Is it possible to use two of these wonderful creatures in the one scene withouth problems being caused by the pbm's. So far I have created a second instances of the blMilWom.obj (renamed to blMilWom2.obj), created a second instance of the Vicky2 Cr2 to point at the blMilWom2.obj, deleted all the rsr's in the geometries and character folders and let Poser build new ones and still no joy as regards the abilty to morph both figures seperatly and differently. Also some of the morphs transfer from the one working figure in the scene to the other non-working one. I have also used MorphManager 4 to strip the pbm's from the body section, but the problem persists. If it is not possible to use two full Vicky 2's together the solution might be to strip one of them of all the pbm's and transfer the morphs as ordinary morph targets body part by body part (by exporting from Poser and loading to the stripped V2 as morphs). Apart from the body pbms what else should I strip to make it work. I like the new joint setup but I want to be able to use two or more Vicky's together, even if one does not have the pbm's but has them as individual body part morphs. Alternativly can I use Cr2Editor to transfer the joint setup to Vicky 1. I wish Daz had given us a second Cr2 with the pbms as ordinary morphs so as to be able to use 2 or more models in a scene. Any help would be much appreciated. Thanks STORM

Select one of the vicky's and give it a a name other than "Figure 1" etc. Do that for both if you have to. The OBJ's and RSR's have nothing to do with the problem, it's CR2 cross-talk. -Jeff

Tried that as well Jeff, still no joy named Anna and Beta in the Hierachy Ed and object properties still the problem persists. Thanks anyway.

Looks like that's one of the bugs that slipped through Beta testing?

The only way that IU can see is this:- Copy Vicky.cr2 (or whatever it's called) to Anna.cr2 . Open it with a text editor. FInd where her figure-name is kept, and change it to Anna. Go right through her and replace "Figure 1" by "Anna" in every slaved morph in all 52 of her segments. Save the file Anna.cr2IN ASCII MODE. Copy Vicky.rsr (or whatever it's called) to Anna.rsr .

I will try it Anthony, Thanks

Storm,
Here is my ERC tutorial. At the top you'll find a line that reads "Click here for the cross-talk update". In the second paragraph of the update is a linked file that will solve your problem. Also, in the FAQ section for V2 of DAZ's website is a link to the question/answer "When I load a second Victoria 2 model into Poser, it copies all the body morphs of the first one, what's wrong?" If you look there you'll also find a link to my website, directing you to the link I've provided above...

Jeff,
Fixed names were at one point thought to be the only solution. That's not the case anymore [refer to above]. The problem is Poser's, not Vicki's. I know for a fact that this was addressed in beta.

Anthony,
Incorrect. That way is not solving cross-talk. The way described is essentially a new figure (even if only a new cr2). Load two of the same figure with that method = same problem.

-Rob
rbtwhiz.com

This isn't a problem with Poser because EMC is not truly supported ;-) I'm surprised an experimental method was used on a new product like this. Everyone get out your Null-Loaders ;-) -Jeff

Thanks Rob

Jeff,
Umm... right, ERC isn't truly supported. ;) Every feature currently in Poser was "experimental" at some point, correct? ;) That being the case... only those directly involved would know at what stage of "experimental" it truly is, right? ;) Not everything is done in the public eye (in fact, most are not).

John,
Your welcome.

-Rob
rbtwhiz.com

Hi Rob!- I'm glad you are in this thread, as I have a question. I'm going to include the null-loader with the next Supermodel Vickie upgrade (she will also be back to :1), but I noticed that when I reopen a file that worked O.K. originally I had cross-link problems, have you run into this? And when you load the figures, should the null-loader's body part BODY be selected or the body itself (in which case Poser indicates "No Actor")? Questions, questions...

Jim I just tried Rob's Null Loader, I changed its name to "Figure 1" after loading and selected body + No Actor and it worked fine. I have it re-saved as "Figure 1" in the Library. I also tried the other solution on the Daz site of loading Vicky 1 first and then the Vicky 2 models and that worked as well, although I think Rob's solution is better as it is not a memory hog! Thanks again all of you Now for my next problem.....;o) John

I guess I did it wrong then, as I had the body part "BODY" selected. I'll have to give it another shot.

Rob, Are saying that setting up unique characters, with their names as both the figure name and the controlling figure name and using only one instance of each figure will not work? Or that it is not a solution that you prefer?

Will this "Null loader" work on a Mac? //Mike

OK, thanks Storm. //Mike

bobsmash All that Rob's Null loader is, is a cr2 file that references no geometry when you load it (I do not mean to belittle it in any way). As a result there is nothing to see on the screen, but for the program it exists and that is the key point. Poser believes there is a figure loaded and this - if you name the Null loader Figure 1 (space between Figure and 1)- stops the cross-talk between the subsequent figures loaded and allows you to have 2 or more Vicky 2 figures loaded and fully working. These subsequent figures load as Figure 2, 3, etc. I don't use Mac's but if a Mac can use a cr2 file then it should work. Otherwise do whatever one does to convert a cr2 to Mac format and try it. BTW Rob's Null loader comes in at only 4kb unpacked compared to over 20 Megs for the Vicky 2 cr2. Very nice going. STORM

Jim,
If SMV uses the same naming convention or less than the current list of actors for the Millennium figures, there is no need for an additional null-loader. The one I've posted will work. Files that are saved into the library (and pz3 scene files) from within Poser do have problems with cross-talk. This is because of the way poser currently handles the assigning of instances to figures and groups. Until a better solution (one beyond what Poser is currently capable of) is implemented, or a utility is written to handle this, manual editing of saved scenes/figures will be required. It's as easy as global replace, but only for those brave enough to tackle cr2 editing. The problems being, when a new figure is read into the scene, the next sequential instance is assigned for like figures and actors. Figure 1... 2... 3 and so on. Such is also true of :1, :2, :3 etc. When you save back to the library poser doesn't re-assign ":1" to the actors, it uses whatever the currently selected figure/actor instance is. That said, when you go to load that figure into the scene (the cr2 already has some instance other than "Figure 1" and ":1" assigned) poser sees that the null is occupying ":1" and that the new figure isn't competing with it so it lets the figure load with whatever instance the cr2 indicated. If that's the case, the null will not work. The null has to occupy ":1" and the loading figure has to request ":1" for poser to re-assign it correctly. Body itself, the root (no-actor) should be selected.

Dean,
Fixed names work... but as I said above, your not really solving cross-talk with that method because you essentially have a new figure (although it's almost identical). Load two of the same fixed name figures and you land back into the same spot... cross-talk.

Mike,
Yes. There is nothing PC or Mac specific in the cr2.

John,
No offense taken. It's true... It's a stripped down cr2 consiting of version number, a list of actors and figure section (parent/child relationships). Pretty simple.

-Rob
rbtwhiz.com

Or Vicky 2 is just not a lesbian?

Mesh, Its a joke, I know, but implied in it - this has something to do with V2 in particular. This is not the case. The problem is with Poser, not V2 / V1 /P4NF / Mike. If you have two figures (any two figures) that have FBM (or other ERC) there can be a cross-talk problem. V2 just brings the issue to a head with the greatest number of users. Rob, OK. As I have said, my way IS to cast from a library of many individual characters. So unless I are doing twins or clones and get lazy, there should be no duplication. I really like the JCJ. I added eyes to my list. But imagine the cross-talk potential with this.

Rob- O.K., that is pretty much as I understood it. The only difference on SMV is she is called "Supermodel". not "Figure 1", but as your CR2 doesn't actually mention "Figure 1" (Only part (whatever):1) assume it makes no difference. It did seem fine until I opened the save later. And I wasn't planning to "use an addittional null-loader", I was planning to use yours (!) in the Zip, with proper credits (and instructions), of course. Incidently, if you do a global search and replace in Wordpad on Vickie's (or SMV's) CR2 it will take about 10 minutes, Word will do it in a couple of seconds.

Jim try EditPad Lite (freebe) it does the stuff in seconds, it flies through the Vicky 2 cr2 and it has great search and replace tools. I got the link from Rob's site some time ago (another thing to thank him for!) and I have to say it is the best text editor I have ever seen and I have tried a few. http://www.jgsoft.com/ John

OK THEN ITS TWO LESBIANS THAT DONT SPEAK THE SAME LANGUAGE? hEH hEH

Jim, welcome to the ranks. I saved a character CR2 from Poser. Opened it in MM4 - copied in some JCM. Did the find-replace in EditPad - new character name Saved it back to Poser library. I noticed that MM4 had increased the file size by 2 meg I opened it in Poser and saved it back. Reduced the size alright, but It put Figure:1 back everywhere but the single name slot. all JCM back to being controlled by Figure:1 Poser can be perverse. The EditPad step pretty much has to be the last one. If all your characters are Figure:1 controlled and you burn the Figure;1 cast member with the Null-loader everything should work. If. But, much as we may wish, it is not a "fire and forget" solution. I do not see anyway to protect a user from having to be aware of the problem and its fixes.

Anthony, Incorrect. That way is not solving cross-talk. What I intended in message 5 above was, after that, load one Vicky and one Anna. Of course the crosstalk will come back if you load two Annas! [Poser] put Figure:1 back everywhere but the single name slot. What is this now?? Can't I even change the pointers in the slaved morphs and trust that they will stay changed!? What's going on???? Where can we download this "null-loader"?

Rob has a link from his site for the Null-loader. Anthony, it will stay changed as long as you do not resave the character to Library from within Poser. Or, if you must, you redo the find-replace in EditPad, and put this back in the character folder. Poser does not change the entries when you open the CR2 (at least I don't think so) only when you save it. I thought I had it fixed for good, when I changed the pointers, too. It was a nasty, nasty surprise that the fix would not stick. We will have to wait for Poser 5 and hope they can and will fix this. Things can get interesting with a Null-loader if you have a lot of clothing and more than one character. More than one clothing item could be Figure:2 and if there are morphs...

To try to fond the "null loader", I just called text search in the Poser Free Stuff for "null", and nothing showed up.

Anthony, Rob gave the link in his first post above, see "Here" or http://www.rbtwhiz.com/rbtwhiz_EMCPrintable.html The file is linked on the page of his tutorial. John

As to the clothing cross linking, I know Poser will put references to the master figure in the CR2 if you save the set with it conformed, I've always assumed if you save it unconformed, with the master figure deleted, it dosen't, but I've never actually checked that. I normally set up JCM on the clothing to work from its own joints, not the joints of the master figure. Gee, this is getting complicated!

Jim, I think it is as you say. But just which Figure even a solo item of clothing is? It probably depends on what was in the window when it was brought in, even if everything else is deleted before you save it. I've not done CM, but I imagine it must be setup in EditPad and the user told to not save it back from Poser, unless prepared to also use EditPad on the saved CR2. It is scarry to even think about trying explaining this to customers, some of whom have trouble with the lashes. I have only gotten one question about the technical part of my JCM and am not sure if cross-talk was involved. There was a language barrier and no reply about my explanation. Way too much of this is like dropping rocks into a bottomless well.

Attached Link: http://www.daz3d.com/pages/faq/faq.html

Mmmm! So, if this affects P4, we should expect an update from CL to cure it (4.1?) If it only affects Vicky2, we should expect an update from DAZ (v2.1?) Either way we are the customers and shouldn't have to mess about trying to get round problems inherent in software we paid good money for (Legitimately bought software works both ways CL) The products are pretty darned good, so I'm confident we will see these companies look after their user base. Mmmm!

This all comes from the way that morph slaving was implemented, e.g.:-

                     valueOpDeltaAdd
                                Figure 1
                                BODY:2
                                SuperHero
                        deltaAddDelta 1.000000

Everywhere else that a part of one model points to a part of the same or another model (ordinary parenting, IK-parenting), it does not state the name of the model pointed to but indicates which model by the colon-and-suffix number on the part name. It is a pity this was not done here also. But that is "water under the bridge" and we are stuck with the method that we have, for now.

Anthony, It is fortunate that we can alter the controlling figure. It is unfortunate that we cannot lock it. If it were restricted to just within the same figure, there could be no CM. The function may be best used with props and clothing, but we may come to value it. But you are right that for figures, it would be easier if it were understood that it is always within the same figure.

If it were restricted to just within the same figure I was not wanting to restrict it to within the same figure. I can parent or IK-parent something to a part of another figure, and the (model that the parent part is in) is indicated by the colon and number on the end of the part name.

I also noticed yesterday that if you open a figure in Poser that is non-figure:1 and then save the set it will make it figure:1, and point all the internal references to figure one. But it has to be the first figure opened after Poser was started, if any other figures were open first, even if deleted, another number will be used when the set is saved. This can save the labor of editing the CR2 back to figure:1, which is part of the requirement to get the null loader to work, incidently. Anyway, I think the easiest way to fix this mess in Poser 5 would be not to allow cross-figure-linking at all, which would of course mean full body morphs would not work as intended, but this would be a small sacrifice to make. JCM morphs should still be allowed to run from the same figures parts. That is to say figures would only reference themselves. CL could perhaps even to this as an update to Poser 4 and PP.

Anthony, I sort of suspected that the colon & number offered options, but I have not built any props that needed any of this, so I have now pushed it. Doen't Poser also change this number on resave? I have not done extensive testing, but, if the colon and number are absent I think JCM will still function. Jim, The find/replace in EditPad is just a few steps, not really labor. The trick is to not have a "find" that conflicts with the geometry name or the name of any textures. "Figure:1" ":2" ":3" does not involve anything but ERC so it is safe to change globally.

Here is part of the read-me included with the SMV JCM upgrade: ***************************************** The cross talk problem: If you try to load two Supermodel Vickies in the same scene you will get cross-talk, where moving the joints or full-body morphs on one will run the conforming joints or full-body morphs on the other. It must be emphasized this is not a Supermodel Vickie problem, or a Vickie problem, it is a Poser problem, and it existed with the Posette, we just never noticed. The answer, for now, it to use the inclosed null-loader. To use it: Restart Poser if any figures have already been opened. Open the null loader set With the "Body" part of the null loader picked from the bottom menu (Poser will respond with "No Actor") load Supermodel Vickie. If you load additional SMVs or other people, do the same thing. It is also important that SMV be figure:1, internally, as she is now. If you want to change anything on her, and resave her as a new set, make sure you do this as the first figure since Poser was started- if necessary save the set, restart Poser , open the figure and resave the set. Does this make any sense? Null loader and instructions provided courtesy of Robert E. Whisenant (aka "rbtwhiz"), visit his site for more information. http://www.rbtwhiz.com/rbtwhiz_EMCPrintable.html

Jim, It is hard to explain, isn't it? I understand it, yes. You have made the choice to fix on Figure:1, it seems. You are using CM on the skirt? Or does a JCM in a clothing figure respond to its own joint control, if the joint itself is set by the figure it is conformed to? Wait. A skirt cannot have buttock/thigh/shin to be conrolled by the figure. It can have one of them , but you still have that damn hip section in the middle. How do you do an automatic skirt morph without CM? If you do have CM and everything is set up for Figure:1 , what about a second SMV with a skirt? Its control would be Figure:1 ? Does the Null-loader allow Figure:2 (or whatever it is, depending on how much clothing comes in on SMV:1) to control the CM on a skirt conformed to it? By the way, do you think anyone is reading over our shoulders? I get almost zero feedback, how about you?

Well! Jaager! Maybe, it's because you are taking about stuff that's way too much technical for most users! :-) I myself tend to stay clear fo JCM or whatever you call it! I Program in Assembler a lot(yes! I'm a dinosaur! LOL) So, when i enter the "poser universe", i like to just have Fun with it! Although, to you Wizzards! This is probably Fun! :-) But, i still enjoy reading your exchanges! They kinda stimulate my Brain! :-) Later! Marco

Dean/Jim, Sorry... I'll respond with answers later this evening when I get home from work. The connection here is horrible... and the time I have left here today needs to be directed toward an ongoing project (otherwise I'd answer right now :)). Please bear with me if there is a lapse with responses, I'm in the process of re-building my workstation (home/studio) on Win2K and it isn't playing nice quite yet. For what it's worth, I am following the thread... as I can. -Rob

Rob, I was not meaning you. I trust you to be there. I was asking about "civilians". At least Marco is keeping a eye on us. In my mind, part of this is us exchanging ideas and part is a panel discussion in public. I hope that there are some users who are at least understanding that there are questions to ask. We certainly understand that the only way to get this stuff is to do it. It is a lot harder to explain than it is to do. It is logical when you understand the rules. But first you gotta understand that there are rules. It is not only creating the ERC morphs and channels, it is also using them. I have a product with ERC and Jim does also. It would be nice if we could help more than one person at a time. It is about heppin' , not a star turn.

Anyway, I think the easiest way to fix this mess in Poser 5 would be not to allow cross-figure-linking at all, which would of course mean full body morphs would not work as intended No need for that. All we need is one of these:- (1) Poser 5 to ignore a slaved morph's master model name and to look for the master channel in the model specified by the colon-and-number at the end of the part name, same as when a part looks for its IK-parent. That colon-and-number to be automatically updated accordingly when the model is loaded onto stage, by the same mechanism that keeps those colon-and-numbers updated elsewhere when a model is loaded onto stage. (2) Poser 5 to update the slave's master name correctly when the model is loaded onto stage, so there would be no need for fudges such as text-editing and null-loaders.

Jaager If these forums equate to the University of Poser then this thread is a rare Post-Doctoral research seminar. There are many of us reading and learning. Please keep it going. John

I do clothing JCM to run from the joints of the same figure, it seemed the logical way to do it. When you conform the item and move the parent figure the morphs work, even though the dials indicate the clothing's rotations aren't moving, of course. As to skirts, sometimes I do hip/buttock, sometimes (mostly) hip/thigh, sometimes you can get away with just hip mesh and the rest as phantom parts, depends what moves me that day! Poser will not do-inbetweens more than one part away, (that is, the joint parameters fall off zones and stuff only effect the next part back), so that comes into the calculations.

Jim, the JP seem to be somewhat directional also. With joint controlled joints I see a difference. What I mean is: I wished to link the rButtock to the rThigh, so that by moving the master, the slave would follow in the way I defined. The thigh is a bigger target and it has higher limits than the buttock. When I made the thigh the master and the buttock the slave, the joint blending between the buttock and the hip weren't right. If I made the buttock the master and the thigh the slave, everything is cool. I have to go to the thigh to get extreme movements finished. But this way I am sure I get a 1:1 between the thigh and buttock in the buttock's range of movement. (It is a method to fool proof the buttock-thigh, collar-shldr, abdomen-chest, neck-upneck-head movements.) For the skirt, I was thinking of a long skirt and a CM that would allow the bottom of the skirt to follow a kicking foot. There could be no conforming for this since there is no practical way to do a shin group in a skirt. But a shirt morph controlled by the xrot channel of the figure would do this. I am thinking that using a null-loader(s) on a figure that has this and is the second or third character in a scene could get sort of interesting.

Sure you can have shins in a gown, as long as you don't have any mesh! When I did the Hi-Style gown (which is floor length) I put phantom shins in it, and ran the JCMs that do the actual work off them. I was sort of suprized when it worked, but it did (and does). I haven't noticed the directional JP, but I haven't tried any interlinking like you mention. I am a little afraid of doing this because, as you already mentioned in another thread (or maybe it was this one, I'm loosing it!) that messes up using existing Poses. I would really hate to not be able to use the two sets of Digiport Poses that were done with SMV, as the Poses are soooooo good!

So you have rotation channels for a non existent shin that control morphs? Are the values for the shin channels on the conforming skirt set by the corresponding channels on the figure? If so, for a long skirt , you can just call the whole thing 'hip' and setup phantom buttock-thigh- shin groups and run JCM off of these. A lot more work to have to actually do the morphs than letting the JP do the morphs on the fly, but much better than what we have now. With JCJ and poses - it's not so much that you lose the poses as that you have to adapt them. An example: I have the butt:thigh as 1:1 and the bend limit on the Buttock as 40. So if I bend the buttock to 40, the leg goes at total of 80. The thigh dial shows nothing. (Unless I double click on the thigh - then the numbers show up - whatever That is all about - and then the thigh adds this value to the one it gets from the buttock. Weird. I think this means that a JCJ figure can be used to setup poses for a regular figure- if you d-click on each of the slaved elements before saving the PZ2 file. Oh, and IK on - does it too I think.) If I apply a regular pose the buttock takes its value from there - the thigh - the buttock value plus whatever value the pose has for the thigh. The fix is to add the dial values from the pose for the two parts divide by two and reset the buttock to this - setting the thigh to zero. If the value is greater than 80 - this additional value is the thigh value. ( Do this for all three rot channels and you get the pose back). While SMV poses may be correctly done, a whole lot are not. If you have a pose with a negative bend for the buttock and a positive one for the thigh you still add them with their values and you get the pose, but you are no longer trying to crack the femor in the middle. I look at some I did and wonder what I was thinking. But I do not intend to inflict JCJ on anyone. When you start throwing JCM into the mix, the 'any old way to get there' method can backfire. I have some gravity JCM that do the opposite of what you want when the figure is learning against gravity. I have figured out a way to get it to do what you want from within Poser, but the pose has to be correctly done. The JCM tied to the Xrot of the hip are going to be a particular problem. I would be surprised to find any pose with the hip Xrot correctly set re:gravity. There has been no need for it to be so.

Yep, the phantom shins pick the value for rotations from the parent figure and supply the action that runs the JCM. I think Nerd did a dress that everthing below the waist was run by JCM, it must have been a ton of work! I started setting the dress I mentioned that way, and gave it up as a task thet would take more time that I wanted to put in it. I think the Posette can sit in his, SMV can't quite sit in any I've done, though she gets pretty close. I also fit my stuff a lot closer than some of the others, this has a lot to do with how good your joint settings and morphs have to be. Anyway, I haven't even thought about doing gravity JCM, SMV is really just setup for outer space!

A large proportion of what I have done and am doing for Millie is meant by and large for use with a nude figure. Clothing would restrain most of the breast JCM. The skirt, My guess: once you do one skirt with phantom groups the script should paste into any other garment. One should work for anything. If you mean the morphs? Yes. Clothing morphs are difficult. Any morph that is limited by or has to match something else is a lot more work than one created out of whole cloth. To do one that anticipates a leg bend can not be much fun. In a metaphorical sense, I keep a candle burning in front of a icon of MASA for MTMirror. Not having to duplicate a mirror image of a morph by hand is wonderful.

Does MT Mirror work on the same body part, or just handed parts? I do a lot of mirrored morphs on handed parts right in Max- I attach the part to the other hand and duplicate all the operations. Was driving me crazy (it is a short drive!) when I was doing the elbow JCM, which is why there isn't any elbow JCM in Supermodel Vickie's update. I find the CR2 editing pretty easy, actually. I've done enough now that I can make a good guess on the inital setting, and I get the setting right that second or third try most of the time.