[TSM and BVH]

Well I suppose that if you did the null constraing in an action so that it didn't completely over ride the SM rig all the time, then you probably could use another action on top of the mocapped one to animate on top of it; however, If your thinking of using this to try and clean up the mocapped data using the TSM rig, then I wouldn't recommend it. I'm not 100% sure how well they would blend as we've never tried doing something like this. Again if you can provide more info on the mocap data then we could probably help you out a lot more.

[TSM and BVH]

Hmm.. well if I'm reading your question right you want to have BVH mocap data control the TSM rig right? If this is the case it can be done; however, it will ignore TSM's setup. To give you the best advice on how to do this we would need to know the mocap system your using to explain how to link it to the TSM rig; however, if the mocap system is a commercial one, and the driving null information is not available publicly then we might not be able to help you.

[Rest Pose]

You can use a pose or action to get the arms to the side.

[Disney's Chicken Little trailer online]

You know... I like and own both "Cats don't Dance" and "Emperors new Groove;" However, when I saw this teaser I thought there was no hope for it. Then I thought back to Iron Giant and remember how bad the trailers for it were, and how great the film turned out to be. I only hope that Mark Dindal has been given the creative freedom he had on the other two movies, and is able to pull off some of his style in 3D... We shall see...

[Frequent Buyer program?]

William, looks like your CD is "in demand"...my order got held up for a bit... ! However...my Jeff Lew DVD showed up...

Cool, I've herd nothing but good things about Jeff's DVD's. He's a really nice guy, and use to work a few towns away from here a few years back (North Adams, MA). As for the hold up, Steve had a quadruple bypass, and so some of the orders got a little delayed as others tried to fill his shoes. Hope it wasn't to much of an inconvenience. (Side note: He's doing a lot better now)

[Frequent Buyer program?]

I actually ordered the Face bundle a couple of weeks ago and have been going through them and am finding them very helpful. It is interesting to me to see how different people address the same subject.

 

When I look at Bill's modeling techniques vs. Colin Freeman's vs. David Rogers' I can see several similarities and several differences. Bottom line is that there is no one way to do something. So I'm taking a few elements from each and combining them to my own style which seems to be working so far.

It's true that there are several ways to get similar results when modeling; however, One of the questions I always ask is how animatable something is going to be. Some techniques can make very nice looking still faces, but when you go to set them up for animation they can suddenly run in to limitations. The face modeling CD was intended to explain why splines were placed the way they were for the purposes of animation, not just making a decent looking modeled face. Does this mean it's better than other techniques? Well I am a bit biased to my technique as I know why I did what I did; however, the truth is no it's not necessarily better than any one elses so long as they achieve the level of animatable detail control a user wishes.

 

I would still like to get several more CDs (materials, secrets 2003 & Non-linear) but between those and trying to update to V11 I'm going broke and it seems like no one wants to part with their used videos!!!

 

Again feel free to talk with Steve. If you already have some of the CD's in a bundle, he'll have record of it, and I believe he might remove them from an existing bundle that contains those CD's reducing the price of that bundle; however, you need to email Steve to get the full scoop on this, because as I said I'm really not the person who knows how this stuff is supposed to work.

 

Maybe it was just wishful thinking that Anzovin would have a buy 4 get one free deal .

Bundles. Most stores have a buy a bundle of something, at one time get a bonus of something else; however, if you buy over time, the only thing that might help is trying to see if Steve will sell you a bundle minus the Cd's you already have... Again I'm not sure about this, thus "Contact Steve" (steve@anzovin.com)

 

And while I have your attention, Bill are there any plans to update some of the older videos to the newer software versions? In particular Non-linear & face rigging.

 

Well I'll say this.. there are plans for more videos and some remakes of older videos in the future. I don't wish to start rumors as to wich CD's these will be yet as we'll assess the demand a bit closer to production time; however, I know that there has been discussion of potentially remaking the NLA and Face Rigging CD's. It should be noted that the techniques in the existing CD's will still work, but the new CD's would cover other potential techniques.

[Frequent Buyer program?]

Being a newbie, I'm guessing that's the hard part

Actually to me modeling is one of the easier aspects of the animation process IMO... But then thats just me...

[Frequent Buyer program?]

Chris, do you own the "Face Modeling CD" by Young? Is it suitable for the older beginner ?

I tryed to make it suitable for every one; however, diferent people learn diferently. Some people might find the pace a bit slower than they might like, others might find it to quick. I just tryed to keep it clear. Interestingly this seems to be a very popular CD over at Umass Amherst right now, one of the TA's has been touting it as the CD every one in the class should have. Was kind of flattering.

I believe this is the same Bill Young who designed keekat...neat...

Well I didn't design Keykat as it was created by the people at Paw Island Entertainment Inc., but I did model him.

[Frequent Buyer program?]

I love all the Anzovin training CDs that I have gotten so far and would love to get more but I'm getting poor buying all these. Maybe there is a frequent buyer program from Anzovin that I don't know about (hint, hint)?!?

Well I know we offer the bundles to keep costs down for people; however, I don't think there is a "frequent buyer" program. I think you might want to contact Steve (steve@anzovin.com) directly if you want to make a custom bundle. I'm pretty sure he has done that before, and it might be able to save you a few bucks; however, I'm not really the authority on those matters.

[Modelling a face]

Are you referring to the "Model a Face with Bill CD"? I just ordered it today...

I didn't want to respond, but looks like the original poster isn't. So I'm going to say I belive he is talking about the CD. Hope you like it.

[Orc]

Well, I'm pretty happy with how not tampering with bias handles is going for me so far. And as all my experience with biases has ended with very weird results, I have concluded that they are basically either very difficult to work with successfully, or I just don't know how to use them. Besides, for the meantime, I like being able to tweak my geometry in big swoops. I like selecting points on either side of the face, or just in groups, and messing with them. With bias handles, that sort of thing gets kind of tricky.

 

Thanks for the crits! Stand by for next post!

 

-Andrew

Bias handles in their free state can be either a blessing or a nightmare. There are times where being able to adjust the magnitude, alpha and gama all at once can be very useful, but most of the time I find it much easier to work with one aspect of the bias handles at a time. There are hotkey's that you can hold to lock out certain aspects of the bias handles, but for the most part I find my self adjusting the bias handles by simply dragging on the attribute I want to change in the properties panel.

 

Magnitude adjusts the weight of a specific handle

 

Gama most of the time adjusts the spline in and out agains the surface

 

Alpha most of the time adjusts the spline back and forth along the plane of the surface

 

There are times when alpha and gama can reverse their roles, but most of the time this only happens around 5 point patches

 

If you want to see what these attributes do better, lathe your self a sphere, select a point, and start playing with each of the values. It will become a lot clearer what they do, and they shouldn't be feared like the black plague

[Orc]

Seems to me there are three rules:

1) start simple,

Sounds like good advice

2) tweak endlessly, and

IF you plan your model a bit, or have some experiance under your belt this can be avoided, but it's also prety good advice.

3) don't touch the bias handles.

Ok Buzzer is going off on this one. You shouldn't touch your bias handles till you've got your model fairly well shaped with the CP's, but bias handles are always worth tinkering with if you feel comfortable with them. They can easily save you from wasting splines in areas that don't need them. Sure there once was a problem where it was difficult to animate bias adjusted CP's, but seeing as thats no longer a problem then there is no reason not to play with bias handles.

[New character WIP]

Wow Jim,

This guys is really looking good. The hair has great form to it, but looks a little flat specular wise. The ears specular also seems a bit hotter than the nose and forehead area, but the eyes! Wow! They say so much tired yet wise and concerned with almost a hint of arrogance. Good stuff. I'm guessing since "Mighty Mouse" he's porcelain all the way.

[Orc]

Vern's explanation is quite good, but I'd like to share my idea on how to think of channels.

 

A channel is best thought of a a gray scale image that represents something other than gray values. typically there are 2 types of channels most people will deal with Color Channels and Transparency Channels. When you are dealing with an RGB image as Vern pointed out you are dealing with 3 color channels a red color channel, a green color channel and a blue color channel.

 

How many values of gray are available per channel is dependent on the bit depth of the image. A typical RGB image has 24 bits of data which when divided by the three colors turns into 8 bits per color channel. 8 bits can only represent 256 values, thus each color channel has only 256 values of gray. A 48 bit image typically used for film is usually made up of three 16 bit per channel RGB values. An image isn't limited to only 3 color channels though, a typical image used in publishing houses are made up of Cyan, Magenta, Yellow, and Black color channels. In some circumstances publishers can use even more color channels than these to represent special colors know as spot colors.

 

The primary thing that all color channels have in common is that when they are displayed together on your screen they will mix their channel information with the other color channels to make up the final color image you see on your screen. Also know that despite images having various different kinds of color channels they will all be converted to an RGB representation of the final image on your monitor because a monitor can only display RGB information; however, other mediums such as printers can use these channels so the key thing to remember with color channels is they are only limited to the colors of the device that is going to reconstruct the image.

 

Transparency channels are different from color channels in the respect that their gray values represent levels of opacity. White is typically thought of as opaque, and black is typically thought of as transparent. By picking gray values between black and white you can vary how transparent or opaque any part of an image is. Some people call all transparency channels alpha channels; however, the alpha channel is a single transparency channel that can be saved along with the typical RGB information. When dealing with an 8 bit per channel image this adds an additional 8 bits of channel information to the standard 24 bit RGB file. This brings the total bit count to 32 bits, and is typically referred to as an RGBA image. In film when they add an alpha channel to an image it adds an additional 16 bits of channel data to the existing 48 bits bringing the total to 64 bits of RGBA channel information.

 

Photoshop deals with 8 bit per channel information quite well, and it has some functionality with 16 bit per channel image work; however, it looses it layering functionality when it is in 16 bit per channel mode. This is where programs like Cinepaint step in. They tend to handle high channel data with layers better than photoshop, and carry a much smaller price tag as well (Free).

 

Vern mentioned layer masks. All these are are Transparency channels that are assigned to a specific layer in photoshop. Also note that the photoshop format is one of the few formats that supports an unlimited amount of channels per image; however, if you have to many channels your image will be so large that you may have trouble opening it.

[RE: TSM]

It's our intent that there will be. We always try and port our A:M products to Mac as well, but some times it takes a while longer; however, TSM2 has one major difference over prior versions. We're developing it in C++ as a true plug-in, and aren't 100% sure whats going to be involved in porting it to the Mac. I'm not sure how well the OSX SDK is documented, and what C++ software we will need for the port; however, we plan on addressing these issues.

[Image Controlled Materials]

The plug-in that you're pointing out only works on PC's, and is restricted to projection maps only; however, if you have a PC and a Projection map will suit your needs then this isn't a bad option. If this won't suit your needs, then you're going to have to render out your procedural, and then reapply it as an image map. This does destroy the infinite resolution of the procedural, and it will take a while to re apply it, but it's currently the only way to achieve this effect in all versions of A:M.

 

This will probably make it into a future version of A:M however, as the basic structure for doing this is already in place, it just a question of when will we see it.

[jack daniels]

Yeah, I knew you were going to get weird artifacts around the decal given your prior image. There currently is a problem with alpha channels on transparency maps only trimming the areas that are pure black. You could have made a gray scale image and done a semi-hack to make the black look more like the underling material, but the problem will go away all together when the alpha problem gets fixed. (And yes for any one wondering the problem was reported with a sample project)

[TSM and work flow]

I'm just wonfering outloud, a bit, here but, would it be possible for Steelshark to use TSM on a single, 2-legged, segment of the wyrm and then string several instances of the TSM'd model together in a chor. to produce the whole worm? Each subsequent segment could then use Translate-to, with lag, and Orient-like constraints to get wyrm-ish locomotion.

He could also constrain the assembled wyrm(s) to a path and then animate the path to get the inch-wyrm effect.

 

Mike Hart

Indeed he could do it this way if he wanted, but this can't be automaticaly setup by TSM currently.

[Rendering .tga's or .mov's???]

You can always load the image sequence back into A:M and apply it as a camera rotoscope, and render it out as a quicktime.

 

It's always safer to render out to an image sequence, but if some one can spare the potential extra rendering time if something goes wrong then there really isn't any reason they couldn't render to a .mov file; however, just make note of what compression method you are using, because many codecs are not lossless, and if you plan on doing any post work this can result in generation loss.

[jack daniels]

You need a transparency map in addition to your color map. This will prevent your decal from accepting the transparency from the materials on the bottle.

[beta 6]

You'll be able to turn off the smart force keyframe on the Action tab from the tools options dialog in the next beta.  Thanks for the feedback.

Noel

Hmm.. Personally I wouldn't like to see this as such a global feature, but something allowed on a bone by bone basis. For example 99% of the time auto zero slop would be very useful on things like feet that are constantly contacting an object that stops them dead in their tracks like the ground; however, for almost every other bone in the body this feature would negate the benefits of moving holds as many people have pointed out. What could make this feature even cooler is if it could be turned on and off for any given bone at any point in time so that when your character starts to crawl on all fours the hands could have this simply turned on at that point in time, and they would gain the ability to work more like feet. Like wise some one might want to put their character into a handstand, so you enable it on the hands, and disable it on the feet.

[El Burro]

I can tell just looking at the image that most of the normals in the nose are pointing in the wrong direction. The 5 pointers might be pointing out, but some of the other patches aren't. All the patches surrounding 5 pointers have to all be facing in the same direction as the 5 point patch or the ones facing in the opposite direction will cause the black creases at the edge of your 5 point patches. Remember you might also have to save and reload the project after you flip the normals to get it to render correctly.

[TSM and work flow]

The easiest way to describe what TSM does is to compare it to a marionette. If some one were to build a marionette they spend quite a bit of time creating the actual puppet, but after they are done building it, they still have to build a controller and attach all the strings to the proper joints. In this kind of process this is where TSM would step in. It not only creates the controller, but it also attaches the strings to the proper points in the frame. In order to attach the stings to the puppet you need to ensure that the puppet has the appropriate frame work underneath it. This framework of the puppet is the equivalent to the skeleton of your model. TSM provides prefabricated parts you can insert into a model to help save you some additional time, and help it know where to attach the "strings" (constraints). Now you'll notice no where in this process did we talk about how the joints are linked together for bending. On a marionette you could use swivel joints, cloth, yarn or several other substance and on some puppets put a piece of clothing over the joint to help hide the joint and alow for a more natural looking bend. This processes, for the most part, is the equivalent of Smart skins, Cogs/Fan Bones or Weights on your characters joints. TSM can't tell how every model is suppose to deform around every joint. So this aspect of setup has been left to the users.

 

I hope this description helped.

 

As for the centipede, TSM is modular, so you can add as many legs as you like; however, the linking of leg pairs is not something TSM currently can do as it is an aspect of rigging a character that will differ from person to person on how they would want to handle the problem. Do they divide up the pairs in increments of 2, 3 or 4+? Also what if the user wants to later pull one of the leg pairs out of the setup to animate them separately for something like a dance, or to temporarily work as hands? These are questions that go beyond the limits of what TSM presently can handle. Some of these might be able to be addressed in a future release, but we'll have to see.

[input /output error]

Hmm... sounds like a bad CD. Contact steve@anzovin.com and he'll help you resolve the problem.

[Facial animation musings.]

From: "Paul V Harris" <paul1@scottishtales.com> paul1@scottishtales.com

Date: 2004-3-29 08:17:46

 

You know what I'd like (dream of) is a scripting language that allows you to

place targets in a chor, then tell character "a" to go from marker a to

marker b using action x, taking y frames to do it. With a face pose why not

the same:

 

1: pose a - hold for 10 frames

2: pose b - hold for 10 frames - play sound file a.wav

etc.

 

Yours Aye.

Paul V Harris FSA Scot

Scotia Publishing, Limited

Highlands Rising Comics

St. Andrew's Society of Schenectady

<a href='http://www.scotiapublishing.com'>http://www.scotiapublishing.com

> From: KenH : Ken Heslip :

>

> I've been dreaming up different ways in which I'd like to

> animate the CPs on a face if I could implement it in a software

> package. The best one IMO is the following:

>

ming.

>

> Opinions and other suggestions welcome.

.

Dope sheets. They already exist, or existed (Haven't used the old style dope sheets in a while). The new dopesheet system for lipsync using the dictionary files could also do something like this, but you would have to append the dictionary file to contain poses situations that match specific words or letters. I've not delved into modifying the dictionary file, but I'm pretty sure this could do what your looking for; however, I will say that this is going to lead to a much more mechanical animated look. Trust your eyes not the numbers, as animation is an art. It's very much like playing music. A computer can play a piece of music with perfect tempo if you enter it note by note, but it will lack the humanistic qualities that make music worth listening to.