I’ve been looking around the UE community and the web in general, and it looks like if I wanted to create my own motion captures and animations for my game,(stuff like swimming, fighting, doing specific tasks like, say, digging) I could do it with little more than just an ordinary web cam.
I’m wondering if it’s worthwhile? Can I create quality motion captures with home equipment or is a much bigger deal to get even the simplest animations than meets the eye (my eye, that is)?
Again, I don’t really mean anything for cinematics or anything too complex, just ordinary actions like, if I wanted my character to dig a hole, or scratch their back, sit down, do a hand shaking motion… etc.
Should I get into this stuff or, unless I’m planning to invest a lot of money, I should just stick to buying moccaps on the marketplace?
I started using the mocap market place stuff and it is worth getting just so you can see how it is done and to use it as a reference. So even if you do decide to do your own animations I think it is worth getting.
I kind of got frustrated in that I couldn’t get everything I needed from standard mocap data. So we decided that we did need to do a certain amount of animation ourselves. We use Maya LT and, although expensive, it does do the job. You can use blender of course.
For me, doing the animations is quite fun and I think it really adds uniqueness to your game. We actually have gone down the route of making up our own animation cycles rather than trying to capture data.
There are many small things that you can’t get generic capture data for. For example your gun firing sequence and non biped stuff like a tank moving its cannon around.
Just an opinion but mocap is like buying a box of crayons where you don’t have to make the crayon you want from scratch but by using a different combination it’s easier time wise to shape the final result your looking for.
Being an animator the question is not so much should or should you not use canned animations but can you based on your choice of authoring software suitable to the task at hand as say a web designer or a programmer would have stock assets at hand to cover the repetitive tasks.
As an example my authoring tool of choice is Motion Builder which allows me to manage a large library of canned mocap across many different formats which so called animation programs like Max and Maya can not do. Using anything else besides MB would be like a coder using note pad to code.
So sure I by canned animations just to add to my kit and from there I use my eyes to tell me if what I see is going to work or needs further editing to taste.
this is a great reply.
also to keep in mind, price vs. what you are actually using. if there are 150 anims, and, you only need 10 … is it worth the cost? you need 50, still worth it?
I just picked up a mocap setup for my studio and, yes, its going to save us a ton of money, and, add a massive amount of flexibility to the entire “what anim do we need next” process. Spell casting? Not a problem now! general anims? yeah we got those, but, so does the marketplace for a fair price. Really this is only saving me on the specific anims, not the general. Why spend time making a wheel when you can just buy one (or crayons … lol), tho, thats assuming your wheel doesnt need 500 custom spokes lol.
Just so happens, the price point was better just buying my own equipment then buying a ton of anims. This is not the case with all studios / dev teams.
The ideal is not so much about buying what you need but about collecting usable assets over time as any artist would collect raw material. Personally I have 10’s of thousands of mocap clips that I can window shop for the perfect walk cycle and shape it to be my own.
Once again though it really is about authoring and how to manage a large volume of “data” in bulk form and Epic’s market place is not the only place in town to buy cheap raw animations or even download free art as part of the collection.
true - very true sir …
your experience and time spent learning and gathering is probably about “infinity” times more then mine haha. Im just calling it like i see it as a no0b to the mocap / animation world.
Just saw this post and thought i’d give my oppinion. When i started playing around with ue4 about a year ago, i had the excact same question, and i choose to try and do the mocap myself. During that process i found that the equipment needed was either too expensive or the capture quality would lack so much that it would require a ton of mocap cleaning which is an entire discipline of its own. But since i went that path with ipisoft “pocket mocap” using 6 ps3 eye webcams and motionbuilder and a lot of determination, i eventually ended up learning quite a lot of mocap cleaning techniques in that process. So to sum up my feelings now, is that doing mocap on a low budget seems really hard, and the disciplines required to get decent animations out of it are quite a lot, so this will take away time from other areas you might be interrested in too, so in the end, it’s a matter of what you want to focus on and what area you want to develop yourself within. Would i do this again, if i had this knowledge i have today. I think yes, but that’s because im more a generalist than a specialist and like to get my hands into a bit of everything.
I think Jonas speaks volumes into this area. I have followed his videos for a long time, and, i have see some serious progression in his work in the process. Now, with that said, I believe there was a break in his videos where he even talks about the complexity of doing your own MOCAP. I believe he talks about (maybe in a comment thread) how he needs to figure out the equipment required and all that.
Optimally i would save up for some mocap equipment that can deliver reasonable results, and so far the only experiences i have are with ipisoft. With that you will however get quite bad quality data, well depending on how the recording environment are setup and lots of other factors. I would love to try out how it is to work with ie. a vicon or optitrack, but as an amateur, that’s clearly out of range for me at the moment. So since im clearly interrested in animation and in particular motion capture, the only viable option i see, are to keep using my ipisoft solution and maybe invest in perception neurons, all though i’ve read quite varying reviews on it… like typical chinese quality (breaks easy and stuff like that). Hopefully someone else which already has a suit can shed some more light on the PN’s.
Back to the question of is it worth trying to do own animations/mocap. I’d say it all depends on your interrests or goals. As FrankieV said, it’s possible to gather quite a nice library of animations by surfing around various sources, so if you mostly want to produce some results and not delve too much with animations specifically, then i’d go with marketplace assets/other sources.
As a small Indy group we have been looking at the so called off the shelf MOCAP solutions for some of the more complex player interaction and although at first glance the “so called” low cost solutions “seems” to be reasonable they all have far to many hidden costs that in the end it’s still more cost efficient to sub out to a studio who already have the necessary resources with out the additional running around to patch work something together.
The big buzz at the moment seems to be the availability of low cost facial capture solutions but the deal beaker for us is the hidden cost of hiring skilled actors who have the talent to extract the emotional context, performance, that can not be duplicated via hardware solutions.
Another deal breaker is caused by the almost impulsive nature of Indy development of what has to be done today and what you hope to get around to doing tomorrow where there is really no time to do all of the things you would love to do as a while your at it just because it’s cool.