Another question about D/S. While I do understand that D/S does not support Magnets or Magnet sets (wyrmmaster's, William the Bloody's) much to my disappointment, I don't understand why non-magnet morph sets such as Posermatics NBS product or his NGM product don't work either. Could someone please explain if possible. Thanks in advance. NBS product (Nudity): http://market.renderosity.com/softgood.ez?ViewSoftgood=33884 NGM product (Nudity: http://market.renderosity.com/softgood.ez?ViewSoftgood=36815

Attached Link: DAZ|Studio forum at DAZ

It looks like it should work in DAZ|Studio (and doesn't SF have one of these sets, MS? The reference might be burried in the hidden section of the FOOL thread). The only reason it might fail, and I can't see any reference to it's using the technique in the readme, is if it used a pose to change joint parameters - that certainly wasn't working in DAZ|Studio, though I haven't checked for a while.

(I could ask her ;)

RHaseltine: As you said "It looks like it should work in DAZ|Studio" but I've tried both sets in version 0.9.20.1 of D/S with no success. It's like you never clicked the thumbnail. I had thought ok if no mags then these should work. I sure wish D/S had something that would accomplish somehow what you can with mags (clothes fitting morphed figures, morphing figures etc.) the V3 Head Sculpture pak (or something similar)is greatfor the head but one for the body would go a long ways. D/S would be my main 3D App if it wasn't for this lack. Anyway........

MorriganShadow: You said "I could ask her" Would you please :-)

Message edited on: 04/25/2005 13:40

:D I've asked, and I've found someone who has that set, and he's going to see if it works for him, or if he can find a workaround in D|S tonight. As soon as I find out, I'll post the info here :D MS

Ok, as of right now, I can't find anyone who can make it work. However, they are submitting it to DAZ as a bug report, and it may, in future, work perfectly. ;) MS

Thank you for checking this out MS.

Update: Although there is still an error popup, the morphs do work, but not perfectly. The collar/chest morphs work, but not the fbm's. And, he's gonna submit a bug report, just to make sure ;)

Thanks Again MS. I tried again in the latest version (0.9.20.5)after installing today and did get some response using the collar/chest morphs with very weird results and no response using the body dials. Sure hope the bug report produces results. :-)

Hi, My guess is that the problem of compatiblity could be in the ERC relation between the morphs because if you select each part and activate the morph they work ok. I'm sure that is the syntax of the erc that could be the problem. BTW the PBM in the body of the latest version of NBS are targetgeoms intead of valueparms but I had make a valueparm version and it doesn't work either.

:D Thanks for that info, Posermatic! Hopefully, if enough people beg DAZ for a fix for it, it will be included in a future release! Cheers! MS

MS Any suggestions on the most effective way to beg DAZ? :-)

Basically, every time something doesn't work, file a Bug report ;) It may be the product, it may be the program, but either way, they'll either let you know what the problem was, or they'll fix it :D

Having been directed to this thread from one over at DAZ, with a request to include a sample image or two ...

The NBS and NGM products can be made to work, and work fully, in D|S - one just has to overcome some mistakes in the Injection files which Poser allows for but D|S doesn't (and they are mistakes - Poser is coded to allow for a lot of mistakes and figure out what was really meant, whereas D|S's Poser-format file importer has not been, by and large.)

Here, for, example, are two purely D|S renders (and these were just quick, with no time on lighting or anything) of a V3 character with both NGM and NBS. She has 0.75 NGM-D in both, plus two NBS morphs applied in the first and one different one in the second, all dialled using the Body dials only - so yes, it can be made to work.

I'd better just note that I'm using the version supplied in the update to the product here at R'osity, so if there have been other changes for the R'otica version the following may not apply.

The main problem with the NBS product in D|S is a mistake in the ERC code in the injection files for each of the chest and collar morphs. In the files provided, where the control channel is named, the 'dial name' of the morph has been used instead. Thus instead of e.g.

valueOpDeltaAdd
Figure 1
chest:1
MIC_01

we have
valueOpDeltaAdd
Figure 1
chest:1
PBM-ArmPsh-Dw-1-Bt-cH

It is the actual channel name that must be used - the MIC_... - and NOT the dial name. If this is corrected (and I've done a few manually to check), then the collar dials slave to the chest and the chest dial slaves to the BODY dial as they all should.

Now, Poser (well, Poser 4 anyway, which is all I have) actually fixes this itself, so if you have Poser then the workaround is to load the NBS V380 in Poser, inject the NBS morphs, plus the NGM morphs too if you have them and want them, save the character with the morphs injected, and then load that CR2 in D|S. When Poser saves the NBS/NGM morphs in a CR2 it corrects them and puts the channel names rather than the dial names in the ERC code, so the imported character works properly in D|S.

If you haven't got Poser, then you actually have to go through all the Injection files (all seventy-whatever of them!) and correct them manually in a text or Cr2 editor ... which is a right pain, as I had to do that to the original NBS release before the update. What makes it even worse is that there isn't a completely simple relationship between all the dial names and channel numbers - sometimes the MIC channel e.g. the collars are on is different to the channel the Body dial is on, etc.

So a corrected update would be nice. :-) I do believe vendors should write correct code in their files, not rely on Poser's capacity to sort out mistakes. With the correct code, D|S will be able to use the product properly.

One other thing. The Injection files also contain another syntax error. Every single one has 6 closing curly brackets at the end when only 5 brackets are still open at that point. Hence when you inect, D|S flashes up error messages, as well as writing a long line about the error for every single one of the 70+ injection file in its log. As of the last release of D|S, this does NOT stop D|S loading the injections, so if the ERC channels are fixed as above but the brackets aren't, everything will still work. However this also should be fixed, both for neatness and simply to stop the annoyance of the error messages and the bloat in the log.

DavidGB, thanks so much for those tips!!! It's great to know that there is a workaround, even if it does take forever, lol. Also, I'd let the merchant know that there is a way to make this work in D|S AND that there is a userbase for the product to work in D|S ;) There's always a good incentive for that ;) MS

DavidGB, Thank you so much for your time, energy, wisdom and workaround for us users of Poser/DS & just D/S. I also hope the merchant (who posted above)will consider the changes you've suggested, as it would make their products useful to a much wider range of buyers. These products if functional (as in Poser) would add greatly to D/S, as D/S (in my opinion)lacks greatly in giving the user the ability to morph figures (no magnets etc.). At this time I don't feel I can recommend them to D/S users. As MS said "....there is a userbase for the product to work in D|S." If fixed could increase sales $$$ :-)

I have IMed the merchant to point him at this thread again in case he doesn't notice it. By the way - if anyone has the V3 NBMs and Stephanie Petite, if you copy the channels (e.g. with MorphManager) from the Victoria 3 80 CR2 to a SP Cr2 and inject the NBS morphs, they work fine on her too as long as (in most cases) you keep the morphs down to a max of 0.5 or so. They'll also then work importing that SP into D|S with the same problems and solutions as above. If you are prepared to pay, then you can actually create morphs in characters and clothes without Poser for use in D|S, at least so long as the content came in Poser rather than native D|S format (i.e. CR2s and .obj files etc which most of it does, not Daz files). Utilities like MorphMasterPro (or the free cut down morphmaster) and the Tailor will work directly on the CR2 without it having to go into Poser first (or at all), and then you can load the output CR2 with the morphs in into D|S. As for creating morphs in Poser with magnets for use in D|S - be advised of a problem I discovered when doing this. Poser has some automatic function for smoothing the mesh at the boundaries between body part groups which D|S doesn't. What this means is that if you create a nice looking morph in Poser with magnets that affects one group right up to the boundary with another group but doesn't include the other group in the magnet's effect, actually Poser will be smoothing out the boundary, but in D|S there won't be any smoothing and you can get a hideous disjoint at the group boundary. MorphMasterPro and the Tailor, however, also don't have any smoothing, so morphs created in these apps show up in D|S just the same as when you created them. After being bitten by this a couple of times, I either do my morphing in MMPro, or finish off in MMPro tidying up the group boundaries if I did use Poser magnets to start with.

Interesting. Again. Thanks for the info.

Hey David great work! I will start right away checking my files. As for the example on the ERC entries, as long as I recall the chest do point to the actual MIC channel name but can't remember of the collar files (my memory is not as good as it used to be). Well had to go to the drawing board, I'm going to start with the NGM package since it is the recent (and fresher in my memory) one. Posermatic

----NGM is now DS compatible and I will submit it as an update tomorrow. The problem was indeed in the ERC relation and in the incompatibility between PML(Poser markup language) and the DAZ Studio version of it. 5 lines are needed in order to be able to control the morphs through just one dial and those lines must be in each body part and in each morph you want to control. The fourth of these lines most point to the internal name of the morph in order to work, if you fail to do this the ERC would not work, to PML is just black and white. If this were a code error as Dave implies, the NBS package would not work in Poser either so this is not true but Dave is correct stating that not all the INJ files point to the Morph Injection Channels (MIC_) but this also is not an error he is just forgetting that the Injection files load in an specific order and that they load pretty fast, so you could trick Poser and make somethings not meant for it work. To PML is still B&W but to us is gray :) Why trick it? because for me at the time of coding was easier to keep track of all the ERC linking. Believe me, when you have 73 dials spread in 3 body parts (that's 219 files!) pointing also to one of them and then to the body and also do the hide/show/rem files (900+ files, yes almost a thousand files!) you start thinking in alternate possibilities! So I decided to make the collars point to the morph dial (the 4th line) instead of the MIC name, as they load, to Poser there is no MIC name, it is the morph dial name so no problem for us either. You could see that the chest INJ files points to the MIC name because that was the final part and was easier to keep track so no trick was needed. I have received a couple of emails asking me about the possibility of porting NBS to DS, and I have tried to do it because customers are first, but since I don't have any reference of how a DS file is and DAZ has been really hermetic regarding this I quit trying, and because of this never thought that the ERC linking in the collars was the code that DS couldn't process until Dave find it out. Dave, NBS doesn't need to be "fixed" because it doesn't work on DS and certainly I had write the correct code for the application for what I make this product. You need to fix things when they don't work properly for the platform advertised for, which is Poser 4 and 5, so if it doesn't work in a platform that is not advertised for, that is not an error my friend. "The Injection files also contain another syntax error" if it were a syntax error it wouldn't work in poser either and let me tell you that brackets in PML are very strict because a missing bracket in a cr2 and your figure wouldn't show at all. The 6 brackets do work on Poser because they are out of the open-close relation in the file, yes the file would be cleaner without it but they don't harm anybody ( well except DS :( But no hard feelings on this Dave, and as in my other post let me tell you again Great Work! ------What will happen with NBS? Because I had no much time left every day (daywork etc...) reworking NBS from the ground up to make it DS entirely compatible is impossible to me right now. So I will take Dave's idea of the cr2 saved from Poser, it's a quick&dirt alternative but it works well and I could make it fast so no long time to wait before everybody could do renders in DS. I will double encode it and post it as an option for the Daz Studio users this wednesday! Posermatic

This is great stuff, Posermatic!! Thank you for taking the time to do this!! MS

Thank you Posermatic & DavidGB. This makes what I consider to be two very valuable tools, for those wishing to pose female characters in a much more realistic manner, an even greater asset to users of not just one app, but both.
Thanks Again,
I'm a very happy camper, :-)
Strixowl

Message edited on: 05/11/2005 10:22

Alf,

I'm glad you are fixing NBS and NGM as they are great morphs. But what you write above is nonsense. And that worries me. It worries me because between the two products I've had to spend hours fixing them to work properly even in Poser, never mind D|S, and if you believe what you wrote above, then that means that when your next product comes along I'm going to be weighing up whether whatever it is is worth both the cost and probably hours more of my time fixing it rather than just the cost.

Brackets

You made a mistake. Plain and simple. In your injection files, above the deltas, you open 7 brackets and close 2, leaving 5 open. After the deltas you then close 6, which is 1 too many and a bracket mismatch. Look for yourself and count them; or use a text editor that checks bracket matching, and it'll tell you the same. Or D|S's Poser-content importer script, written at DAZ who invented injection poses and know a thing or two about them. Poser is notoriously fussy about some aspects of syntax being absolutely right, but equally notoriously lax about other aspects being wrong, and this is one of those instances. It's been letting you get away with it, but that's no reason to go on doing so.

It's a mistake. Everyone else's files I have here have matching opening and closing brackets; yours don't. Please don't invent spurious excuses for it really being right. It's not right, but it's a mistake easy to learn not to make again. If there's 7 opening brackets, there should be 7 closing brackets. Period.

ERC Channel names

ERC was discovered and passed on to the wider community in 2000 by Nerd and rbtwhiz. Nerd, in his instructions, says to use the internal AKA channel name - the name after the targetGeom in a morph channel - for the ERC line naming the control channel. rbtwhiz pointed out that ERC will also work in Poser if you use the external AKA dial name, but strongly recommended using the internal/channel name and not the dial name.

That stuff you wrote above about speed of loading and tricks is nonsense: your products worked in Poser because Poser will identify and slave to the master channel whether the internal or external name is used. There's no difference in functionality with the exception of the proviso below.

The reason for the recommendation to always use the internal/channel name was because of a potential conflict - it could happen that the same name is given to one channel as its channel name, but to a different channel as its dial name. In that case Poser always gives priority when assigning an ERCed Master channel to the channel with the matching channel name over one with the matching dial name. If everyone always uses the channel name for the control channel in the ERC code, then that avoids the problem. Plus when Poser saves out a CR2 containing ERC code, Poser always uses the channel names for the control channels even when it was given the dial names when it loaded or injected the character.

So - Poser gives priority to using the channel name over the dial name; Poser always writes out the channel name and not the dial name; and the 'fathers' of ERC say to use the channel name and not the dial name. And guess what? Every single example of ERC code I've looked at in figures and injection files by DAZ or any other merchant where the dial names differ from the channel names uses the channel names not the dial names to name the control channels in the ERC code ... except yours.

For these reasons - and removing the potential source of conflict, I imagine - it's not surprising that the script in D|S that imports Poser content does NOT allow dial names in the control channel of ERC code but insists on the channel name. It's even less surprising when you consider that rbtwhiz, one of the discoverers of ERC, is one of the D|S development team, and I think is the person who writes the Poser-content importer script! And so your ERC doesn't work in D|S - yours alone, out of all the GigaBytes of Poser content packed full of ERC I've loaded into it. There's no tricks involved in Poser meaning that using the dial names lets something happen that wouldn't with the channel names. You are just going against what everyone else does for Poser, and locking yourself out of the growing D|S market at the same time.

Producing this kind of content so that it will work in D|S as well as Poser doesn't need anything special doing. All it needs is producing it properly and accurately for Poser, and not making mistakes that Poser's lax file parser will let you get away with.

I don't mean any of the above unkindly. I'm just trying to make the points forcefully because I just hate to see a merchant as talented as you block himself from potential sales unnecessarily or annoy customers with the usability of the product for the sake of your misunderstanding a few things.

So in morph injection packages, match brackets and use the targetGeom channel names for the control channels in the ERC code - please, it's the proper way to do it and all it needs.

And good quality morphs, but you're already good at that bit. :)

What?! I have come here in good faith and now I'm "annoying customers with the usability..." and also now I'm lazy trying to "get away" with something and also what you start saying that was your discovering no turns into "hours fixing it"? Spurious excuses? Lazy? Try to do a product that at the end need 1000 files and then tell me if that could be considered laziness. Annoying customers? You are the only one in HUNDREDS of users that say that my products dont work. I wonder, if this were true don't you think that there would be a flood of people asking for a refund? My products works fine in Poser and that's all that matters. The user DOESN'T need to do anything to them except installing in their computer and start creating and that is all. Do they have an extra bracket at the end? YES they have Do this affect the usefulness of the product in Poser? NO Do the collars point to the external name of the channel instead of the internal in the ERC relation? YES, I have explained why I did this and if Poser let me "Get Away" with it in order to finish this product, well Thanks Poser! And also want to say that many, many users would thank Poser because of its "flaws" I could finish this product earlier. BTW I'm perfectly aware of the work of Rob and Charles. Finally, I have decided to not post the poser-fixed-cr2 because it's just a so-so solution after all and the DS users would have a crippled version of NBS and I don't want that. As soon as I had the time to do it I will make NBS compatible for the DS users. If anybody want to use NBS and NGM in DS with the poser-fixed-cr2 give me an IM and I would give you the link to this file but remember is a partial solution.