Michel Hafner

The July American Cinematographer has an article about "Dark Knight". Apparently the 35mm negative was printed to IP for color correction and then scanned at 4K for the DI for the IMAX prints. How exactly the regular 35mm prints were done is not revealed. But my question is, why scan an IP at 4K and not scan the ON and do the color correction digitally? What is there to be gained this way to outweigh the loss of detail from ON to IP when creating an IMAX print? The 4K then had to be degrained and sharpened, of course. Also, the IMAX contact prints from the IMAX ON were described as 18K. So the final IMAX prints are from sharpened and degrained 35mm IP at 4K, IMAX ON, IMAX ON via 8K scan and 5.6K scan (?). Should look 'interesting'...

That's pretty much a given for CGI material created in >= 1080p. Film prints can't hold the high MTF while 1080p data with 1;1 mapping to 1080p monitors can.

For anyone interested in Bollywood: Sony has released (Saawariya (2007) on Blu Ray. And it looks fantastic. Obviously direct digital from the DI source this is a gorgeous disc. Unless you hate the film highly recommended.

I watched the Blu Ray and liked it pretty much overall. But it's the second Genesis sourced BD I have seen that has issues with strange aliasing artifacts (the other is "Next"). Some of the aliasing is the normal kind around high contrast full res edges, but there is also very visible wobbling of vertical lines at times. I can't believe that this is the kind of picture coming from the Genesis. Looks like a post problem or disk mastering problem. Anybody know more about this wobbling and its source? There is a featurette about the film(ing) and Lumet makes some statements in it about HD versus film that surely must have made some fellow heads turn purplish and shaking with anger and disbelief. :lol:

It's OT, but the last depressing development on Blu Ray are the releases of "Patton", "Longest Day" and "Gangs of New York", each a complete failure in their own regard. Patton and Longest Day were changed from film into grainless/soulless wax corpses and Gangs was gangraped with edge enhancement and DNR. The first two got almost exclusively great to rave reviews while on the third some reviewers noticed that this was a piss poor transfer. It makes one wonder how powerless or uninterested/unaware directors like Scorsese (!) are, if they can't/won't stop a release like this. I hope Scorsese can at least have it recalled and reissued properly.

It looks like the Blu Ray had some DNR applied, at least in some shots (shots with Marie Richardson).

I think I'll pass. Film high priests and digital acquisition high priests have become equally obnoxious, while the reality of the quality reaching the audience is sobering in most cases.

I would have preferred an answer from the author of the claim, together with some references to papers. Helas nothing was forthcoming.

Brainstorm...

Including Tarkowski, Bresson, Dreyer and Angelopoulos?

No answer. The answer is available in scientific papers about the issue. 35mm film does not have 10-12K resolution by any accepted standards of measuring resolution. It ends under ideal conditions (measuring static test charts with optimal focusing and exposure) somewhere between 4K and 6K. What's going on > 4K is very subtle and can i no way survive a single analogue copying step to an IP.

No, not at all. I said 2K is not as good as EK prints in some regards. EK prints are not available to > 99% of the audience. If you read carefully what I say you will see that I have a far more differentiated opinion of the issues than simple good and bad verdicts. You come across as a rather single minded anti digital person, though. I guess it's this which kept me 'coming back for more'.

It was not built for that so it can not do that. That's the whole point. Because it does not have to be a light cannon it can do a much better job with On-Off contrast. Image size is not everything.

http://us.imdb.com/SearchTechnical?for=Red...o.y=8&Go=Go (ignore the obvious errors, looks like a bug)

I find watching anything on most LCD monitors distracting. Badly implemented display technology (in the consumer section, professional monitors are better). And I completely reject the call for DNR on HD transfers as soon as it goes against the authentic look of a film. I want to see films as they were made, not digitally mutilated versions with no more HF detail, waxy skin and worse. People who dislike grain can use the noise reduction in their display chain (and reduce the sharpening and brightness to correct levels) and/or stick to originally low grain/noise material (film or HD camera).

It looked more like a source problem to me than a DLP projector problem per se (except more contrast would always be welcome). In my experience a 35mm print from the same DI looks even worse and the HD on my home projector better than both.

Compressed 1080p will have film grain and some compression noise. How much of each I would not dare to guess for specific examples witout seeing the uncompressed version. But there should not be a lot of compression noise on well compressed high bit rate AVC or VC-1. As I said, you see grain in 1080p and 2K, just not all of it. There is more in 4K. It's not the same although both often use AVC, Trailers are not hand tuned and use lower (+-constant) bit rates. BDs are hand tuned and use variable bit rate and higher averages with peaks reaching > 40 MBits/s at times.

I caught a 2K DLP screening. For the first 15 minutes or so I thought it looked pretty poor (mushy, no detail, no depth...), then it got better, but rarely looked as good as I would expect from today's DI work. I'll enjoy it more at home from BD disk on a higher contrast projector.

What exactly are you talking about? The quality of the end product or the mere survival of certain technologies and the companies clinging to them? If you talk about the quality of the end product do you mean technical or artistic quality? I would have to disagree that with current digital still photography or digital music recording you are getting a technically worse product than you got with film stills or analogue music recordings, at least as far as the technology itself is concerned (as opposed to how people use it). Concerning the needless complication you got a point in a film world with no digital effects and no need for making anything but prints directly from the OCN. But this world is not today's world. The digital part is here to stay and making prints from OCN for everybody is totally out of the question.

Digital projections are indeed of varying quality as are the sources used. In my personal opinion 2K projectors don't have enough resolution to make a picture that looks as analogue and free of digital artifacts as a print from the OCN looks once you sit close enough that your eyes resolve down to the pixel level (which is realistic wth 20/20 vision and sitting in the first couple of rows/sitting <= ~1.5 screen heights away). In addition the On-Off contrast which is stuck at ~2000-2500:1 these days provides for quite elevated black levels and a flat looking picture with low APL material. Whereas digital projectors for home cinema go as high as real 30000:1 now and do far better with dark material. 2K projectors simply can not make sharp pictures without showing aliasing/jaggies. Sharp as in sharp looking on big cinema screens. Everytime it looks really sharp (as in white letters on black ground during credits) the jaggies show up. If they don't the letters are somewhat fuzzy. One or the other. The same goes for sharp textures and edges on real footage. Digital has flaws as has film. These days you can do digital in transparent ways. The DI can capture all relevant information from the negative. But each transformation step has a lot of potential to degrade the quality and there are many steps from OCN to print from DI. I don't like the DI route with film prints in the end at all. But as long as we have no digital projectors that can really render 4K sources as they are there are valid arguments to continue to make prints (in addition to the economic realities of existing cinema infrastructure and the archival requirements).

How do you know? How do you tell real grain apart from video noise without access to the uncompressed master? Of course you see film grain in 1080p as you see in 2K. Just not all of it.

That's an interesting claim but without further details rather puzzling. Where exactly is that inferiority supposed to crawl in? Is it the digital projectors? But then why would the same projectors suddenly make a beautiful picture with an originally digital process providing the source (whatever he means here: Computer cartoons/CGI? HD material? Dalsa?)? If it's not the projector it must be the source. But the source format on the data level is the same, whether you fill the bytes with information coming from film elements, computer files or digital cameras. So if it's not the source format it must be how the information is put into the source. Is Spielberg saying today's scanners are no good and can't convert the information on the film properly into digital form? What exactly does he mean?

I'm sorry, But by posting here in a public forum you agree your views to be seen by others and you invite others to an exchange about the subject matter. If you don't like an argument, why are you here? Nobody is telling you you can't judge/feel/argue... one way or another. But if I'm asking for some details it's hardly appropriate to pull the authority hammer and feel 'attacked'. We have all our experiences and preferences and try to make sense of them. In your opinion. Sure. I dislike aliasing as well. But it's a take your poison situation. The film print has aliasing too if it comes from the DI, plus additional film artifacts. If there is no DI the standard film print has no aliasing but lack of sharpness and detail, more noise/grain and likely color and contrast deviations compared to the IP and even more compared to the graded negative (if there were such a thing in the analogue world). You will see artifacts in any presentation if you don't blend them out. Your tolerance of one type of artifacts over another will decide your preference. I advocate to do the whole show with DI or at least for the digital master use only the DI parts and scan the rest from the IP (with this rest coming all directly from the OCN). As I explained already. For practical reasons this is usually not done, as far as I know. You are barking up the wrong tree, I'm afraid. I'm neither a digital nor a film fanboy. I have seen pretty much all there is to see from 8mm to IMAX and Showscan, from DV Cam to Dalsa 4K, from EK prints to normal prints and digital projection from 1280*1024 to 4K, from VHS to LD, DVD and HD. I think I have a good idea what looks how and what to make of it. The world is not black and white for me in the format department. What about you? Thanks, but DI is not HD 1080i and a reasonable discussion makes a distinction between different sources and projectors at the very minimum. The quality split is not between analogue and digital but elsewhere. If you prefer your standard prints to a 2K digital projection that's fine with me. I find both far from optimal. The standard prints look simply bad compared to the best prints you can make. And the digital projections in cinemas lack all good blacks and image depth with dark material (the projection, not the digital source!). Aliasing is frequent as well. For films with no DI I like best high quality prints from the camera negative (rarely available to me). For films with DI a 4K projection of the DI.

Not even Kodak makes such a claim (e.g. you need 10K or so to get all information from the 35mm negative). So based on what tests are you suggesting this? Fact is the MTF goes for all intents and purposes to zero somewhere between 4K and 8K for 35mm (and is quite low between 2K and 4K already), So what kind of relevant information do you think you can find in the 8K to 12K range?

Changing ASA on RED merely changes metadata or how the RAW data is interpreted in post or the monitoring output. It changes nothing about what is recorded in the RAW data unless you fiddle with your exposure as well to do 'the right thing' with the ASA you chose (whatever your idea of 'the right thing' is).