There's a document on the NVidia site relating to Mip maps -for shaders called let's get small. There was one item mentioned that caught my eye that might be of possible interest in 2D in some cases.
It briefly mentions the concept of stage makeup and how the concept can relate to 3D -unnaturally accenting certain features so that even though it looks unnatural closeup at a distance it has the effect of defining the features better.
Here's some excerpts and I'll try to add the image example.
--------
Stage makeup can accentuate symbolically in place of scientific realism, either overtly or subtly
Stage makeup looks un-natural at “normal”distances, but reads very well when scaled-down (seen from a distance)Accentuates the key expressive elements and geometry of the face.
-------
I mention it as an idea here since I've heard people complain about loss of detail at distances. Maybe if you need to improve visibility of face or other items at distances a layer of accents -like stage makeup- can be drawn in and faded as you move the object forward. The key seems to exagerate certain elements that humans normally cue into to identify a face or other object. There is more info on the stage makeup site listed in that image about the concepts if anyone is interested; I find that if you understand the underlying concepts of a thing it can often be used in things that don't seem to be related so hopefully this will prove of interest.
Ah, the classic stage technique. The general term, familiar as all hell, is "chiaroscuro" (italian then translates to "lightdark"). I imagine it's derived from old opera makeup, although it might go back before the form existed. It's used for more than just stage. A truly competent beauty makeup will use it to very subtly de-accentuate any percieved flaws in the facial structure. I ain't that great at that part, and don't really enjoy it that much (gimme a big slimy monster any time), but I've gotten many the hug from actresses on shoots from making them look, in their words, "better than they do".
I imagine, as it directly relates to that image, that you could set up your highlight and shading as Styles and adjust them to the shot. If you draw your images with that sort of style (I'm thinking some elements of the old Max Fleisher "Superman") it could be a great wat to make sure that expressions, etc, read properly. From my backgrtound, I really should have thought of that. Good call, Burt, and I think this should really be in the "Tips" section.
Thanks Brian. I really enjoy the forum here as it seems there are people like you with such varied experience. I'm going to play with an example later if I get some free time. Maybe LM can move this if appropriate.
But aren´t you assuming that the character/subject is going to be very small and stay very small in the animation?
The larger the character gets, the worse it will look - like the large photo.
My 2 cents- better to draw it normally (it would be very hard to learn that stage makeup technique for creating one´s drawings), make your Moho animation at a larger than final version size, and then take the files (image files preferably) into a video editing program such as Adobe After Effects or Premiere or other and reduce it to the size one wants. The final quality video in video editing programs is much better than what Moho does - even at smaller sizes.
Toontoonz wrote:But aren´t you assuming that the character/subject is going to be very small and stay very small in the animation?
I'm not sure how useful this is in practice in Moho, but you could always use a switch layer and switch between an exaggerated drawing and a normal one, depending on the view. (You could even use an embedded script to automatically switch between the two.)
Lost Marble wrote:but you could always use a switch layer and switch between an exaggerated drawing and a normal one, depending on the view. (You could even use an embedded script to automatically switch between the two.)
Potentially have to draw every character in every pose twice? No thanks.
I have a hard enough time drawing it once.
That's one of the reasons I suggested tying the exagerated shading to styles. For closeups, adjust your, say, three fill styles to have the same color. For longer shots, adjust your highlight and shading accordingly. Then it doesn't involve multiple drawings. I could be wrong, but it's been my impression that acces to Styles is a bit (completely?) spotty in the script interface. If that wasn't the case, making an embedded script that could control the shape shading, blur, all that cool stuff would be (hopefully) pretty damn easy to implement as an embedded scripts.
The basics of stage makeup aren't that time consuming to grasp. And I think some of the concepts can be translated to animation. The standard guide is Richard Corson's "Stage Makeup", expensive to buy for just a look but it is a very rare public library that doesn't have a copy. Pretty entertainingly written for The Standard Textbook. Also has a lot of great reference for makeup and hairstyles from bygone (or really ancient) days that can be handy if you wnat to make something that takes place in the past.
Theory vs. Reality.
I took that cutey above to test what in reality she would look like in a small Moho render.
Here it is two versions, both in a very small, 160 x 120 pixel frame.
The one on the left shows the woman at full height, the right half the size.
The result (and this is a poor, "30 quality" jpeg image - imagine how good one could see the facial lines at a better quality):
Here´s the 160 x 120 pixel movie: Theory vs. Reality 0.3Mb
(NOTE: When the zippy video website opens you will have to resize the video smaller for it to show in its correct size -
go to "Resize this Video" under the video window and then select "Smaller".)
My opinion: Still ugly (and the exaggerate lines as obvious) as heck. Great technique for the theater, but for animation or movie making????
Maybe it works better with drawings. I look forward to seeing everyone´s animations employing this idea.
Last edited by Toontoonz on Sat Jul 02, 2005 2:50 pm, edited 1 time in total.
I think the bggest problem with that example is that it is, frankly, a pretty piss-poor example of how a proper stage makeup should be executed. In a way, as it relates to the discussion at hand, the execution is determined my the size of the house and the distance of the stage to the audience (or the size of the image in this case). If I'd done that in high school I'd probably be pretty embarassed. It's anatomically shoddy, doesn't blend the highlight and shadow in a believable way, and I don't much like the choice of colors. Okay, maybe I'm being a bit harsh, but I'm in the biz.
This is not to say that executed properly some elements of the technique would be inapplicable to animation. It's all about the simulation of depth where it doesn't really exist, informed by the amount of detail the eye can percieve at a given size. A more representative example might be to show the larger picture with the "made up" face somewhat desaturated or with the contrast reduced. Because that's really the point, it's that you would exagerate some portions of the drawing only when it would be far fropm the "camera". You'd never want to see it like that up close, its bound to look like hell that way. As we can see. It's difficult to adjust this image as a particular example. I'll try to whip up something for a "proof of concept" to see if I'm actually yapping about something useful.
It would take a real good artist to pull this style technique off. (And, frankly, I have yet to run into the situation where I would need to do it. There are easier alternatives.)
For example, I think it is easier to draw whatever characters as one normally would and then after doing the bones animation in Moho take it into another video editing program to make the final video. From my limited experience, Moho starts to have quality problems with rendering anything smaller than about 640 x 480 pixels.
I look forward to seeing the results of your experiments so I can understand your ideas better.
