robcat2075

Playing with the concept a bit more...

A patch is like a surface stretched between *two* pairs of rails like in largento's right-side drawing. That's the situation where the math can be made to work predictably and quickly. A three point patch is really like a four point patch where two opposite rails happen to converge at the same point. Five point patches are a special Hash-only phenomenon. NURBS can't do it. It was regarded as impossible, sort of like finding a new whole number between 1 and 2 is impossible. Martin actually got written up in a math journal for inventing the thing. A five point patch is really subdivided into four-point patches and a center "CP" is manufactured out of nowhere to hold it together. I suppose they could make any number of sides subdivide themselves, but there is no mesh situation that requires a more than 5 sided patch and you're better off subdividing your n-sided shape yourself so you control where the curves are.

Red left is "standard" and most glasses will be built that way now. I've read one article that said more people "get it" if the process is done for blue left, but I don't know if that is true or why it would be except that it's a brain thing.

Scotch Tape was actually named that because William McKnight, one of the leading figures in the 3M company, was of Scottish descent. Are we off topic yet?

If you have your red-blue glasses handy here's a Stereoscopic version of that test with special 3D motion Hint: Full screen it. AnkylTestAnaglyph.mov

The odd thing about European monarchies, to me, is that no one seems to be from the country they are ruling. The British royal family is really German (George I was recruited from Germany when the English-Scottish line ran out), The Swedish and Spanish Royal families are French-Austrian, in Norway they are Danish...

You can scale the camera manually on X and it does flip the image, oddly enough! Try Scale constraining it to a null and Scaling the null by -100% on X. But if your scene is long to render it's easier to put your original footage on a roto and reverse scale the roto and render that. Likewise you can key frame the displayed frame of image sequences to make them run backward.

You won't need to create buffers for something that doesn't have buffers. But I haven't been able to get A:M Composite to work with anything but EXR files, so if I want to include some imagery in a composite project that didn't originate as an EXR (like a photo I took with my digital camera), I need to convert it into EXR first. Or is there a way to get Composite to accept a JPG or TGA?

not much. There are a few interface glitches to avoid. Right mouse button isn't a mac thing but you can plug in a two button mouse and do it i believe. TAoA:M explains the mac equivalent for one button mouse.

Those are not hard. You know how to lay down splines now, right? Turn on Snap to Grid and you can lay down something like this: If the sides need to be super flat you can turn on "Show bias Handles" to tweak them. There's a brief vid in the tutorial link in my signature called "think Ahead" that shows how to build bevels into your cross sections before you lathe or extrude. Also there are quite few primitives in the data folder you can load to use or study.

Yes. It works with EXR images and EXR image sequences. You can load non-EXR images/sequences in to A:M and re "Save As" into the EXR format. EXR has different compression schemes, each with a different purpose. Read about them on pg. 10 of the EXR tech ref definitely read more about EXR at Open EXR.org

Did you DO the modeling exercises in The Art of A:M? Don't skip them. They are there to teach fundamental modeling techniques. #9 teaches lathing which is key to any primitive with a round cross section. #10 teaches extrude which covers the other shapes. What shape are you not sure how to make?

That's golden info Detbear! I wonder, in them 'big apps', when you render a shadow pass do you get the shadows that happen on your objects as well??? That's the way EXR shadow buffers work. Shadows are grouped by light rather than by object. So all the shadows cast by one light are rendered in one image no matter what object they fell on. However, that means you don't get any shadows that are behind/under an object. For those, use the shadow method detbear describes.

That turned out good. I'm glad you got that working.

This won't solve Rusty's syrup problem but this is a test I had wondered about previously. It's animated splines, not a simulation. syrupBH264.mov

McKnuckles. The least popular of the McDonalds Happy Meals.

btw, there really are such cars... I saw one at a show here in Dallas in the 90's, and everyone who came up to it had a look of total bewilderment and disbelief. "Is that really... a car?"

Smartskin would do it best. Do one smartskin key at that angle that you feel is probably the maximum bend that arm is going to do. I think you'll just need to move the inside CPs out (away from the joint) a bit to eliminate the pinch appearance.

Here are some variations on the paint...

It's a BRAND... NEW*... DELAHAYE! DelRedMP4_1200.mov YouTube version of above: Another modeling and material experiment. I was going to call this the "one-hour car", but I spent three hours lighting it. * 1947 model year. Tax, title, and license extra. May not meet modern emission standards. Wide white sidewall bias-ply tires are for display only and may not be roadworthy above 25 mph. Passenger compartment is an extra cost option installed by the dealer. Dealers may not be in your area as the Delahaye company ceased to exist in 1954. Seat belts not available. Always park on a hill so you can start the engine by engaging the clutch as the car rolls down the incline. Never park on a hill as the safety brake is the object of a recall. Interior of car may contain numerous clowns who will jump out at you from the trunk. Open with caution. Always drive carefully. Do not drink and drive. If you need a drink after driving your 1947 Delahaye, a bottle of Mad Dog 20/20 may be concealed in the "glove compartment".

Basically yes. How you tell your software to do that, or if it is even allows this, and whether it expects you to give it pre-squeezed footage or if it wants square pixel footage and will squeeze it for you... that's where you hope the translation of the manual from German to Korean to Spanish to English went real well.

Looks like a great start and I'm glad you got rid of that mole!

Remember that all modern HD formats are square pixel formats. 1280/720 equals 16/9 In other words, a 1280 by 730 pixel screen fits a 16:9 frame exactly 1 is the correct aspect ratio to put in the A:M render options. Now, cramming a widescreen movie onto a standard DVD... that's a different issue. DVDs were made to contain 4:3 Images. Digital NTSC has 720x480 pixels which is not 4:3 if they are square. Those pixels are really about 0.89 aspect ration which skinnies up the 720x480 image to fit in the 4:3 television screen. Widescreen DVDs (not blu-ray) have to use those same 720x480 pixels to store their 16:9 image. They do that by telling the widescreen display to fatten them out to about a 1.18 aspect ratio. that stretches the 720x480 image to fit a 16:9 frame.

In order to put my CG animation into the NLE where I worked I had to compress the brightness level to fit into levels that are specified by something called CCIR 601. Basically, TV had to use the top and bottom of its potential signal level for things besides showing full white and black. So white on your TV happens from a signal that is less than full on and black from a signal that is less than full off. The more than white and less than black parts of the signal are reserved to flag things like frame synch . Digital TV imitated this and restricted picture levels the a range of 16 to 235 within the 255 levels of 8-bit imagery. Even within that range there are color saturations that are not allowed. A red of 235,16,16 would be too red. A:M has a "video Safe " post effect to limit color, but it doesn't limit brightness. I did both operations in After Effects instead. It may be that your DVD burner expects only footage from video cameras which automatically restrict themselves. I'm not sure that this is your problem, but if your whites and reds are over bright, that's an indicator. I wonder how one might do this compression in A:M alone? hmmm....

One thing to check is what res does your DVD burning software regard as "normal". Most will take almost anything and scale it since they are re-encoding it anyway. But scaling is time consuming.