry and polygon flow. From what I've seen, 50% of it is just artistic choice, especially with such areas like the ridge you describe (although I have to say V3 looks like utter crap there!!!)

    But speaking of geometry, if there's something I'm a little worried about, and Ambientshade did warn me about this, is that I have zero definition for particular muscle groups like what V4 has. I do have the prominent stuff like the pecs and ribcage, calfs, quadriceps, gluteous, mastoidious, rhomboidious, trapezius, and soforth, but not detailed stuff like the serrator in the abdomen, indivudual six-pack muscles, or smaller muscles in the arm for example. This was a deliberate choice, however, since 1. I prefer girls with curves, 2. I found V4's polyflow g-d d-mn bloody moth-rf---ing annoying most of the time, and 3. you could probably still get the definition you want want with any mesh, if you're creative :D



  • I have noticed that other figures besides the daz ones lack realistic definition ie there should be an impression of an underlying bone and muscle structure. The sm figures had definition but it was very rigid looking giving the appearance of the person being a beetle when muscular was turned right up. If there is no soft tissue definition then the poser figure looks kind of puffy. I think this was the main criticism of apollo max. especially his back which had virtually no definition. Anton, who made apollo argued that he had done a survey of all renders in the gallery and noticed that almost no one did renders of a subjects back. However i make a comic so obviously I need a good back as well as the front.

    For expressions there was a guy called Schlabber who used to be the master of posing back in the old poser 4 days. He taught me that you can apply the expression pose at frame 40 then just wind the dial back until you get the frame where the expression looks realistic.



  • @erogenesis
    It is all a matter of what you need for the task and making design choices to meet.
    Sure unlike some animals people have no muscles in their face to move their nostrils. I think the edge loop at the nostril may come in if one needs to really change the shape of the nose, or want to animate the classic she-werewolf.
    There are plenty other figures that can do this, so why bother to add one to that lot?
    Same with the definition of particular muscle groups. If PE is made for representing saucy young lasses in lascivious situations, let that be where she is the figure of choice. There are plenty other options if one needs an old hag or a weightlifing champion. The time of the 'one figure for everything' is over.
    I like that for example in the adman figures. They are a deliberate choice for a character type that is not the average.



  • I had a second look at the mesh and I don't know if Morks meant that but something about the area around the nostril indeed didn't feel right. And like I said its probably 50% artistic choice. The ploys weren't flowing like I actually would want... so I went and redid it again lol. Its loads better now. I also moved around some polys in the ears because the earlobes weren't smoothing naturally. But eh, i need to leave it at this now... really haha! In any case the mesh is loads better than the original and the UV map is even better. I need to send it off now to my texture people, they've been waiting a bit long :D

    Will post a pic soon.

    PS: I re-read @morkonan 's post again and I think he actually meant what I said in my post! oops. anyway, I think we're all clear now



  • @F_Verbaas said in New Figures for Poser any update?:

    The time of the 'one figure for everything' is over.

    Eh YES! DAZ also gave up on the unimesh idea quite fast with Genesis 2, and Pixar has unique meshes for just about every character they make. And then there's always Poser's new subdivision tech for if we really want to go bananas.

    PE is indeed for making pictures of the average woman, and a bit of muscle or age or fat can definite work on her (There's enough polys in the face to get her to look pretty old), but she's not meant for a full blown cave troll, ork, or Miss Steroids Universe. That's what Genesis is for...



  • @estherau said in New Figures for Poser any update?:

    I have noticed that other figures besides the daz ones lack realistic definition ie there should be an impression of an underlying bone and muscle structure.

    I took that a step further and modeled her bent. The original mesh is actually in fetus position. When the model was complete, I straightened her out to a T-pose. So her knees and elbows should be fairly well defined in bent positions and there's hardly any texture stretching as a result.

    For expressions there was a guy called Schlabber who used to be the master of posing back in the old poser 4 days. He taught me that you can apply the expression pose at frame 40 then just wind the dial back until you get the frame where the expression looks realistic.

    hehe that's it yes, although now you wouldn't have to do the whole frame thing, but just dial it in until you're happy. Along those same lines, many of Aiko's morphs are VERY useful at low values, in fact, most of Lali's expressions up there in the green top are Aiko's


  • Poser Ambassadors

    Those expressions are wonderful!


  • Poser Ambassadors

    @erogenesis said in New Figures for Poser any update?:

    The original mesh is actually in fetus position.

    Well now, that's something I've never heard before. Mind if I ask what prompted that approach?

    And when you say you straightened her out, do you mean prior to rigging?

    I do admire your technique with facial expressions. They're some of the most realistic I've seen among all the figures. Did you achieve them by manipulating the stock expression morphs that come with V4 or are they custom sculpts? I'm assuming - judging by her shoulders - that the example images up above are from your earlier custom V4 version of this character, correct?



  • @AmbientShade said in New Figures for Poser any update?:

    @erogenesis said in New Figures for Poser any update?:

    The original mesh is actually in fetus position.

    Well now, that's something I've never heard before. Mind if I ask what prompted that approach?

    I sometimes have to work with bends like -160 degrees, so with V4 the mesh stretched horribly, including the testures. So I thought it would be better to model the mesh with the bends 'built in' so that all bends in either direction don't result in crazy polygon shearing. So if you'll bend PE's knee to 90, you'll see all the ploys line up very nicely. She was built for bending, basically.

    And when you say you straightened her out, do you mean prior to rigging?

    yep

    I do admire your technique with facial expressions. They're some of the most realistic I've seen among all the figures. Did you achieve them by manipulating the stock expression morphs that come with V4 or are they custom sculpts? I'm assuming - judging by her shoulders - that the example images up above are from your earlier custom V4 version of this character, correct?

    thanks :) they are a combination of V4 standard, Aiko and my own morphs. And yes the one on the green top is my early V4 rig (and that top and jeans are also V4 mesh extracts because I couldn't afford content at the time lol. That was back in 2012



  • @Biscuits said in New Figures for Poser any update?:

    Those expressions are wonderful!

    <3 :)


  • Poser Ambassadors

    The default for Poser figures has always been to model in "T" Pose resulting in the what we have called over time "The Poser shoulders" when bending the arms down.

    A later alternative that was less accepted (because all content was already in the "T" pose) was to build in the "A" pose to get partially rid of the mesh and texture stretching. => "Little One" came in an "A" pose.

    Ero took it a step further, and I suppose he build and UV_Unwrapped his figure with the arms down, the forearms bend, and the legs in a sitting pose.

    Well, UV-Unwrapping in a "bend" pose is something I have not done yet, but a very good method to get rid of texture stretching when bend.

    Good thinking.

    Then "unbend" and rig in a "T" pose or close to V4 to get the V4/M4 clothes to better fit.



  • @erogenesis said in New Figures for Poser any update?:

    @morkonan hey. ok so I read your posts a few times to make sure I understood what you were trying to say, I hope I understood it.

    Honestly, don't think its even remotely an issue. One thing I've noticed in Poserworld is that expressions are the most overrated most misunderstood aspects of poser figures. And this results in artists either overdoing their expressions (almost like Hitchcock-type 50's over-acting), or artists fearing to touch them resulting in totally lifeless characters. While the secret is that: people just don't use their face as much as you think.

    You're absolutely right - Most expression morphs are "caricatures" of expressions. In fact, almost universally, I have to dial-down any expression I apply to a character, especially any that involve a customized face. They're almost all too gross in their depiction of facial deformation.

    If you're content with it, then that's fine - It's your mesh! You're putting in a lot of great work, so it's not my place to suggest you take on even more of it!

    ... A woman's private parts has more 'expressions' than a face, I'm not even kidding. I have 300% more dials for her gens than her face, and that's not because I am a sex maniac.

    :) Well, that depends on whether or not you took the trash out this morning, like you were told to do... Honestly, though, the human face is quite remarkable and it's a critical component of human communication. Even ever-so-slight variations in expression, even if something looked upon like a "single component" of the face, can communicate a great deal. ("Universal Expressions", something like "Universal Language" is a "thing", at least as a popular topic. It's not an unquestionable theory and there are certainly exceptions. But, for a great many expressions, there appears to be cross-cultural analogues that are recognizable, at least.)

    In any event, without facial expressions, it's darn unsettling for a human to attempt personal communication. We "need" all those ticks and jumps in expression in face-to-face communication. (ie: Any standard in-game model facial expression will write that particular tale. Feature presentations, however, have very different standards. We're sort of dealing with Poser as a "feature presentation" still/animation tool when it comes down to meshes/definition/topo/animation/etc. At least, one that is available for public mass-consumption, at any rate. I can't run Weta's studio on my poor machine... :( )

    But speaking of geometry, if there's something I'm a little worried about, and Ambientshade did warn me about this, is that I have zero definition for particular muscle groups like what V4 has. I do have the prominent stuff like the pecs and ribcage, calfs, quadriceps, gluteous, mastoidious, rhomboidious, trapezius, and soforth, but not detailed stuff like the serrator in the abdomen, indivudual six-pack muscles, or smaller muscles in the arm for example. This was a deliberate choice, however, since 1. I prefer girls with curves, 2. I found V4's polyflow g-d d-mn bloody moth-rf---ing annoying most of the time, and 3. you could probably still get the definition you want want with any mesh, if you're creative :D

    One of the hallmarks of Poser is, even if we don't always like it, mesh-morphed characters. This allows users to get a great deal of use out of their characters and allows artists to create interesting characters/products for those meshes. A "curvy girl" may not need topology that supports "lean muscle", but a character morph for that mesh might very well need it or the morph-creator may need to do a great deal more work to create a custom morph.

    So, for instance, if anyone wants a "thin" or "svelte" looking character, can they easily use your mesh to create that? Yes... no... maybe? As you correctly point out, you might want to consider your mesh's utility for certain things. It's sort of like the popular claim "I can recognize xxx Poser mesh anywhere, no matter the morph" sort of thing. Well, why is that? Often it's just because morph creators usually don't make elaborate/extreme morphs, just push a few verts and call it a day, leaving a lot up to the texture. This isn't "bad", by the way, since all the other junk, included deformers/jcms, expression morphs, body morphs, etc, all rely mostly on the mesh not deviating too much for any custom morph in order for the final effect to "look good."

    I agree, V4's mesh can be aggravating. But, for a very versatile mesh that is very capable of portraying a broad range of body-types and customization, it's pretty darn good. There are issues, here and there, especially with some unnecessarily cramped geometry, (or even absent bits) that make certain types of work problematic. (ie: Trying to smooth out an area and puckering or linear-deforms start to crop up, unwanted, little density for certain deforms, yada yada.) But, overall, it's nice for a low-budget, hi-d mesh.

    That's another issue with some meshes - Low-d vs hi-d and "intention." The sorts of stuff many people do focuses on hi-def work. For that, a "game model" intended for easy and low-profile game animation will not work well. With Sub-D incorporated into Poser, "any" figure mesh can be "hi definition" but that's really only in polycount - Without the information there, already, the sub-d isn't going to "add" definition. And, I suppose that's what I was sort of thinking about when I posted. :)

    No criticisms here, just rambling on, I suppose. When one is enthusiastic about something, one tends to pay more attention to it and is more willing to nit-pick or theorize about "details", I suppose. :) So, if I didn't have plans of rushing out, buying your figure on release-day, and then doing it an unspeakable injustice by daring to customize it, I wouldn't be posting about topology... :)



  • @morkonan said in New Figures for Poser any update?:

    "I can recognize xxx Poser mesh anywhere, no matter the morph"

    It is interesting, but many morph creators almost do not touch/do not change from morph to morph most critical in recognition region - eyes orbits - from top of brows to tip of nose. Especially zygomatics (cheekbones).