Bluebox vs. Chromaflex

[Zutho]

Hello everybody,
I`m from Germany and I love this board.

To get my Bachelor of Arts, I have to do a so called Honours Research Project.

Well, I think I`m going to write something about conventional bluescreen methods and a modern/alternative/new bluescreen method, the chromaflex technologie.

Waht do you think about this:

Theoretical Part:

Basic Issues
a) Why Bluescreen?
b) Bluscreen vs. Greenscreen?

Explaining the two methods
a) Bluebox (function, affords (color, size), setdesign, lightning, actors, trackingpoints, problems, costs)
b) Chromaflex (cp. bluebox)

How to get the objects
a) Masking (garbagematte, transparency/alphachancl)
b) Keyingmethods (how do these keyers work: Ultimatte, Primatte, Diamond, Mattenee, Keylight)

Basic problems
a) Lightning
b) Shadows
c) Trackingpoints
d) Size

Comparison of both technologies:
a) Production process
b) Account analysis
c) Code of practice
d) Advantages: room/space
e) Advatages: time
f) Application Areas


Practical experience:

Doing some Bluebox and Chromaflex shots
a) long hair
b) glas
c) blue jeans

Keying the same footage with diffrent keyers
a) Primatte (Shake)
b) Keylight (Shake)
c) Diamond (Combustion 4)
d) Matteene (After Effects)
e) Ultimatte (After Effects)

Gaging the allocation of light and color (waveformmonitor and vektorskop)
a) Bluebox
b) Chromaflex (optimal conditions)
c) Chromaflex (bad conditions)


Conclusion:

- Which technologie works best?
- Which technologie is the cheaper one?
- When should I use which one?

[russ_c]

Let me know if you need any personal opinions!

I have shot both on ultimatte blue painted screens and on chromaflex in a studio environment at work.

Russell

[Zutho]

Thanks Russell,

I think I´m going to need as much opinions as possible
Feel free to tell me as much as you want about your bluescreen tests/work.

At the moment I`m trying to find out what research is necessary to get a clear result. I`m reflecting if shooting on DV is enough or if I should try some Beta or HDV or even 16mm, too.

[Hugh]

You'll find that different shooting formats make a huge difference, too....

DV and HDV are both 4:1:1 sampling, so there is a lot less colour information, making getting a good key that much more difficult...

Beta and Digibeta would give a nicer result here....

Film is a very different animal. I've only done keying from film on one project, and the stock that we were using had a lot more grain in the blue channel. Because of this, pulling a key off green was much easier than off blue....

Anyway, just some things to take into account!

[Zutho]

I`ve heard of the differences, but I didn`t work with all of these formats, yet.

Do you think, there is generally any difference between DV and HDV?

Do you think I should concentrate on these two formats and say: From this it follows that ... If Chromaflex is better than BlueBox in using DV, it`s also better in using Beta/Digibeta?

thx for your advice

[Hugh]

As far as I am aware, HDV is just a higher resolution version of DV...

I'd probably recommend doing your tests with the highest quality camera you can (DigiBeta/HiDef) - then you can know that any problems are down to the keying technique rather than the camera...

[Zutho]

I read the same about HDV and DV. In the strict sense DV = DVCAM = HDV (higher resolution).

I think you`re right in suggesting to take the highest quality camera.
Concurrently it would be interessting to compare consumer DV cameras (1chip) to professional DV cameras (3chip) to highest quality cameras (DigiBeta/HD).

However it`ll cost a lot of money to borrow a HiDef cam. The others are no problem, I think.

[Hugh]

If you can, I'd still go for DigiBeta, otherwise, well, you've got to work with what you've got...

And your comment about DV = DVCAM is perfectly correct... DV is the compression. MiniDV and DVCAM are recording speeds (a 60 minute MiniDV tape will fit 40 minutes of DVCAM - it leaves more tape surface per second)

[beaker]

Add these few things to your practical experience:

Setup with really crappy lighting with hot spots all over the place and bright shiny metal bars behind people on the ground and have people wearing blue clothing on a bluescreen and a green dress on a greenscreen. Oh, also use film that isn't FX stock and has a high iso (blue channel has grain the size of watermellons) or HDCAM tape(crappy 3:1:1) instead of a Thompson Viper. Even better shoot it on film and then telecine/transfer it to HDCAM and capture it on the local G5 and let me key that. One more thing, shoot with a really wide angle lens toward the ceiling(where there is no blue/green screen) with people jumping in the air that are motion blurred.

Then test the keyers on that .

I'm kidding, but this is the reality in many commercials and movies are shot this way because either the Tard in charge didn't hire a vfx soup till the end of the project or they hired a people who don't know what they are doing.

-deke

[Zutho]

If you can, I'd still go for DigiBeta, otherwise, well, you've got to work with what you've got...

And your comment about DV = DVCAM is perfectly correct... DV is the compression. MiniDV and DVCAM are recording speeds (a 60 minute MiniDV tape will fit 40 minutes of DVCAM - it leaves more tape surface per second)

DigiBeta: I`ll try.
DV = DVCAM (Compression & Lenth): I know. But thanks a lot for your suggestions and help!

... but this is the reality in many commercials and movies are shot this way because either the Tard in charge didn't hire a vfx soup till the end of the project or they hired a people who don't know what they are doing.

film:
thx. I don`t know if I can do it with film `cause film is probably a bit to expensice. But I`m going to try to take an old 8mm camera, shoot the wohle keying material and try two methods.
- scanning the material (analog -> digital)
- screening the projection with a dv-kamera

clothing:
- blue jeans infront of a bluescreen is in planing
- green dress: I think, I`m not going to do this, because I want to have the same shot with all of the tree methods: bluebox, greenbox, chromaflex

fx stock:
what do you mean with that statement? I don`t clearly understand that point.

key:
- you can key all my footage, as all other people here on that board, too.
- when? I`ll tell you.

bad lightning:
- chromaflex: this method seems to work even is there is a bad lightning.
- bluebox/greenbox: without a good lightning, you`re going to have real problems keying that footage (as you know). so i don`t know if I realy should compare chromaflex (good lightning) vs bluebox (good lightning) vs greenbox (good lightning) vs chromaflex (crappy lightning) vs bluebox (crappy lightning) vs greenbox (crappy lightning).
Well, in reading the last lines, I think you`re not completely wrong.

Any comments?

[Zutho]

I`m looking for some recommendations (books, articles, websites...)
Every suggestion makes me happy.

I already own these books:
- The Art and Science of Digital Compositing, w. CD-ROM (Ron Brinkmann, Morgan Kaufmann Publishers) (1. Juli 1999)
- Digital Compositing for Film and Video (Steve Wright, Focal Press) (Januar 2004)
- Shake, w. DVD-ROM (Marco Paolini, Peachpit Press) (July 2003)

[borrell]

DV and HDV are both 4:1:1 sampling, so there is a lot less colour information, making getting a good key that much more difficult...

Not true. As we all know. DV NTSC is 4:1:1 and DV PAL 4:2:0. But HDV is ALWAYS 4:2:0 both in NTSC and PAL land.

As far as I am aware, HDV is just a higher resolution version of DV...

Not correct. DV, DVCAM, DVCPro and EVEN DigiBeta all use DCT-based intraframe compression schemes - kinda like JPEG compression.
What makes DigiBeta better is its 10 bit 4:2:2 compression vs DV's 8 bit 4:2:0

HDV is 8 bit, 4:2:0 MPEG2 @ 25 Mbps.

[Splash]

DigiBeta: I`ll try.





fx stock:
what do you mean with that statement? I don`t clearly understand that point.



Any comments?

Stock is the characteristic of film (grain, sensitivity the colors and others characteristics). A non FX stock can be much grain in determined channel like the blue and you need to make a keyer using the blue channel ( bluescreen) or the stock have low sensitivity to a determined color like green

[odysseus]

The basic problems You've listed, consist of an amount of other variables not mentioned with what You've listed, and in some occasions, they make a big difference if not noticed in good time. You might know about these already, but incase You don't, or if someone else doesn't, I'll throw in a couple of words. Anyone happen to know better, please correct me so I don't unnecessarily mislead anyone.

Your list:
Basic problems
a) Lightning
b) Shadows

I'll grab the two out of four to say something and a few words on some other matters.

a)
Lighting always has to do with several issues. One is the equipment (=in my country usually the budget) at hand. One is the camera and the lenses involved. One is about acceptable possible alternatives in the ways of using them. All of these vary over time and there is only so much time, so the better You come up with variations, the better You come up with whatever the situation. In my experience, the more inventive one gets, the better he does. This has to do with not only lighting and related stuff but the surface coverage of the screen with the needs of camera maneuverability and that sort of things. You always have to think in some causative way: You do that and it will cause this, so how do You come up with a result saving the day? Be awake, be inventive, be curious, do not fall into the pit of premature optimism.

Since A and B tend to involve pretty much the same issues, I'll separate this A section into the camera itself yet not going too far out with it because I'm not in a good enough base knowledge to have a final say. You've heard about resolution and You've heard about compression. The more pixels the better, the less compression the better. But one more thing You have to bear in mind is the capability of each piece of optics that'll be used and each camera type, because of whatever the primary choice (usually made by Your producer). You'll have to make the best out of it no matter what.

There are apertures lenses work poorly (creating smudged, noisy artifacts), there are apertures lenses perform at their best, and this in addition to the resolution or compression rate is what really counts. Always aim for the best You can, but remember the variables involved! This means to say, find out the worst and the best scenario with each camera if You can. Someone always knows. It'll save You a lot of effort otherwise thrown out of the window. Don't trust on "hearsay", trust someone who's tried it in a live situation.

On b):
Shadows happen in two ways: the ones cast by lighting and thus falling over whatever the receiving surface (usually tends to to turn out noisy), others happen within the subject that's to be keyed out (unnoticed problem areas lurk in the dark). In many cases I have worked in, people have been told to make sure the screen is well lit. So concerned as they become with this, all the information gathered about color temperatures and whatever, they forget the subject and ways of variating the lighting. Should I even say, variating in unspeakable ways. An example of this "unspeakable" is the use of soft filters to get rid of some seams in a poorly set up screen or other minor physical artifacts that'll always be there. Do NOT use soft filters, because You can create the effect later on. You aim for the best result On Set. Also, all the lurking-in spills that tend to "crawl in" in the dark est spots in Your character /talent (ears deep inside, under eyebrows, neck etc), these are areas that may come in with later trouble but You might not see the problems with Your bare eyes while at set and yet, the spills are there. Ofcourse, there are tools for spill removal, but the better material You get in the first place, the better and easier the result in the end of the day.

Depending on the DoP (Director of Photography that is) he might want to set up his key light first, then work out the rest as he best sees. When doing screening whatsoever, I've came to a conclusion that it's best to start with an evenly lit screen, THEN wash off all the spills with your fill, THEN set up Your key light to a level reasonable with further tweaking capability within whatever the material You're shooting. And keeping in mind, ofcourse, the detail level at each level of lighting with Your contrast ratios. It may sound complicated, but it's quite straight forward. This is a moment of truth. Once You're asked for some advice, it has to do with first hand experience. And it is You who has to come up with it. Read about it from whatever the source, it is not a real situation where You have tens of other things to mind about.

About the size of Your screen:

Size matters. Usually when people come up with the idea of using some screening technique, it has to do with replacing the backround but nothing else. They are so driven by this that they forget the reality of surface coverage while dealing with FOV (field of view) at the same time and they drop You in the middle of some solid impossibilities. "The talent will run for some 400 meters and then jump on something." Ok. How long is the coverage of your green/bluescreen? "What do You mean it's not large ebough? Aren't You the post production guy?" Yes. yes You are. And You have think about these things in advance.

There are a lot of other things to think about. But those were the most important ones for starters.

Wish You good luck with Your thing