Forum Coordinators: RedPhantom
Poser - OFFICIAL F.A.Q (Last Updated: 2024 Nov 26 5:46 am)
Also, is this on offer as a freebie or for sale? And is it a re-rig w/ weightmapping or a conversion of the existng rig?
It would be sure grand to settle on an unofficial "official WP v4" so we could see continued vendor support and vendor conversions of their existing products.
Quote - I am curious PhilC, have you heard about Phantom3D's project to convert V4 to WM as well. Would love to see it though, but I think a lot of folks would like to see a "standard" WM V4 to make sure there is no confusion.
Poser Pro 2012 has the tools in its Joint Editor to be able to replace existing jointing with the equivalent weight mapping, however at three joint rotations per body part and 50+ parts that is a boat load of work.
Sounds like a job for Python but no method existed. So we asked for one, and SM came through. Still under test but I'm happy to say that it looks like it is working. Way to go SM :)
Yes if I'd known before I could have saved a number of folks a lot of work but I only got the beta update today and wrote the script this evening so I'm posting now. There is no point in me posting the script yet because it will not work in your version yet. I most certainly will when the Pro update is released. The script is only 20 lines of code and once the method is known its a no brainer to write so I'm not being particularly clever here :)
So for now I suggest that we concentrate on the best method of distributing weight map data without infringing on any CR2 file copyright.
Quote - So for now I suggest that we concentrate on the best method of distributing weight map data without infringing on any CR2 file copyright.
According to the linked thread, Daz accepts the method discussed in this thread. But apparently any kind of auto-conversion of existing zone joints is not acceptable to Daz.
I'm excited that SM has added this new Python feature you mention, but it sounds like it represents exactly the kind of auto-conversion that Daz finds unacceptable. :unsure: Would you legally be able to share this script-converted V4, even in the acceptable injection pose format?
===========================sigline======================================================
Cage can be an opinionated jerk who posts without thinking. He apologizes for this. He's honestly not trying to be a turkeyhead.
Cage had some freebies, compatible with Poser 11 and below. His Python scripts were saved at archive.org, along with the rest of the Morphography site, where they were hosted.
Quote - So for now I suggest that we concentrate on the best method of distributing weight map data without infringing on any CR2 file copyright.
So you're saying that Poser Pro 2012's 1 click conversion isn't as 1 click as we thought. Still doesn't answer the question I posed.
Quote - Would you legally be able to share this script-converted V4, even in the acceptable injection pose format?
Just distributing the script would suffice I would assume.
Quote - > Quote - So for now I suggest that we concentrate on the best method of distributing weight map data without infringing on any CR2 file copyright.
So you're saying that Poser Pro 2012's 1 click conversion isn't as 1 click as we thought. Still doesn't answer the question I posed.
Quote - Would you legally be able to share this script-converted V4, even in the acceptable injection pose format?
Just distributing the script would suffice I would assume.
Just distributing the script would work provided it will give the exact same result every time and therefore would be a reliable base for development of weightmapped content. (I assume scripts being mathematical in nature 'should' yeild predictable results, but what user side variables might there be)
The only downside to this is it wont fix V4s underlying rigging issues and dependence on magnets to function correctly.. or am I wrong? if so I am quite used to it... :(
Quote - Just distributing the script would suffice I would assume.
The script would definitely be a good thing. :laugh: And your unanswered question becomes especially important. Is the Python method just reproducing the same joint results? If so, that's certainly useful in itself. But could one use those converted joints as the basis for improvements, then distribute the improved joints, if the Python method doesn't somehow automatically improve joint handling?
If the Python methods can't improve the joints, it sounds to me like the real benefits of weight mapping can only be realized on Daz figures by following the approach Diogenes has been using, and doing it the hard way, by hand, from scratch. :sad:
Edit: Cross-posted with meatSim. :laugh:
===========================sigline======================================================
Cage can be an opinionated jerk who posts without thinking. He apologizes for this. He's honestly not trying to be a turkeyhead.
Cage had some freebies, compatible with Poser 11 and below. His Python scripts were saved at archive.org, along with the rest of the Morphography site, where they were hosted.
How could you prove that any particular weight map was produced by a scripted conversion? Its a no brainer to add code to apply a small random factor to each weight. Unless you knew what that factor was you could never prove where its source came from. Then if the user takes that as their starting point and made a few minor adjustments .......... "tweaking those weights" ...... would, in my opinion, have been achieved.
I do however accept that the weight map data must be injected into the figure.
I see the work flow as:-
Use script to convert existing joints to weight maps. This just speeds up the process already available to Pro users in the Joint Editor. (If you are on the beta team I've now posted that to the RDNA beta forum.)
Starting from that base fine tune and enhance the weight mapping as required.
Create some form of injectable weight map pose that may then be redistributed.
I guess I'm a little bit confused by the role this script plays, Phil. PP2012 already has the tools to do an autoconversion to weightmapping, assuming you're not looking to do anything more than an "equivalence." Is the point here that this will be more robust/accurate? or that it will be pretty much the same in effect, but maybe available for P9 users as well as Pro users? or just what?
______________
Hardware: AMD Ryzen 9 3900X/MSI MAG570 Tomahawk X570/Zotac Geforce GTX 1650 Super 4GB/32GB OLOy RAM
Software: Windows 10 Professional/Poser Pro 11/Photoshop/Postworkshop 3
There are many figures that would benefit from having weight-maps. The Poser content library is briming with them, and SM is not likely to complain if we get the best of them upgraded.
With a sript to help, the many content creators, who have figures already made, can convert them, and then refine their work to provide improved and completely Poserized figures.
Remember, DAZ is NOT the only content provider with figures.
lmk
Probably edited for spelling, grammer, punctuation, or typos.
Quote - I see the work flow as:-
Use script to convert existing joints to weight maps. This just speeds up the process already available to Pro users in the Joint Editor. (If you are on the beta team I've now posted that to the RDNA beta forum.)
Starting from that base fine tune and enhance the weight mapping as required.
Create some form of injectable weight map pose that may then be redistributed.
That sounds good to me, and I like any tool that could make all of this easier. :woot: I'm just saying that the process looks like exactly what Daz is forbidding in their statement released at RDNA. Maybe they couldn't prove derivation, given a random factor or adequate manual tweaking. Maybe their stated policy is a bit silly. :lol: But it just seems like the same thing.
===========================sigline======================================================
Cage can be an opinionated jerk who posts without thinking. He apologizes for this. He's honestly not trying to be a turkeyhead.
Cage had some freebies, compatible with Poser 11 and below. His Python scripts were saved at archive.org, along with the rest of the Morphography site, where they were hosted.
Quote - I see the work flow as:-
Use script to convert existing joints to weight maps. This just speeds up the process already available to Pro users in the Joint Editor. (If you are on the beta team I've now posted that to the RDNA beta forum.)
Starting from that base fine tune and enhance the weight mapping as required.
Create some form of injectable weight map pose that may then be redistributed.
So just clarify what the script does in noob terms...
It automates the process of going through each axis of each joint and pressing the 'merge to weight map' button, but does not modify the results from what one would get if they actually went manually through each joint and axis and pressed the button?
Yes, it just does the tedious donkey work.
Rigging is more art than science, it will always take a human touch to set any form of jointing values for a 3D figure model. Weight mapping is not magically better than any existing method. It does however offer other options previously not available.
There is also nothing that magical in the script either.
http://www.philc.net/forum/viewtopic.php?p=17943
You'll need the Pro version and the update when available so take it at face value and subject to change.
Yes any figure.
No not this script however, Poser Pro 2012 has a menu option to generate weight maps in clothing to match the figure. I'd need to check if that is in Poser 9 also. There is a Python method available should you want to script the process. The option is also built into Wardrobe Wizard.
copyToFigure.CopyJointParmsFrom(copyFromFigure)
Quote - I guess I'm a little bit confused by the role this script plays, Phil. PP2012 already has the tools to do an autoconversion to weightmapping, assuming you're not looking to do anything more than an "equivalence." Is the point here that this will be more robust/accurate? or that it will be pretty much the same in effect, but maybe available for P9 users as well as Pro users? or just what?
Really? Do tell!
Yes you can use the tools in the Joint Editor to create equivalent weight maps for each joint of each body part.
Scale + 3 orders of rotation multply by 50 body parts @ 3? mouse clicks per joint ~= 600 clicks.
600 clicks @ one per second = ten minutes of tedium.
Or run the script.
Then use the result as your basis for revised/improved bending.
Phil, thats a great script. You have a knack for coming up with them.
But from my communications with Daz, nothing you make afterwards with it is distributable because it used the original joint zones.
I am no lawyer, so don't take what I am saying wrong.
I am just posting this here so people have been forewarned.
For personal use, they can not do a thing about it.
People that want to use it for that, will definently save more than a few mouse clicks.
Some things are easy to explain, other things are not........ <- Store -> <-Freebies->
From what I got from Daz. If you do a conversion using the original joints and zones, you are violating the legal agreement for the character.
If you use you own rigging and zones, you have not violated any intellectual property rights.
Some things are easy to explain, other things are not........ <- Store -> <-Freebies->
The main problem is that even if you bought a Daz3D character, you do not own it.
Your purchase only gave you license for personal use.
If you use any part of it to create a product for distribution that could be considered to compete with it, you are in violation of the EULA for the character.
That would include free items as well.
Some things are easy to explain, other things are not........ <- Store -> <-Freebies->
How can it compete with itself?
You would have to own V4 and load it into the scene before applying a weight mapped injection pose. Does that not enhance the perceived value of V4? I fail to see how it could detract.
Consider a remapped V4. One has to provide the new UV data in a form that can be "injected" into the original. One must similarly own the original version before one can use the revised version.
I've not heard any argument that this violates any EULA so why now with weight maps?
Yes I probably could answer my own question but that would mean going back to the beginning, the start of it all and I do not propose to make that exodus.
Quote - The main problem is that even if you bought a Daz3D character, you do not own it.
Your purchase only gave you license for personal use.
If you use any part of it to create a product for distribution that could be considered to compete with it, you are in violation of the EULA for the character.
That would include free items as well.
I'm a little confused. If this is the case, why isn't it a violation of the EULA to create clothes for a DAZ figure? Don't you have to use the original bone structure and rigging info to create a conforming outfit?
Yes but the clothing is said to enhance the V4 figure not detract or compete with it, and I would agree with this.
In my opinion weight mapping will enhance the V4 figure not detract or compete with it.
How that data is created is, in my opinion, immaterial ...... providing that the data is supplied in a form that can only be used by someone who has already legally obtained the V4 figure.
Phil, I didn't write their EULA, and I don't enforce it.
Section 4 of the agreement basically says you can not make anything 3D from it.
Daz told me that you can not use any part of the character, in part or in whole, to make another product.
From the answers I did get, they seem to be ready to enforce it as well.
EClark, thats a good point. I will ask them again, and add that into the mix.
I am just going by what I got back from them after it was discussed at Daz.
That does seem a little conflicting, doesn't it.
Some things are easy to explain, other things are not........ <- Store -> <-Freebies->
Quote - How that data is created is, in my opinion, immaterial ...... providing that the data is supplied in a form that can only be used by someone who has already legally obtained the V4 figure.
Phil. I agree with your opinion.
But if Daz decieded to sue one of us over weight mapping V4, our opinion would not stop it.
Im not saying that they would, but being sued has finacially ruined more than a few people even though they won their case and were not convicted of anything.
Even if the info/data could not be reverse engineered, if they could show that it was created from their product, we would loose.
CYA comes to mind.
Some things are easy to explain, other things are not........ <- Store -> <-Freebies->
Yes I understand and agree that one can not take any part of the V4 mesh and create a redistributable model from it. For example you can not cut off the scalp area and offer that as a skullcap for Hair Room hair.
But consider a morph target. That must start with the original geometry. It is then "injected" into the figure. It enhances the figure, it does not compete with it and again one must own the figure before you can use the morph.
Weight map data is essentially a look up table of vertex indices against the degree by which they are displaced when a joint is rotated. You can never recreate the geometry from them. They are useless without the original figure.
Quote - Weight map data is essentially a look up table of vertex indices against the degree by which they are displaced when a joint is rotated. You can never recreate the geometry from them. They are useless without the original figure.
I agree with that as well, but if they can show that a certain product, free or for sale, is a direct conversion of their product. You are in fact in violation of the EULA.
Reverse engineering and violation of intellectual property are two different things.
Some things are easy to explain, other things are not........ <- Store -> <-Freebies->
I'm going to draw a line under this for the time being. I'll conclude with some house keeping.
My reason for posting was to:
Bring to folks attention the exciting developments that are in the pipeline. They are not confirmed, they should be taken at face value.
I knew that a number of folks were working on this subject and if I delayed it could very likely mean that much of their work might be wasted.
The method by which weight maps are generated and distributed is a potential minefield. Far better to thrash this out before the tools to create them efficiently are made generally available than to wait and have everybody trip over themselves in the, "Is this the I want to be sued sign up line?"
My thanks to those, particularly shvrdavid, who have very eloquently provided counter point opinion and information to enable the salient points to be discussed openly and intelligently.
Phil, you always do great work and have contributed a ton of stuff to the Poser community (Both in product and knowledge). For that I thank you.
I just want everyone to make sure that whatever they do is legal, that is why I contacted Daz about it.
If anyone is in doubt, ask Daz (or whoever is the owner of the original), to examine your product and provide a written waiver or consent on it.
Better off safe than sorry.
Some things are easy to explain, other things are not........ <- Store -> <-Freebies->
Here's one important thing to also consider, adding custom morph injections is not a violation of the EULA, as you require the original mesh to do use it. You are not distributing a mesh or a figure, you are saying "take vertex #x and move it here" Without a frame of reference or any knowledge where vertex #x is, you can't even use it.
From the way I understand how the WM will be distributed it will be similar. It will refer to a specific vertex or group of them, say how those interact with the others around them and go from there. If you took all the data from the "WM INJ" file and tried to reverse engineer it into a figure, you would get...well, nothing. I can tell you how point A interacts with a million points, but until you know something about where they all are in relation to each other (something the INJ file would not have) it is worthless.
Besides, DAZ has already spoken and said that doing a script/INJ form of distributing WM's is perfectly legal. (Reference: http://www.runtimedna.com/forum/showthread.php?65006-Distibution-of-weight-mapping-legal-issues.-Confirmation-from-DAZ )
In the end, DAZ has already spoke and has said that they are done with Poser when it comes to figures, so expect new figures from other sources to take her place anyways.
Er... that thread says precisely what you're denying:
QUOTE
Having discussed this internally, the view is that as long as the new
weight map can be separately shipped without needing to include any
proprietary DAZ data (models, weight maps, material definitions, etc.),
then it's fine - note that this would preclude converting a DAZ figure
to weight-mapping and then tweaking those weights. The distribution
method proposed here http://www.renderosity.com/mod/forum...ead_id=2836045
looks as if it should work.
/EOQ - bold print added for emphasis
The distribution method is one thing. The starting point is another. You cannot distribute a weight-mapping that starts with DAZ's existing rigging and converting it.
Edit: I'm not saying that makes sense. I don't think it does at all. But nonetheless that's what DAZ has said.
______________
Hardware: AMD Ryzen 9 3900X/MSI MAG570 Tomahawk X570/Zotac Geforce GTX 1650 Super 4GB/32GB OLOy RAM
Software: Windows 10 Professional/Poser Pro 11/Photoshop/Postworkshop 3
Quote - That then begs the question of, "How will they be able to tell what the starting point was?"
The weight injection process seems to require that the zone joint data be included in the pose. The example Nerd posted, presenting the format which Daz has accepted, contains that zone data. So unless one ran the auto-conversion and then adequately altered the zone joints by hand, the weight injection pose could show what that starting point was.
===========================sigline======================================================
Cage can be an opinionated jerk who posts without thinking. He apologizes for this. He's honestly not trying to be a turkeyhead.
Cage had some freebies, compatible with Poser 11 and below. His Python scripts were saved at archive.org, along with the rest of the Morphography site, where they were hosted.
Quote - > Quote - That then begs the question of, "How will they be able to tell what the starting point was?"
The weight injection process seems to require that the zone joint data be included in the pose. The example Nerd posted, presenting the format which Daz has accepted, contains that zone data. So unless one ran the auto-conversion and then adequately altered the zone joints by hand, the weight injection pose could show what that starting point was.
That's not really an answer though. If all of the map data was altered joint by joint after auto conversion, how would you know what the starting point was? I don't think you really could prove it unless they missed one or two joints and had matching data in those joints. And before someone jumps on me for trying to breach a EULA, I haven't even messed with the weight mapping yet and doubt I ever will.
Unless one took steps to alter the zone joint setup after running an auto-conversion, that data would end up in the pose. Unaltered, it would clearly show the starting point. If one is going to bother to manually alter all of the joint zones, one might as well go ahead and set them up from scratch (or from whatever starting point the Setup Room or a .phi conversion might give you), IMO.
Not an answer? Ehh?
I don't think Daz could ever, absolutely be able to identify the starting point in every single case. Steps could be taken to try to make such identification difficult. But embedded zone data in the pose is a way that Daz could identify the starting point under certain conditions.
===========================sigline======================================================
Cage can be an opinionated jerk who posts without thinking. He apologizes for this. He's honestly not trying to be a turkeyhead.
Cage had some freebies, compatible with Poser 11 and below. His Python scripts were saved at archive.org, along with the rest of the Morphography site, where they were hosted.
Mainly Daz has always stated that using scripts for content creation from their IP (models) is a nono. You can use a Daz figure and manually place polys on the surface to create a form fitting out fit. You man NOT take that same figure into Blender or similar program and use a shrinkwrap script to create the same outfit.
I know it doesn't make sense, but that is their prerogative. The poly by poly method would more than likely follow the figures topology rather than the scripted method. I feel the former method would be a EULA violation and not the later.
Attached Link: http://forum.daz3d.com/viewtopic.php?t=72992&postdays=0&postorder=asc&start=0
It really depends on usage. For personal use you could cut the head of a figure and scale the result for a bodysuit as long as it is for your own use. Sell that suit or even give it away as a freebie and you would be in violation.Like I said, it's confusing and I'm not sure they know what's legal or not. See above link for more info.
Sorry to post in more than one thread, but this is good news. A script is now available to share weight-maps (and animatable origins) legally through a proxy CR2. Our most profound thanks to Cage, Python Maestro. :)
LMK
Probably edited for spelling, grammer, punctuation, or typos.
Nobody who knows most probably may post here. Maybe you'll get an answer from ThinkCooper in the RDNA-Forums, but I doubt it.
A ship in port is safe;
but that is not what ships are built for.
Sail out to sea and do new things.
-"Amazing
Grace" Hopper
Avatar image of me done by Chidori.
This site uses cookies to deliver the best experience. Our own cookies make user accounts and other features possible. Third-party cookies are used to display relevant ads and to analyze how Renderosity is used. By using our site, you acknowledge that you have read and understood our Terms of Service, including our Cookie Policy and our Privacy Policy.
I am curious PhilC, have you heard about Phantom3D's project to convert V4 to WM as well. Would love to see it though, but I think a lot of folks would like to see a "standard" WM V4 to make sure there is no confusion.