Rodney

You know... immediately after seeing that sequence I thought you'd pulled a fast one like that but it was so quick and clean that that I didn't even want to go back to look and find out. To me that was a slick movie... like in magic... where you are vastly interested in the process but don't realy want to know because it'll lessen the joy of it. I was trying to figure out how you could get away without showing legs in the 'exposure' but you probably did about the best that can be done. I guess what you said about the bed (and leg room) really sunk in. Nicely done Mark.

Well, don't give up yet. You've still got 15 days to go!

The obvious answer is yes, but the real question is how to get from here to there. My primary thought would be to further reduce variation. There are two areas of concern with regard to a render: - What the renderer performs. - What the user sets. A problem to consider is that all renders are not created equal. An optimized render for one case might not be optimized globally for all cases. The use case model suggests that optimization be made for the most likely case first with consideration for other likely cases. Those cases that are less likely or not likely in any case would be set aside, bypassed or jettisoned. Disclaimer: I am not a programmer either.

Very nice. I had missed this when originally posted so glad you post those updates in your main topic (Thanks!) When I grow up I want to model like this.

Looking good! Love the style.

OOoo.... I like that. Nitpick: I wish the brush looked just a little more brush-like but I can see how difficult that is with the text overlaid. (I assume the pivot in the middle of the image is an artifact from the screencap and isn't suppose to be there) Nicely done!

Try saving the Model out as an external model and then try 'Ctrl-S' again. It should save the Project (thereby saving the Model) without the dialogue property coming up. Keep in mind that if you have your Model as part of the Project (and not as an external file) what you really are trying to do is save the Project. So save the Model manually first then A:M will know you want it update it/save it separately each time. The current version does have an Auto Save plugin but I don't use it enough to comment on it. From what I've seen it works well enough.

None that are specifically A:M-related but basically any 'behind-the-scenes' presentation these days will contain elements of matte painting, compositing and related themes. The older ones to look for are mostly from Disney who pioneered the process on the one hand and Ray Harryhausen as the pioneer of miniatures. The entire history of CG is basically the quest to juxtapose artificially created images in with those that are real with the goal to have the audience not perceive (or care about) the difference. There is so much information out there it helps to have a target to shoot for and perhaps a list of expected obstacles or limitations. Since we are talking in terms of compositing it'd be good to consider rendering in passes as well. For instance, if you have a long shot where the camera begins at a close-up but then pushes back and out and up leaving the characters and presenting a view of the surrounding mountains what you'll need is obviously more than a 2D matte but... you can model the closeup and the environment separately and then composite them together separately. In this way you can concentrate on what is important in each area without losing the focus the entire scene needs. For rendering purposes you might simply duplicate those Chors and never again (until compositing) shall the twain meet. Your question was a simple one and I've run off on another tangent. I suppose the real answer is "No, at the moment I cannot think of any particularly useful links."

When troubleshooting its always good to eliminate unnecessary variables and this represents one. It would be better to start with a new Chor and then delete everything anew rather than modify an old Chor that might have some setting you don't want. If you've got Final Cut Pro then you are well into a very effective compositing workflow so you may want to stick with that. As I mentioned before, if wanting to use A:M for straight up compositing and post effects there is also A:M Composite. Robert does a nice demo of how to use A:M Composite with FakeAO in this video tutorial. The primary Post Effect I use when composting 2 or more images is 'Mix'. We can get some nice transitional fades by animating the percentages between the images. One of the benefits of using A:M Composite is that we can use the resulting images created with it 'live' in A:M. (i.e. save or render out a complex composite and see that automatically update/effect a scene or character in A:M). Granted, this level of use would require much experimentation and at this point is far beyond your current need in compositing.

Sometimes we can tell a lot by what we don't see... Not that it matters (esp. if there are no lights in the Chor) but I do note that both Layers are set to receive shadows. It also appears that at one time the Camera might have been set to save out Light buffers. Not really a guess at the problem but more of a ruling out for myself... if a light Buffer'd image was somehow thrown into the midst of the sequence... that might account for the odd frames.

Can you share the Project file (without any of the images/models/assets should be fine). That alone might be able to shed some light on the subject.

We will assume you deleted the default lights in the Chor then manually. When performing a straight composite in a Chor it is generally better to use Rotoscopes than Layers. Layers work better for when you are moving images around, scaling them, etc. If your clouds are all of such then I see why you used Layers but generally I would still use a Rotoscope for the overlay in the front (Just make sure to set the setting 'On Top'. If there is something off in your original sequence then you have found your solution. You have to scrub through your sequence and remove/replace the frames that do not belong. Not sure how they got there but I postulated as to one reason they might be there in my last post. If for instance you rendered a 24 frame sequence with shadows on and then rendered again with shadows off but this time didn't render all 24 frames... well... that'd result in any of those frames not rerendered still having the shadow. Without having access to the project itself there is a lot of guesswork here. From your images though it sure looks to me like that one frame has a shadow and for that to happen there would need to be a light shining at some point. Other objects in the scene (the bench for instance) are also casting shadow so there was/is light there at some point. As this matches with why you wouldn't be getting this same effect in Final Cut Pro (i.e. in Final Cut Pro you wouldn't have any of A:M's lights) my diagnosis is still that you are casting unwanted shadows with lights. BUT... the most important thing is to make sure the sequences you are using are rendered numbered right. You might also try to render with Multipass set to OFF. (with no multipass and no blur A:M will not attempt to blur between frames) I need to reread your descriptions... Do I understand that you are only using images in A:M or do you have your models in front of the layers and background?

There is a lot we don't know about your setup but if you set your background plane to receive no shadows that should keep the shadow from appearing. I'm not sure why it would appear on only those frames except to say that perhaps you rerendered some frames or terminated a render before it re-rendered the whole sequence. You should also note that if compositing your images within the Chor the lighting will cause cutouts to cast shadows by default. One way to get past this is to drag/drop your lights onto specific models thereby creating a Light List for which objects will be lit and shadowed. Another way is to remove the lights from the Chor which you are using for compositing. To bypass all of the lighting issues you might also use A:M Composite via the PWS, which does take a little experimentation to learn.

If you want to select every CP you can use the comma key to select everything on that particular spline or the slash key "/" to select every spline/CP connected to that spline. I suppose a lot depends on what you mean by 'every'. *** I note that you mentioned before that you had peaked all of your splines. Note that you'll want to unpeak those splines (shortcut key O) or else you wont see all of the desired changes in the Bias for those splines. ***

There use to be a plugin called 'Set Bias' that filled in the gap where it comes to adjusting Bias In/Out settings simultaneously. Unfortunately, that plugin is no longer supported nor available.

I'm not sure about copying per se but... there is a quick way to duplicate Bias values (i.e. duplicating a Bias Handle's settings across multiple splines/CPs). Select the splines/CPs you want to set making sure the last one you select is the one whose settings you want to duplicate on all of the rest. Nudge the Bias Handle back and forth (or adjust the setting in the properties panel up/down) and all of the others will be reset. Almost the same thing would be to select the group of CPs/splines you wish to set parameters on and then adjust the Bias Handles or Properties (or via the Manipulator Properties window if you've toggle that icon on). All of the settings should take on the same values. Note that this same methodology can be used to Translate... moving whatever is selected around simultaneously and... bringing them closer together or further apart when using the Scale settings.

Happy Birthday Patrick! I always have enjoyed your comic book art!

All the best to you Jeff! Happy Birthday!

Top notch Mark. I love the style and detail... lighting and texturing too. That flower should be great fun. He has great character already and we haven't even seen him move. I can already envision it trying to attack our heroes. Ooooo... that'd be good. That'd be a great progression from the old 'Flower Power' tute!

Try Keekat... I seem to recall he has a Blink pose slider.

Drilling down to the splines in the PWS is not so much for selecting them but for applying things to them... such as the old style fur/hair. In that case we drag/drop the hair material onto the spline in the PWS.

That's one of the great things about this forum. Whereas one person... perhaps even many or most... may forget, it is a very rare occasion indeed when everyone forgets.

I'll add more fodder to the fire and suggest this also... One reason you might want to use Models dragged into Models over other methodologies (which are generally preferred) is that applying Model to Model will only replicate the geometry whereas in an Action or Chor you'll also be replicating Constraints and Bones. Therefore... a very quick way to create complex geometry/meshes out of simple geometry... without any Bones or Constraints... would be to continually drag/drop premade models from your Library (or the PWS) into another new/empty/ready-and-waiting Model. Disclaimer: As this is not an optimum method for modeling only basic/simple models should be used. Dragging/dropping complex Models into other Models will have you staring at the Desktop wondering what happened. (If the method was used more often it might be worthy of a report)

That was a lot of text to lay down so let me summarize: The downside to dragging/dropping a Model into another Model is that the location/orientation is controlled by the original Model whose origin is located at coordinates 0,0,0. (Example: Dragging/Dropping the same Model into a new Model multiple times will result in that same Model being created multiple times in the same location.) The benefit to dragging/dropping Models into an Action or Chor is that the location can be more finely controlled. For instance, a Model dragged/dropped into an Action will be located at the mouse's cursor. A difference (i.e. further benefit) to be found in dragging/dropping Models into a Chor is that raytracing can be used to drag/drop Models onto the surface of other Models. Note that this is the default behavior of drag/drop in a Chor and is also what allows us to drag/drop characters/objects onto paths in order for those paths to automatically direct a motion.

There are a lot of things you can do here but... there is good reason to collect your models in a Chor versus in a Model. One is that you want to maintain Bones and such in their correct locations with the proper naming. If you feel you must drag and drop a Model into another Model you can do that (at least you can in current version) via the Project Workspace and you should be able to in a Modeling window in current versions also. As a test here's what you might be able to do in earlier versions if you can't otherwise copy/paste or drag/drap from Model window to Model window: Create 2 (empty) Models. In the first, lathe a simple shape Now in the PWS tree drag the first Model into the second Model. Drag it a few more times. Opening and inspecting the second Model will reveal the first Model has been copied several times into the second. The down side of this method is that the model will be placed at the 0,0,0 coordinate so after dragging and dropping you'll need to go into the second model and move things around appropriately. (Note: There is probably a method to offset where the model gets placed but I haven't experimented with draggin/dropping back and forth between modeling windows enough to know. There have also been some copy/paste methodologies that have been updated in recent releases as well that I haven't explored) A better place to combine Models would be in an Action or a Choreography. In an Action the subsequent Models that are dragged/dropped in maintain become instance of the first. The model will be located at the mouse's position upon dragging/dropping. I find this methodology particularly useful for dragging/dropping in Lights. The benefit to using an Action is that you can then subsequently drag/drop the Action onto a Model in the Chor and it will magically add these Action Objects to any Model you apply the Action to. Doing something similar as the above in a Chor is even more powerful because the Chor was specificially designed as a collection place for multiple models, constraints and actions. By dragging and dropping everything into a Chor and then Exporting out as a new Model you can quickly create very complex models with Bones, constraints and... well... anything that you can drag and drop into a Chor or onto the shortcuts within the Chor. Note that in order to open a model (say from a Library) it doesn't have to be dragged/dropped anywhere. It can also be double clicked. Similarly, a model that is on your desktop (outside of A:M) can simply be dragged/dropped into a window in A:M. The Libraries primary give the user a means to preview the Model before opening. Those with good memories and organizational skills may find the operating system is 'library' enough for their purposes. One of the benefits of A:M is that there are many ways that we can establish workflow. (i.e. workflow can be highly personalized) The hierarchical and compartmental nature of A:M helps us create, leverage fully optimize that workflow. While there is considerable benefit to be found in the tried and true methodologies it can be useful to occasionally experiment and explore new/other methods too. The underlying principle that drives drag/drop that is fundamental to A:M is resource reusage. We can leverage this principle best when we establish shortcuts that point back to the original resource. Then when we update the original we automagically update the shortcut. When we use an Action or Chor as a means to build/modify we gain considerable advantage over dealing with only the original sources. After all, there is a reason we call these assets... re-sources.