Regarding phosphors...
 

I'm spinning this off from the thread asking for opinions on the Panasonic monitor...

<<<-- Originally posted by Jeff Donald : [...] Most people prefer the phosphers (colors) that Sony monitors produce. I used to use Panasonics a lot, but clients could see the difference and I got tired of explaining it to clients.
Jeff -->>>

After doing some research so that I could further understand this statement, I have a few more questions. :)

The two main phosphor types, in the US at least, are SMPTE-C and P-22. SMPTE-C and P-22 phosphors differ in the spectral content of the individual phosphors. Sony makes monitors using both types. I'm assuming you are suggesting that clients would prefer P-22, as that is the phosphor used for consumer televisions.

P-22 phosphor, I am understanding, will yield a brighter picture at the expense of linearity of response, accurate color reproduction, and consistancy from one monitor to the next.

SMPTE-C is created to much tighter tolerences, and an SMPTE-C monitor, properly calibrated, can be reasonably expected to look exactly like any other SMPTE-C monitor.
This is arguably the only way to do serious color correction.

Of course, the question I need to answer for myself is, "Do I ever need to do serious color correction?"

Leaving that question aside for now, let me ask this one instead; If the main argument against a SMTPE-C phosphor display (aside from the cost) is that it will not look like what a client expects a 'teevee' picture to be, why use the production monitor at all for client screening purposes? Why not buy a 27" - 32" consumer TV, with its super cold white point and other atrocities that most people associate with a "nice picture", and let them view the material on that?

Is it just the added caché in having them view that industrial looking box set in amongst all the other nifty devices?

lyd

In one of your other posts you noted how your consumer set would not adjust properly to color bars. Many, many consumer TV's have that problem. In my house I've a Sony WEVA flat screen (about $1,200) and it adjusts pretty well. Most $400 consumer TV's aren't. If money is tight get the $400 to $600 (13" - 14") production monitor. It will look good and have full adjustments, blue gun, under scan, Cross pulse and many other non consumer adjustments. Get a little extra money and get a 27" - 32" consumer TV for clients to view your work. The production monitor and consumer TV will have similar colors. A high end SMPTE-C phosophor monitor will never match your client TV. I do some high end color correction work in After Effects with special plug-ins. It also helps to work in 10bit or 16bit color for that type of work.

Jeff

I agree that most consumer TV's can't come very close to proper adjustment. I wasn't suggesting this could replace a good production monitor, just that it might be a more palatable way to show work to a client. If I understand you correctly you are agreeing that is a possible approach? In any case, I agree that for my situation the thing to do is just as you suggest. I will be looking at the Sonys mentioned in the previous thread.

Speaking more generally, though, I don't quite follow one statement you made, "A high end SMPTE-C phosophor monitor will never match your client TV."

Isn't that the point? Don't you strive to be as exactly close to the industry standard as possible during production, because that way even if doesn't look "right" when viewed (i.e., exactly as it did on your production monitor) it will it least look "as expected" ( i.e., the same way all broadcast, DVD, whatever, material looks on a given set).

Or am I really just putting way too much emphasis on this issue?

As always, thank you.

lyd

Color reproduction fidelity involves a broad spectrum of issues going from camera spectral properties and electronic masking matrixes, through display chromacity values (phosphors for CRT, direct view LCD, Plasma, FED..color splitter properties in LCD, dlp..projectors) and of course involves the different display settings. On the phosphor issue. since in 1953 NTSC introduced the NTSC primaries based on the available phosphors (saturated green) at that time a lot has been changed.. Later on, the SMPTE and EBU phosphors where introduced as standards for pro monitors with lots of disputes on how to cope with these newer pro phosphors.. even between EBU and SMPTE.(e.g. how to get the real scene colors displayed ...still something else than the diff between pro and consumer displays) In addition because of the brightness struggle for consumer TV and CRT projectors still more formulations (P22..)got involved. The main changes related mostly to the "green" (saturation and hue). All by all, if a high end consumer TV is well adjusted it can be close to the ultimate subjective fidelity especially if not too pure colors are involved

lyd,

Most of the work that's talked about here is not for network broadcast. I think the majority is probably going to VHS, DVD, Web and some local origination (cable, small market broadcast). Corporate, wedding etc falls into the above categories. Use a production monitor that will match (or come closest) what your clients will view on. Your correct that if your doing network broadcast or PBS documentaries that you'd better have a damn good production monitor (and waveform monitor/vectorscope).

However, while high end consumer TV's can closely match color, contrast, etc of a moderate cost production monitor, they lack crucial features. I would not use a monitor, if at all possible, without blue gun, underscan, and cross pulse (H/V delay). Another problem is the size of consumer TV's. Small tube sets (13") usually are poorly made and lack crucial adjustments. Larger sets (27") have more adjustments and better control and color but are too large for comfortable viewing.

So what works? A good, small Sony production monitor and a 27" Sony consumer set for your clients.

Jeff

There are two things that relax the phosphor and color fidelity issue:
-only studio people can see the the color diffs between the real scene and the displayed image.
-only the editor can verify the difference between a (pro) monitor and a consumer TV.
Consumers only see the picture on their set, and when the colors are subjectively OK, even when they are far off w.r.t. the real scene colors, the enduser will be happy.

This all makes good sense.

One thing still nags at me, however...

Say I am going to produce some content for limited distribution on DVD. Jeff's suggestion is to "Use a production monitor that will match (or come closest) what your clients will view on."

Does this imply that I should not use a production monitor with a 6500k white point, as most clients will view on a consumer TV with a 9000k or higher temp?

Say I do this, and color correct or generate effects or whatever so they look subjectively "good" and "accurate" on this monitor. Now, on some small percentage of consumer TV's that happen to be close to my production monitor, this will look great. But on the rest, that are adjusted differently in tint/hue/picture/whatever, it will not only look different from what I intended, but different from what that viewer, who has adjusted his TV the way he likes it, expects. Does this make sense?

Dre's points on things that mitigate the color match issue are well taken, but how about an example like this:

I have a scene prominently featuring the well-known MacDonald's Golden Arches sign. Now, if this scene looks as close as possible to perfectly accurate (relative to the reality) on a production monitor calibrated to the NTSC specifications, that will be true to any other monitor so calibrated.

But, if a random consumer, call him Joe, has his TV set up differently, it will not look like the original. But it will look like what he expects. It will look exactly the same as it does when he sees it in MacDonald's commercials, another DVD, etc. It may not be reality, but Joe isn't looking for reality, he is looking for consistancy.

If I make that scene look "accurate" on a monitor set up some other way than the standard, it will not look like Joe expects, and he will think, "Something is vaguely wrong with the color on this DVD..."

Alternatively, Joe can't tell that the golden arches he is looking at now are slightly different than the golden arches he looked at five minutes ago, and it is all a non-issue, which is I think what Dre is suggesting....

I guess I am obsessing over this a bit, but given that it is so easy to do this one way or the other, it seems worth making sure I understand which one is correct.

Thank you for your patience.

lyd

Subjective color fidelity perception is indeed a complex matter, more related to psychophysics than to colorimetric math. The most important in the color perception is indeed consistancy, like you mention. For home viewing, people want to get the colors, brightness,contrast,sharpening...like they are used to see on their TV on broadcast programs. Correctness of whitepoint settings, phosphors... are almost unimportant. So try to realize WB, saturation levels...which are as close as possible to broadcasted images. Skin colors and the green of grass and trees (woods) are known to be critical and are the colors the human brain remembers best.. Artificial objects (MC Donalds) are seldon remembered, unless you "see" them for several years ...If I had to solve that problem, I would use my TV at home (on which i am used to see broadcasted images) and, without changeing any setting, see if my work matches the colors,contrast levels..(even audio levels, balance..) which I am used to see and hear.

Thanks. Again, enlightening. I begin, finally, to see not only how I need to approach this but where I have been slightly off-base in this whole line of questioning.

While this discussion has firmed up my understanding of various important relationships in ways that will be very useful, it has not in fact answered what I only now realize was at the root of my questions.

This is because the answer is "it does not exist", or "only experience and a trained eye can do that".

The question was, "What is the technical setup and procedure that will, without any other external references, guarantee my final product will look just like professional material on any given monitor."

I think I expected there could be a "do like so, and so, and then so, and that is the way it is done" answer. I believe I underestimated just how subjective this process actually is.

Armed with understanding this thread has given me, I am prepared to go ahead and do what you could've dismissed me with in the first place (and thank you so much for not doing so!), "Get to work, create some material, and test it all over the place. Then do it again. Repeat as needed. Eventually you will start to get a feel for it."

I'm an IT professional by trade, and I suppose I tend to assume that anything can be completely described by a well written white paper. I'll have to stop that. :)

lyd