Something for morning then. I'm nearly there myself. :)

My clarification on "too far gone" is that maybe there is too much V3 content (god, I said that?!) that may not work with a renewed and improved V3. Such improvements may go to the wayside if noone can use current additions with them.

I was going to ask about the release expectations of VickyPro. First I've heard of it actually. Is this the next version Victoria from Daz3d?

About Point 1:

To a certain extent, I agree. It seems that to gain the overall potential of Poser, one must go beyond the basics and then beyond experience to some sort of mastery. It would be great if the advantages gained by years of experience and tweaking were already available rather than requiring the tedium to get the desired results.

Being a Cinema4D user, I have reaped the benefits of a responsive development team and quality user-company relationship. With each new version, the features that are most desired by the users are added - usually with astounding appraisal.

We are definitely getting good products/content for the cost. And I never scream about it - except in the case of Zygotes' recent human internal organ package. ;) But if improvements are evident and seem to be easily remedied, as you related to the shoulder issue above, why are they not introduced?

The question that I ask myself often is this: Okay, Poser is the only character package that has tons of pre-made content: clothing, hair, character morphs, and so on, but why do I have to live with the shortcomings that have been voiced for years and the small inadequacies that seem to never be addressed? Content is what makes Poser, but the content must start to achieve certain levels or Poser will forever remain a hobbyist's plaything. Obviously, I project more onto Poser than what many users would. :)

If improvements in Poser require more complexity, then interfaces should be derived to reduce it! This is definitely what I see has happened with the Material Room in P6. That is the correct approach. Have a "I'm not a nerd" interface for the non-hardcore users and a "Yeah, I'm a nerd" interface for the hardcore users who can't help tweaking and twiddling. Make it simple for the users who don't want to be bothered while availing the more detail-oriented users with a more complex interface. All around, this lacks in the Setup process.

Message edited on: 04/28/2005 02:59

Attached Link: http://www.runtimedna.com/messages.ez?forum_id=45&Form.ShowMessage=134358

I've been watching rigging and animation training videos for Character Studio, Motion Builder and Project:Messiah over the last 18 hours, also reading reviews and forums. Pretty damn cool. But....for now.....I'm just going to 'park' that information. CL might be working on a substantial improvement to the current Jessie under the current rigging system. Jim Burton is working on Glamorous Jessie, lookin' good. Maybe CL will read this thread, plus all the other imput about a new rigging system and beginning working on a new Poser rigging system. There's a good solution in EJ. While all of that is boiling up, time to go finish my current animation (attached link). ::::: Opera :::::

I just got hold of poser 6 and I have not looked into what python can do with poser figures. Curious labs says that you can now access all of a poser figures propertys, etc with python (from web site). I wonder if a new rigging system could be made with python.

OK, speaking as a guy who is currently trying to create such a Python-assisted rig - I can categorically tell you that (a) Poser 6 does not give you access to everything in a figure and (b) there are some inherent problems with Python-assisted rigging in Poser. To fill you in on my current gripes, have a flick through the list below... 1. There is still no way to access all the rigging parameters via PoserPython. That is I cannot read or write rigs without reading/writing CR2 files (with all the parsing and file-based functionality that goes with it). 2. While one can set a script to be run when the figure is loaded into the scene - said script file MusT be in the default runtime. This means one cannot have a Python-assisted rig in any other Runtime bar the one under the Poser 6 directory. 3. Testing Python-assisted rigs is arduous at best, a bloody nightmare on average. One cannot edit the script "on the fly" but must make tweaks to the script in a separate application, unload & reload the figure in question, and then see how the changes worked out. Given the average figure loading time, this becomes tiresome very quickly! 4. The way one must structure the Python scripts to get this working is incredibly counter-intuitive to an artist (it's not the easiest to understand for me, and make my living by programming). As such, it is not very likely to be used... 5. Unless one is only controlling deformers, the script will need to be applied to the clothing as well. Even with deformers, there needs to be a custom pose/script for adding the clothing articles to the deformer's "influenced actors" list. This fact lessens the appeal significantly (i.e. without clothing support - why bother?). 6. Currently the method of accessing the dial values for an actor in the scripts is quite cumbersome and as such doesn't lend itself to an easy reading of how the script will work. What NEEDS to be done is for Curious Labs to add in expression support &/or the ability to embed python scripts into the figure CR2 (from a coding point of view - this is pretty much one and the same thing). Hell, should Curious Labs need direction on this issue - there are a tonne of examples out in the world of open source and I have done it for two custom applications requiring statistical formulae entered at runtime...

Hmmm embeding python code into the cr2. Um not up on cr2 files... but can you include comments into the cr2 file? The basic idea would be to have the python code commented out in the cr2 file. Then instead of using the poser library to load a figure into poser you would use a python script. It would look at the first comment line which would have um a tag (pscript) and look for the last comment line which would have a tag (/pscript)and then run the embeded script between the two after loading the figure.

Hmmm, two things wrong with that scenario - (a) It is a hack, and like all hacks, means it could be broken by any update CL decides to make; and (b) one cannot launch a python script from the library hence making it harder for the end-user to load your figure. Combine that with the fact that one cannot run pythono scripts outside the default runtime and that one cannot determine the exact location of the loaded CR2 to pass to the "pscript" reading function - and the idea (while not bad) is dead in the water due to faults in Poser (not in the idea itself). Sad really, because (done right) Python integration can make an application incredibly flexible. Look at all the cool stuff ockham has done with what they gave him. Imagine what would be possible with the right hooks into the Poser engine!

Attached Link: Project Human Beta

Quote Helgard= "Yes. Sixus1 can afford the bandwidth. And Sixus1 have a good team leader/project head. Sixus1 is not a community project, they are a company, and they make money. They may give their figures away for free but they do make money on the add-ons to pay for it. Personally, as much as I like and respect Sixus1 and the fact that they have made their own humans, those models are not advanced enough to beat Vicky or Jessi or James or Mike." There are several points that I would like to address. The original point of the first line of humans was mostly that we wanted figures that we could use to make characters sets for that were our own to avoid any problems with EULAs, licenses, etc..... And I am actually insulted when you say that Sixus1 is not a community, but merely a business. I would love for you to say that to people who come to our site, to the new friends that I have made there. To people who wish each other happy B-day and send each other Gift Certificates to people that they have never met and barely speak the same language that live on the other side of the world that they are part of nothing more than a business. Our community may be small, but that doesn't make it non-existant. We decided to make Project Human Open Source for those that want and desire human figures that they can make their own. We will be supplying bandwidth, a website and the base files for free. The download is for the final BETA, I do believe that Les is going to be tweaking some final JPs before the offical release, but if anyone would like to take a look...the Male and Female are both available at the link above. Also, the figures are completely uncompatible with DAZ or any other Poser figures, so we can freely give the permission for you to use them as you please. No one is saying that you have to use are figures and we aren't saying that they are better than everyone else's...they are different and they are going to be there for everyone to have the option to use. We hope that the new Project Human will foster freedom and that in turn, a new level of creativity. --Rebekah-- P.S. Sorry to get slightly off topic....

Knight: rats! I think I'll drown my sorrows in something sweet um like a bunch of krispy kreme donuts. Carbs and sugar are good for you!

Sorry, my apologies. My post, upon re-reading it now, sounds harsher than it was intended. I did not mean that you do not have a community, or that you were a business intended purely on profit. I do hold you and Les in the highest regard, and I am not trying break down your Project Human endeavour, or trying to say that it is not a worthwhile contribution to the community. I apologise for the tone of my post, and wish you all the success with your figures. Helgard

Thanks. :) --Rebekah--

Maybe we are going about this the wrong way. There are a couple of dozen rigging systems out there that technically supercede the current boning rig that Poser uses. but the added spinal joints aren't a function of a new rig; that is actually the modeller's task, as they are the ones who arrange the mesh sections. Boning in Poser has to take that into account. Perhaps what we need to do is actually look at just how far we can shove the current rig forward. One model with the added torso flexibility. Once that particular part of the figure is tinkered into solid function, move onto the next problem area. This could be kept 'in the family' during development, and presented as a show piece once completed. If the current rig proves itself sufficient for the actual =armature= functionality, then it may be that we need to actually focus on other areas. Say perhaps advancing the boning to more of a 'woody' type armature, and actually developing a method of skinning the armature so that a secondary boning system articulated 'muscle' structures. Similar to JCM, but built in as a true function, not a hack. Or a redefinition of what a 'joint' actually is. Not so much throwing the baby out with the bathwater, but seeing if maybe there actually -is- a better fix than nuking the system

would this need fresh mesh, dale?

I don't think so (even though the query was not directed to me). If the current rigging system can support better armature functionality - which to some degree I believe it can - then there should be investment in that direction. This is why earlier I mentioned Cactus Dan's new CD Morph - which reacts to bone rotations. Brilliant! Poser already has something rudimentary to this - bulges. But they are very generalized 'muscle' structures. In reality, they are just a way to avoid crimps in the mesh during deformation. The key (as in all rigging systems) is to attain the type of deformation on the mesh that mimics the system desired - in our case, this is usually human musculature.

Here is a WIP of how Glamorous Jessi stands now, next to stock Jessi. Still a way to go, but I can see progress. ;-)

Main difference in her set up is I'm using JCJ (Joint controlled Joints) instead of V3's JCM (and Stock Jessi's nill) to help the bending. This is partly in a hope it will make for easier transferring to clothing, which can be extreammly hard to match when you have to recreate the morphs.

I only can work with the tools I have.

I agree with what Anton said, too much too fast could kill Poser. It is already EXTREAMLY complicated.

Message edited on: 05/04/2005 21:35

As always, Jim, you really are a master. Damn, you're good. 8-)

Hmmm - JCJ is an interesting concept. Anyone (Jim?) know of an article that talks about this or will I need to experiment a little to work them out? Watching closely :) EK

Lol, I just sent an IM asking for the same thing. Can't find one decent tutorial on the subject.

I'm guessing it is just ERC with the target being a rotation/scale/translation channel rather than a morph channel. But given the relative inflexibility of controlling when ERC kicks in - I am curious to see how it works for joints.

Exactly. I create a ghost part and then ERC link it to a body parts rotation. I remove all the unneed channels (scale, other rotations, etc) from the new ghost part. I've been doing it for awhile, most of my newer skirts have JCJ to hook up a working thigh part (in addittion to a buttock part), for example.

So far I've linked them directly to the body part, I may have to hook some of them thru an intermediate channel to set a delay in where they start, I did a lot of that kinda stuff when I set up the cheongsam I sell at DAZ.

Big advantage in setting them up is you get instant feedback, it is just like working with magnets. I run them manually to work up the ERC ratio. They also seem to work much more "naturallly" than morphs, it always seems like the JCM morph is fighting the bending fall off zones.

Message edited on: 05/04/2005 22:41

Agree completely on the effects of the JCM's. Combined with the complications that arise with clothing - I tend to avoid them for all bar the essential parts (i.e. thigh squash, knuckle creases etc) and even then try to make them as minimal as possible. "Ghost parts" - are they creatable within the Poser setup room or does one need to do some CR2 trickery?

Yeah, the only thing you have to do pretty much is make sure that the group name in the clothing is the same as the group name of the bone (for example if you create an OBJ group named "skirt" you also have to name the bone "skirt"). Otherwise Poser puts unassigned polygons into a group called "Figure 1 Setup" and it becomes a separate prop.

I'm still going to argue that adding just a few more bones will help immensely! The meshes are already dense enough. Some refinements to grouping and boning can make a visual difference... as in the difference between regular action figures and superposeable action figures. As an example, look at Jame's knee and butt: If the knee had 2 bending points, it could accomodate the type of bending that real knees do (think of Japanese monks sitting between their feet). With a buttocks group, a leg which is going back ought to be able to morph to indicate the clenched muscles.

Elbows which have 2 bending points won't look like broken soda straws. This is a skeleton for James. In order to fit within the skin, and stay there, the knee has to work like this, with bone crunching bone: If the knee had two joints, the bones could act like real bones with a nice tidy patella fronting them. (Note that I don't use precanned poses, or limits, but with 3 1/2 years of life drawing as well as rock climbing and dancing, I have a fair idea of how the human body moves.) Carolly

Good points, Carolly. So as a very rough spec for re-rigging, we are talking about something like this: 1) Fully articulated hands and feet (with the feet having a global 'toe bending' morph associated with the ball of the foot, for ease of posing the toes globally [like Judy has]. Perhaps extra controls like the fingers 'clench' and 'spread', to curl the toes globally and collapse them inward towards each other, to facilitate using high heels or tapered front footwear). 2) Adding a second bone at the knee, to move the actual hinging action to the backside of the joint, so that instead of a single hinge that pinches itself at extreme fold, there would be a double hinge with either locks to prevent the two joints from folding past 00 degrees with IK enabled (or a reset pose to correct 'shattered knee syndrome'...good for some fight scenes, not good othertimes). 3) Increased spinal flexibility. Currently what we have is hip, abdomen, chest, and neck (and in some cases, neck 2). That basically puts a hinge at L5 (lumbar5, at the top of the pelvis), around T12 (bottom of the ribcage), around C7-T1 (top of the ribcage), and the neckhead, which usually sits around C1. With maybe another bone around C3-C4. There should be at least 1, possibly 2, added bones between ribs and hips. The thoracic part of the column is the most rigid, but even that has some side to side bending of a few degrees (but that -should- be doable with morphs that are linked to the model section's controls). 4) One extra bone in the elbow region to perform the same function as the knee, in preventing mesh breaking by creating a 'space' for the mesh to fold into without overlapping itself. 5) One or more added bones into the shoulder area to allow for more realistic posing and motion at extreme angles. 6?) Add a bone for each breast in the female, to provide an articulation point that would function as the mass center (just a thought; this may or may not be feasible, or desirable) 7?) A bone for the jaw, so that you have actual articulation of the face, rather than morph based deformation (again, this may or may not be desirable or better). What else can people think of? Once we have a possible spec regarding what we can do with the current rig, then we can isolate the trouble areas and look at specific solutions. Oh bloody hell. Soft IK. Being able to specify a joint as being the 'end' of the IK chain (and all other children taking their cues from that designated joint) would be invaluable for the animators, at least.

The current IK system does this - albeit Hard IK. The goal and chain from it are specified in the figure{section of the CR2. I think your 7-points are a good indication of what needs to be done. A jaw bone would be desirable. It alone would remove a large number of morphs that used to achieve the same thing. If done properly, the results would be more anatomically realistic. I think the problem with the knee isn't that it is a space-hinge joint. It is a sliding joint (the patella). If you study how your knee bends, the patella slides up and down with attached ligament (patellar ligament). If you have "Human Anatomy for Artists", you can see this on page 35.

To add: the space-hinge joint idea is a good one since these joints are not point bending. They are bones rolling along their wide surfaces. The design, as it were, accomodates the large muscle mass and strong ligamentation attached to these joints. If there is a 'point' at the radius of bend, it is not centered at the specific bone-bone interface.

My point is this: in order to improve the rigging and realistic anatomical contortions of human beings requires some studying up on human anatomy. In order to accomplish this means knowing the structures that culminate in those exterior features we recognize as proper.

Message edited on: 05/05/2005 12:22

1. Increased spinal flexibility. Currently what we have is hip, abdomen, chest, and neck (and in some cases, neck 2). That basically puts a hinge at L5 (lumbar5, at the top of the pelvis), around T12 (bottom of the ribcage), around C7-T1 (top of the ribcage), and the neckhead, which usually sits around C1. With maybe another bone around C3-C4. There should be at least 1, possibly 2, added bones between ribs and hips. The thoracic part of the column is the most rigid, but even that has some side to side bending of a few degrees (but that -should- be doable with morphs that are linked to the model section's controls).

Along that note, it would also be neat to include dials that split poses between several levels. For example, when you turn the head of a figure left or right, divide the turn between the head and neck sections as it would happen naturally.

Message edited on: 05/05/2005 12:28

One thing to remember about doing things like double jointed knees (and V3's buttock parts)with the current Poser setup is that childern parts can only bend their parents, they can't bend anything past that. When you set up a figure (like V3) with buttock parts the thigh bending (now) can't extend into the hip, you have to stop your fall off zones short of the hip mesh, else you get creases where the bending stops abruply. I don't think the answer is more bones, myself, it is better control on the bones we have. ;-) E.K.- I did a mini-tutorial on making ghost parts in the Clothing & Characters forum at PoserPros.

Well, this was just a starting point...=anytime= you start talking tech, there tends to be a pendulum swing into the hyperbolic regions. The high end bone-muscle-skin systems would be nice to play with, but I doubt most of us could afford it, and fewer could ever master it... :P although I will argue about the torso needing a couple of extra articulation points to get a better bend sweep to it. And an actual jaw bone. However, taking the list as a rough, and assuming they can't pull something like softskin dynamics and sliding joint capability out of the hat affordably, what changes to the current rig would need to be implemented? Some alteration of the parent-child logic? Fall-off zones that can overlap, or interact with each other? If the current rig just won't hold up longer, how could things be set up to keep the backwards compatibility? Some type of metatag in the cr2 or obj of newer content to tell the system to use a new rig, and lack of that tag enables the old system? We all should remember that P6 was said to be a service release, with the major changes to the code slated for P7 (and with the new version of Firefly, we got some significant goodies anyway). CLSteve would almost have to be asking regarding the changes for the next version. The animation & rigging system is probably the last untouched artifact in the program (save for the Kai Krause interface, and even that has been spruced up a tad). They might just want enough people clamoring for something a little more capable, to justify the time expense in coding it... :)

I really don't think the jaw is necessary or particularily doable ... It is like the webbed fingers thing that no one has been able to do at all.

Jaw is easily do-able, but the problem comes when you have to set up the morphs for MIMIC. It can be done, but much easier to do the jaw bone movement with a morph like the P6 figures. And the webbed hand has been done. I have one that I downloaded as a freebie from here some time ago.

Any idea where the webbed fingers are since I know a pile of people were looking for it over the years :)

Attached Link: http://market.renderosity.com/softgood.ez?ViewSoftgood=37163

The jaw in that skull is a separate articulated part... it can chew, or laugh or whatever, but there are no phonemes for mimic. (The skeleton is in the store, and works quite nicely, but a tweak or two is needed for close fits.) There are other models with a separate jaw, too. I'm thinking of horses and such. So, the boning can't be a problem. In fact, if the jaw WAS a separate piece, we could have more life-like movements even with skin on top! Carolly

Carolly, I did an experiment with a jaw bone, and it means that each phenome for mimic has to be spread over two morphs, just way too much hassle to make them speak. It is not impossible, and I think Little Dragon has actually done this on one of his figures.

Since Little Dragon has added mimic capabilities to just about everything under his fingers... including Mike the TV... maybe we should get his opinion as to whether it is worth the effort. It will be more work, but will the results look better or offer more flexibility or realism? Or will it just be a pain in the kazoo? Also... is mimic compatibility all that necessary... aren't there other speech programs? Carolly

Maybe the question should be wether or not Mimic 23 (which in the minds of most who want actual speech out of their figures has become entangled with DAZ and Poser as 'the standard'), with a custom configuration file, could deal with it efficiently.

Well for skeleton's it is good but when you get into skinned models there is a problem. The old sea serpent at DAZ has a jaw set up but it basically sort of fakes things and you cna't really do that with human's. I am curious on if it can be done but I just not sure.

I'd say that the jaw is the least of our concerns. Shoulders, elbows, knees, back, and neck must be the main areas of focus since these are in desperate need of fixing.

I had a whole dissertation set to post, thought I posted it and of course it got munged.

We want better rigging of characters for what purpose?
Poser can do some great stills.
Is ANIMATION is the next paradigm for Poser?

If so, do we need to shift our mindset and say, we want a better program for the ease of use in animation.
If that's the case we want characters that are going to be easily posed and rigged for animation. Is this the same model target goal as the artist who needs a 3D mannequin to paint over?
If we get the nicely bent elbows and knees, etc.,so be it.
Will this shift mean more DYNAMIC type clothing?
Will this shift change the types of users that move on and upgrade to Poser 7?
How will this affect Poser's position in the market?
The animation mindset may 'scare off' some users. They may not wish to upgrade. They may shift to DazStudio (familiar content). Some may say why stick with Poser and 'trade up" to the more PRO packages. (Who knows how the market is going to be in the next three years? There may be irresistable deals for the other packages in the future.)

Dont you hate these crossroads?

Rigging affects both posing (stills) and animation. An incorrectly deformed knee is still incorrect (and more painfully visible) in a still than in animation. At least in animation, some of these incongruities can be glossed over by motion. Not so in stills. They are ever-present for all to see forever. And no one's gonna tell me that Poser elbows, without master modifications, aren't sharp needly things that look about as real as Mickey Mouse. The application is improving. Its feature set grows. The current set of figures is very high quality and continually moving towards realism. Yet you can spot a Poser figure a mile away (unless it has been lovingly caressed for months by a dedicated master). You can spot it in the bends, in the shoulders, in the deformations.

dlfurman; Why not? Poser's days as just an artist's Woody on a CRT are long gone; they vanished when P4 came out, with transparency and the animation support it had. The number of people who actually create a scene -just- as a reference piece to take into Photoshop or Painter and create 'real art' are actually pretty few and far between. The dynamic cloth and hair add to the incentive to experiment; I wasn't overly interested in animation myself until I saw what Phoul could do with P4 and Vue 4. Several people have been attracted to the animation pallette by the requirements of the dynamic props. Like kuroyume said; improvements in the rigging would benefit everyone (I certainly wouldn't miss the mesh breaks), and if the app can turn out better animations than before, then it will be that much more likely to be used. Everyone knows the tightrope that CL is actually trying to walk; ease of use for newbies, and enough meat for a pro to sink their teeth into. Lights, cloth, and hair have all been addressed; the only subsystem that has been fundamentally unchanged is the animation setup. If we can have our cake and eat it too, lets.

Nature Stages tools inside Poser, improbe the render speed.

Attached Link: http://www.renderosity.com/messages.ez?Form.ShowMessage=2248430

Yes the Great Agitator (self named) is back. See the attached link. All of this stuff, plus an enhanced rigged figure. Would you want Poser 7 to be 1st part the transistion of Poser to a new level of application? What I mean is Poser 7 will have NEW code and from THIS version, better enhancements come forth? We come back to this question (some ideas in the attached thread) how much are you willing to forego? Can't get something for nothing.

Attached Link: http://poserpros.daz3d.com/forums/viewtopic.php?t=40497

Talking about knees, here is a link to the latest WIP on Glamorous Jessi.

5 JCJs, 2 JCMs, 1 Delay ERC in each leg, see what you think of the results. All automatic operation, of course.

Message edited on: 05/07/2005 18:58

great reads, guys.

Good grief! This is complicated stuff. It looks as though, from what I've seen in other places, that some things could be improved without needing to change the internals of Poser. A figure with multiple groups/bones at a (human) joint which can be posed with one dialset. The later figures acquired buttock groups. Some of the movement of the (human) hip joint should be in the thigh-group, some in the buttock-group. The shoulder area is more complicated, but the same could be done. The tech is already there to have a set of dials on the "Body" which control specific joints, instead of jumping between body parts. It's a big job, I don't know if anyone would want to try it, but can't somebody hack the mesh out of Vicky for the knee area, and just see if it could work with new grouping and bones.

Antonia Tiger, The problem with "hacks" is that they wind up getting fixed. Can we say CrossTalk?