The new enemy of the HD Formats is Noise Reduction! (1 Viewer)

[Ed St. Clair]

Hopefully our hearts, if not our minds, are in the same place!

 

[Douglas Monce]

Your memory of it and your ability to recall specific color pallets, grain structure or other details of a film print are not exactly the same thing. You are NOT going to remember that kind of detail with any accuracy.

If I were to color correct a shot on Monday, then try to come back on Wednesday and try and match that shot, there is no way that I could do it with out directly referencing the original corrected shot. Believe me I've tried it. The human brain just doesn't have that kind of recall. Not to mention the fact that the eye adjusts for color, meaning that something that looked white to me on Monday, may not look white on wed because of other factors in the shot that effect the eye.

Also the theater you saw the film in may or may not have had a new lamp in the projector. The lamps do change color over their life span, even if the required 16 foot-lamberts are maintained on the screen. The color may or may not be accurate. Then there is the fact that rapidly manufactured release prints are not always accurate to the answer print. You could see the same film in 5 different theaters in one day and never see the same colors twice.

Doug

 

[DaViD Boulet]

I didn't relaly mention color-balance as one of the things I recalled from my experience (color balance is not what I meant by color vibrancy)...



And I agree with you that color-balance would be one of the hardest things to remember/compare from a theatrical memory. Wouldn't contest that.

But when I watch a film print that's vibrant, saturated, and crystal clear to the point that it gives me goose-bumps, and the DVD comes out low-contrast, muddy, and murky-soft, I know something's wrong.

 

[Stephen_J_H]

Doug, keep in mind that the 35mm print that David was referring to was an independent film. It probably looked as good as he described it because a limited number of prints were made, hardly the same thing as a massive 5-6,000 print run of the average Hollywood blockbuster. With smaller print runs, the prints are usually fewer generations away from the ONeg, and are therefore more likely to conform to the answer print. I recall in my days as a projectionist having to run as many as 5 prints of the same film, and even in that small sample, none of them looked identical. The best looking print I ever saw (and keep in mind that outside of IMAX, I've never seen anything projected in 70mm; hoping to correct that around Easter with a trip to Seattle to see LoA) was Bubba Ho-Tep, again because of an extremely limited print run.

These same limitations work inversely when it comes to video transfers. Limited resources usually mean that you can't get the best colourists/telecine suites, so you're stuck with what comes out. As RAH has indicated in various threads, it takes an awful lot of work to get video (even HD) to look like film, and a "straight" transfer is not going to yield that, as you well know, based on your experience.

Did I mention I love playing Devil's Advocate?

 

[Douglas Monce]

I was lucky enough to see Lawrence of Arabia in 70mm on a 65 foot screen in its 1989 release after the restoration. I distinctly remember being able to read the time on Peter O'Toole's watch in the scene at the start of the film where he is working on the map in the "dark nasty little room". I was most impressed.

Beyond that I doubt I could remember any meaningful details about the look of the film other than particular lighting choices or camera angles that struck me as interesting.

Sometimes a limited run can work against a film too because they may not be able to afford the better labs. There are so many things that can effect the presentation. By the way I do was a projectionist at one time.

Doug

 

[Mark Zimmer]

It depends. Some films I don't remember the theatrical presentation all that well. Others stand out and make an impression. I very much remember saying to myself when I saw EYES WIDE SHUT theatrically "people are going to lambaste this when it comes out on DVD for being terribly grainy." So I remember it's supposed to have a lot of grain, which the HD DVD reduces substantially but at least doesn't wholly eliminate.

 

[Douglas Monce]

Well and again this brings up the question, did they really reduce the grain for the HD DVD, or is the grain just less obvious when its not on a 40 foot screen?

Doug

 

[Stephen_J_H]

Another question it raises is how much grain was added as a result of release prints being several generations away from the ONeg?

 

[DaViD Boulet]

Which raises another valid question: What's the "master signal" that the HD medium ought to reflect... the negative or the print made from it? Some director's might feel that the final negative is their "artistic work" and that the dupe prints are 2nd generation removed from it. Others might view the final prints themselves as the final work... and they might take the visual changes (such as additional grain and color-timing differences) into account while working on the negative so that the print represents their intended look.

It should never be blindly assumed that one answer is always the right one.

 

[Mark Zimmer]

All good questions, and unfortunately few of them are answerable from our vantage point.

 

[DaViD Boulet]

True, but those are EXACTLY the questions a studio should be asking each time they do an HD transfer. If the director is alilve, ask him or her. If not, do some research. Lots of information exists for those industry sources who care to find it about what a director was after with the negative/release-print appearance. Some directors actually leave flaws in the negative's image (like wires, make-up, color errors) knowing that they'll be obscured or can be corrected at the dupe stage. These things aren't "secret", they're easily found out for any studio who cares to do a little research.

 

[Douglas Monce]

Generally the answer print is regarded as the final approved version. Although many directors are involved in the the telecine and do things differently there than they did with the answer print. So in that case both versions are correct.

But if we are talking about an older film, even if the director and the DP are still alive, the answer print is generally considered to be the primary source of information about the look of the film. Again memories of the way they look, even the memory of the director, are just not that good. If the answer print is not in good shape that I would imagine that a fair amount of guess work would be involved.

Doug

 

[Michel_Hafner]

A negative is never a final master because it's a negative (inverted). The camera negative is the source with the most information but it's not a finished master because every negative must be color and contrast corrected into a positive so you can actually watch it with a non inverted picture. Camera negative projected directly is not only the 'negative' of the intended image, it's also quite odd looking (usually very flat and with color balance problems from edit to edit).

 

[JeremyErwin]

Realistic skin tones are very difficult to render with CGI. The face should be a good reference point for any reviewer-- detail, color balance, a bit of luminosity. It's often what the DP focuses on when determining lighting, film stock, etc.

But a diet of video-games and animated film may train the brain to look elsewhere, to ignore the clayey bits, to focus on things other than realism.

 

[Douglas Monce]

Also color negative film has an orange contrast mask over it, which is why when you hold a negative up to the light, it looks orange. The orange mask is compensated for when the positive is printed.

Doug

 

[Michel_Hafner]

Latest DNR victim I have seen: Eyes Wide Shot.
Very bad call, WB. You did so well with "The Shining". And now this.
I'll stick to my HD D-VHS recording off the air.

 

[Gary Murrell]

I haven't checked out my BD copy of EWS yet, that would suck!

-Gary

 

[DaViD Boulet]

Well "duh" about the fact that the "negative" is... well... a negative.

I think what I meant was that often the negative is used as the final analog-domain-source. Colors and luminance can be inverted digitally after scanning. That's the flip of a switch, but it doesn't change the point that often the negative element is scanned and used directly, and is the "source" by which one maintains fidelity in terms of film-grain etc. (and color too, once its inverted).

 

[Felix Martinez]

I just popped it in, and I totally agree. Sad.

 

[Douglas Monce]

I'm not sure that the camera negative is used very often. Rather the internegative is used that has all of the color timing information from the interpositive. Once the answer print is approved the camera negative is generally stored away literally in a salt mine.

The only time the camera negative (which has no timing information) would be use is if the internegitive or the interpositive is in such bad shape that they are no longer printable. In that case the timing information would have to be recreated based on the lab notes.

Doug