Michel Hafner

Yes, the Sonys reach 30000:1 with cheating, Epson claims 60000:1 with cheating. These machines have native <= 10000:1. The JVCs native 10000:1 to > 40000:1 depending on model, color temperature and zoom lens 'position'.

DLP and SXRD cinema blacks need to get better indeed. What is the same as saying the On-Off contrast of these digital cinema projectors needs to go from ~2500:1 (current state) to a multiple of it. But film print's blacks are not exactly stellar either. They are usually in the same ball park, stuck somewhere between 2000:1 and < 10000:1. The best blacks are available from CRT projectors (for (very) small screens) and LCD variants (SXRD,DILA) for larger screens. Available now for home consumption: Native and real 30000:1 and more (JVC).

No. Anyway, there are many examples out there on the net with the same stills from the DVD and the HD, the DVD upscaled to 1080p for comparison. There is no comparison if the HD is high quality. If they would look basically the same HD would look basically the same as SD. And we all know that is not the case. Either that or Blu Rays suddenly lose their HD detail when they are mastered from the studio tapes. That ain't usually the case either. But there are old transfers around with barely 720p detail and some films are just not that sharp to begin with. So the difference is not always big. Even some upscaled SD has been put on HD disks. That's looks marginally better than DVD, of course. Appreciating HD detail depends a lot on viewing conditions, though. Some examples here: http://chidragon.thedessie.com/bdcomp/

Hm. I have seen maybe 300 HD discs the last 2 years and there was not a single one where the resolution difference between the HD and the DVD was not obvious to brutal. No, I'm not saying I watched the same 200 on DVD in parallel. But I know how DVD looks on my system, even the best DVDs. It was not an LCD monitor, though, but 1080p projection onto 3.5m wide screen.

(you must have sit practically next to me if you were in the middle) Only the 'good' 4K RED shots were visibly better than any 2K material, but it shows the potential when people go for 4K and sharp lenses and optimal post production. There was also one shot that looked barely 720p. And most shots could benefit from more contrast from the Sony projector. The degrained 16mm was amazing. First time I really liked the look of degrained material and saw no artifacts at first sight.

I loved almost all of the footage too. Some shots were jaw dropping. My favourite shot was a black man addressing a crowd in a hall, shot from below. Noiseless 70mm clarity. What a difference to all the examples shown on the 2K DLP.

You can have the Blu Ray for much less. It's supposed to be a good transfer.

The 2K DLP screening looked fine. Neither very soft nor grainy.

The Blu Ray of this from the DI files looks remarkable. Practically noiseless, squeaky clean, free of aliasing, no DNR, no obvious sharpening artifacts, well compressed. Images are graded flat at times and skin tones are sometimes unnatural looking (yellowish, greenish, golden...). Grading/shooting decision or Genesis problem, who knows. Overall a demo disc for a certain look different from film and regular HD camera as well. The opening shots in nature are gorgeous.

The 35mm parts on IMAX prints are 4K scans from the 35mm IP (!), degrained and sharpened. The IMAX sourced sections are 8K scans for sfx work and the rest contact printed. So the highest and lowest resolution is second generation IMAX versus second generation 35mm with digital processing. The difference can't be subtle at all.

I watched some Metropolis from PAL DVD lately and was tempted to mute it as I found the non stop orchestral score going on and on distracting and drowning the images. It's supposed to be the original music but I really prefer the solo piano version of it playing with the film, or no music.

One thing that is often forgotten is that what you actually see on the screen, even if projection were lossless, has very little to do with what happens on 35mm in the >= 3K and <= ?K range. That range is marginal for what you see. Much more relevant is what kind of modulation and noise you have in the range upto 3K, which defines how sharp and detailed the image looks you are seeing (on normal cinema screens, not IMAX). Film acts a bit like SACD with DSD recording, while digital cameras are PCM recorders. So whether 35mm is 4K or 6K or 8K is largely of academic interest. What you see in a cinema changes a lot once you see 4K with a MTF staying high till 4K is reached.

I'd love to get my hands on newer studies. Are there any? What is the consensus (or lack thereof) on the DCI stem tests?

I was talking about spatial resolution, fine detail, lack of noise/grain (that is the part that makes people talk about 70mm look). I can't comment on the shading detail compared to 35mm or 70mm, native or via digital projector.

No. I have seen Red footage with 4K projection and my eyes tell me if this was footage with 1K effective resolution (16mm or worse) or rather excellent 35mm resolution if not 70mm like.

I smell a misunderstanding here. There is a postulated loss of 30% modulation on MTF test patterns. This is not the same as a general loss of 30% resolution (any kind of resolution). And it does not mean either that a normal print has 10% resolution left in the cinema. :lol: The numbers are not mine, They come from scientific studies. A rebuttal where necessary must come from scientific studies as well. I'd like to see some, if they exist. Current graphs as here (www.digitalpraxis.net/zippdf/di-guide.pdf) show a pretty steep loss from ON to IP and then it tapers off. Hence my question why to scan the IP for an IMAX deliverable. Visual examples are here: http://digitalcontentproducer.com/mag/vide...inemas_special/ Quite telling, isn't it, unless you want to argue these examples are not reflecting reality because,,,

Don't you mean 35mm has a long way to go till it has, cough, cough, IMAX or only 8K resolution?

What has this got to do with digital projection? It's what's on the film elements, The (analogue/digital) projection makes it even worse, of course. There are different kinds of resolution. There is spatial resolution, there is color and gray scale resolution. And there is color gamut and dynamic range. They all interact to some degree. I was talking about spatial resolution. How you resolve fine SPATIAL detail in textures, how you resolve egdes. The MTF refers to the loss of resolution in this regard brought about by loss of modulation of frequencies. It's not directly about the other types of resolution. And it's not about latitude or dynamic range either! The shadow and highlight detail argument is about dynamic range and latitude, not spatial resolution and MTF. Don't be so polemic. That projection and typical real world conditions cause further loss of spatial resolution and modulation is a given. Does not change anything about what's going on when you copy from one film element to another. I know why I cherrish my home cinema where I call the shots concerning ambient light, dirty lenses, misfocused lenses, babbling, cell phone wielding and popcorn munching patrons and other potentially disruptive interferences to my viewing pleasure. :lol:

We are talking about different things. I'm talking about spatial detail, not shading detail (how many different shades of colors between color x and y, what primaries, what gamut). Prints can show colors the 1080p can not because it's limited to HD primaries. Agreed. And normal prints can show more colors in between 2 colors as well since consumer HD is limited to ~8 bit per primary. The prints are not, But they are noisier and lose HF detail which degrades the color as well. It's not a clear cut comparison in overall quality. But in the spatial detail department the 1080p has an advantage with luma compared to regular prints. You can scan at 20K if you like, That's not the point. That oversampling is beneficial is very old news. I'm not interested in what the people scan at. I'm interested in the differences on the optimally derived 4K master versus an optimally derivd 6K master from the same 35mm elements. Project them at 4K/6K split screen. Any relevant difference visible? Project them at 6K (or a zoomed in section in 4K). Any relevant difference visible? But this thread was about watching Wall-E, a digitally created motion picture, in 1080p direct digital versus some print in a cinema. And in this case it's silly trying to deny that the 1080p has better luma detail and sharpness than some normal print.

I would think Lucas and Spielberg get it their way too. If not out of legal reasons or respect then at least because of likely future collaborations.

Does it bother you to see some sources? As much as you hate sources, would you happen to have one to support your claim and prove ARRI wrong? After all the results are coming from measuring the prints, not just applying some hypothetical formula. Now, even if it were 30-15=15. Do you really think this is not visible on IMAX? Sounds to me like you are mixing up spatial resolution/modulation and shading detail/dynamic range. I was talking about the former, not the latter. Why should I have to chose? Wrong assumption, You are very good at moving the topic elsewhere when you hit a dead end. We were talking about spatial resolution, not how many shades of gray or color you get between black and white, and how much darker black is than white. Of course the latter is affected by the former as well. No modulation = no shades, no contrast.

Huh? You are mixing up production and media standards with display standards. 1080p is very much a production standard (HD master tapes of film transfers of the last couple of years are 1080p24, not 1080i60) and on Blu Ray as well a media standard. The film data is stored as compressed 1080p24, not 1080i60. The player outputs 1080p24 or 1080i60 depending on what your display can accept and what you tell it to do. Apart from that whether it's 1080p24 or 1080i60 with correct pulldown, after reverse pulldown the 1080p24 is the same if the 1080i60 is not more filtered than the 1080p24 (which it might be). > 95% of current Blu Rays of feature films are 1080p24. Most of these are original 1080p24, not 1080i60 converted to 1080p24. Don't know what Vertigo has got to do with this. It's a Vistavision film restored to 70mm. With 70mm and 35mm prints. I stand by what I said, Good 1080p HD is sharper looking and more detailed than standard prints, especially if made from 2K DI. The reason is obvious. The HD can keep the MTF of the DI minus some compression losses and color subsampling while the print had a film out stage and a copying stage added, removing >= 30% modulation. That's visible. The print might be about the same for some saturated colors, but not luma. With traditional non DI prints compared to 1080p from the negative the print has no chance.

Of coure it's stagey. The film opens with a shot of the city from above which is intentionally and very obviously a painting or CGI creation, signalling that what's coming is stylised, unrealistic and artificial. at least on the surface level. The same story was filmed by Visconti (Le notti bianche). A comparison should be interesting. Another film in the same vein is Beineix' "Lune dans le caniveau". Both films got heavy flak for their 'style over substance' approach. And both films are HD demo disc dream material (although Beineix' is not released n HD yet). And although Saawariya is technically refined Bollywood, it's still Bollywood. Song and dance is included and important.

It has less detail and sharpness compared to the camera negative. And no, not marginally less, IMAX is not television. IMAX is the (spatially) most resolving film system we have, a frightening loupe. The difference between IP and ON on IMAX is quite visible. It's also visible in 1080p, by the way. On IMAX everything on the 35mm negative is very relevant. And the percentage is not say 3 or 4% but a lot more. The factor for the MTF is 0.7, e.g. 30% loss of modulation! (http://efilm.com/publish/2008/05/19/4K%20plus.pdf). I would also suspect that if they did the 35mm version optically, out of convenience they used the IP, although it should be quite feasible to use the ON for the 35mm parts of the IMAX version and get the same colors with better spatial resolution.

Blu Ray is mastered 1080p for almost all feature films. What's that fascination with 1080i? And yes, good 1080p projected with a 1080p/2K projector often looks sharper and more detailed than standard prints projected in standard cinemas, on the same screen.