Andre LeBlanc

Adding magenta can negate *part* of one problem, and add a new problem. Yes, for areas with green spill, you negate it. However, just as it's difficult to control spill, it's also difficult to control exactly where the magenta hits your subject. So areas that don't have green spill to begin with now have magenta in them. This creates the unintended compositing problem of having an additional color that you didn't want to begin with. Perhaps the colorist you spoke with has some ideas around this issue, but just a heads up that it can be a problem.

Sorry if this was already posted... Some stunning color photographs taken in Russia circa 1910. The photographic technique here is described as using "a specialized camera to capture three black and white images in fairly quick succession, using red, green and blue filters, allowing them to later be recombined and projected with filtered lanterns to show near true color images." http://www.boston.com/bigpicture/2010/08/russia_in_color_a_century_ago.html

These are brilliant! Coming from a VFX background, this one had me rolling on the floor...

Interesting idea... although I think it would get a little bit nutty on this project! After giving this more thought, it seems like the 75% case will be the Plexiglas with the greenscreen backdrop. Ensuring a strong enough key on the actors to get a reflection on the Plexiglas. And when it's not the 75% base case, it sounds like there's a multitude of more unique solutions (reflective materials, etc...) when we need to get really specific.

Very good idea with the projectors! Seems like an obvious solution for the lighting fx when you mention it, but i hadn't even thought of it.

Thanks for the input! I think the idea of putting a camera behind the screen to catch reverse action is interesting. On key shots where we'd really expect the reflection, it could be matchmoved to the original plate. I also like the idea of using reflective material when the actor is not right in front of the screen. This might make for an interesting shot where we only see the actor's reflection, and not the actor in the foreground (a clean shot of the screen). It seems like there's about 10 different solutions... each one getting you about 50% there. To answer Robert's question, yes, it's actually a window that doubles as a video screen (it's supposed to be modern OLED technology). And yes... it will eventually shatter. Is the story that obvious? :) I'm not 100% sure what I can get away with, but maybe I should be selective. Only worry about the actor's reflections when we're close to the screen, and it's prominent in the shot. In those instances I could consider all the options discussed above (green screen, reverse camera, reflective mtl). Otherwise, maybe highlights and spec hits will do. Any thoughts on mimicking the light that would be emitted from the screen? The only time I see this being critically noticeable is if we're on a CU of the actor's face watching the screen as it changes. In that case, I'd expect almost a dithered effect to the light hitting their face. Might be as simple as waving a cookie cutter pattern in front of the light.

Hi all, I'm doing some visual effects tests in the next couple of weeks for a sci-fi project, and was looking for some thoughts/ideas on techniques. The scene I'm testing takes place in a room with a large digital screen (the screen is maybe 5ft wide by 3ft high). It's essentially supposed to be an oversized security monitor giving visuals into different rooms in a building. An actor would stand in front of the screen, controlling what he/she sees on the screen with a controller. The contents on the screen would be entirely done as a post production comp. But some of the things that concern me are 1) reflections and spec hits, as the screen is supposed to be glass and 2) light being emitted from the screen. I've had numerous suggestions involving Plexiglas with a green sheet behind it, or even the use of a mirror or mylar as the screen where we'd only see the actor in the reflection. I have an extensive compositing background, so how to put all this together is less of a concern then making sure I get all the right elements. Trying to fake reflectivity and spec highlights in post always falls apart... especially if the camera's moving. And of course, there's the issue of the light that would come from the screen! Anyone have any ideas or previous experiences they'd like to share? Thanks for you input!

Hi all, Thanks for the input! There is a Hollywood location that offers similar services forpanes up to 4ft x 8ft. It's also made from plastic resin, and is absolutely clear-- unlike sugar glass http://www.alfonsosbreakawayglass.com Talking to them, they'll make anything custom, so it's worth looking into. I'd still use the real glass for setting up the situation, but for the actual 'hero' break, something like this might make sense when worried about safety.

Yeah, I think there's several good ways to cheat this. Now I'm thinking you wouldn't even have to be looking directly at the window when the first break happens. It could be a cheat. Intercut with plenty of medium and close shots. I definitely want this to play safe! That being said, has anyone had any good experience with sugar glass? I was thinking maybe it would work for some medium, or close shots, but every time I see it used, it has a milky or yellowish tint to it. It also has a very brittle breaking effect-- the opposite of what I'm looking for.

Definitely tempered glass-- thanks for pointing that out. In terms of thickness, single pane would be better for practical purposes. The one downside I can see with tempered glass is the whole window shattering at once, rather than a gradual breaking, or poking through. This sounds like a good thing to test thoroughly before shooting :)

Hi all, Anyone have experience where a character breaks a large piece of glass? I'm planning a short film where a character is ramming his fists (or a blunt weapon) against a large window. We're watching the action from the inside, and the character is on the other side of the glass. Over time, this window has to break enough so that the character can climb or walk through the opening. There's a few givens: 1) We'd cut away from the action several times, so there's a chance to cut back to the breaking at different stages. 2) The window is large. I'd say about 5ft W x 4ft H. 3) The glass is clear. 4) We're always on the inside. The character is on the outside. I'm concerned about safety, and of course flying bits of glass going everywhere. I guess that some of this could be done with CG, and clever editing and framing will help, but I'm really hoping for some practical methods too. Anyone have any thoughts or experiences they could share? Thanks!

Having used both Mac and PC extensively, I can say one of the biggest reasons to use a Mac is to be able to properly interface with small post production facilities for doing your digital transfer services, such as coforming EDLs, etc... The majority of them are using a Mac with FCP in my experience. I ran into a lot of problems when trying to get files back and forth (even simple EDLs) from my PC world to the post production company's Mac world. Coming from a very technical background myself, and having no real bias of one platform to the other, I would use a Mac for all my future editing projects. A Mac is basically just a PC with a different operating system. They even use Intel processors now. In terms of software, I can say whole heartedly that I like *both* the FCP suite and Adobe's suite, and I can think of several reasons to use both for different situations. You can get both these suites for the Mac now. Plus, there's no reason you can't have a Mac with both OSX and Windows running on it if you think you might need both.

Agreed. Hopefully Quicktime will resolve this issue with their encoder at some point!

I've now managed to reply to my own post 3 times... After hours of troubleshooting, I believe I've solved the problem. After scouring message boards, it appears that the problem resides in the conversion from YUV color space to RGB color space, and more specifically, the way quicktime handles this conversion. Most compositing/3d packages use an RGB color space, while most editing software uses YUV. Quicktime will do the file conversion for any file going into Shake, or coming out of fcp, and in this case, it was clipping some of my color values (it's not doing the most accurate YUV to RGB conversion). The workaround was to use After Effects S3 to do the conversion. After Effects has its own built in converter, and therefore does not rely on quicktime to do it for us. It was a bit ugly, but I end up converting the DVCPRO movie files to a targa sequence in AE, and then importing that targa sequence into Shake to do my compositing work. It's a bit ugly, but the only way to avoid quicktimes YUV to rgb conversion. I guess that's it for this thread. Hope someone finds this useful! Or maybe someone else has a more elegant workaround for such a problem?

Also, I am aware that fcp does its own gamma correction, while Shake expects you to provide your own lut, thus sometimes creating a brightness discrepancy between the two programs. This appears to be a different problem: what looks like some sort of compression going from shake to fcp. Could there be a problem from going between YUV space to RGB in quicktime?

As an update, I also notice these same subtle differences between my plate and comp when I view them side by side with quicktime. Is it possible that Shake is doing something to my footage?? That seems unlikely to me, but...

Hi all, I've been doing some minor post production compositing on a short film we're finishing, and I'm noticing I'm getting some strange artifacts at one of the stages: My workflow is editing in FCP, and doing any compositing work in Shake. The footage was originally shot on Varicam. What I do is bring the raw plate into Shake (which was originally shot on Varicam), do my comp work, then output the new version as '8-bit uncompressed 4:2:2'. If I re-import the new composited shot back into shake, and compare it frame by frame with the plate, they are identical. No artifacting, no increase in noise etc... 0 difference. But when I bring these same 2 shots into FCP and compare them, there's subtle differences. The new uncompressed 8 bit 4:2:2 movie has more defined noise, as if I ran a gamma curve on it, and some strange 'blocky' artifacts in dark areas. What is FCP doing to my footage? What step could I be missing in my workflow? Should I be bringing my new composited 'uncompressed' shot into Compressor and converting it back to DVCPRO format? Currently, my project settings in FCP are set to read the footage as DVCPRO. Any help or clarification is greatly appreciated. Thanks!

I'm sure people can become colorists from all sorts of varying backgrounds and disciplines, but I've personally witnessed people who started as compositors go on to become colorists. Most compositors I know have some training in core arts: painting, photography, etc... I would think similar training would be helpful for a colorist, but again, I'm sure it varies case by case.

It this for real??? http://www.slashfilm.com/2008/04/25/uwe-bo...to-death-match/

They may have done it with a set extension in post. In other words, everything screen left looks like it could be a composited effect on top of the original plate. It's a pretty common thing, and with all the blur with this particular camera move, you could really get away with it. Not saying that's definitely how it was done... but it would be possible to do it that way.

Hello all, I've seen numerous slow-motion videos on youtube of light bulbs exploding. Sometimes it's obvious how they're doing it, and other times it isn't. Is there a way to explode a light bulb safely, when it's still on? This is for a specific effect I'm trying to do of a typical lamp shade light bulb exploding-- say 60 watts. Anyone ever seen this done, or done it themselves? Any info greatly appreciated!

I just realized you guys replied to my original post from 2007. In the end, I just did one short, and am leaving #2 for later this year. The reason I separated them in the end was that they were so different in subject matter (one taking place in the 80s, one current, one comedy, one dark...). Grouping them became a logistical nightmare in terms of even just getting into the right mindset. Also... there was the whole issue with money, as was mentioned in the replies. In the end, I'm glad I separated them this time. Maybe there will be a day down the road where two projects are small enough and similar enough that I can do them back to back, but right now wasn't the time. Thanks for the replies!

OK, cool, I've got it working now... and a small pile of improperly formatted DVDs :)

Unfortunately, this doesn't seem to be working for me. I've tried exporting a movie out of FCP at pretty much every aspect ratio, and whenever I export to DVD and view it on the television, it's slightly squished vertically, and cropped on the left and right. Something's wrong somewhere in my methodology. Should I be exporting a quicktime movie that matches the original specifications of the footage (DVCPRO 720p, square pixels). Or should I be down-converting in FCP to an NTSC resolution movie before I bring it into DVD Studio?? It seems like I can be making a mistake at many different levels. Any thoughts? Thanks again!

Hi everyone, I'm fairly new to the Mac world, and have been trying to use Final Cut Pro suite to create a down-converted letter-boxed SD-DVD from my DVCPRO 720p footage. Current work-flow was to import my final HD movie into Compressor, and export a file using the MPEG-2 preset for SD-DVD. From there, I've been using DVD Studio Pro to export to DVD. The final DVD is letter-boxed, but looks stretched, and the sides of the image are still slightly cropped. Anyone have experience exporting an HD file from FCP to an SD DVD? Any help is greatly appreciated!