Ai Modelling Tasks

[Springheel]

Ok, so far we have the following tasks that need to be done, before we can start doing any new animations. Domarius has volunteered to do everything that anyone else doesn't sign up for, but with only one person working on these we won't be done until a year from now. Any other volunteers?

 

1. Current models need to have heads separated.

http://forums.thedarkmod.com/index.php?showtopic=4331

 

assigned to: Domarius.

 

2. Current models need to have hands replaced with low-poly hands.

http://forums.thedarkmod.com/index.php?showtopic=4330

 

assigned to: nobody

 

3. Models need to be resized.

http://forums.thedarkmod.com/index.php?showt...=4222&st=25

 

assigned to: Domarius

 

4. Weapons need to be removed from models.

http://forums.thedarkmod.com/index.php?showtopic=4332

 

assigned to: nobody

 

5. Robed characters need feet added.

http://forums.thedarkmod.com/index.php?showtopic=4333

 

assigned to: nobody

 

6. AI models need to be split into submeshes for armour detection.

http://forums.thedarkmod.com/index.php?showtopic=4335

 

assigned to: nobody

[Domarius]

Also I'm a bit busy with Uni and a job right now

 

Okay I'll really try, to help things get moving. No one else is available and the project is kinda going to come to a stand still without this stuff.

[Springheel]

Where is BT lately? I thought he was back a while ago but I haven't seen him post on any of these topics. I was kind of hoping he'd be able to do #2 and #5.

 

Solis, do any of these look like they're something you could handle?

[SneaksieDave]

Looks like we've pretty much lost BT. Two posts since July; not a good sign. Guess we have no character modeler again.

[PinkDot]

I can't take any of these jobs now, cause I have very poor experience with character modelling. I mean - I can model some character already (I've done one monk for my Dromed abandoned mission) but I've never done rigging and all that dirty stuff which I'm not aware yet . But this is something I'm going to learn soon anyway and I hope to contribute to TDM some characters in the future.

 

Well, I could try to do the modelling jobs but then somebody would have to rig it for me unless I learn it by myself.

[ascottk]

I'm the next most likely candidate. I'd like to get the gen_texture project off the ground now that Gil & I have the organizational issues fairly resolved.

 

Removing weapons should be fairly straight forward.

 

Is splitting up the mesh necessary for armor detection?

 

Should we wait for the head separation? That might throw any of the other tasks off. I have ideas on how to do that but I have to do a little research. Dom, have you got anywhere with this or has uni taken most of your time?

 

BTW, maybe you shouldn't have "nobody" since "nobody" is also the nick of a forum member (unless "nobody" has that task assigned to him). Maybe use "none". Also links to discussions about these tasks might be helpful.

[Springheel]

Removing weapons should be fairly straight forward.

 

Agreed.

 

Is splitting up the mesh necessary for armor detection?

 

Apparently so, if we want to assign different material types to different parts of the model.

 

Should we wait for the head separation? That might throw any of the other tasks off. I have ideas on how to do that but I have to do a little research. Dom, have you got anywhere with this or has uni taken most of your time?

 

Head separation shouldn't really throw anything off--although we should either do that *after* the resizing, or be sure to resize the separated heads by the appropriate amount to avoid bobblehead syndrom.

 

BTW, maybe you shouldn't have "nobody" since "nobody" is also the nick of a forum member (unless "nobody" has that task assigned to him). Maybe use "none". Also links to discussions about these tasks might be helpful.

 

 

I originally had N/A, but for some reason the forum wouldn't let me post with that. The 'nobody' person isn't a team member, so I don't think there's a problem. You're right about the links, I'll do that when I get a spare moment.

[ascottk]

Apparently so, if we want to assign different material types to different parts of the model.

I can reassign materials without separating the mesh. But we'd still have to retexture (maybe not, I'd have to play around the materials).

Head separation shouldn't really throw anything off--although we should either do that *after* the resizing, or be sure to resize the separated heads by the appropriate amount to avoid bobblehead syndrom.

I'm not sure how max handles bone resizing but when I tried that, the bone orientations were thrown off. Was that picture in the resize thread doom units (8 units per square)?

You're right about the links, I'll do that when I get a spare moment.

Thanks

[Dram]

I can reassign materials without separating the mesh. But we'd still have to retexture (maybe not, I'd have to play around the materials).

 

What he meant is you need to have it in submeshes, like you do with a static basically.

[ascottk]

What he meant is you need to have it in submeshes, like you do with a static basically.

 

Well you can have submeshes with materials assigned to each, then again you could have several materials on the main mesh (which ASEs can't do).

[PinkDot]

then again you could have several materials on the main mesh

AFAIK it doesn't work with AI, only with static meshes.

[Springheel]

Well you can have submeshes with materials assigned to each, then again you could have several materials on the main mesh (which ASEs can't do).

 

Which would be easier, do you think? Given that there are already a few separate submeshes--the visible chainmail and visible flesh are both separate submeshes, IIRC.

[ascottk]

Which would be easier, do you think? Given that there are already a few separate submeshes--the visible chainmail and visible flesh are both separate submeshes, IIRC.

 

Once I reassign materials it's only a matter of detaching those from the main mesh in 3ds max. FYI I have more experience with the physique modifier in max than I do with the skin modifier so rerigging might take a while. It's either that or learn maya which I find to be unintuitive. I wonder if deep exploration converts physiqued max meshes to skinned maya meshes?

 

As I ramble on . . .

 

When I import an md5mesh into max the bone orientation is horrendous. But when I convert my max files to maya then the orientation looks good.

 

Another option is having me separate the meshes with the bones then convert them to maya for rigging for those ppl who had more experience with that.

[Springheel]

I think we should keep all our rigging in Maya, since that's what apparently works best with D3 and is what our animators are using.

 

At the moment we have several different tasks, each of which will mess up the rigging. Does it make sense to just do all of them and THEN rerig the character, or does that have to be done after each task?

[ascottk]

I think we should keep all our rigging in Maya, since that's what apparently works best with D3 and is what our animators are using.

 

At the moment we have several different tasks, each of which will mess up the rigging. Does it make sense to just do all of them and THEN rerig the character, or does that have to be done after each task?

 

I would save rigging for the last.

[Springheel]

Well, if you're interested in doing the materials, that would be great. You'll have to confer with Domarius about any issues involved with working on two different 'issues' simultaneously--I suspect only one can be done at a time on a particular AI without messing things up.

 

I don't know where the original thread has gone, but basically what we want to achieve with the submeshes is to be able to divide the model up into four different material categories based on their appearance: flesh/cloth, leather, chainmail, metal. Each one would give different levels of protection from arrow damage.

[ascottk]

Well, if you're interested in doing the materials, that would be great. You'll have to confer with Domarius about any issues involved with working on two different 'issues' simultaneously--I suspect only one can be done at a time on a particular AI without messing things up.

 

I don't know where the original thread has gone, but basically what we want to achieve with the submeshes is to be able to divide the model up into four different material categories based on their appearance: flesh/cloth, leather, chainmail, metal. Each one would give different levels of protection from arrow damage.

 

Awesome news! I took my T3 Cabal character rigged with physique, used deep exploration to convert to a maya binary, open that in maya, & the rig was working! So here's my thoughts:

 

I can convert the md5mesh & md5anim to max, make the changes then rig it with physique, convert to maya, create def files, then import the MBs to doom 3. I can easily (well not easy) use the meshes and anims from vanilla d3. This means we could scrap oddity's bones & be compatible with vanilla d3. It's just tweaking from there.

 

What did I just get into

[New Horizon]

You got into a lot of work my friend, but something that I think will be extremely beneficial to getting the mod together faster. If our models can be made to use the animations from doom 3...then we're going to be on a lot more stable ground. Great work ascottk!

[Springheel]

I can easily (well not easy) use the meshes and anims from vanilla d3. This means we could scrap oddity's bones & be compatible with vanilla d3.

 

Anytime anyone suggests scrapping things I get nervous.

 

What does this mean exactly? We'd lose the ability to use any of our current animations and ragdolls, but we'd be able to replace them with any existing D3 animations?

 

If true, do we get much of a benefit there? Other than the standard walking, running, and possibly sitting animations, what else does D3 have that would be useful for us?

 

edit: ok, I see a few other animations that might be useful--turning, several conversation ones, pain animations...maybe there are enough to make it worth it?

[ascottk]

Anytime anyone suggests scrapping things I get nervous.

 

What does this mean exactly? We'd lose the ability to use any of our current animations and ragdolls, but we'd be able to replace them with any existing D3 animations?

 

If true, do we get much of a benefit there? Other than the standard walking, running, and possibly sitting animations, what else does D3 have that would be useful for us?

 

edit: ok, I see a few other animations that might be useful--turning, several conversation ones, pain animations...maybe there are enough to make it worth it?

 

We could just do the vanilla d3 head bones & use the bodies of tdm. As for having enough for it to be worth it, I don't think we have hammer/sword drawing/swinging/etc animations in vanilla.

 

But I think this'll work great for the thief character. I'll try him out first to test things out.

[Dram]

Yeah, being able to use D3 anims woould be super useful, as they were pretty good. They also have cower etc

[New Horizon]

Yeah, being able to use D3 anims woould be super useful, as they were pretty good. They also have cower etc

 

Yes, and really...it just makes sense for us to use what has already been established and tweak from there. No need to reinvent the wheel. The D3 settings already have a lot of the wrinkles ironed out, so it saves us a lot of work.

[Springheel]

You know me, I'm always in favour of saving work.

 

Just so I'm clear though, this would mean redoing the existing animations that we don't have D3 versions of, right? (searching, drawing swords, etc).

 

And it would mean that any existing animations for basic D3 AI (not monster types) could easily be added to our character models? Sounds like we would probably gain more than we lose if so.

 

Would it have any other impact? AF models are actually separate from the animated version, so would they be affected by this? (actually the D3 versions would be an improvement)

[New Horizon]

Would it have any other impact? AF models are actually separate from the animated version, so would they be affected by this? (actually the D3 versions would be an improvement)

 

Yes, the d3 versions would be a huge improvement. I found them to be extremely well done. Honestly, I can't think of any major drawbacks to this.

[Solis]

Unfortunately modeling isn't my forte, however I can probably do #2 and #5. I don't know how to set it up for UV mapping or rigging however, so once I modify the basic models someone else would need to finish those steps.

 

Are we going to have some kind of system so we can see what models have had what done to them? Perhaps we should do this in steps so that two people aren't modifying the same base model and we end up with two different versions that have each had a single change made to them.