what happens when i pose my figure with knees bent all the way? or even just bend them to SITTING POSITION? i start seeing very unsmooth knee joints. and what about elbows, same thing. They look ridiculous. That's one of the reasons, i can't really create anything realistic. That's why, every character turns out to look like a doll, instead of a real person. Standing position with no deep bending looks great and realistic, but as soon as i bend the elbows or knees, if it's deep enouogh it will look aweful. Is there no way to fiix this? so i have to get Poser 7 to get more realistic body part bends? And by the way, why is it, i've seen characters and galleries with characters made in 3D Max, and they are able to achieve realism in their knee and elbow bends. but Poser characters, everytime i see an image with bent body parts, they look like crooked plastic. I understand maybe Poser creators/DAZ3d creators never figured out the algorithms necessary to create realistic bends, but did they ever try to improve that? I hope in Poser 7 they did, and i also hope my collection of V3/M3 clothing and characters would work in poser 7.

I said unsmooth knee joints. Yeath that too, but look at the back of the leg and calf, right where it bends. that looks so much like plastic ,very unreal

There are much more robust rigging systems in higher end 3d applications then in Poser.

It's one of those cases where you really do get what you pay for.

Some figures bend better then others, and some rigging is better then others. With a simpler rigging system like poser has, you can only do so much.

This is a limitation of Poser's joint system (and the fact that it's not a professional studio application by any means).

When a real person extremely bends body parts, several things occur naturally:

• the muscles flex

• the muscle, skin, fat push out of the way on the two colliding body parts (that darned Newton thing).

• skin slides and wrinkles

The only way to achieve these realisms in 3D CG is to have a skin/muscle/joint system with dynamics calculations.  Think Shrek and ILM here.

The only other way to achieve this is to fake it all the way by using morphs, possibly along with JCMs, and other forms of deformation (magnets or such).

It is really not worth trying to design such complexity into Poser figures.  Would you be willing to pay several hundred dollars for a figure this well designed (to compensate for the effort required)?  I'm sure most wouldn't.

Quote - ... And by the way, why is it, i've seen characters and galleries with characters made in 3D Max, and they are able to achieve realism in their knee and elbow bends. but Poser characters, everytime i see an image with bent body parts, they look like crooked plastic. ....

And my Toyota Camry is never going to win the Nextel Cup, either. Not sure why, but I don't think it's entirely the fault of my driving skills or the manufacturer.

When the human body bends, say at the knee, the entire shape of the region changes because of the change in position of the bones involved (the bones in the calf twist slightly, even when bending the knee straight back, and the kneecap floats and rotates forward), the muscles under the skin swell and contract, and the skin stretches, compresses, and even twists slightly to accommodate the underlying change.

Poser characters have neither skeletal bones nor muscular under the "skin". In order to represent a realistic shape change during a bend, the program would have to have multiple armatures (what Poser calls "bones") for each joint, and a mathemetical model for the muscular changes that take place under the skin, so it would know how the polygon vertices should adjust. It would also probably have to be able to sub-divide the polygons, increasing the mesh size dramatically. While that is, technically, doable, it amost certainly isn't cost-justifiable to spend the programming man-hours necessary to create that type of product and still be able to sell it for a few hundred dollars.

Also, algorithms of that complexity might take a really long time to run on an average computer. Imagine what it would be like to have to wait the same length of time to do a knee bend as it now takes to grow hair or mold a piece of dynamic cloth. It just isn't possible to make a small, hobby-sized program do all the great things we want it to do... yet, anyway.

Computers keep getting more powerful, and new algorithms are being developed every day by the people with mucho dinero to spend on their systems, and all of that filters down to us, eventually. Poser is a bunch better than it was a relatively few years ago, and a few years from now it will be even better still. It will probably never be as good as software that retails for ten times as much, and that may be the best we can hope for.

In the mean time, lots of folks are working on morphs for characters to improve the look of bent joints, and those are the current best bets for good looking bends in Poser.

Captain Jack

Consider the fact that higher end applications tend to rely on something called "weight mapping" which is a more advanced form of falloff between body parts, coupled with the ability to tweak  a model's polygonal structure after posing to achieve a more realistic look. This is something Poser 6 and under just isn't capable of doing easily (you may be able to get the look down with magnets after posing or with predefined morphs).h  Whether this is something Poser 7 corrects or is capable of helping improve the look of remains to be seen. 

If Poser has things from High applications, Poser will be a High application, not only in its specifications bt its price too. I think poser is going better and better and its price is still very good. I think this is the way poser havs to take.

"If Poser has things from High applications, Poser will be a High application, not only in its specifications bt its price too."

I don't think its that simple.  How many more copies do you think poser sells than other high cost programs?  If they sell 12xs as many downloads as a different program that costs 8xs as much they are still making more money. 

If the owners of Poser were to go this route too quickly, it would disenfranchise the current userbase (most likely left behind the dust of pricing) and, considering the level of respect it gets in professional arenas, would sell a far less portion than expected or required.  It would need to have features at the level of or beyond current systems that can do this (whether built-in or third-party).  It would also need to establish itself at the higher echelon.  No one moves from hobby tool to professional package in one go.

I'd say that they would need to go to about $3000-$5000 and definitely include some form of non-polygon-based geometry, a much more modern joint system w/ muscles, true dynamics simulation, and proper, more-powerful, built-in content editing tools (there is currently no easy way to do morphs and no way to do master-slave dials).  By that point, the only people interested are studios.  And studios will be a hard sell with all of their proprietary and existing solutions in this range.

You think poser would have to sell to all of its existing userbase at that price to do those things? 

If poser sells more copies of its software than other companies they would not need to charge each user as much for the same cost of development.    Unless we can analyze how many copies of poser they sell each time they upgrade there is no way we can say how mcuh they woudl need to charge for a new feature. 

If poser has a larger userbase then they can add very expensive and complicated features without increasing the price very much for each copy.   I just don't know if they do have a larger userbase than other high end programs.

I was thinking about the Poser buyers like me, not e-frontier. The thing I want to say is that I don0t mind if poser has specifications like max and character studio or things like that because I have no cash to buy it.

open your favorite graphics program and exercise some art skills...presto chango, ugly folds are gone!

From what I've read, ZBrush is a "relatively" inexpensive application that's started to get some fairly good professional attention. It's also a much more focused application than Poser so they can afford to concentrate on improving what it does. I don't think EF could afford to do the same and please everyone that wants high end posing, rendering, cloth, etc. etc. They have to carefully consider how much they can do and still maintain a good sized user base - which means maintaining a reasonable price. Maybe Daz Studio's open api and ala carte pricing is a better bet and some 3rd party developers will do some impressive things down the road. If EF keeps Poser and keeps developing it gradually, it will become more capable but it will always be behind the whiz bang big bucks applications. If someone with deep enough pockets had wanted Poser, they've certainly had enough opportunities to buy it. I'm beginning to think that at the lower end, the 3D character development is going to be in games oriented applications. Aside from Daz and a maybe couple of odd balls, there remains a gap between that and the higher end stuff.

    There are still a few reasons to go, I am expecting rigging (maybe weightmaps) to be one of the reasons to upgrade. How many poser users are scared to go into the joint editor as simple as it is now?  When poser adds weight maps or something of the like, how many users are going to be willing to go back to the library and start rerigging all the products they have?  If poser gets too complicated how many people will drop it for Daz Studio bare version for simplicity?

editing the joint parameters, limited as they are to spherical zones, is not gonna help folks with bad knee bends in p4/5/6. that's currently fixed by something else, which unfortunately is beyond the scope of this forum. it may be partially addressed in P7, but will be challenging for many users, who are still turning off shadows, using nostril plugs, failing to read the manual, and committing other user errors due to a failure to understand the "complexities" of poser, which started out as one of the simplest 3D apps. of course, it's only gonna get more complex with time, as it continues to outgrow the original intent of its creator.

Attached Link: Hedgehog rigging

Check out the videos on this link for a truly impressive custom rigging setup. I think it's in XSI. Of course this little catroon hedgehog isn't set up to be realistic, just to bend exactly how the artist wants him to bend. Achieving realistic bending is a whole 'nother bag of beans, but no less complicated.

(the main ones to check out are numbers 3 and 4, but the others are cool too)

I understand and sympathize with tebop's frustration, but also realize that unless you're willing to fork over a price for a more complex Poser in the range of 3DS Max ( which I did and regret ) and is so process intensive that it forces a state of the art home pc to labor at a snail's pace, we'll have to find some work arounds to solve these issues.

One solution that comes readily to mind ( besides post-work ) might be DAZ's creation of figures pre-modeled in various kneeling, limb flexing and sitting positions that can then be adjusted into a wider range of  poses that will eliminate polygon shear/warping. Alternatively, they might offer "limb kits" that allow for the replacement of standard figure legs, arms and pelvises that can be integrated into a modified figure with a couple of mouse clicks, using the same process Poser has always offered. Just a thought.

limb kits is a really good idea mogwa.  like replace body part with prop, except another body part with different rigging. would be really cool if the replacement of the part could be animated. 

i also think a lot of this could be cleared up if we could animate the joint parameters, like sliding the zones and the joints around as we need them.  unfortunately no keys for them : (

the limb kit idea could make that sorta possible though.  just same limbs with the different parameters, switch out as needed.  wouldn't be a smooth, but would probably work pretty well i think.  i never did a replacement though...does the joint that was there insist on being the same? 

another possibility could be that the body parts are actually always there, but are hidden until needed.

maybe the switch in geometry could be handled the same way the "genitals" morph is.  actually i guess that probably is not animateable either though....so would have to be transparency.

good ponderings on matters that bug the h@$$ out of all of us.

by the way nancy, what is that fix?  beyond the scope of this forum is cool, but where is the discussion on that?  that would be really interesting.  gracias

obm890, thanks for that link. It was very informative and interesting.