Daz 3D is part of 
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.Daz 3D is part of
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
Yeah, that doesn't allow legacy authoring. I do recognize that isn't what you asked, but my first goal would be bidirectional support. That said, I see some indications in the script API that you can do legacy authoring, and getting that to work with the existing code would still require converting the Legacy figure to weight mapping in order to get the ... stuff. That should be fine, as long as the converted figure is removed from the scene when it's done, but it does require that another copy of that Legacy figure be in the scene, unconverted, to act as the target. So, it may be possible. I'll still need to look into it.
More specific to your question, presently, the script relies on the ability to pose the figure as part of generating the transfered morphs. So it would be important that the conversion to weight mapping be accurate. I've noticed that this doesn't work so well with hair figures. You have to, basically, re do the weight maps. This is not insurmountable, but it would end up being a whole separate script (which is about what I expected). I will look into it in more depth, but it probably will still be an addon, even if it's unidirectional.
I'd prefer bidirectional, because V4 still has significant following, and that's worth supporting.
The test crew will hear from me today. I hope I've ironed out the major issues, and will be ready to submit for products review by the end of the month. But I haven't been planning on that last. Still, I've just got a couple of hours worth of tests to do, and then I'll be ready for more beta. I really don't anticipate trouble, but the tests are necessary. I've already caught an issue with the way I formated Genesis 8, originally. It's incompatible with the clever solution I invented for baking morphs. But it's an absurdly easy fix. Took 10 minutes for G8M. I still need to G8F, but that's allI should need.
I was going to do these tests with Hiro 5, but I do have Sadie and Sam, so I'll do that, instead, I'll get back to you.
I'm packing the script up for initial testing now. I did discover a couple of issues, and I'm going to have to fix those. They'll largely relate to Genesis, Genesis 2, and Big eye morphs (Toons). The last was expected. A limitation of the method I'm using to do the transfer. I need to add yet another helper file and another code loop to fix the ton problem. The others are related to reversible heads. Genesis, and G2 to a lesser extent, FBMs include the head shape, and you are expected to reverse out the head shape by dialing the head morph negative. The way the script generates FHM and FBMs will not work with these morphs, so I need to add a new option for transfers. I hope to have all that done this week.
A few notes, which I'll repeat in the email.
The UI has changed. Right now, nothing on the Options tab is of use to you, so don't bother with it. Some of it does work, but it is automatically engaged where needed.
The Head and Body options in the Morph selection are only used together, and only by themselves. if you feel the need to click the On option, you should not use Head or Body. Eventually, the UI should enforce this, but it doesn't now.
To transfer a character with the source figure's shape "baked in" you need to first do a basic transfer. When that finishes, you would run the script again, locate and check the head and body morphs, and rerun the script. This is important.
I forgot to do something, so the script is not ready to bake source shapes for characters that don't have separate head and body morphs, or have reversible heads. So limit initial tests to characters with separate heads.
Here's something to help you separate the head and body...
https://www.daz3d.com/the-executioner-for-genesis
(You've been working too hard. Take a break, man
)
I'm just so thankful and excited for this!!
Thank you thank you for your hard work singular!
I am looking forward to this HUGELY. I have tried and failed so bad to transfer morphs. PLease save us all :) <3
I'm not sure we want the head and body to be quite that separate, but I could be wrong.
FYI, I probably won't be saying much until Saturday. I'll be busy tomorrow, finishing the additional code, and (hopefully) responding to test results. Friday, Good Friday, everything is closed, and I won't be able to get online much at all. That said, I will get round to responding to any PMs by Saturday, if only because I need to make the descision about adding additional testers then. But mostly because I shouldn't (knock on all the wood) have any more pressing issues to deal with.
I hope to get to that point tomorrow, since all I have left to do is put some code back in that I removed, and create the conditional trees to make sure it's used when needed. Today, I've spend most of the day getting the UI to not let the use enter invalid combos. That Transfer/Head/Body option stuff above. The looping on that was so simple, it was turning things on right after turning them off. So, I can at least say, with confidence, that Saturday's release will not allow the user to get that part completely wrong. and if the user does the one thing they shouldn't do, the script will give a warning when it's run, and stop.
You know, I thought I'd posted Saturday to say I'd be late with the next release. I found I'd written a post but not posted it.
I'm sorry.
Anyway, the next release is about done. It's fully ready, and it incorporates some of the feedback I've gotten. I'll be in touch with new volunteers today, as well. Again, sorry. My bad.
Watching this with anticipation! I have some G3 characters that I'd love to get over to G8 :D
hi, new to this thread.. its quite long.. I just want to transfer the face morphs from a G3 to a G8. Is there a link to a product or script avaialbe ? Or even an up to date tutorial? Thanks
Nope. If you read thru the thread there is one being worked on, but not there yet.
If you think you are on the test mailing list, and haven't heard from me, PM because I've obviously messed up. The revised release went out yesterday.
The tuts in my sig will mostly do. The resultant morph will be the faces on G3's body, but G3 and G8 have bodies so similar, it won't be very noticable. Otherwise, I hope to have something in 6 to 8 weeks. There's more testing to do, and Daz can take a month or more to decide the product passes muster.
The next Beta is comming soon. Like the first, there will be known issues. In this case, those issues are as follows:
UI Bug causes list items to write over each other.
Possibly same UI bug prevents execution on Mac.
Mac progress UI pops uner main GenNext widow, making it appear as if nothing is happening
Mutant X has strange arms after transfer to G8M. (Can't seem to replicate effect with any other morph, not tested in transfers to other figures).
Progress UI only reaches 100% on single Morph transfers.
IIntent of UI operation is not conducive to user assumptions. (I'd have liked to have already redesigned this, but given the problems with the UI and Macs, it seems politic to get a solid handle on what is causing the Mac issue before redesigning the UI. I know some of you found it poisitve and easy to understand, but over all feedback was that the options were confusing, so that needs fixing. All of this refers to the same part of the UI.
Catastrophic failure in $.11 Beta. No logging data, so no clue if this because GenNext of 4.11. Current plan is to build a test environment around 4.11, and see is it's possible to determine the source of the issue in the same way the Mac issue was found (disabling things, even if it breaks the script, and then seeing if the failure mode changes. (The issue on Mac is similar, there is nothing logged, and no error, the difference is that 4.10.0.123 doesn't die when the script runs on Mac.)
Poor fidelity in the shoulders when transfering from G8 to any T-Pose Figure. This is slight alignment issue with the meshes, and will have to wait until I have a working monitor to be fixed. Trying to evaluated tests in Studio is hard on a 5 inch screen. Modeling would be impposible. (The new monitor assembly is en route). should have it by next weekend, at the lateset.)
Things Fixed:
Transfers from G*M to G2M failed. The G8M clone was missing.
Transfers from G2M failed. This was due to a bug that was not consistent on my system. It should work now.
Directory mapping transformed to Daz's iStudio's internal standard. This was largely true, anyway, but there were some Windows assumption built in. There probably still are a few left, but they don't impact operation on the Mac.
As a side effect of the above, support for Connect products, and library indepence. GenNext should run correctly wherever it is installed. Please note, this has only been tested on one Morph inside the Mac test environment (I don't use Connect, personally, but it seemed worth using to test with). I may not have resolved all of the possible bugs with file casing (Genesis8Female.dsf on Maunal of DIM libraries, genesis8female.dsf in Connect) Also, yes, the scrip is running on Mac, but only with the offending (critical) part of UI disabled, and having hardcoded exactly what Morph to transfer.
Feedback required on:
I've gotten good support in terms of getting examples of issues spawned by transfers from odd figures to G8. But I need more info on events occuring when transfering from G8 to older figures. I know this isn't the primary, or even the most likely use case. Still, GenNext can do it, so I need examples of what haappens when it does. I discovered the G2M issue above, while checking to see if the problem reported with transfer from G2M was symetrical).
The original intent of the script was to transfer a whole product at once, with JCMs, MCM, and other character specifice properties intact. However, (between G3 abd G8) this inot necessary most of the time. It is possible to make this the optional behavior, and limit the default behavior of the script to only transfer those Morphs explicitly marked by the user. Which would you, as users prefer? This is a choice between "One Checkbox transfers everything, unless otherwise specified," or, "You must mark exactly the items you want to transfer, unless specified otherwise." Personally, I think the latter is better, but I thought the current UI was fine, so my POV is not a good meter stick. Some lements of how the script works will have reamin the same, despite being less than optimally clear. But they are odd (but important) use cases, and there's no real simple way to do them. GenNext's Design philosphy was always the opposite of GenX's. I don't want to give the user hundress of chices, most of which will never be used. Nor is this meant as PA tool that I'm sharing with the public for wider profit. It's meant to help you move the morphs you want to the figure you want them on. I need to know what you expect that to look like, and what you want, pruely as users, so I can make sure the script gets close to exepectations.
For those who are looking for something more tech heavy, I have it on fair authority that someone is on that. Since they haven't been public (that I'm aware), I won't out them. I have some experience with how difficult the really simple things turn out to be.
All that said, I also need to know if my assuption is wrong, and you mostly expect to have access to everything the script could do. I won't promise to add all that before release, but I can adjust to the idea.
EDIT: Soory about the typos. 5 inch screen, 1080HD display.
Always great to read an update, Singular. Keep on trucking and thank you!
I have just cleaned out all my G3 morphs transfers to G8 thinking a prior version of SingularBlue's scripts might of been the cause of a mesh distortion problem but his scripts weren't the cause (also to my relief too, although that leaves what I think is a bad set of bent knees sculpted on the original G8M).
I was going to start my beta tests with his latest scripts but now it looks like I should wait for the next beta to soon be out.
Thanks SB.
I am bloody exhausted. I had to walk across Newark to get the replacement part for the laptop. But it got got, and installed, and I am--thus--happy.
The Beta release has been delayed because I don't think it's a good idea to reinforce a failure* if I don't have to. I have A) figured out a UI design that might reduce what had become a multi step process back to a single step (by "step" I mean transfers. It's not a one click to transfer), and B) figured out why the script hung on the Mac. (basically, the Mac script engine doesn't always do what the PC engine would do, so a design that works on PC might become corrupted on Mac, meanwhile a code segemet that passes muster on Mac throws a fatal error on PC).
After installing the replacement monitor, I finished the changes to the UI. However, because the UI now does a bunch of things it didn't used to do, I need to refactor part of the back end. I hope to make some of that more compact, because I think there's a handful of redundancies in the pipeline now. In other places, new code loops are needed to handle cases that didn't exist before.
It shouldn't be as much work as it sounds. Still, if I don't get it done by end of day, tomorrow, I will be releasing the file I had ready on Monday.
*by rienforce a failure, I mean the old UI is a problem, and I need people to react to the new UI without getting used to the issues that exist now. I don't want the new changes to suffer because users have gotten used to doing it the other way.
I finished yesterday, but I had no time to test. So I got the old bit and started to pack it, when I noticed some code seemed to be missing.
This monring, I looked for the code and couldn't find it. So I finished testing, rewrote the missing code, tested more, realized a code fix was in the wrong function, so everything it did got undone at the end of the run, fixed that, debugged, tested one transfer on Mac (I only loaded one purchase on the Virtual Hactintosh).
It's alive.
I know the transfer entire product checkbox doesn't do anything. I haven't figured out the best way to code that option. I think I'll need to take a code loop and make it into a function and add a new loop to check for the status of that checkbox. I'm still tired from the move. (and repairing the computer, and removing all of the clearly dead code, and unused functions from the script (now I need to find all the code that isn't obviously doing nothing, and get rid of it). I've decided to procrastinate, and just write in the What's This for the UI elements.
How's your day gonna go?
Is this also going to support V4 to G8F or anything like that? Just curious if it's across all the Genesis bases.
V4 is NOT a Genesis base .......
I had hoped to, but it's taken nigh on 18 months (I admit to falling off for almost four of those, but still) to get this far. I still need to learn how to read V4 and to make it actually work will involve a completely different order of operations (because this really is a script. It does what it does by leveraging Studio tools. The tools it uses don't work directly between "Poser" formats and Genesis formats. So V4 will need to be converted, which will probably break the rigging which will ... Yeah, it's a lot of stuff. I will attempt to make that happen, but probably as an add-on. Also, since Daz has confirmed(?) that G9 will not be out this year, that support might also be an add-on (assuming this becomes a salable product. If I end up giving it away, I'm not gonna update it in separate pieces) instead of an update. I really haven't decided, on that because reasons. Anyway, I wouldn't get any hopes up for Generation4 (or less) support before next 2020 (if ever).
You are the bomb, Singular. :) Thank you for all your hard work. :)
Thanks for keeping us updated on the progress. Take it easy and take care of yourself, don't stress too much.
Understood. I know it's a lot of work and it is greatly appreciated.
Unfortunately, I think it's time people realize Gen4 is falling off the back end... going back past Genesis is just too hard for less and less of a market share or benefit. For people heavily invested in Gen4, they will decry this, but their thinking will unfortunately be clouded by personal investment. The reality is that if it was good enough in Gen4 to carry forward, the real solution is to remake it from the ground up for newer characters.
I agree. I have converted most of my v4 collection and rarly, if ever, buy any more items for that generation. And as stated by @Joe Cotter if it was good enough to carry forward PA's please remake it from the ground up for newer characters. Also, when I find iteams reworked for newer generations, they tend to have more fuctinallity and morph options.
One can always convert Generation 4 morphs up to Genesis 3 with GenX2, and then move them up the food chain...
I agree with both jimmyzpop nd WandW. I use both G3 and G8. If I want to use a V4/M4 morph, I will just use GenX to convert it to G3.
I keep seeing everyone mention how well GenX2 worked, but I always had problems with body morphs on M4--nipples and navel were always in the wrong places, and pecs never transferred correctly. When comparing the shapes used to M4, the chest and navels don't match M4's placement. So my question is how are you guys managing to get good transfers.
A good example is Angelo M4, the transfer to Genesis is questionable, and tranfers to G2M and G3M keep getting worse. The pecs are horrible, they bascially start folding into the upper abdomen.
@SingularBlues, can't wait for this to be finished.