OK, folks, I've never used CrazyTalk, but I think IT'S TIME.
See for yourself:
http://forum.reallusion.com/Topic2363-71-1.aspx
CrazyTalk 5
Moderators: Víctor Paredes, Belgarath, slowtiger
Wow!!! That's brilliant.
Have you seen Daz 3D Puppeteer? It's similar in concept.
http://www.youtube.com/watch?v=3AZRYJC9RV8
D.K
Have you seen Daz 3D Puppeteer? It's similar in concept.
http://www.youtube.com/watch?v=3AZRYJC9RV8
D.K
I've been using CT for over a year, and it's a lot of fun and I think it's definitely even possible to integrate it into AS work if someone wants to, but the actual nuts and bolts may drive AS users crazy.
It has a lipsync program similar but vastly inferior to Papagayo -- the real problem is the timeline. There isn't any way of audio scrubbing, so the best thing you can do is spend long tedious amounts of time trying to get the audio right -- or juse a simple (and I mean *very* simple) audio track.
And the timeline has issues when it comes to facial expressions. Examples done by RI staff are great -- but have been tweaked and worked on for weeks. Getting a good, expressive face of even 30 seconds takes many many hours of work, and in the end you could do a far better job in AS (IMHO).
Until I found AS I was sold on CT as a way of getting 2D animation -- but now that I've seen what AS can do I'm not even going to upgrade from CT 4.6 to 5.0 (at a very minimal cost). I just can't justify the work versus the final output.
It has a lipsync program similar but vastly inferior to Papagayo -- the real problem is the timeline. There isn't any way of audio scrubbing, so the best thing you can do is spend long tedious amounts of time trying to get the audio right -- or juse a simple (and I mean *very* simple) audio track.
And the timeline has issues when it comes to facial expressions. Examples done by RI staff are great -- but have been tweaked and worked on for weeks. Getting a good, expressive face of even 30 seconds takes many many hours of work, and in the end you could do a far better job in AS (IMHO).
Until I found AS I was sold on CT as a way of getting 2D animation -- but now that I've seen what AS can do I'm not even going to upgrade from CT 4.6 to 5.0 (at a very minimal cost). I just can't justify the work versus the final output.
At the risk of sounding very dense, can I assume that you feel these difficulties in animating CT 4.x are still going to be inherent in version 5.0? The video makes things look so easy...mkelley wrote:I'm not even going to upgrade from CT 4.6 to 5.0 (at a very minimal cost).
Thanks DK for the link to DAZ. Very interesting!
But I suppose that, again, it won't be as easy as it looks... ? !
The CT 5.0 Pro beta is downloadable -- while all of us 4.x owners were sent the link I don't think it's "secret" or anything (it expires January 15th) so I suspect you can get on the site and download and try it out yourself.
Indeed, I highly recommend you do so -- because you will quickly see if you like it or not.
I only tried the 5.0 demo twice -- both times convinced me it was nothing special -- let me clarify that. It was nothing particularly special compared to the 4.x version. The most they've brought new to the table is the real time "puppeting, but in practice the feature is awkward and unnatural to use (just IMHO, of course -- try it out yourself to see what *you* think).
CT is a wonderful program for folks who just want to "play" -- I used to use it to create some stuff for my grandkids and they were thrilled by it. It's strictly a talking heads program or, more properly, a talking "head" since using it for multiple characters is next to impossible without a lot of work. Rather than say much more, take a look at what I did with it:
http://www.youtube.com/watch?v=JqmQBUUun7Q
http://www.youtube.com/watch?v=bxojLh4PHFY
Note that neither of these things are very good -- the animation is extremely limited. I suspect I could have done both of them about ten times faster (and better) using AS. But for someone with little talent Crazy Talk is great, because you can put something together knowing almost nothing about art or animation or, indeed, anything.
For a while (a short while) I had delusions I would use CT for the heads of characters I'd create in AS. It's doable, no question about it, but using Papagayo and having some talent you can do *much* better, truly.
Using the heads in CT you can get some very limited turning and nodding, but once again it's nothing you can't do in AS with either bones or switch layers. The most significant thing about CT (and one I wish could be implemented in AS) is the ability to overlay "emotions" onto the lipsync.
I'm no graphics programmer, but it seems to me something similar could be done in AS -- start with a neutral set of mouths, for example, and then (using same vectors for translation) allow a new mouth shape to influence the phoneme shapes. That way you could define a smile, a frown, a grimace, etc. and these would then influence your stock mouth shapes in such a way that you could "emote" on the time line with little effort.
We do this all the time in 3D, and I don't see why it isn't possible in 2D, but I'm fairly sure this can't be scripted. But it would be ENORMOUSLY powerful, this offset ability. It's all I really long for in AS.
Indeed, I highly recommend you do so -- because you will quickly see if you like it or not.
I only tried the 5.0 demo twice -- both times convinced me it was nothing special -- let me clarify that. It was nothing particularly special compared to the 4.x version. The most they've brought new to the table is the real time "puppeting, but in practice the feature is awkward and unnatural to use (just IMHO, of course -- try it out yourself to see what *you* think).
CT is a wonderful program for folks who just want to "play" -- I used to use it to create some stuff for my grandkids and they were thrilled by it. It's strictly a talking heads program or, more properly, a talking "head" since using it for multiple characters is next to impossible without a lot of work. Rather than say much more, take a look at what I did with it:
http://www.youtube.com/watch?v=JqmQBUUun7Q
http://www.youtube.com/watch?v=bxojLh4PHFY
Note that neither of these things are very good -- the animation is extremely limited. I suspect I could have done both of them about ten times faster (and better) using AS. But for someone with little talent Crazy Talk is great, because you can put something together knowing almost nothing about art or animation or, indeed, anything.
For a while (a short while) I had delusions I would use CT for the heads of characters I'd create in AS. It's doable, no question about it, but using Papagayo and having some talent you can do *much* better, truly.
Using the heads in CT you can get some very limited turning and nodding, but once again it's nothing you can't do in AS with either bones or switch layers. The most significant thing about CT (and one I wish could be implemented in AS) is the ability to overlay "emotions" onto the lipsync.
I'm no graphics programmer, but it seems to me something similar could be done in AS -- start with a neutral set of mouths, for example, and then (using same vectors for translation) allow a new mouth shape to influence the phoneme shapes. That way you could define a smile, a frown, a grimace, etc. and these would then influence your stock mouth shapes in such a way that you could "emote" on the time line with little effort.
We do this all the time in 3D, and I don't see why it isn't possible in 2D, but I'm fairly sure this can't be scripted. But it would be ENORMOUSLY powerful, this offset ability. It's all I really long for in AS.
I've had Daz 3D for several years and never looked at it much. After your post I opened it up and. WOW!DK wrote:Wow!!! That's brilliant.
Have you seen Daz 3D Puppeteer? It's similar in concept.
http://www.youtube.com/watch?v=3AZRYJC9RV8
D.K
I was surprised I could do all that stuff. I have had Poser and use that for character references and so I have lots of resources to use in Daz 3D. Thanks D. K.
Bob
.
.
.
I keep saying this... I'm like a broken record sometimes... but if they open up more of the interface to scripting access, features like this can be developed OUTSIDE of efrontier or Smith Micro to create add ons and tools to extend the functionality.
I personally believe that would better than hoping for or asking for new features. Let those of us capable create cool new stuff no one might have thought of before.
Look at all the plugins for photoshop or AE. Adobe didn't have to create all of those. That's my dream for the future of AS anyway.
I could imagine this kind of thing being done using actions in AS. A set of predefined poses or actions that can be "displayed" and "previewed" with out keying until you click "record". Then it could create a brand new action.
It all is pretty straight forward... it's just "math". Averaging out the position of bones based on proximity to a cursor. At the moment the current scripting functions would make this really difficult... but not entirely impossible.
-vern
I personally believe that would better than hoping for or asking for new features. Let those of us capable create cool new stuff no one might have thought of before.
Look at all the plugins for photoshop or AE. Adobe didn't have to create all of those. That's my dream for the future of AS anyway.
I could imagine this kind of thing being done using actions in AS. A set of predefined poses or actions that can be "displayed" and "previewed" with out keying until you click "record". Then it could create a brand new action.
It all is pretty straight forward... it's just "math". Averaging out the position of bones based on proximity to a cursor. At the moment the current scripting functions would make this really difficult... but not entirely impossible.
-vern
You're dead right there Vern. I learned Flash programming because I wanted to introduce some nteractive elements into it. At first I offered to pay people to do the programming for me but the results were terrible. Same story here with AS. If AS was opened up so users could develop third party plugins things would just steam ahead. Personally, I don't mind going down another learning curve as long as there are ample learning resources available.
D.K
D.K
Look at Youtube.
Sold for 6 billion to google. It is just a "fancy" Flash gizmo created by a couple of guys who wanted an easy way to share video online. It uses the same Flash/actionscript technology available to anyone.
I wish there was a way to create some kind of "packaged" scripted solutions for AS. With an installer of some kind. Maybe I could just create an installer using some kind of zip installation creator?
It would be great to offer an easy installer for custom tools so people wouldn't have to do funky things to make them work.
-vern
Sold for 6 billion to google. It is just a "fancy" Flash gizmo created by a couple of guys who wanted an easy way to share video online. It uses the same Flash/actionscript technology available to anyone.
I wish there was a way to create some kind of "packaged" scripted solutions for AS. With an installer of some kind. Maybe I could just create an installer using some kind of zip installation creator?
It would be great to offer an easy installer for custom tools so people wouldn't have to do funky things to make them work.
-vern
I deeply appreciate mkelly's skill with Anime Studio.
However, now that I have tried Crazy Talk 5, you may consider me just as deeply impressed with it as he is UN-impressed with it.
I strongly disagree with his view that it is difficult to get good results with it.
I even more strongly disagree with his view that one could do anything remotely similar with Anime Studio... unless you had a team of geniuses working around the clock!
mkelly, you can prove me wrong--post an Anime Studio project with anything remotely like this kind of facial animation.
I also disagree with his view that its limitations mean it can't be taken seriously.
Granted, it doesn't support head turns beyond a precious few degrees. This is obviously because they provide only one human profile for all head fitting. If you turned the head more than a few degrees, the fidelity to the front-on model (which is good!) would break down.
I also disagree with his suggestion that it ought to support multiple audio tracks.
As a matter of fact, I even think that its support for music is stupid and ill-advised.
This is a facial expression/lip-synching tool, for godsake.
This is NOT a tool for disc jockies, film composers, or rap artistes.
/rant begins here/
I would go so far as to say that I see similar absurd expectations being placed on animation tools across the board. Why can't Anime Studio do video editing, fade-in-fade out, multilayer video composition, bitmap retouching, multi-track audio mixing, and offer a solution to world peace, all at the same time?
Answer: because it's an ANIMATION TOOL.
You want to do video editing? USE A VIDEO EDITOR.
Want to do advanced video compositing? USE A VIDEO COMPOSITOR.
Want to do multitrack mixing? USE AN AUDIO WORKSTATION.
I don't understand where the problem is with this!
OK, enough of that.
/End rant/
Back to Crazy Talk.
My prediction for 2008 is that most of you using Anime Studio on Windows will be using CT5 to generate reference pictures for your vectors.
And those of you using AS on other platforms will be spending some of your time, somehow, on a Windows machine, using CT5 there.
( Note: for the record, my only relation to Reallusion is as a paying customer. )
However, now that I have tried Crazy Talk 5, you may consider me just as deeply impressed with it as he is UN-impressed with it.
I strongly disagree with his view that it is difficult to get good results with it.
I even more strongly disagree with his view that one could do anything remotely similar with Anime Studio... unless you had a team of geniuses working around the clock!
mkelly, you can prove me wrong--post an Anime Studio project with anything remotely like this kind of facial animation.
I also disagree with his view that its limitations mean it can't be taken seriously.
Granted, it doesn't support head turns beyond a precious few degrees. This is obviously because they provide only one human profile for all head fitting. If you turned the head more than a few degrees, the fidelity to the front-on model (which is good!) would break down.
I also disagree with his suggestion that it ought to support multiple audio tracks.
As a matter of fact, I even think that its support for music is stupid and ill-advised.
This is a facial expression/lip-synching tool, for godsake.
This is NOT a tool for disc jockies, film composers, or rap artistes.
/rant begins here/
I would go so far as to say that I see similar absurd expectations being placed on animation tools across the board. Why can't Anime Studio do video editing, fade-in-fade out, multilayer video composition, bitmap retouching, multi-track audio mixing, and offer a solution to world peace, all at the same time?
Answer: because it's an ANIMATION TOOL.
You want to do video editing? USE A VIDEO EDITOR.
Want to do advanced video compositing? USE A VIDEO COMPOSITOR.
Want to do multitrack mixing? USE AN AUDIO WORKSTATION.
I don't understand where the problem is with this!
OK, enough of that.
/End rant/
Back to Crazy Talk.
My prediction for 2008 is that most of you using Anime Studio on Windows will be using CT5 to generate reference pictures for your vectors.
And those of you using AS on other platforms will be spending some of your time, somehow, on a Windows machine, using CT5 there.
( Note: for the record, my only relation to Reallusion is as a paying customer. )
If you like it and enjoy using it, fine. But I suspect that after you've played with it for a few months the limitations will really bother you and you'll realize it's not really an animation program that will allow you to do much.
Put it another way -- when was the last time you saw a movie that just consisted of a single talking head?
You can't tell a story (or at least, not much of one), you really and quite literally have a "talking head". This sort of label is applied to those TV shows which feature panelists who sit around talking, and is a derogatory term because the shows themselves are very limited. Well, the same thing is true of anmation.
For extremely short (under 30 seconds or so) periods of time it's very amusing to watch a single head, which can't do much in the way of movement, "talk". After that you'll want to move on and do a lot more with your animation -- it's kind of the same thing that happens when one first learns to play a melody on the piano with the right-hand. You get all excited and think you are creating music, which you are, sort of, but in a quick order you realize that you want to do a whole lot more, and learn to play the left hand as well.
And you can see this time and time again on the Real Illusion forums -- folks there quickly get tired of what CT can do and want to try and do more (which always means they move on to other programs -- usually the 3D program RI offers at a reasonable cost).
The greatest thing about all of this, as I mention, is there is a free demo and everyone can judge for themselves. But really judge -- use if for six months and I'm fairly sure you will decide there are far better things to use to animate with.
Put it another way -- when was the last time you saw a movie that just consisted of a single talking head?
You can't tell a story (or at least, not much of one), you really and quite literally have a "talking head". This sort of label is applied to those TV shows which feature panelists who sit around talking, and is a derogatory term because the shows themselves are very limited. Well, the same thing is true of anmation.
For extremely short (under 30 seconds or so) periods of time it's very amusing to watch a single head, which can't do much in the way of movement, "talk". After that you'll want to move on and do a lot more with your animation -- it's kind of the same thing that happens when one first learns to play a melody on the piano with the right-hand. You get all excited and think you are creating music, which you are, sort of, but in a quick order you realize that you want to do a whole lot more, and learn to play the left hand as well.
And you can see this time and time again on the Real Illusion forums -- folks there quickly get tired of what CT can do and want to try and do more (which always means they move on to other programs -- usually the 3D program RI offers at a reasonable cost).
The greatest thing about all of this, as I mention, is there is a free demo and everyone can judge for themselves. But really judge -- use if for six months and I'm fairly sure you will decide there are far better things to use to animate with.
