[Steve Bentley]

Ok thanks. I only get that listing of the thread once I’ve clicked on reply, but I do see the edit button now.
Cheers!

[Steve Bentley]

Yes and no.
Projectors are always one of the cardinal formats (HD, SD or SD widescreen, 4k or if you are really lucky 8k or 16K (imax). But how you send images to those projectors can vary greatly. Some projectors will accept a format other than what they shoot out (an HD projector will sometimes accept a 4k signal but only show a scaled down HD number of pixels). Some of the 4K projectors have a scaler built in and can scale HD input up to 4k (and do it surprisingly well). Usually the aspect ratio of the footage input must match one of the projector’s output formats

If the screens are blended or overlapped or in some matrixed configuration often a larger image will be divided up by some fancy software or black box and then distributed to each projector so each projector gets it’s chunk of the overall image. Whats important in this instance is not what the final number of pixels on screen will be, but what the black box can handle pixel-wise and how it distributes the image.

You can also do the latter scenario with multiple synced media players and manually divide up your large content into specific movies, one for each projector and feed each projector separately or each projectors media player separately.

More often than an LED set up, a projector array will often blow up the feed simply because no one wants to go to the trouble of making massive content (I’m with Walter, a 20k pixel wide movie is pretty normal at our shop too), or solving the tech or cost problems of sending massive content to multiple projectors in sync.
LEDs tend to have the matrixer or multiplexer as a piece of hardware that comes with the screen and accepts a standard resolution and outputs resolution specific to that screen. Projectors aren’t that unique.

[Steve Bentley]

Have you tried putting your “alpha” for the smoke in the transparency channel of the texture? (usually you have to invert it though – white is “clear” in the trani channel.)

[Steve Bentley]

Hey Jim, How did you do that edit to your post? I’ve always wondered why in a forum such as this I couldn’t fix a boneheaded error of my making (so ya I end up wondering that a lot!)

[Steve Bentley]

I ask this question when I have a layer’s attributes open and there are lots of visible keyframes to snap to, so it feels like its not snapping to what you expect but is in fact snapping to something.
Possible?

[Steve Bentley]

I’m not sure what you are after here. There is no animation (except the motors) and because of the dynamics on everything I assume you are using gravity alone to keep the car on the ground.
Interestingly if you turn the project gravity off it floats off the plane and tumbles. That might be the motor “torque” but I didn’t know it would do that.
I have a feeling the drift you are seeing is the “noise” that is in any dynamics system. Its what keeps the objects moving along and keeps them from running into each other and sticking. So over time that noise could be a butterfly effect that sends the car off course.
Is there a reason to use motors instead of keyframes on the position? Using keyframes would eliminate the noise problem and you might not even need the dynamics. But the dynamics might be the point of it all (exhibiting Newtons laws of motion for instance)

[Steve Bentley]

You haven’t mentioned the reason for no green or blue.
Red is actually the best color to use but as soon as you get humans in there, not so much. (Vulcans or Andorians or Orions -Orionians? are good to go though).
There are lots of ways to get around the no green or blue problem. Give us some more info.
If all the characters are stop motion there’s no reason that you can’t use a false color for their “skin” that would key on red and then color correct it back to flesh in post.
Screen colors: pink, cyan and yellow can work but aren’t ideal because those colors of light are made of pairs out of the red, green, blue triumvirate. So its not really one unique color you are dealing with anymore.

[Steve Bentley]

The first thing is that sooooo many media players can’t take content larger than HD or 4K and this spec is way above both. So while the screen may in fact be that resolution, doing the spec’d resolution might be overkill.

Second, LED’s, especially rental ones on stage are still fairly low resolution (6mm pitch or so). So if you do the math that comes out to 150 feet or so long. Does that seem like a reasonable length – have you seen the screen? Because an audience is so far back they often don’t have to pop for a screen with anything higher than 6mm (or even 8). But even with a “high” resolution screen at 2.5mm pitch, that’s still over 50 feet long. Again does that seem right? I’m just trying to see if they are giving you correct information.

It could also be that their media player takes the spec you suggest and scales it down to fit the screen.

It could also be that from your 6k file they will cut off sections and make separate movies for individual media players or feed the whole thing into something like Green Hippo or Watchout which will cut the section up for each screen. But I’m pretty sure that even these two programs/hardware combos can’t take larger single content than 4k.

Anyway that’s my skeptic speaking.

As for size of content, AE can handle that size with ease. You can work with proxies to speed the work flow and you can watch the content play at 1/2 or 1/4 rez or make smaller temp movies to check the animation.

LEDs can indeed take pixel for pixel content and often have to because they are rarely a standard aspect ratio (especially convention rental configurations). In fact its often better to give a larger resolution so the matrixer can super sample the content to reduce ringing between pixels once it gets to the screens “dots”. But its more common to feed a standard format (4k) into the screen and have it blown up; that content will often have large areas of black where the odd screen aspect doesn’t match the 4k format (so the screen uses the upper half of the 4k frame and the rest is black)
If the screen is in fact that size and that many pixels then the information they have given you is correct and very doable in AE. Your limitation is your desktop screen size. If its going to be an ongoing thing for you it might be worth your while to invest in a multiple-head-to go from matrox or one of Radeons Fire card arrays that will let you do multiple screens (think: 6 screens horizontally) with a massive amount desktop realestate. Otherwise its hard to see the content fully AND at 100%

[Steve Bentley]

The image you sent is obviously all 3D, so I ‘m not sure what needs tracking. Or is the 3D flyover from another project that you don’t have the c4d files for?
With all the tracking points from boujou you should be able to tell where the roads are in 3D space. You can make splines that run down the lanes of the roads and “attach” the cars to those splines and make them point the right way and follow the curves with Jim’s suggestion. The splines will also help you tell whether you got your tracking solve correct.
Because the original animator would have only made this one interchange, once you set up the overhead tracked shot with splines on all the roads, those splines will also be in the right position for all the other shots too, you just have to find the right camera position. You should be able to put all your different camera solves with markers into one project (with each set of tracker solves and markers made children of a unique null) and then move the parent nulls around until all the markers line up in 3D space. Even though you have many different camera angles you are still tracking around one interchange, one common 3D environment and identical geometry.

[Steve Bentley]

My answer falls under the topic of “when you have a hammer everything looks like a nail”.
I had been doing range mappers in another software package when I answered your post and totally forgot about the range expression.
Here’s an answer from Jay Roth that I took from a post a few hours after yours. (saves me typing)

I think you’ll want the linear expression hooked up to a expression slider on a null.

The linear expression works by taking one set of values doing the math to interpolate them to another set of values. So, you would want to put this expression on the range selector of each of your text layers

range = thisComp.layer(“YOUR NULL”).effect(“Slider Control”)(“Slider”);
linear(range, 0, 10, 0, 100); // Change the first two numbers to be the keyframed values you want for the slider for each layer

Essentially what this is telling the range selector is “As the null slider goes from 0 to 10, change the range selector value from 0 to 100.