[Steve Bentley]

There’s a version of Unity that is free (and there may be one for Unreal as well) that you could do a file format coversion for and then render from there.
What’s the final usage?

[Steve Bentley]

Are you using 3D exchange to do the transfer? or the built in exporter?
I’m kinda shocked that there is not direct file format converter for C4D. All the other big boys are there (and quite a few small ones as well)

What about trying to do a file format conversion to something like 3DsMax. You could download a trial version to see how big that file is. From Max you might be able to save the file back as a 3Ds format (What Max used to be called) and then you can import that into C4D.

[Steve Bentley]

Ok. Thats a shame about polytrans – it covers so many formats I thought Iclone might be one (but its pretty new).

When you bake an animation in C4D, I’m pretty sure that the vertexes are each given their own animation paths. But I’ve never seen a cached file reach the mega giga proportions you are talking about.
Now to be fair we’ve never stripped out a c4d cache file to see whats in it (we will be now though just to see), so I can’t say for sure that’s whats going on, but given that you can scrub forwards and backwards on mesh hierarchies, deformers and dynamics, particles and cloners I figured that’s what they would have to be doing.

You could bring the mesh and the animated skeleton into C4d separately and then link them together but then you will have to redo all your weight painting and that really depends on the model of how laborious that would be. (unless there is a tool, built in or aftermarket, that lets you export the weights from IC.

Have you talked to iClone support?

We don’t use iClone – can you render out of iClone? could you get a third party render engine that would get you there without having to transfer the assets?
If you can’t render out of iClone what is the pipeline they expect you to use?

If you bring your model into C4d and do a quick rig and then save it, what’s the MB level then?

[Steve Bentley]

What about doing your main video the way you normally do and then put that video in Flash or an HTML wrapper and then you have the ability to overlay a counter (which are crazy easy in either platform).

[Steve Bentley]

Ok, at the risk of elevating the ire of the community and lengthening this topic….
I hadn’t actually watched that youtube tutorial in the initial post until now (since you mentioned it again).
Where do I start…. I haven’t yelled at TV for a while but that gave my lungs a good workout!

I never like being negative in these forums because it doesn’t help anyone, and because there’s never one right way to do something. And often, when investigating even dubious techniques, even us seasoned pros can learn a trick or two.

Buuuutttt…. There is just so much questionable information in that tutorial. So much so that the instructor doesn’t seem to know how to use AE in some areas. At one point he decides to use another layer as a matte. Since forever we have this thing called track matte that lets you pass the layer below through the layer above in either alpha or luma mode. While his technique is not wrong , it sure is a long way round a simple every day occurrence, and his technique is one we haven’t used since AE was called COSA. So that makes me wonder where the rest of the info has come from and his level of expertise.

Anyway getting beyond that little nit to pick.

The first issue could be classed a preference, but using full-red to check a key – any full chroma will overload your eyeballs and your monitor. You will not be able to see subtle issues in the matte edge with your cones screaming to turn down the volume and your monitor being pushed to the limits of its gammut. In addition, any two full-chroma primaries next to each other (red/Blue, Red/Green, Green/Blue) can vibrate in the eye (especially red/Blue – I can explain why if you like but its not important here)

His method for recoloring the screen to make it easier for Keylight to see…. Anytime you add a color correction or level trim or compression, or even change the color space, you are losing or transposing information. You can’t make more of what was there you can only crush it or invent it. Its like yelling at a person with a hearing aid. It kind of overloads a delicate system put in place to detect subtle things and you end up warping or replacing the information.
When we do film transfers we go to great lengths (enormous lengths, these companies spend month getting LUTS just right to tease the most out of the footage) to ensure that as much of the color of the screen is preserved, because even in a $1500 an hour grading suite you are loosing or shifting information and every time that happens the key gets worse.
There are subtle little roll offs between the green of the screen and the person in front of it. The keyers are designed to pick up those subtle differences and work in magical ways to determine what is screen and what is human; pushing and pulling on feathered edges and gentle blurs, even keeping but desaturating greenscreen pixels and letting them play as foreground pixels. By changing the color of the screen you are forcing colors into those subtle areas in ways that are more heavy handed than a keyer would and without control of that roll off. A keyer doesn’t care whether it sees a particular color, it only cares about the chroma separation between that specific color and things that are not. You can use keylight with green or blue or red, or beige or skybluepink if you want, but use it with the color that came on the virgin footage. You are also killing the grain structure when you stomp on the screen color – this is great for reducing the noise, but noise is detail.
Will his replace color technique produce a sharp key with less issues? You bet! Will it be the most accurate key and keep all the little feathery wispy details and show you his outline right to the last pixel of him? Not on your life. If you used that technique with something like water being poured into a glass (all on green screen) you would end up with a posterized mess.

Then there’s the whole matte math done with, among other transfer functions, the Lighten. Instead of describing how lighten works across a range of uses, you just have to use it for a while in compositing to realize it has a very narrow niche. Try using it instead of screen or overlay or a few of the other blend modes that are used daily. It creates harsh roll offs which look sort of posterized. When applied to mattes or the product of mattes its not going to be the best and will surely wreck those subtle rollsoffs the keyers were designed to deal with. Then he adds a blur to overcome the harshness all his monkeying added. You could have probably added the blur to the initial keylight matte and then crushed it and come out the same. But again, its not going to preserve all the little edges that make a comp look real vs pasted on top. If you have harshened an edge and then softened it, you have changed that edge irrevocably, an no CSI type software is going to be able to resolve the original details again. ( I have a friend in Forensics who says CSI is is favorite comedy show, I’m starting to feel the same about this youtube video).

I’m sure there are situations where some of those techniques might be useful but to put them forwards as “here’s how you pull a key”, is doing a disservice to all the things keying software can do when used properly. And I think this shows in his final comps. Especially the guy with the glasses at the end. I don’t have any clients that would accept that boiling bubbling key where the stem of his eyeglasses disappears.
Notice I’m not posting this rant to his youtube page. I only spouted because you are having troubles, and this video you brought forward is going to steer you wrong.

[Steve Bentley]

There’s this one
https://aescripts.com/true-comp-duplicator/

There are a few more that have slightly different functionality depending on your need.

[Steve Bentley]

Good point Dave,
Richard, If you have a lot of nested within nested comps you can really go down the rabbit hole because each of those increasingly inner comps probably also has to be unique. Perhaps the script is a better idea.

[Steve Bentley]

There’s a script that does this for you but I just duplicate the layer in the comp, tell AE to find it for me in the project footage list, duplicate it there and then option or alt drag it back ontop of the one in the comp (which has to be selected for the alt drag to work). Now you have two versions, each unique of that original comp. Changing whats inside one will not affect the other.

[Steve Bentley]

Hi Richard, I’ve complained about this multi length tab thing for years. Closing multiples was a pain due to size differences and missing that microscopic close X button (Adobe does realize that while they have twiddled their thumbs monitors increased their resolutions right?).
But its much better with the new, right click “close all but this one” feature. In previous versions (gain back to at least CS4) the tabs in my comps were the length of the name of the comp and not a uniform one. They did have a max length though and if you named something too long they might all appear to have been the same length with the extra text of the name hidden. But “Comp1” was well below this limit so those ones would be shorter.
What drove me nuts what that horizontal scroller above the tabs that was so thin it was hard to hit if you were working at speed. The new listing on the far right replaces it but I’m not sure if its any better.

[Steve Bentley]

Yes! Video Copilot’s VC reflect script. I think it’s free. Use it to make shadows instead of reflecting. You can rotate the angle, set to the skew to compensate for the reflection plane across which the “shadow falls. Feather the shadow in length and blur the edges. Its our new goto fake 3D shadow tool.