I'm editing an actor's reel that will be delivered as both DVD and Quicktime. It will not ever be broadcasted.

Should I be concerned if the color levels drop below 16 or go above 235? I know such levels are illegal for broadcast, but that won't be done. Do "illegal colors" cause a problem for DVD's as well?

THANKS MUCH.

Peter,

If you burn it yourself, just make it plesant to the eye.

However, I have done this and it looked grreat on my lcd, but when played on aCRT TV it was bad news!!! On the plasma it was ok just a bit to bright at times!!

color safe is insurance really.

Really depends a great deal on the hardware being used to playback. They're all different. Some will pass thru full bandwidth RGB (0-255), some will only pass thru 16-235. In this case, it will playback but all "illegal" colors will be clipped. It's much safer to put a knee in your gamma curve during post, to bring those illegal colors into range. Generally speaking, an NTSC CRT type of monitor will only display 16-235, regardless of what you feed it.

I'm bumping my own thread b/c this is an issue I'm running into again.

The rule of thumb is to use 16 to 235 for DVD's, AFAICT. But this can create an image that looks less contrasty than most commercial DVD's, IMHO.

So I decided to import a commercial VOB file into Vegas and see what it looks like on the waveform monitor. Well, low and behold, the DVD for "Syriana" has frequent places where the colors are illegal. Either below below 16 or above 235. I in no way think the "Syriana" DVD is an anomaly, I think it is the norm.

So this begs the ?, "If professionally authored commercial DVD's can have illegal colors, should we follow suit with the DVD's we author?"

Thanks much for any input!

I pay scant attention to legal levels when making DVDs. I will limit with graphic files in Photoshop if I happen to be working on them anyway, and sometimes will adjust in colour correction when I'm going through a sequence, but I'm far from anal with it.

Never saw any issues, or got any complaints.

I'd be more concerned with keeping an eye on the Vectorscope - the further you overshoot a target (especially red), the more likely you are to have bleed or blooming, ESPECIALLY if hooked up composite from the DVD player instead of component, S-Video or HDMI.

Shaun, doesn't the issue with legal colour have more to do with broadcasting equipment, rather than consumer viewing devices?

Colour interlaced with luminance (ie. RF or composite) can cause SIGNIFICANT bleeding and blooming if you over shoot legal. This is different than the 16 - 235 limit and is measured using the Vectorscope and making sure none of the 6 colour vectors (again, ESPECIALLY red) overshoot the target reticles in the scope. This is mostly allayed with colour separation cabling (component/YUV, RGB, S-Video et al).

And yes, I occasionally see combo DVD/VHS players hooked up via RF! Non-legal colours WILL bleed like mad.

Shaun,

So it seems that levels below 16 will not cause an error, per se. But you run the risk of having them remapped to absolute black on the display, depending on how its set up. Is that correct?

Which brings up the ?, how do you set black levels when creating a DVD?

I burned a DVD with SMPTE bars and 1 to 254 grey scale ramp and played it in my DVD player. The TV is a CRT connected via composite. Super whites and blacks were clearly visibly brighter/darker than broadcast level white and black.

I realize you said to use the vectorscope identify overly sturated colors, esp. red. And thanks for that advice. But what do you do w/ black level?

Thanks very much!

I'm editing an actor's reel that will be delivered as both DVD and Quicktime. It will not ever be broadcasted.

Should I be concerned if the color levels drop below 16 or go above 235? I know such levels are illegal for broadcast, but that won't be done. Do "illegal colors" cause a problem for DVD's as well?

THANKS MUCH.

Watching on CRT vs LCD vs Computer LCD will give three different results no matter what you do, because they all have different color spaces. Stick to HDTV, because I'm assuming most people will watch your auditions on a computer/LCD. It doesn't matter what your color levels are because -

Whatever best presents the actor (including looks) should be highlighted. That is the only criteria. Nobody cares about what else is going on in the frame, really.

Sareesh,

Thank you for the reply.

I should have really started a new thread instead of reviving this one from a while ago, as I'm no longer working on the audition project.

I now have some dramatic material that I want to burn to DVD, and am trying to figure out the best color levels to choose.

I am at the stage where a little information is a dangerous thing. For example, restricting color to studio swing levels of 16 to 235 is commonly recommended for DVD authoring. This sound logical and I sound smart when I mention it.

But the DVD's authored this way can look a tad washed out. And AFAICT, commercially released DVD's do not abide by this rule, or at least violate very often.

So I'm trying to discern what I should and shouldn't do color-wise to make the nicest looking DVD's.

And AFAICT, commercially released DVD's do not abide by this rule, or at least violate very often.

They probably figure that people will watch their DVDs many times over. And as time goes by, more and more people will watch them on a modern digital TV or a computer. So, from a long run point of view, it makes sense to use the full range of colors available on a large number of monitors already and potentially on all monitors available in not so far a future.

I mean, you could say the same of HDTV.

Should it stick to the old color range simply because a large number of people are watching the programming on old TV sets with a converter box? How annoying would that be?

Then there is the question of how many DVD players, whether hardware or software, are not capable of converting from the full color range to the TV range? I do not know the answer to that question but many certainly can do the conversion.

I'd say, don't worry about the old technology too much. Soon CRT will only be shown in museums. You want your audience to get the best picture they can get. If some of them are still using old technology, do you want to "punish" those who own digital TVs, especially as their number has been rising for a few years now and will continue to rise as time passes. Clearly, that is the thinking of the commercial studios.

S
I realize you said to use the vectorscope identify overly sturated colors, esp. red. And thanks for that advice. But what do you do w/ black level?

For DVDs, I'm not even REMOTELY afraid to push blacks down to 0. In fact, I do it often. I was just trying to indicate that the 16-235 suggestion is only PART of what constitutes "safe" colours. Whites I try to keep inside of 100 IRE (or SLIGHTLY above on SMALL amounts of "real estate") but NEVER on broadcast material.

Watching on CRT vs LCD vs Computer LCD will give three different results no matter what you do, because they all have different color spaces.

CRT and LCD have the same colour space but different response curves (gamma). Computer LCD has both a different colour space AND gamma.

the only way to judge is by knowing your target market, and then watching your DVD on the television/lcd/computer screen/DVD player that that market mainly caters to.

There's a huge difference between LCD/CRT/PAL/NTSC/HDTV/Computer screens. Sometimes the results are unacceptable, regardless of your best efforts, simply because they handle color differently. It also depends a lot on your source material and/or master.

Who's your main audience? Burn it for them. If the rest aren't paying your bills, forget them.

Thanks for the perspective. I'm in the US, so that's the market. There have to be certain parameters that are used when DVD mastering for the US market.

When looking that this issue more deeply, it seems that many of the rules I've been told are not actually followed, and that not following them makes a better looking DVD.

The most notable is "Limit color levels to 16-235." The blacks look like very dark brown on some tv's when I do that. So as Shaun mentioned, I'm going to be more willing to crush the blacks down to around 0.

I will check for overly bright color, esp. red. But I don't recall ever having the vectorscope trace hit the graticule color boxes while color correcting material. When looking at color bars, yes, but not when CCing a real life image.