I'm not being mean, but they all look like plastic cartoons to me. Nothing remotely real in the group.

As you say, a picture = 1000 words.

What were you thinking Basic? Am I missing something?

edit*

yeah there's only so much one can do with 8K polys. they make for decent low-end game res characters though. but even games are using upwards of 60K now. Maybe not quite that high yet but getting there. 

~Shane

I'm commenting on vilters' post.

Where's the picture? LOL

edit: Ok, I had to refresh three times to see the image.

Laurie

ED...

if you were talking about my response to your post...

I had always thought more polys=more detail and better morphability.

Am I the only one who can see vilters post of a picture with six figures in it just above me "not being mean" post?

I can certainly see the difference between the outer figures in Vliter's render and the inner.

So Vilter's is using "low level" morphs of the cruder low-poly meshes here, in combination with Poser's smoothing feature... am I understanding right? I think I can see where he is going with that... and where that could have its uses... sure.

ExistentialDisorder, relative to your excellent explanation of subdivision etc... how does Poser's smoothing feature fit into this equation?

ah, yes you would be correct.

but subD's are just as morphable. the only real question is their practicality. There's a much steeper learning curve to them which doesn't usually equate to their benefit, and they aren't really used a whole lot in the "real" world, by most artists. 

~Shane

...and how does this stuff figure in terms of taking, say, a V4 head into ZBrush, modelling it there and bringing back into Poser?

This is an aspect I don't quite get yet, I think....

...as in, does ZBrush introduce new polys?

somehow I thought vilters was talking about subDs vs low poly... ? wasn't he? at least a few posts back?

I can see a good bit of improvement in his image in the female. I don't see much difference between the males but he said it's still a wip. 

it's still the same poly count either way, it's just arranged differently. he would get better shape and flexibility at a slightly higher resolution.

i can't say how poser does it specifically, but generally, smoothing during rendering is just increasing the mesh resolution. it all gets triangulated when you render an image. (quads become tris. this is why game characters use tris instead of quads). whether or not poser increases the mesh resolution and THEN triangulates, I can't say, cause I don't know the details of how poser does it. I'm sure someone else can ansswer that. 

zbrush only introduces new polys if you divide the mesh in zbrush, but if you do that, you will "break" the mesh. It does something with the vert order in the mesh so even dropping back down to original level before exporting back into poser won't fix it. your morphed head will "explode" in poser if you divided it at all in zb.

usually, if you're comfortable enough in zbrush, you can morph the mesh without adding more resolution to it. otherwise, duplicate the mesh first, divide and morph one copy, drop down to its lowest/original subdivision level, and then project your morph onto the original, un-subdivided mesh, and export that out of zb and back into poser. but that's kind of an awkward way of going about it. 

~Shane

Ah, yes... that would make sense, regarding the triangulation during the render process.

But quads are mostly better for bending / posing right? ...and for defining the lower level shapes to start off???

Except with the Poser cloth room? Think that was looking for triangles ideally?

I have been starting to do some modelling... and have been sticking to quads, and transferring those within the exported objs into Poser, as I'd read that was best... somewhere... although I'm not modelling human figures... and am not anywhere near trying that, yet.

Anyway... this is going wayyyy off topic... sorry!

Right now, I'm just looking forward to the folk who already have this stuff in the bag, putting out some great new male figures ;-)

i think there's something wrong with the way i explained that smoothing part. feels like i'm missing something. Lack of sleep maybe.

~Shane

I am in bed now, but I never thought I had to explain the purpose of the picture.

YES, the figures are LOW RES at 8.000 Poly's.

But?? Do you see any straight lines between the vertices in the 2 middle rendered ones?
No, Posers smoothing option took care of that.

Secondly; to lower the breasts of the female figure by that amount with a high res mesh, without getting into trouble would be quite an achivement.
And even with the lowering I showed, the mesh still bends and twists normally.

@ mc
A morph, MUST have the same number of vertices. You can take the head in whatever app you have to morph, but you can not add or remove any vertices.

So yes, i use Lower Poly meshes, morph them, and use Posers smoothing to take out the roughness of the meshes.
And YES, i go back and forward between Poser and the morhing app to fine tune untill I have a reasonable morphed mesh and shape while maintaining the bendability and the twistability that I constantly check during the process.

Main purpose of the picture was:
You do not NEED a high Poly mesh so get a SOFT surface.
Posers smoothing does that for you.
And the Smoothing is fully controllable by setting a specific crease angle to each group.

You can set the smoothing level HIGH (180) if needed, or you can leave it a default (80) on groups that do not need the smoothing.

Fully adaptable and controllable, bringing even a LOW poly mesh to a soft status.

@basicwiz

Your comment of "plastic cartoons" was the biggest compliment you could give.
And you did not even know it :-)

You did see that only the inner 2 where area rendered?
The outside figures where dafault?
The others only had the shader but where not rendered? => Showing off the blinn node?

Default, morphed+shader," rendered, rendered", morphed+shader, default.

@ ED

yes, you are seeing what I wanted to show.

Cloth room;
Quads give you a harder cloth like leather or so.
Triangles are better for soft cloth like silk.

But here, poly density is of bigger importance.
= More Poly's = softer flowing cloth. => At the cost of longer calculation time.

quads are always better than tris when modeling organics. if you need to triangulate for whatever reason, wait until you've finalized your model. of course sometimes tris are unavoidable in certain areas depending on what shape you are trying to achieve. in that case, try to hide them in areas like ear holes, nostrils, crotches, etc. anywhere that is least likely to be affected by animations. but if you're planning your topology then you can almost always avoid tris. 

if you're building hard surface models, then it doesn't really matter if you use tris or quads. it may even benefit you to use tris, as that is less geometry, and optimized geometry makes your software - and your art director - love you.

~Shane

Hi, estherau,  I am working in him , making the genitals and ending the materials and some of the morphs.

My male figure has, without genitals, around 36.000 polys.

Here a wip render with a jumping pose and 0.5 of the strong morph, with a 0.6 of the disgust expression.

Quote - But?? Do you see any straight lines between the vertices in the 2 middle rendered ones?
No, Posers smoothing option took care of that.

That's a render setting. I'm talking about actually morphing a figure, not rendering it. The more polys you add to a surface the smoother that surface becomes - or has the potential of becoming. 

More verts along an edge is what allows a straight edge to bend into a curved edge. Without more verts that edge will never curve. 

i'm not sure why you think that more polys would make it difficult to adjust the shape of the figures breasts, or any other body part. That just defies the logic of 3D modeling. But I don't know what your workflow is, so maybe it has to do with the way you create your morphs. It's definitely not the norm. 

~Shane

adh3d I like your guy so far--good work!  :0)  Not being a modeler myself you guys are loosing me with all the talk about polys and lo and hi res--etc.  I f I'm reading correctly some people want low poly figures for some aspects of their artwork--maybe a crowd where the camera is not going to be looking at them close up without gaging their computer.  For portrait work one would not want anything too low res though. 

Frankly I don't care abouit how many polys something has, the only thing I care about is whether a figure looks good or not and if it is something that I'd want to use to make a picture.  All this modeling talk is beyond me.  For the record I like the Genesis figure, but I want to be able to continue to use Poser and I don't think that Daz and SM are going to come up with a viable option to satisfy everyone.  Daz had the best looking male figures for Poser, but now that Genesis has come along we need alternatives if we want to use Poser for our work--and I don't want to be stuck forever with the Generation 4 of Daz's figures and not have anything new to work with.

Hi, he looks really good.  I think his wrists taper a bit much.  it was M3 puny wrists that I never used to like about M3.

Love esther

Quote - somehow I thought vilters was talking about subDs vs low poly... ? wasn't he? at least a few posts back?

 

I can see a good bit of improvement in his image in the female. I don't see much difference between the males but he said it's still a wip. 

it's still the same poly count either way, it's just arranged differently. he would get better shape and flexibility at a slightly higher resolution.

 

i can't say how poser does it specifically, but generally, smoothing during rendering is just increasing the mesh resolution. it all gets triangulated when you render an image. (quads become tris. this is why game characters use tris instead of quads). whether or not poser increases the mesh resolution and THEN triangulates, I can't say, cause I don't know the details of how poser does it. I'm sure someone else can ansswer that. 

 

zbrush only introduces new polys if you divide the mesh in zbrush, but if you do that, you will "break" the mesh. It does something with the vert order in the mesh so even dropping back down to original level before exporting back into poser won't fix it. your morphed head will "explode" in poser if you divided it at all in zb.

usually, if you're comfortable enough in zbrush, you can morph the mesh without adding more resolution to it. otherwise, duplicate the mesh first, divide and morph one copy, drop down to its lowest/original subdivision level, and then project your morph onto the original, un-subdivided mesh, and export that out of zb and back into poser. but that's kind of an awkward way of going about it. 

 

~Shane

That's not ENTIRELY accurate. While most poser figures will break (and currently any figure exported from Poser based on my tests), the actual OBJ files of Ryan and Alyson and James Johnson and Miki 3 should all stay welded and not break when subdivded in ZBrush.  You may lose the Group names on export from ZBrush but UVMapper can get those back as long as you work on a copy of the original obj and not the original itself. I do this all the time. :)

Vilters' illustrations show how brilliant he is at making optimum use of limited polygons.  Those extremely Lo-res figures are useless to me because I don't have the skill to use them and I don't have the inclination to learn it, but I have great respect and admiration for his tour te force. Our goals and needs are different, and efficiency should be measured relative to achieving a goal.

As for Vilters, it is about efficient use of polygons, for me it is about efficient use of time. I have a powerful computer and limited time. I like variety, I hate it when I make a group scene and the figures look too similar. So I use different models for group scenes, including the Daz Gen 3 RR meshes and the Poser 4 figures. Around 40000 or slightly less polygons gives plenty of scope for variation, but I don't hesitate to use the G2 figures either, and they have over 100 000. For a main character in a scene or a story the ability to produce subtle facial expressions is critical to me, and for that you need polygons. I don't mind making the morphs in Blender, but I can't do it when the resolution is too low. "Edges don't bend". Exactly! I gave up on trying to use Posette as a main character for that reason.

I look forward to the new figures. I love Bella, and I am very impressed by the look of ADH3D's new male. If PhilC supports a male figure for WW and OBJ2CR2, I will use it.

I think Vilter's point is that if you had displacement mapping you can get a kind of micropolygon look and get any amount of detail, but I'm not sure how that would work with bends and morphs, and also none of these displacement add-ons are routinely made commercially. I use a lot of commercial add-ons as I want to spend my time making my comic and not making the things to put in the scene for the comic.

Love esther

you should never use more geometry than what your model requires. that's what optimization is all about. 

if you can make a low res figure do what you need it to do then that's always a good thing, because lower resolution means less processing and faster rendering. 

there's no reason for a figure to weigh in at 100k polys when you only need half that or less to acheive the desired results.

so there's nothing wrong with using a low res mesh, if that's what you prefer or require for your purposes. the only real issue is that in order to achieve a higher level of anatomical detail, more fluid bending for posing and animation, etc, you need a higher resolution. 

Teyon - my use of the term "breaking" was in regards to exporting a single object - like a head - and then getting the "exploding head" when applying the morph.

have you tried pulling a model directly from the geometries directory for the figure instead of exporting? this seems to be the desirable method to retain poser groups. ZBrush will still see all your groups as polygroups, so you can hide various parts you don't need within zb. Just apply in poser as a full body morph. From that point, you can break the morph up into individual bodypart morphs for more control, if thats what you need. Using this method, I've never had a problem with any poser figure I've tried to morph. I don't have UV mapper so I don't know how regrouping in that would work.  

~Shane

Quote - > Quote - Forgive my probable naivety on this matter... ...but it would be good if each figure came with both a high poly and low poly version perhaps??

If that were the case, could they share the same bones?

DAZ has done that.  They had "RR" (Reduced Resolution) versions of V3 and M3, and maybe others.  The RR versions could use the same clothes and textures as the regular versions.

And yes, I think it was a good idea. 

If it is the case that reduced resolution figures would be interoperable with clothing etc. for their hi res counterparts, then it seems like a good idea to me if this was done as a standard?

I guess V3/V4 were using older bone systems though and weight mapping isn't in the equation...

...would this still be possible in the latest figures, I'm terms of rigging, and how easy is it for  figure creators to make a low or mid res version of their mesh?

I agree Esther. My story requires me to make a lot of things for the scenes and I enjoy that, but If I can use something ready-made and it is affordable, I'll do it.

Quote - "However, is weight mapping not reasonably beneficial to posing figures for animation?

I had thought this was one of its merits perhaps? e.g. more fluid movement?"

Hi Weight mapping has nothing to do with animating the character that is more a function of how well the Ik solver functions and the applications keyframe editing abilities.

The best system of animation is a NON linear one where you can smoothly mix different motions
Here is an Example with the nonweightmapped v4
MOTION MIXING

Cheers

Ah, many thanks Wolf... a whole other dimension for me to explore there.... cool ;-)

Quote - I agree Esther. My story requires me to make a lot of things for the scenes and I enjoy that, but If I can use something ready-made and it is affordable, I'll do it.

Me too, absolutely.

This, for me, is a substantial portion of the value of these 3d staging / rendering software tools (Poser and Vue being my current mainstay ones)... the ready-made content... an integral part of the time saving functionality I need to allow me to portray an idea in the limited time I have ;-)

@Teyon

I tried to use the Alyson Lo Res as my Low Poly figure.

Unfortunately she has a welding problem at the back of her neck.

She was my hope on a somewhat higher Low Poly figure, but NO GO because of the neck welding problem.
Such a shame.......

Click to enlarge

Vilters...

I guess I just don't "get it."

Nothing you have posted is even in the ball park of having any realism about it. Why are you putting this forward as being the solution to the problem under discussion? Everything you post seems to add support to the idea that more polys are needed to get a realistic figure.

What is the point you are attempting to make that continues to go over the top of my head?

About 40.000 vertices is the lowest threshold to depict the human body as a Poser mesh with some semblance of realism.

Thats not just an "opinion" taken out of thin air, but the result of 12 years of optimizing and improving meshes for Poser.

If you want to get lower, you need displacement maps which at least in Poser defeat any slight performance advantages you might have achieved by using fewer vertices.

8 to 15K meshes "as is" are only good for background "decoration".

For a convincingly human form, you need more detail.

Especially for a mesh that has to morph between different ages, bodyshapes and genders like a Poser mesh has to do.

Of course you can take a 4000 poly roughly human shaped blob and then paint an uber-detailed 10.000 x 10.000 displacement map in ZBrush and make it look like a 100.000 poly mesh, but that is just not practical in Poser.

People always forget that Poser is NOT a high end app that is used by PROFESSIONAL CG ARTISTS to create PROFESSIONAL CGI.

It's a HOBBYIST tool used by HOBBYISTS to create HOBBYIST ART, so it's goals and the ways it achives this goals are entirely different.

You can't just transfer high end techniques over to Poser and expect them to work in a completely different environment.

Here's the best PP 2012 can do with 4000 polygons without a displacent map.

What I'm taking from this thread overall is that there are a lot of different aims and usage scenarios amongst the Poser user base.

The central question I see is whether one catch-all figure is really feasible... or whether a good variety of different figures is best.

Personally I'm inclined to think the latter... but with the caveat that individual figures would still be good to have a fairly wide range of adjustability... and have a relatively neutral, zero-mode, relative to their underlying and still pervasive "character".

It sounds like high and medium/low res versions of each figure could also be desirable to different needs... especially as the high res versions grow to substantially more polys.

The issue then, with a diversity of base figures in one's runtime, is portability of clothing and hair etc... but I guess the newer, dynamic clothing and hair features will counter for this more and more, in ongoing future iterations of Poser.

EDIT: ...and of course utilities like Wardrobe Wizard and the Poser Place Outfitter.

As someone who started out in Poser 9, I only got onto looking at conforming clothing / hair  after exploring dynamics. I've only really started using conforming clothing to any extent as a result of playing with the Poser Place Outfitter in fact.

Hair, I've been sticking to hair props, after playing with the hair room at the start... as I found some excellent hair props by Neftis and one or two others, that work pretty well so far, for the static images I've been making... and I think it takes more time to get good results in the hair room than I've had to put into it so far.

But I'd like to see the hair room develop and already there are people like Carodan ( see this thread ) that I see getting some excellent results with it...

;-)

17K version for comparison.

Much better, but still not nearly detailed enough for what I'd call "realism".

The problem with different brand figures is the same as with scale modelling.

Even if two models of different brands are meant to depict the exactly same real world object, the different "styles" of those models will be painfully obvious once you put them next to each other. You will not just notice the flaws that are unavoidable because the object is a model and not the real thing, you will also see the differences between each other because the two model makers made different decisions in what details to keep and what too loose.

So I couldn't put MIKI next to Vicky in a render because neither are "100% perfectly photorealistic" and the small "flaws" caused by different modelling and rigging styles will add up and distract me.

That's why I use only figures from one brand (DAZ) and from one generation (3rd). Because they all share exactly the same "flaws" and therefore those flaws are easier to ignore and I can easier suspend disbelief when looking at those meshes.

So a single figure that is not capeable of easily morphing into other genders, races and ages is absolutely useless to me.

I either need a complete "family" of figures all done by the same artist to guarantee the same "style" or I need a highly flexible figure like the DAZ figures are.

That's another reason I think coming up with a proper replacement for Genesis is practically impossible.

A cobbled together group of meshes done by different artists is exactly what I DON'T want for Poser.

I think that's a good counter-point to what I'd just posted JoePublic... I see your point there too.

I'm actually putting Vicky4, Miki3 and Antonia in a render together currently... it will be interesting to see how that works out for me ;-)

Yes, there is likely an extent to which they can and will "clash"... and I suspect as someone who comes from a background (art-wise) of traditional drawing and painting (oil on canvas) my ability to suspend disbelief... or rather the extent of photo-realism as opposed to a more "painterly" effect I'm perhaps, ultimately, aiming for (certainly not yet anywhere near getting to)... is slightly different to yourself.

But, to concur with what you're saying there, and expand on that, a core issue I find, is not maybe just so much in terms of getting different figures to gel together and look relatively convincing...

...but indeed to get the variously sourced props, hair, etc, that I am encompassing those  figures within, to gel with them and add and not detract from the overall appearance of the image...

;-)

Even though Poser isn't a high end application, it is used professionally by graphic designers and also used for animated ad work.  Sometimes a render is used as the base of a image, other times it's used for final output.  If you read The Secret of Poser Experts, you will see many professionals examples of how they use Poser in their work flow, from everything from book covers to product packaging.

I guess I'm missing the point, computers have advanced and many people are running 64 bit systems and Poser functions well with many high polygon figures without significant slow downs--even the 32 bit version, for me at least,  took a lot of stuff in a very complex scene before I'd experience a slow down.  I just don't see what advantage there is to use low poly figures--particular for people who don't animate.  Professionals use Poser for book and magazine illustration--any artwork where you'd be using close ups and want to get a realistic look would demand high poly figures and hi res texture maps.  I don't always use smoothing in my render settings because it has a tendency to distort other things in my scenes.

i will definitely agree that there needs to be a new "family" of figures that are as customizable as the DAZ figures have been. So far that really hasn't happened. I'm working on changing that, but it takes time and a lot of planning so that it's done the right way. 

~Shane

Wow ED, that is exciting!  I can't wait to see what you come up with.  It's great that someone with your talent is interested in this.

Love esther

I would totally second Esther there, Shane ;-)

On a side note, out of interest... on the topic of resourcing community-oriented projects... how much are folk here aware of this sort of thing:

http://www.kickstarter.com/

Thanks Shane!  Not since the G2 series came out has any owner of Poser tried to upgrade the "stock" Poser figures.  SM has stated on another forum that they are working with an artist to create a new male, wm premium figure (not sure if that was a reference to what you were doing Shane :0)--but the more choices we have the better).  Miki 4 is also coming out as well.  I do hope the scaling issues in Poser are completely resolved in the future; I kind of feel the issue has fallen through the cracks a bit.

If Poser is going to continue without adopting the Genesis technology people need to have figures for it that they will want to use as much as the Daz series of humans.  I don't mean to be pessimistic here, but if there isn't more Poser specific quality content I don't think Poser as an appllication will survive.  I really want Poser to survive--like Photoshop I view Poser as a venerable old friend who continues to be an asset in my image creation arsenal.

Oh believe me when I tell you the issue of scaling is now very much front and center. I can't really go into it but yeah...the engineers know just how hosed scaling is.

where's the hype?  where's the hoopla?

I have no connection to SM or anyone who is working on anything SM is developing. 

The idea that they have someone developing a figure is great, but at the same time it kind'a makes me wonder if I'm wasting my time. I don't have the resources that SM has, to put into a project of that scope in a way that would be deliverable in a timely or consistent manner. That's not to say I can't develop a professional figure that would be comparable or competitive, I'm just saying I can't do it as quickly and completely as they could. 

So I don't know. I'm developing Lucas, and a female counterpart for him. I have my own ideas of how to go about it. How similar they will be in comparison to anything SM releases, I have no idea about at this point, but if they're too similar then it will be kind'a pointless for me to continue with it. I guess that's just something we'll all have to wait and see, which project gets completed and released first. 

~Shane

Quote - where's the hype?  where's the hoopla?

I don't think there's a shortage of cheerleader outfits in the marketplace... surely we can assemble a decent squad to cheer each member of the new male team onto the field of play?

;-)

Quote - I have no connection to SM or anyone who is working on anything SM is developing. 

The idea that they have someone developing a figure is great, but at the same time it kind'a makes me wonder if I'm wasting my time. I don't have the resources that SM has, to put into a project of that scope in a way that would be deliverable in a timely or consistent manner. That's not to say I can't develop a professional figure that would be comparable or competitive, I'm just saying I can't do it as quickly and completely as they could. 

So I don't know. I'm developing Lucas, and a female counterpart for him. I have my own ideas of how to go about it. How similar they will be in comparison to anything SM releases, I have no idea about at this point, but if they're too similar then it will be kind'a pointless for me to continue with it. I guess that's just something we'll all have to wait and see, which project gets completed and released first. 

 

~Shane

Figures are like snowflakes - everyone is unique.  The key is to have clothing support via WW or Xdresser.