Antti Näyhä

Takayanagi in Variety:

"On 'The Grey,' we tested several film stocks and processes and ended up choosing Kodak 5229 forced processed. We used Panavision XL2 and Platinum cameras."

(I assume he means ”push processed”.)

Another anachronism with a touch of irony is the use of speeding up footage to make it look old - thing is back in the day if something was shot at 18fps, it was also projected at 18fps so the pace was natural.

Actually, silent films were frequently projected at a slightly higher speed than they were shot. This was the artistic intention. The speed difference just wasn’t usually as huge as 18-to-24.

Kevin Brownlow: Silent Films – What Was the Right Speed? (1980)

(See the table at the end of the article.)

was any projected film historically ever sepia colored?

Several sepia/amber/golden hues were used on black & white film prints, even though the exact chemicals might not have been the same as in photograph tinting. Check out the articles about Film tinting and Photographic print toning at Wikipedia.

Edit: sorry, just realized that you may have been talking about sepia-tinted color prints.

I have certainly seen installations where unencrypted DVI was going straight into the projector (I know this because they were monitoring it).

OK, I believe you… but DCI servers / media blocks don’t have DVI outputs nor do they decrypt content.

So what you saw must’ve been something else. Perhaps a regular PC or a hard disk media player connected to a D-Cinema projector? This is sometimes done at festivals or special screenings, if a DCP isn’t available.

As said before, any DVI/HDMI device can be directly connected to a D-Cinema projector. However, any cinema that’s properly equipped to handle ”alternative content” (= anything that’s not DCI content) has an external scaler/switcher connected to the projector for more flexibility. The scaler has all kinds of video inputs, both analog and digital.

Such cinemas are also able to directly feed their sound system with digital audio from non-DCI sources. For example, the popular Dolby CP750 D-Cinema sound processor has both coaxial and optical inputs for AES3 / S/PDIF audio. Venues with the older CP650 processor (for 35mm compatibility) commonly use a DMA8Plus adapter box to achieve the same thing, although it only accepts up to 5.1 channels. In either case, no need to fiddle with individual amplifiers.

I usually also install a box that splits a HDMI signal to separate DVI video and S/PDIF audio outputs. This way, you can conveniently connect any HDMI device to the D-Cinema system with just one cable.

Reasons why films aren’t distributed to cinemas on Blu-Ray instead of DCP:

1) Reliability. Blu-Ray is an optical media, and as such it’s just not reliable enough for professional use. It’s extremely prone to scratching, fingerprints, etc. For some reason, optical discs always work at home and test screenings, but will fail in a middle of a screening when there are paying customers in the screening room… Ideally you would always rip any optical disc losslessly to a hard drive before screening, and project it from a computer instead of a Blu-Ray drive.

2) Piracy protection. Phil: DCI content is not decrypted by the media block. The cables you are referring to carry encrypted content, which isn’t decrypted until at the very last stage deep inside the projector. Recording the content is by no means trivial – you are severely underestimating the paranoia of the studios who wrote the spec. ;) Ripping a Blu-Ray, though, is indeed trivial.

3) Quality. Yes, a good Blu-Ray can look (and sound) very good on a properly calibrated D-Cinema system. It’s certainly the best way to enjoy a Blu-Ray movie. But it’s still not quite up to D-Cinema quality due to intra-frame compression, much lower bitrates, narrower color space, etc. And Blu-Ray mastering quality varies a lot. There are a lot of Blu-Ray movies out there that I certainly wouldn’t pay to watch in a cinema.

Im very fond of dark color and details in the shadows, but in this new tweak the shadows are crushed into solid blacks

Also, the blacks are not really crushed (as in ”clipped to absolute black”) in the Blu-Ray screengrabs you showed us. The shadow detail is still there, it’s just graded very dark. Check for yourself in Photoshop.

Same here. I watched it on a calibrated D-Cinema system, having seen 35mm prints of both the 1982 and 1991 versions before. Perhaps if I saw a side-by-side comparison, I might have seen a difference in the color cast… but just watching the Blu-Ray on its own, the only thing I noticed was that how good the whole thing looked.

Remember that if there’s just a consistent overall color cast difference between two copies of the same film, our eyes will adjust for it somewhat. Just try staring at a white sheet lit by 6500 °K light for a while, and then quickly move to a room lit by 2700 °K incandescents. Everything looks yellow now, because your eyes have auto-adjusted their white balance to 6500 °K. That’s why side-by-side screengrab comparisons tend to exaggerate color cast differences.

Personally, I was never a fan of the 35mm-to-15/70mm DMR process. If the film was shot on 35mm, I’d rather watch a good 35mm projection of a good print. Or a 4K digital projection.

That said, some of the optical, non-DMR 35mm-to-5/70mm blowups I’ve seen have looked really good. When IMAX came up with DMR, they introduced all sorts of digital grain reduction and sharpening. I understand that those things might be unavoidable to make the footage watchable on a 130-foot screen, but I still prefer it on a smaller screen – with nice, natural grain instead of DMR artifacts.

Most of the DMR releases of Hollywood blockbusters were always 15/70mm only.

Actually, I suppose that there never was such a thing as a 8/70mm DMR. IMAX DMR is a trademarked process which has only been used for making 15/70mm prints and digital IMAX DCP’s (although I suppose those are two quite different processes really). So the Disney 8/70mm blowups were just… well, blowups.

Sure the 35mm shots were put through DMR after being scanned at 4K resolution from the timed master positive. The IMAX shots were probably 15-20 minutes worth of footage 2nd generation contact prints at all IMAX theatres, or scanned at it was at least 8K for SFX work and then output back to film and contact printed.

Brian was talking about Indy 4. There never were IMAX prints of that one, so when he said ”35mm simply blown up”, he probably meant actual 35mm projection on an IMAX screen. No wonder that was soft; it was just a 2K DI film, to make matters worse.

While some IMAX theatres are running a smaller frame size, believe it is 8-perf., the vast majority of real film installations run 70mm 15-perf.

All film-based IMAX projectors are 15/70mm. 8/70mm is a separate format, used in science centres etc. and not branded by IMAX, even though a lot of 15/70mm ”science/ride” films also had 8/70mm reduction prints made.

Most of the DMR releases of Hollywood blockbusters were always 15/70mm only. Disney made 8/70mm prints of The Beauty and the Beast, Treasure Planet and The Lion King, but those are the only feature films for that format as far as I know.

Regarding The Dark Knight Rises, the producer said in December that there will be 40–50 minutes of IMAX-shot footage. Also, there was some talk earlier that Nolan is considering shooting the rest of the film in 5-perf 65mm – I don’t know what’s the latest news on that but it sounds good. B)

At least this article reveals no DI details – only the fact that 6K HMI’s were used. ;)

So it seems they also shot some 8/70 mm, and even some Super 35 for the 1.40:1 sequences…

Does anybody know about the post process? Somehow I doubt that they went the Nolan route, but I hope that they at least didn’t ruin their 15/65mm footage with a puny 4K DI (like they did in Transformers 2).

No 3D. 25 or 27 minutes of 15/65mm footage, the exact amount depending on who you ask.

Opening next week on 200 IMAX screens a week before the ”regular” premiere – those will be both 15/70mm and digital, so be careful!

It looks like the 15/70mm preview screenings will all include the Dark Knight Rises preview, too. (The link only mentions USA and UK, but it will also be playing at the Kiwi IMAX, so I guess that means all 15/70mm screens globally then.

IMDb says it’s estimated at $40M.

I read somewhere that they used Vantage’s Hawk V-Lite 16 1.3x anamorphics. The low squeeze factor still gives you the flares, but there is much less distortion in the out-of-focus areas compared to 2x.

For example, since the projector lens is usually very far away from the screen compared to its distance from the film it's projecting, the focal length of the lens must be measured in meters, right? So how is it able to focus so closely and yet project a focused image so far away?

Projection lenses are usually within a range of, say, 35 mm and 150 mm. You have to remember that the projected image is also several meters wide.

And on the topic of film projectors, I'm interested in understanding how film projector shutters work. Is it just like how the film was exposed, where there are 24 frames projected ever second, but each frame is only shown for 1/48 of a second, followed by 1/48 a second of black, and then another frame, etc.? If so, then over the course of a 2 hour movie, are we technically watching 1 hour of dark screen, if for every 1/48 a second of visible frame there is 1/48 of a second of black screen?

Yes, exactly. The 48 Hz flicker is indeed visible to some people, and because of that, better projectors use three shutter blades to achieve a 72 Hz flicker.

From a color correction standpoint, there would be a few side-effects to correcting and projecting at 7-11FL.

Thanks for the insightful post, and sorry for hijacking the thread! It’s just that ”it’s too dim” is still the most common complaint that I hear about 3D. So I’m still wondering about the side-effects to grading at 3.5–4.5 fL and projecting at 7–11 fL anyway.

I think Technicolor still makes small handfuls of over/under prints for release in obscure areas, but they never have been touted as being anything worth looking at.

Technicolor is actually marketing an improved version of their over-under system now, as a cheaper alternative to digital 3D. Not sure how it’s selling, though. Quoting a salesman: ”According to independent research in North America and Europe, audiences rated Technicolor 3D quality and viewing experience comparable to digital 3D”

I think that correcting and projecting at 11.5 FL thru the ExpanD glasses would look amazing.

Yeah, I guess that goes without saying! B)

I absolutely do realize the inconvenience of providing several DCP versions, each graded for different light levels. That’s kind of against the spirit of the DCI spec anyway; the idea was supposed to be that a single screening copy should work equally for all venues worldwide…

But I was just wondering if we should project common DCP’s at higher light levels or not. I know those are graded for a much dimmer projection, but I can’t really say I’m seeing any drawbacks when increasing brightness to 7 or even 11 fL. What kind of problems should I be looking for? Gamma, contrast, saturation…? Someone explained to me that increasing projection brightness increases perceived saturation and lowers perceived color temperature. Would you agree?

I think that 3D has not had a real chance to be widely accepted since there is no real 3D projection standard in terms of brightness. We researched this completely before doing our 3D color correction for Captain America. The lab recommended a projection level of 4.5 foot lamberts as a good starting point. Since Marvel had a tough time with Thor, which was color corrected at 4.5 FL as far as I know, we color corrected Captain America at 3.5 foot lamberts.

The lack of standard light levels for 3D is frustrating indeed!

I’m working at a cinema where we can currently reach 11.5 fL for 3D, measured through glasses. Both our projectionists and our audience seem to much prefer the brighter 3D image on pretty much any film, compared to the 4.5 fL ”standard”. We’re using XpanD 3D and a 1.4 gain screen, and with this setup we’re not seeing any artifacts from the added brightness.

Now, I do realize that ideally we should be projecting all films ”exactly as intended”, including matching the projector’s brightness to the light levels that the film was graded on. And I know that at least Disney explicitly says not to project their films any brighter than 5.5 fL – which is still way too dim, in my opinion. I guess that one of the reasons behind that recommendation is that over-bright projection causes problems on RealD silver screens, that are unfortunately popular in many countries including the US… On our screen, however, something like 7–11.5 fL for 3D looks just fine to me. If there are any problems related to gamma, saturation etc., the added brightness seems to by far overweigh them.

This seems like a great opportunity to discuss this with a Hollywood pro, so here goes: Shelly, did you test Captain America at high light levels (let’s say from 7 fL upwards) on an XpanD system? Did you see any serious problems, and if you did, would you care to elaborate? Would you object to see your film projected like that?

I’ve heard it’s 5.5 fL.

It says in the article:

The Wild Oceans footage was filmed out from a 4K DI using a CRT film recorder (the only type that is available for 70mm film)

Now, does that imply that 4K is the highest film-out resolution for 70mm?

I like how the test made sure all the IMAX material was 4th generation; as we are well aware most IMAX content is 2nd generation contact printing.

I for one wasn’t. That would certainly explain some of the results.

BTW, I saw Wild Ocean in 15/70mm, and it was far from being the sharpest example of IMAX I’ve seen. While watching it, I actually thought that some of it must have been 8/70mm and/or 10/70mm. For what it’s worth, IMDb says it was indeed shot on 15/70mm but went through a DI (4K?), which is probably the reason why it was softer than usual.

Scratch that last bit: as it says in the article, both clips were actually scanned at a whopping 11K. It's just that the Wild Oceans clip went through a 4K DI.

Sorry to dig up an old thread, but this seemed to be most relevant here:

LF Examiner: 4K vs. 15/70 shootout

It seems like 4K holds up surprisingly well on IMAX screen after all. Please note though that this is about 4K projection, not 4K acquisition. Both test films were shot on 15/70mm; one of them was a photochemical print and the other was scanned at 4K.

Don't know about series II. Most projectors in my local market are 1.4K, and they look it.

OK. I guess that’s what you get for being early adopters. Anyway, full-res 2K 3D has already been here for a while (even before Series 2).

Sounds like with your digital remark you're trying to start a war. Normally I'm down to play

So I’ve noticed. ;) However, I’m a pacifist, and I was only pointing out that the digital-originated bits of TF3 are ”true 3D” in the sense that they actually used stereo cameras – as opposed to the 35mm bits, which they post-converted.