Machinima: Triggering animations? New player models? Lip-sync?

[Exavior]

Greetings all,

I am a designer out of Los Angeles, and I'm working with a Hollywood improv comedy troop doing a live-on-stage machinima project using the HL2 engine.

So far I have been focusing mainly on the creation of sets and other "minor" tasks, working my way up to the real editing that will be required to get this ready for a live performance.

I have a couple of questions that I have not been able to get answered elsewhere, and I was hoping maybe someone here could help.

Can anyone think of an easy way (using hl2dm) to get the player mouthes to open and close (much like a puppet) when a key is pressed/released?
Ideally I would love to have microphone input do this for the actors, but I'm not sure what is possible at this time.

How hard would it be (using hl2dm) to use other character models as players? The armor clad warriors it uses at the moment are not too fitting to the theme of our project.

Also, would it be difficult to bind other keys to player animations, such as wave?



Thanks in advance for any help!

 

[ghost_asassasian]

it would probably be possible but it would require a lot of digging and code searching to have a key activate mouth movement because they dont have preset expressions they have to make each expression in face poser.

 

[Faulk_Wulf]

Correct me if I'm wrong (and I know you all will ). But doesn't CS:S have a "cheer" command? (In 1.5 this was a fist pumping motion.) At anyrate if a Cheer can be done by a press of a key, couldn't other animations?

 

[Vegeta897]

Go over to hl2w0rld.com (replace 0 with o)

They know 10x more than anybody on HL2.net. They are the Gods of Source editing.

And this is a hl2.net fanboy speaking.

 

[Exavior]

Yeah, their site is a great resource that I use for tutorials. I do like this forum, but I have a hard time ever getting answers to more "advanced" questions.

I guess I'll hit them up over at the other forum. Thanks.

 

[Iced_Eagle]

Also be sure to email VALVe... They are more than happy to answer your questions

 

[Exavior]

Somehow I don't see valve going out of their way to help modders whos final product would not be lucrative to them.

So it looks like I'm stuck at sarcasm filled forums for now.

At the moment it looks like I will need to create a simple multi-player mod for this project, since hl2dm does not have the ability to trigger the type and volume of animations I will eventually need. Sadly, given the overwhelming number of kiddies out there asking for mod help, anyone who has real answers is either jaded or no longer reading question threads.

 

[botman]

"Sadly, given the overwhelming number of kiddies out there asking for mod help, anyone who has real answers is either jaded or no longer reading question threads."

I'm not sure that's true. But, in most cases, someone won't respond to your question unless they know the answer to your question.

What you are doing is unique (but not impossible). Since no one has done this sort of thing before, you won't get many people responding with "Okay, here is what you need to do...". You are pretty much going to be on your own.

You would probably do better by asking more simple questions, like "How do I change the rotation of a bone in an animation to allow a player to contol a limb independantly of the animation?", or "How to I change the position of the jaw without using the facial poser application?". Break it down into smaller problems and solve each one independantly then bring everything together to create the system that you want. This will work much better than saying "How do I make it so that players control characters like marionettes?"

botman

 

[RabidJester]

The animation triggering isn't going to be your problem



The trouble will be with the lip syncing. All of that is done and contained within the scene files. The last interviews on the topic I believe said that they didn't know if real time sync up would be possible.

I hope this works out though.

 

[Exavior]

botman::
Thanks for the suggestions. I figured it would be best to ask broad questions and have those who know routes to possible solutions share them.

There are modders out there who have done some of these things before, the problem is that many are not on this forum, or reading/posting here. That is why I am posting this same question at a couple different forums, in an effort to find conclusive answers. Don't get me wrong, all the help I have received here is greatly appreciated, and I am thankful for it.

At the moment I am the only one working on the design/technical aspects of this project, and guidance/aid of an experienced modder would be a godsend. Since the final application of this is a serious project (not a weekend mod project) the end result needs to be professional quality work, and I don't feel that I have the C++ experience to deliver the quality required of such a high profile project. My background is in 3d modeling, texturing, lighting, mapping and scripting. Which is already quite a work load. So I hope you understand my frustration.



RabidJester:
Real time lip sync would be great, but we can live with just puppetry style mouth open/close on key press/release. Real time sync would be great though, but I don't mind setting my sights a bit low to start with.

Thanks again for the suggestions so far.

 

[ChildeRoland]

I have a related question, however, I am not yet as far in the work as the poster.

I am trying to create a mod where the player can fly around and set
waypoints. Then, NPCs will follow these points and complete certain
assigned animations and actions (shooting, saying a sentence, pointing, etc). I have a general idea how to do this, but I am not
sure where to start.

There are so many different AI/NPC (CAI_Actor, CNPC_Companion, Monsters, etc) classes, which would work best for this type of work? Also, does anybody know a good resource or tutorial (probably aimed at mappers) for how to set waypoints and doing scripted work for HL2??

 

[crackhead]

i had an idea but its not too brilliant. if you made the animation of the mouth moving in face poser then save the choreograph file and add it into the map and everything then have a trigger and a game_ui entity. then link them to trigger the game_ui. then in the flags tab in game ui uncheck freeze player so you can still move around for different camera shots. then in the outputs tab you can use wasd and right click and left click to trigger things so you could have freds mouth moving choreo scene triggered by w and then bobs mouth moving scene triggerd by a and you could even have d and s stopping the scene so you can make the animations longer and then press s when hes stoped talking so that you dont have the sound file still going when the animations finished or the animation still going when the sounds finished. also if you did it like this that would mean you could only have like 3 ppl talking at a time but you could then move into a different trigger to trigger another game ui etc. or maybe it is possible to edit the code to alow for more buttons to trigger things. i think this would be the easyest way to achive what you want and if your ok with it not beeing lip synced then the best solution. however you could still have it lip synced if you know what their going to say before hand. if you didnt understand or would like me to explain all this in more detail and in steps then just say so. p.s i love your maps

 

[crackhead]

i could also make an example map showing this and send it to you if you want.

 

[crackhead]

ok i have planned it all out an know how you can do it exactly. i have also made a custom tutorial map using this technique that i will send you if you give me an msn adress or email adress. i have just used random speaches. the only draw back to it is that inorder to get the lips to move a sound of a character speaking has to be loaded. so you will have to think about how to get past this. maybe somehow mute the voice or mute the sounds from the actual game and only have sounds that are beeing dubbed in. but apart from that factor everything works perfectly for what you want. in the map there is more detail on how to have more than two ppl speak and stuff. and if you want it to be perfect if you know what the actors are going to say before the live production then record all this onto the computer and load it into face poser and then create the choreo scenes using the lip syncing on that so that it will look perfect. but it only lip syncs automaticly for the certain characters. i will also think about how you can get past the problem of the speaking.

 

[Exavior]

The only problem is that would only work with NPCs. We need multiple actors to be able to log in as players. Also, they will be speaking in real-time, not playing back recorded files. In the end what is really needed is a mod. I'm still in the process of searching for a modder with the skills needed to pull off such a mod, but I have had little luck.

/Exavior

 

[crackhead]

why does it need to be real time? i think this is the only way you are going to be able to do it

 

[Hells Angel]

Exavior,

If you wish to model these and create your own animations the only way i can think to get a proper voice sync would be to use 3DS Max along with "Voice-O-Matic" a software that converts Wav's and creates an Instant voice sync to it. When used with "Facial Studio" it creates a characters face with lip sinc to what you want in a matter of seconds. Although this does not create a body or hair. Hope this answered your question because i was trying something like this with UT not so long ago

Hells,

 

[crackhead]

ahhh the ut modding days that brings back sum painfull memories