[Steve Bentley]

Can you upload the file and we’ll take a look? What version of AE and Duik are you using?

[Steve Bentley]

Its all just multiple elements being darkened or fading out at slightly different times – and all part of a pullout camera track, each element having it’s own alpha or using silhouette as an alpha to reveal the next scene, and with depth of field to being used to draw the eye. I think this looks more complicated than it is because of the liveaction elements used.
While the elements could have been all shot with a dolly out and similar timing,I think it would be easier to just shoot them at a higher rez than needed so you have the source pixels to work with and don’t have to soften the image when they scale up (in the end zooming is just scaling when are working in post; from a pixel perspective). You can scale footage up to about 130% before it really starts to degrade – depending on your compositing software.

[Steve Bentley]

You say you have stabilized your face, that’s correct (you might also need to stab the rotation too), but have you also tracked the footage you want to comp onto? Even a “locked off” shot can still move a little, more so if it’s film-based from gate weave.
You didn’t mention which tutorial you followed. Without knowing what you have done we won’t be able to steer you in the right direction

[Steve Bentley]

That’s the pure RGB you are seeing. If you take the PNG or the movie into AE you don’t get that (assuming you interpret for an alpha).
Perhaps what ever you are looking at it in can’t handle an alpha or you just opened it in pshop or media player. With these you will only ever see the RGB on its own, but in pshop you can apply the alpha a a mask to produce the final result.
You are doing everything right, it might just be how and where you are viewing the render.

I would not complain if someone gave me this element to work with in AE. It’s alpha is correct for the look you are after and composites perfectly.

[Steve Bentley]

Ya the killer renders times are the volumetric shader.
But maybe you don’t need as many particles or vertexes to work with. There is a huge amount of overlap going on.

The file you sent had the matrix generating the matrix only and that produced no render (for me at least). I set the matrix to produce thinking particles, deleted the matrix’s material and then linked the particle geometry and the matrix attribute to the same TP group.
That worked but again slow.
BUT
I’m not sure you need true volume from each particle. The sheer number of vertexes in the object is the volume you are really interested in (I think). What about using mograph to map a disk or sphere to each vertex of your object, then putting a circular grad in the transparency or alpha channel on each disk iteration with a base full blast luma channel so that each disk glows with a fall off, and have each disk point at camera. This would give you the puffs that would build up the density and would cut the render time down significantly because no pyrocluster is needed. You could also have an effector make the disks toward the center of the object scale larger than the ones at the edge to increase the “volume” density at the core of the object. You could then glow or blur the whole thing in post to “blob” them together if you are still getting gaps. The disks could overlap their neighbors by a huge amount to smooth things out. You may get some intersection artifacts from this. There is a way to fix this too.

All the math that’s going on for pyro is to get the volume tracer figuring out where the edge of each puff’s envelope is, and that’s proc intensive. But is that needed?

I’m sure your tube is just to test with but you should make it editable so you get the right number of vertexes you are expecting from the matrix or the cloner. Otherwise you get the default number from the “render perfect” version of the primitive.

Like this?:

12537_12534testwithdisks.c4d.zip

[Steve Bentley]

Those are ratios based on the y dim being set to 1 for easy math, so – the x dim:y dim(set to 1). So just divide the x pixels by the first number to get the Y pixels.
So 1920/2.35
Remember though that some formats. (like 2.35) often include a letter box. So while the image area might be 1920 x 820, the frame size may still be 1920 x 1080 with the area beyond. The image area just being black horizontal strips.

[Steve Bentley]

There is a post about this on this forum. I was part of the thread if that helps narrow the search.
The trick is to make the headlamp geometry not visible to the light that shoots into the world. (make sure to exclude the geometry in the project tab of the light). Then add a light or a glow effect that makes the headlight len s light up properly for you. That light should have all other geometry in the scene excluded except the lens geometry.
Often you will need a spot light to throw the light in a beam and a point light with a fall off to light up what’s near the light.
Don’t rely too much on the GI. Get the lighting right first then you either render really fast without GI or render with GI and everything looks even better.
Some 3D lights start at a infinitely small point so often they need to be placed behind the light fixture so the cone has some diameter as it exits the lens. Just exclude the fixture geomtry in the lights project tab.

[Steve Bentley]

The creator probably just animated the scene normally and then rendered with a higher frame rate. When played back at a slower frame rate it gives the slow motion effect.

[Steve Bentley]

I see whats happening.
First, your alphas look good.

I assume the problem is when the light turns off, you see near-black letters when you don’t want to see anything. Or you see a big square of black when the letters turn off.

There’s two ways to solve this. Have your friend use a Screen function when he supers this footage over his own. Light is additive so its an appropriate way to add a neon sign to a shot. When the light goes out, even though the black letters remain and are part of the alpha it will render properly due the Screen function not letting the black composite on top of the background shot. (“screen” lets pixels of the supered layer “add” to the brightness of the background layer – that is a gross simplification of what the screen function does but its close enough for this). So black pixels won’t add any brightness. You could also use “Add” or “Lighter Color” or “Lighten”, but for glowy things, “Screen” is usually the best.
If you use the “screen” function you don’t even need the matte – rendering your glowing letters element on black with no matte and you will get the same result.
Your “off” letters are slightly pink, just a bit brighter than black so they will still show up a little with a screen function even when the glow is turned off.

The second way is to turn off not just the glow (when the sign is not lit) but also turn off the layer the glow is generated from (the text) during the off-blink, so nothing is left in the alpha for those frames. Then using a “normal” blend mode, the alpha will control what gets composited on the background. If you look at the alpha with this method you will see white glowing letters and then nothing (all black) during the off-blink. Think of the alpha as a mask – the white parts let the RGB channels pass through, the black parts let nothing through.

But I think you would still want to see the structure of the light when the sign is off and that’s probably what you were going for with the dark but not black “off” letters. When a neon tube isn’t glowing you can still see the tube and the holders for the tube and the metal base that is a backer for the light or the structure of the channel letters, as they are called. At night this isn’t as important but can add some good realism. If there are no other lights in the scene to light the underlying structure it’s acceptable to just “turn the neon light off” and it’s a good enough illusion.
For extra realism don’t forget to have other objects close to the sign in the background footage get some of the light that would be coming from the lit letters. This can be done with simple masks. It doesn’t have to be much but it really grounds the sign in the shot. The flashlight that lights a page of a book you are reading is also bouncing off the page and lighting up the room too.
I think what you have here is correct, but perhaps not what you were expecting.

The only other comment is that neon or nighttime signage tends to blow out and over exposes at the core of the text. Like a lightsaber, the center tends towards white or a hot version of the actual color. Go watch Blade Runner (the original). This sign doesn’t really look like its glowing because its not “hot” enough. You can also try duplicating the sign layer and screening it again over its twin and the background – this will keep adding to the glow and core brightness. Also try Hardlight and Vividlight blend functions as they might be better to kick the brightness up. You can also add a Glow effect to the sign layer – but I don’t like the Glow effect’s roll-off and prefer just adding multiple copies with a blend function that works best.

Keep in mind that if you deliver this element in a compressed format – like H.264 – the alpha will be thrown away.

Let me know if I still haven’t hit the heart of the problem.

[Steve Bentley]

When you say “flip the other direction, do you mean reverse the normals? If so, you can do that. In face mode, just select all the ones you want to “flip” and in the Mesh menu select normals/ reverse normals.