lesbentley: well, I hadn't looked at the toes for some time since I was going to give them some proper attention after finishing with the hands. But thanks for the tip! It's quite possible that I would have forgotten to fix this.

PS: Don't congratulate me on the foot scaling just yet. I might yet run into trouble when trying to get them to work without propagating scale like I did the hands.

On the subject of the hands. Don't pay too much attention to what I am about to say, I'm only brainstorming, and with JPs there are so many things affecting other things, that improving one thing often makes others worse. That said, it seems to me that if the origin of the hand is aligned to point straight down the middle of the forearm (and parallel with the with the palm), the scope of the twist can be extended from near the wrist to the elbow, without too much distortion of the forearm. It's just a thought for your consideration.

P.S. deleted a previous post, because the values I gave were rubbish.

Thanks again for your thoughts, Les!

My take on this is that hands don't really twist. There's a perfectly functional forearm twist, and all I could achieve by this change would be a duplication of that. I think that little bit of extra convenience for the user is probably not worth redoing the wrist JCMs and risking further problems with the hand rigging. If this leads to confusion, maybe I could just hide the dial for the hand twist?

Quote - Yep! I'll try to do something similar for the feet, and then all that's left of my agenda for 0.9.124 is making some JCMs for the toes and of course copying all the new stuff into the master CR2.

Don't forget about the tongue!  I think you were going to fix the tongue JP's for version 124.  Unless  they've already been done.  In which case...

I'm diggin' the improved hand scaling!  :woot:  Very few figures in Poserdom scale well.  Antonia is one of the few figures which does scale well.  :thumbupboth:

Quote - My take on this is that hands don't really twist. There's a perfectly functional forearm twist, and all I could achieve by this change would be a duplication of that. I think that little bit of extra convenience for the user is probably not worth redoing the wrist JCMs and risking further problems with the hand rigging. If this leads to confusion, maybe I could just hide the dial for the hand twist?

Your probably right. It was just a random idea.

Quote - Don't forget about the tongue!  I think you were going to fix the tongue JP's for version 124.  Unless  they've already been done.  In which case...

Hmm, my plan was to finalize everything from the neck down in 124, and take care of the head parts later on while making expressions. I guess it couldn't hurt to have a look at the broken actor parenting and try to fix that, but fine-tuning the JPs will definitely have to wait until 125.

Quote - > Quote - My take on this is that hands don't really twist. There's a perfectly functional forearm twist, and all I could achieve by this change would be a duplication of that. I think that little bit of extra convenience for the user is probably not worth redoing the wrist JCMs and risking further problems with the hand rigging. If this leads to confusion, maybe I could just hide the dial for the hand twist?

Your probably right. It was just a random idea.

No worries! Keep them coming.

 Hi odf,

I have a suggestion.  It seems that lesbentley is working on the same issue.  Do you think that the Google Code project can facilitate the sharing of the "code" development?  I feel that we are in a kind of "open source" project.

FYI, after I made the rendered picture of Antonia holding 2 water guns, I've found that some issues with her neck, it doesn't turn well.

Quote -
I have a suggestion.  It seems that lesbentley is working on the same issue.  Do you think that the Google Code project can facilitate the sharing of the "code" development?  I feel that we are in a kind of "open source" project.

I had thought about that. But since Poser files aren't modular, there's no point. A CR2 that someone else creates is useless for me because I have no practical way of distinguishing random changes Poser introduces from the meaningful ones. Pose files are much better, if they can be made. Failing that, they'll simply have to tell me.

What we would need for communal projects like this is some kind of specification language which would allow us to write modules for certain aspects of a figure which would then be interpreted and assembled to form a single CR2 file. We'd also need some powerful ways to extract just the information we're interested in from a CR2 that Poser (or D|S) wrote. I'm using some tools in my workflow that represent a tiny, tiny step in that direction.

amy_aimei

Thank you for the Beach Volleyball set.
 
I am having an issue in Poser8 though.

Here are the bottoms when I load them (MAT added for contrast).

And here is what happens when I conform them to Antonia.

Doesn't happen with 'Fit To' in DAZ Studio so I'm guessing it is a Poser problem. Just not sure if it is my copy of Poser or a wider issue.

Quote -
I am having an issue in Poser8 though.

Doesn't happen with 'Fit To' in DAZ Studio so I'm guessing it is a Poser problem. Just not sure if it is my copy of Poser or a wider issue.

I think someone had the similar issue with the older release of my conforming clothing.  You may try to set Antonia to zero pose first, then add the clothing.  Please let me know if it helps or not.

Thank you for testing it in Poser.

Used the same process for both applications.
Load Antonia. Zero her. Load the Bottoms.
She comes in both apps in the same pose. Apply Fit To in DS works fine. Apply Conform To in Poser and I get the results above. I even tried setting her pose to the standard T pose before loading the Bottom  with no luck.

Edit -
I have reinstalled the old set (dated 13 March) and they conform without any problem. It is therefore something that was updated between the two releases.

Hope this is an easy fix - I know nothing about clothes creation.

Quote - I have reinstalled the old set (dated 13 March) and they conform without any problem. It is therefore something that was updated between the two releases.

Sorry, ignore that bit.

Tried the old set again and applied a color for contast and found that the old set has issues as well but in a different place.

Does it help in Poser if you zero the shorts, memorize them, then conform them?  Upon being conformed, a figure will have its memorized settings restored.  If the memorized values contain any posing or scaling, the figure may not fit properly upon being conformed.

Not sure what difference it will make to anything, but, the SR3 for Poser 8 has now been released. I had a quick play and it loads, renders and shuts down faster. I'll leave you guys to see if anything relevant to yourselves has improved.

CHEERS!

Quote - Does it help in Poser if you zero the shorts, memorize them, then conform them? 

Indeed it does. The zero and memorize has solved the problem entirely.

Many thanks cage and again to amy for the clothes.

Quote - Hmm, my plan was to finalize everything from the neck down in 124, and take care of the head parts later on while making expressions. I guess it couldn't hurt to have a look at the broken actor parenting and try to fix that, but fine-tuning the JPs will definitely have to wait until 125.

Whilst there are lots of things one might do with the tongue. It's important to realise that the tongue (or more correctly the jawLower) is one place where Antonia is seriously broken. I think imediate remedial action is called for. Refinements to the tongue, if they are to come at all, can wait till later, but the problem should be fixed so that tongue1 is at least usable.

The problem is that any rotation of the tongue1 actor causes the same rotation of the jawLower. Apart from this being totally contrary to real annatomy, it also means that the tongue1 effectively can not be posed. The problem is not with the parenting, it is a problem with the JPs in the jawLower actor. The solution is quite simple. Delete the JP channels in the jawLower that have tongue1 as 'otherActor', namely:

twistZ tongue1_twistz
jointY tongue1_jointy
jointX tongue1_jointx
smoothScaleZ tongue1_smooZ

Quote - "Does it help in Poser if you zero the shorts, memorize them, then conform them? "

Indeed it does. The zero and memorize has solved the problem entirely.

Many thanks cage and again to amy for the clothes.

Amy, it tight be worthwhile for you to do that with the cr2, to save other users from running into the same problem.

Thanks Les! I hadn't realized fixing the problem with the tongue would be so simple. I'll do that then.

Quote - Amy, it tight be worthwhile for you to do that with the cr2, to save other users from running into the same problem.

How can it be done in the cr2?  I don't know.  Please help me.

Well what I mean is, in Poser you could just load the figure, zero it in the Joint Editor, use the Memorize function in the Edit menu, then save the figure back to disk. I'm assuming there is some way to do the same thing in D|S.

What the above is actually doing at a code level, is setting the 'initValue' lines in all the rotate channels to zero. That must be done for a conformer if it is to be used in Poser.

In your panties, the hip and hip2 are OK, but the initValue lines in the rotate channels of the thighs are none zero, and this is what will be causing the problem.

Very informative Les,

I'm not sure I even know where the Joint Editor is! The SR3 for P8 is said to contain 'auto conforming' though whether this only works for P8's own figures isn't entirely clear. They seem to have made other tweaks too, which are said to bring it in line with Poser Pro 2010.

CHEERS!

Quote - > Quote - How can it be done in the cr2?

Well what I mean is, in Poser you could just load the figure, zero it in the Joint Editor, use the Memorize function in the Edit menu, then save the figure back to disk. I'm assuming there is some way to do the same thing in D|S.

What the above is actually doing at a code level, is setting the 'initValue' lines in all the rotate channels to zero. That must be done for a conformer if it is to be used in Poser.

In your panties, the hip and hip2 are OK, but the initValue lines in the rotate channels of the thighs are none zero, and this is what will be causing the problem.

I did not find that issue with DAZ Studio, that's why I never aware of this issue.  I did not modify the thighs, it is copied from Antonia's cr2.  Should it be cleaned up in Antonia's cr2 too?

Quote - I did not find that issue with DAZ Studio, that's why I never aware of this issue.  I did not modify the thighs, it is copied from Antonia's cr2.  Should it be cleaned up in Antonia's cr2 too?

I think sometimes a non-conforming figure will have a pose memorized to help with IK posing somehow.  Not sure whether that's the case with Antonia.

For conformers, Poser really does require the zeroed memorized state.  I have no idea why the code applies the memorized state when a garment is conformed.  Is this a bug, or some obscure feature from a version long ago?  Hmm.  It's one of many things I wish they'd remove from Poser, or at least make the effect a settable option somehow.

Is this why they have included 'auto conforming' in the SR3 for P8 and in Pro 2010!?

CHEERS!

Quote - Is this why they have included 'auto conforming' in the SR3 for P8 and in Pro 2010!?

That's a good question.  They haven't explained what the feature does, in the P8 SR 3 docs, that I can see.  Apparently you have to have PPro'10 to know what it does.  It's a double-secret feature.

I'd like to know what it does, too.  I don't even know how to invoke it, to test. 

Load a figure, load some clothes and see what happens would be the way I guess.

CHEERS!

I have gotten into the habit of ticking both those boxes since I first saw the feature, for the figure and the conforming figure, before I do the conform to function. While I'm a novice compared to most of you in this thread, it seems to prevent alot of poke thru problems that I had before.
of course I have no idea if this helps or has any bearing on the current turn of the conversation.
:)

Quote - > Quote - Is this why they have included 'auto conforming' in the SR3 for P8 and in Pro 2010!?

That's a good question.  They haven't explained what the feature does, in the P8 SR 3 docs, that I can see.  Apparently you have to have PPro'10 to know what it does.  It's a double-secret feature.

I'd like to know what it does, too.  I don't even know how to invoke it, to test. 

It was the fix for using the Freak. Poser users had to do all the copy/ paste each group to get conforming clothes to scale and fit right. Now you just use that option. Very handy and something that should have been done a long time ago. Specially if you like to change scales on any of your figures and still be able to have your clothes conform right.

It was stated a long time ago that the next P8 SR would have this included as it was a PP2010 only option at first. DAZ and SM are, by rumour, working on more scaling issues between the 2 software to get them more compatible with dealing with scales. This was step 1.

So, that volleyball bikini may work right using this option without having to redo the Cr2.

Hope I did not offend anyone for explaining what this option is for and mentioning a figure by a company that is not to be named here.

Glad you did!

CHEERS!

Quote -
I did not find that issue with DAZ Studio, that's why I never aware of this issue.  I did not modify the thighs, it is copied from Antonia's cr2.  Should it be cleaned up in Antonia's cr2 too?

That's odd. I just checked the cr2, and the init value is 0 for every rotation in every actor.

Edit: Just checked the change log. I fixed the init values for thigh, shin and foot in Version 0.9.122, so you must have used an earlier cr2 as your basis.

Quote - Hope I did not offend anyone for explaining what this option is for and mentioning a figure by a company that is not to be named here.

I think I'm gonna let it slip just this once.

My rant a few pages back was more related to the fact that I resent sifting through a million pz2 files to figure out how DAZ did something, not to mention providing potential fuel for their copyright lawyers, and therefore do not want any more suggestions of the kind 'just see how they did it in figure XYZ'.

I tried to fix the problem as described by lesbentley. I edited the .cr2 and set the initValue to 0 for all rotations.  I tested it with DAZ Studio, it acts the same as before, i.e., it works in DAZ Studio.  

Haruchai, could you try it in Poser?

odf, you are right! I'm using 122's cr2 as the base because it is the one that is available at the time I started.
 
Updated version can be downloaded from 
http://sites.google.com/site/amyaimei/antonia/AmyAimei_Antonia_BeachVolleyballSet.zip?attredirects=0&d=1

Quote - > Quote - Hope I did not offend anyone for explaining what this option is for and mentioning a figure by a company that is not to be named here.

I think I'm gonna let it slip just this once.

My rant a few pages back was more related to the fact that I resent sifting through a million pz2 files to figure out how DAZ did something, not to mention providing potential fuel for their copyright lawyers, and therefore do not want any more suggestions of the kind 'just see how they did it in figure XYZ'.

What I like about your figure is that you're not following someone else's specific way in the building of it. Thank you.

:wub:

Oh, man.  I'm having an ERC-JCM nightmare, here.  :scared:

I'm working on creating some conforming clothes for Antonia, so I have Antonia and her conformed clothes all loaded into a .pz3 file.  The .pz3 gets opened and saved repeatedly, during development.  Most of Antonia's body parts are hidden, through all of this, because of the clothes.

Somewhere along the line, Poser 8 is dropping all of Antonia's ERC from her joint-controlled morphs.  :crying:  :scared:  I've seen this happen before, but never with a conforming parent.  In the past it's only affected conformers.  And it's never been this prevalent.  The problem relates to actor visibility at the time of saving the file, somehow.  I'd thought it wouldn't affect actors which were turned off individually, as that seemed true previously.  It no longer seems to be the case.  An actor which is hidden seems to be at risk of not having its ERC lines saved by Poser.

I'm going to see if all of the ERC can be re-inserted with a pose.  Hopefully....  :unsure:  Antonia has a LOT of JCM ERC, and I don't want to have to keep replacing it by hand.  :scared:

Thought I should bring up this problem here, since it will presumably affect others as well, once Antonia has some garments.

Quote -
I'm going to see if all of the ERC can be re-inserted with a pose.  Hopefully....  :unsure:  Antonia has a LOT of JCM ERC, and I don't want to have to keep replacing it by hand.  :scared:

I think ERC insertion via a pose should work fine, just as long as all the channels involved are still there.

If you want, I can write a quick script to extract Antonia's ERC code into a pz2 and send you the result. Would that help?

Quote - I think ERC insertion via a pose should work fine, just as long as all the channels involved are still there.

If you want, I can write a quick script to extract Antonia's ERC code into a pz2 and send you the result. Would that help?

It might help others, when this happens.  I think I've fixed things.  Hoping....

A pose worked to re-insert the ERC, but I ended up with double listings until I saved and re-loaded the file.

A bit of testing seems to reveal that the problem affects actors which were turned off using a visibility pose, or by toggling the full figure or the figure's BODY actor in the hierarchy editor.  Apparently Poser isn't caching the visibility properly when a pose is used, and a visibility pose is a hack, as evidenced by the fact that the Undo function won't undo such a pose.

When a visibility pose has been applied, the hierarchy editor doesn't seem to properly update the visibility settings of the affected actors.  It looks like this is related to the problem, and this seems to be a new twist which has cropped up with P8 SR 3.

So far it looks safe to save hidden actors with ERC if each actor's visibility is individually toggled using the hierarchy editor or the parameters tab.

So: do NOT use a visibility pose to turn off actors with JCM's, unless they will be made visible again before the file is saved.  That seems to be the only safe way, here.  I suspect toggling visibility with Python would be as bad as using a pose, but perhaps not.

All of this may only affect .pz3 saves.  I haven't seen the problem when saving figures to the library.

Spoke too soon.  The above only helps with the currently open file.  Once Poser loads that file back and saves it again, it drops the ERC.

But it's strange.  I have five figures in the scene, one of them completely hidden, on a per-actor level.  Antonia's head and neck are visible.  But the fully-hidden figure retains ERC and Antonia loses hers.  So the problem is seemingly inconsistent.  And it looks like something that will definitely spell trouble, down the road, once more people start saving Antonia in .pz3 form, with clothing and hidden bits.

Poser has the gremilins.  :cursing:

Quote - Haruchai, could you try it in Poser?

To confirm what GeneralNutt said the new file is fixed in Poser8.

Thanks for taking the time to sort this out.

Did I miss the update or hasn't it been uploaded yet?

CHEERS!

Quote - Did I miss the update or hasn't it been uploaded yet?

CHEERS!

It's in the text above -

Updated version can be downloaded from 
http://sites.google.com/site/amyaimei/antonia/AmyAimei_Antonia_BeachVolleyballSet.zip?attredirects=0&d=1

Cheers.

Thanks for that!

CHEERS!

Quote -
The problem is that any rotation of the tongue1 actor causes the same rotation of the jawLower. Apart from this being totally contrary to real annatomy, it also means that the tongue1 effectively can not be posed. The problem is not with the parenting, it is a problem with the JPs in the jawLower actor. The solution is quite simple. Delete the JP channels in the jawLower that have tongue1 as 'otherActor', namely:

twistZ tongue1_twistz
jointY tongue1_jointy
jointX tongue1_jointx
smoothScaleZ tongue1_smooZ

I'm a bit baffled by this. Unless I have tomatoes on my eyes (as we say where I come from), these seem like perfectly normal 'mirrors' of a child's JP channels to me. The only explanation I have is that because there are no JPs set up between the head and lower jaw, but the lower jaw is set to bend, Poser takes whatever JP channels it can find.

Anyway, I've set bend = 0 for the lower jaw and kept those channels in for now. That seems to have fixed the problem. I don't see any reason why the jaw actors should bent, seeing as they only contain the teeth and parts of the gums, but as long as the extra channels do no harm, I'll keep them in until I'm further along with the rigging of these parts.

While I was at it, I've also made the translation channels for the tongue1 actor visible. I'm not completely sure yet whether I'll use ztran or xrot for pushing the tongue forward in the end, but for now, people can use the z- and y- translations for positioning it the way they need it.

Any plans for visemes or phonemes if people want to do any lip synch!?

CHEERS!

Yep, I'm planning to do visemes.

Quote - Anyway, I've set bend = 0 for the lower jaw and kept those channels in for now. That seems to have fixed the problem. I don't see any reason why the jaw actors should bent, seeing as they only contain the teeth and parts of the gums, but as long as the extra channels do no harm, I'll keep them in until I'm further along with the rigging of these parts.

Turning off "Bend" for an actor will make it immune to magnets.  I've had cause to magnet-morph the teeth, from time to time, myself.  :laugh:  Generally, I think your solution is pretty good.  I fixed the tongue on my own copy of Antonia by adding a set of mat spheres for the tongue1 actor and adjusting them until the jaw no longer moves.

Quote - Yep, I'm planning to do visemes.

That's cool, imagine an animated Antonia telling you about herself in a promo video.

CHEERS!

Quote -
Turning off "Bend" for an actor will make it immune to magnets.

I didn't know that. I guess in that case it would be better to remove those channels right away as Les suggested, right?

Quote - I didn't know that. I guess in that case it would be better to remove those channels right away as Les suggested, right?

If you want to keep the parameters in there for some reason, you could deactivate them, perhaps.  The otherActor name could be switched to a dummy name.  I think bend and side-side parameters have a line that can be edited to set the strength of the effect, which could possibly be used to deactivate them.  I don't think Twist parameters support this option, however.  Probably simpler to remove them and re-insert them in a future version, should that be desired at some point.

But I don't think turning off "Bend" is really a problem, although it might be good to include a statement in the readme about activating magnets for the actor.