Quote - From my point of view it's not so much about 'cheap' , rather how easily I can create the variety of character I desire. I'm not exactly flush, but I will pay for what works for me.

JoePublic -Your points about using lindividual rigs for certain meta types of characters do make sense - I'm starting to come around to this idea much more. I'd still suggest you'd need (in each of those Cr2's) a full range of body/head morphs to make them usable (have enough variety  mean).

Of course, at this present time there is nothing quite like this on the market (i.e a complete system of character rigs).There's a mish-mash of different figures but they lack a cohesive approach, which is what I think you need to have. I suppose the Daz approach isn't a million miles away from this idea, but they do seem to milk it for all it's worth with each individual figure they release - they don't sell their figures in terms of being a unified set that all share the same morphs and textures, even though you can achieve this to a greater or lesser degree.

So, have you been working on something like this?

Where it becomes troublesome is in clothing support, which is often crucial to figure success. Most clothing makers, those of us who like to make detailed stuff anyway, and who are trying to pay bills doing this, can only succesfully support one or two figures.

This means that if Aiko, and the Girl, and Elite series and She freak, and hiro, and freak etc, were all using separate rigs (as opposed to being morphs on a single rig) I would not be able to afford the time to make a separate piece for each one. Especially since each of those figures would sell considerably fewer number of pieces. Poserdom as a market is small enough that for people like me, selling fewer number of pieces doesn't mean a difference between making a lot of money, and a little bit less... but a difference between, do I make my bills this month or not.

So, it becomes a catch 22, what direction to take, unimesh concept, and more clothing, or more optimized base figure, better nudes, less clothing. Figure not having enough clothing support is yet another very contentious thing in poserdom, where a lot of people feel snubbed and ignored, while the reality of the situation is, most clothing makers can't keep up.

That's another thing, if poser rigging was more modernized, making clothing would be lot less time consuming too, and we'd be able to make more stuff, for a greater variety of figures..

Quote - Joe, to be fair, in some respects, DAZ simply gave the community what it asked for: "WHY WONT MY PRETTY STEPHANIE TEXTURE WORK ON VICKIE! CANT YOU GUYS MAKE IT EASIER?"

And so they did, with the Unimesh. But the problem with the Unimesh, the way I see it anyway, is that it has to be so polygon intensive to anticipate ALL of the needs that it's like painting in oils, which is really suggestive of moving mud around. It makes this huge, weighty figures unavoidable, which means that unless you have a really powerful machine, you're severely limited in what you can and cannot create. I started to worry about this when I saw how Hiro3 could go from a skinny, smooth anime figure to a detailed, realistic muscleman: the polys needed for such a transformation dont appear out of nowhere, and displacement maps can only do so much.

Well, now we've taken it to the next logical step, I guess. And it's only going to get better (or worse, depending on your point of view).

As for the rigging video that Phantom put up, I have to admit, in all my years of working with the stuff, I never thought to look at the joints to correct that kind of mesh issue. No doubt this is the same problem I see with clothing converted in WW, and now I'll feel more confident about going into those and reworking them for a better fit.

That pretty much sums it up. One size fits all and ease of use comes at the cost of something else.  No matter how you shake it, somene is going to be unhappy and cry bloody murder. So, most businesses try to shoot for the majority. Result of that is a popular, versatile, but  somewhat average product.

Most people and places that do this as a business aren't going to spend exorbitant amount of time doing something merely for the bragging rights of having the most technically advanced product. Once you become a vendor, you realize that most people don't really care all that much about the latest and most advanced stuff.

oh, IIRC = If I Remember Correctly :)

That looks like an awesomly informative thread!  (I only scanend through it so far, will definately bookmark it for a detailed read.

Quote - > Quote - DS and Poser aren't going to split that much. The market is small wenough that they need each other.  Business wise, it's better to be a smaller fish in a bigger pond, then a huge fish in a tiny pond.

Some pulling and tugging back and forth just sounds like a good healthy dose of sibiling rivalry to me. No more then that.

Hopefully not, I havn't even made anything to sell yet, and when I do, I'd much rather make something that works in both apps.

It's not too bad. Especially if you stay somewhat middle of the road.  if you get too advanced in either direction, they you start bumping into limitations of one program, or the other.
like, the non-uniform scaling thing, or with P8 having the dependent parameters now.
To make things work for the broadest audience and piss off as few people as possible, you always end up staying at least version behind the latest tech stuff.

If you go too far forwrd, or too far outside the mainstream, you get the bragging rights, but if you're trying to do this for a living, it doesn't pay the bills.

DS and Poser aren't going to split that much. The market is small wenough that they need each other.  Business wise, it's better to be a smaller fish in a bigger pond, then a huge fish in a tiny pond.

Some pulling and tugging back and forth just sounds like a good healthy dose of sibiling rivalry to me. No more then that.

Okay, just a quick comparison between Poser and DS and some findings. I compared it in DS3 Advanced with Figure Setup tools, and in Poser 7, and PP2010:

When a bone is non uniformly scaled, along a Y axis for example, in Poser, the SmooY parameters stay static - as in do not change the "Y" location, as the bone location moves. This way, very quickly the mesh falls out of the bend influence zone and produces the stairstepping effect.
When a bone is Uniformly scaled (XY and Z) the smooY parameter moves/scales along with the bone in Poser.

The other three JP's, XYand Z rotation move, movements and scaling move and scale with the bones or joints as they move around. This is the same in both programs.

In DS, the smooY parameter moves with the bone even when it's non uniformly scaled (in poser it stays static) - this way the mesh doesn't fall outside smoothing influence zones as quickly, producing a smoother bend.

So, I guess I can see why some people would consider it a "bug". I can't think of a specific reason why make all the other JP's move around with the bone, except for smooY parameter, and only when it's scaled in one direction. When the whole bone is scaled, the smooY does move with the bone.  First impression is, it looks like it got forgotten.

Maybe there is a specific technical reason why this was done in poser, if so, and someone knows what it is, please stop by and explain.  Kitty (and others I suspect) are dying of curiosity here.

I guess a possible fix for this would be to have a separate set of JP's for Poser, where the smooY parameter start and end locations are ERC'd to the bone Yscale. IRRC JP's can be INJected, but not REMoved, so Poser users would have to load a figure, then INJ the Poser JP's.

What I don't know yet is if that's the only thing that affects the smoothing or not.

Quote - Hi Connie,

I knew there must be something going on internally in the Daz scaling, things look smoother. So I wonder if its still possible to have something like a hybrid? Poser as it is, is not going to recognise the automatic smoothing, or more correctly just doesnt have the same ability. I was thinking more along the lines of combining the scaling of several body parts like the chest-collars and maybe even the neck by setting up the erc for it with the D/S setup tools. But still leaving the individual dials so you can make fine adjustmens in individual parts.

Yes if you happen to know of any information on the joint smoothing or run across it in your travels, I would be very interested. I wonder if a PY script could be devised to accomlish something similar? Not that I could ever write it myself but someone like Cage might be able to.

You can make those adjustments inside DAZ's figure setup tools(FST). I usually inherit them from the V4's rig when I make clothing, so I don't mess with them often enough to know the finer points of the differences between adjusting them in FST vs. Poser.

I know they can be tweaked in FST, I've selected them by accident plenty of times, when I thought I was moving bone centers.

I jsut ran a few test movements on the Neck and chest and head smoothing parameters, and DS definately reacts to their movement - on Antonia. On V4, they appear to be locked in place and hidden, so I'll need to hack the cr2 first to show and make them selectable, or pick a different figure to tinker with.

What I haven't done yet is do side by side comparison on whether the mesh behaves identically when those parameters are adjusted in both programs. Just from memory, without side by side comparison, I find it harder to create the sharp, stepped looking dropoff between body groups, in DS. It appears that if something falls outside of a bend zone DS still takes a row of polygons and from a neighboring figure and bends then, rather then leaving a sharp dropoff.

Running a real technically valuable comparison is going to take some time, because i want to make sure I use identical poses and numerical values in both programs, and test it on several figures... so give me a few daysto find a bit of time and patience to get throug it.

Also, I'm going to post some technical questions over on DAZ developers forum, see what other technical info I can shake out, about the differences.

I don't particularly subscribe to the idea that "It's a poser bug" or that the kids are 'badly rigged", or taht DAZ is "abandoning Poser user support" , but it's obvious that some differences between those two do exist.

Quote - > Quote -

You can inject joints (and falloff-zones) into a figure's cr2 just like you can inject morphs.

So ideally, if you incejt an "obese" morph or an "emacicated" morph, you could also inject modified joints and falloff zones at the same time.
Same for clothing, of course.

But I usually re-rig all my morphed figures and turn them into standalone characters as I want the joints to be as perfect as possible.

I just wondered why no-one has attempted this as a commercial option for a base figure (perhaps it has been done?).
Would a rig/shape injection system be workable for the versatile figure concept?

Thing is, for the end user who doesn't know rigging at all, but wants a solution to character design via dial spinning that moves beyond V4/M4, it seems like a good idea.
I'd always prefer to use a good base in this way. I did like Apollo for this design potential, for all his shortcomings.
Could be I'm just dreaming.

While this can be done, it's a really good way to get no clothing support for the figure whatsoever.  Just think how much time it takes to adjust JP's just for a single rig.
Now let's say a fiigure has 50 morphs... you could spend the same amount of time refining the JP's for each morph. If it takes someone only an hour to adjust JP's for a single morph (which would be incredibly fast), it would take 50 hours - a long week to adjust JP's to be at the optimium for each morph.  And this is just assuming each morph is to be used at a value of 1. The moment you start using other morph values, the relationship between morphed mesh location, and JP influence zones change again, and you no longer have the optimum. Theoretically, to get best results at incrimental morph values, JP's would have to be ERC'd to responf to the morph value.

Of course, as soon as you start moving the JP's, your JCM's no longer behave quite right, because in 90% of the cases JCM's are very dependent on the JP values staying static. So what do you do, make a new set of JCM's for each morph with it's own set of JP settings?  So, lets' sau a figure needs 30 JCM's... each morph is likely to need a new set of JCM's (that would be 30*50=1500 JCM's).
Now, I'm very fast at makign JCM's, when on a roll, I can go as fast as one very 10 minutes. We're talking about 250 hours just making the JCM's... not to mention a HUUUUGE file. This is just to use morph values at a setting of 1. Now, that is not the kind of figure I would do clothing for - unless I wanted my house in foreclosure.

This doesn'teven account for any ability to start mixing morphs and get different characters....

What i'm getting at is, it's very easy to take something out of context of how the entire machinery works together (the way Joe Public's statement did), and say, eh, it's an easy adjustment. it is not. You adjust one thing, and it breaks and influences 50 other items. This is the limitation of rigging figures in general, and little more specifically Poser rigging, since it's very old technology. Newer rigs, in more advanced software are a tad less sensitive to this, because of advancements in rigging technology. But even there (like in Cinema, or Max etc) rigs have their limitations, and need to be made at an optimum between purpose and performance.

Quote -
So the Daz Studio scaling will work sorta like universal for clothes too? Now thats interesting, I have not worked with clothes much yet in Poser, a bit, and they do have to be adjusted for scaling differently than the figure to get a perfect fit. That would be a plus if you did not have to do that.

No, you still have to tweak and adjust each JP so that it bends properly around the underlying figure. It's actually a bit trickier, and to me it appears to be more time consuming to do it in clothing then then on a figure itself, because you have more then just aesthetics of the bend to worry about, but the pokethrough from the underlying figure. Including accounting for it's JCM's. Often in clothing you end up with more JCM's then in a figure... if you want somewhat realistically fitted clothing, and not something that lookes like a stretched out tube sock a mile away from the figure body.

Quote - I am looking at adopting some of the Daz style scaling especially for  the chest-collars.  In Brad the collars are just a thin line befor the shoulder, not much to work with for scaling. But if I tie them together with the chest (morph forms) they work great. I'll likely end up with a hybrid Daz-Poser scaling. The example I used in the video is over simplified as well, cause you have to remember that you also have other body parts and x and z scale to account for as well. So it gets more complicated than just setting up the y scale, they all have to play nice together.
cheers.

There's no such thing as DAZ specific scaling rig. The difference is that internally, DS soesn't depend on the neighboring body part smoothing parameters like Poser does, but it happens automatically - internally in the program. If the 'bend' on a joint is turned ON, then it automatically smooths the transitions, while Poser depends more on the parameters you showed in your video. I can ask around and get more specifics about this if you'd like more info.

Quote - Won't the the mesh always stretch when a body part is bent? I mean, I know that's why extra density is added in those parts, but they're still gonna stretch some ;o).
Laurie

Of course it will, on any figure, that's exactly what I was trying to show. That's how poser rigging works. This is not something isolated to Kids or DAZ figures.
However, I'm seeing it used as 'proof' that DAZ figures have more issues then 'other properly rigged figures', and that's just not true.  It's just how poser rigging works. All figures are susceptible to it's limitations.

Now, when it comes to whether one wants to adjust aspecific JP's more in one direction or another, it boils down to a difference of opinions between riggers, and what their ultimate goals are for the figure. This doesn't mean that one=bad and improper, and the other = proper, and good.

Quote - erm no, the pose is NOT irrelevant at all.

if you move the figure into a pose that would break a human anyway, of course the rigging's going to break...!

but what do I know ....

Well, if we're going to use that logic, on real humans, biones are not scalable, so then why are people biatching that kids bones not scaling well? Can't have it both ways.

Quote - if you show me the problem using a pose that would break it anyway, your not showing me anything other than you can break a figure anyway.

Good point. See, it all started with people showing those kings of poses using the kids, as proof that 'they're poorly rigged'. I'm trying to stick to comparing apples and apples.

Quote - > Quote - The thing is that it is something that most users would not even attempt, changing the joints is not for the feint of heart. It is also something that you would have to to do for every piece of clothing, etc. And it is not a one size fits all solution. So what you end up with for one thing, doesn't work in another, etc. That is always what the poser scale bug was. It has always been the joints.

Well if you use Wardrobe Wizard to transfer the clothing to the newly rigged figure you'll save a lot of time.

I just love it when people who never actually do things act like they know how it's done.
Can you show us some examples of your work, pelase?

Quote -
To make it work with current rigging system, one would have to anticipate Z Scaling and make a whole array of ERC'd JP adjustments and JCM's for each joint. That would be a humongous undertaking. Not to mention very cost prohibitive to make a figure like that, and especially clothing for it.

And ahem allow me to channel Joe public and quote myself(LOL)  from the previous page of this thread (See above).
The joint smoothing parameters that phantom showed in his video are just a part of JP's that have been a part of the rigging all the time.

A figure will bend different ways, depending on how those JP's are set. Also those may needs a different setting depending on whether a bone is bending forward, back or sideways, and whether the neighboring bones are bent and/or rotated or not.

One would have to be a real entry level rigger to not be aware of those.

Quote - Using a figure that is in alpha is hardly fair, although I can understand why Connie would considering some of the statements made in this thread. :)

So what really is going on with that neck there? Here is a pic  See no broken seams.

Actually, you have stretched seams on the uderside of the neck, and also, you didn't use any of the head nod rotation.  Try compaing apples and apples.

I'm going to re-download your latest, and show you what it looks like when you compare apples and apples.