Re: [Frameworks] Forbes editorial about Kodak

From: Alistair Stray <>
Date: Sun, 9 Oct 2011 09:40:33 +0100 (BST)

> From: Aaron F. Ross <> > Definitely good points. However, don't forget that any film stock can > now be emulated, given good enough digital source material.   No it cannot, not remotely. I work in post, and have done for some time, with both film and digital source and this is not true. The mediums have a wholly different look, you can make digital look a lot like film, but you'll not get a true and accurate emulation.   >As I said before, the moment that HDR sensors become affordable, then celluloid >will be irrelevant. If you start with 20 stops of latitude in a >32-bit floating point color space, you can push or pull it wherever >you want and the end result will be indistinguishable from footage >shot on the stock of your choice. --   Did you not read my post when you first started overstating the impact of HDR, and misrepresenting its purpose ? Yes the latitude of HDR will give you much more control in post but also you have to think about compression artifacts limiting what you can do. Yes you can work with uncompressed footage on the most expensive cameras, but as I said earlier in reply to another post of yours (also containing gross overstatements and misinterpretation) the switch to HDR sensors is a waaay off still. Surely as someone who works in Maya you can appreciate the difference in file size, and the difference in time that it takes to apply any processing at all working in a 32bit space,(assuming you've used a linear workflow and actually done some post other than tonemapping). Its a huge increase in production cost, even when the sensors become available large hollywood productions will not jump on a completely HDR workflow for quite a while (and when they do it'll
 probably be at 16bit, being as that is good enough). You're also kind of asking all filmmakers to shoot 'flat' to give the post guys all of that latitude too, HDR does have a start and end point you know, reality still has a wider dynamic range. One of the main advantages of digital film making is not really having to guess the final look with a video assist, but to be able to take the footage to a laptop on set and push and pull it there. With HDR that'll be a bit of a time sink without a considerably cash cost. Its not just the sensors that have to become affordable its the whole pipeline that has to become affordable, and also HDR has to be really advantageous at all stages in the pipeline too. Film is actually a much cheaper way to get a high dynamic range when you think about these issues. Most filmmakers here, if they do switch to digital will not be working with HDR footage or processes, their costs would actually increase dramatically.
 They'll be shooting on VDSLRs or P2s, or FS100s or equivalents thereof.   More importantly your final remark that it will be 'indistinguishable from footage shot on the stock of your choice' is completely wrong, it won't.     Also, the end delivery point of film is not, and will not be for a very long time, HDR projectors. You are still going to bring it down to a smaller dynamic range to ship and show it, so why not start working closer to the end result (as we do in DPX and Cineon formats at 16, 12 or 10 bit) to start with. HDR footage will mostly be used for VFX footage, to give compositors more latitude and control in post first with complex shots, its general use is some way off. Basically, even though HDR sensors/footage is coming, the audience are not going to be watching full dynamic range films for a very long time.   Another point, motion, this is very important. The look of movement on film is wholly different (and much better) to digitally shot footage. Maybe you should look at the rolling shutter tests in the Zacuto great camera shootout someone posted a link to here. Shooting with very high speed cameras such as the Phantom solve this problem (but you're not seriously going to be shooting a feature on one of those). Even the most expensive digital cameras suffer from the rolling shutter problem. Yeah, alternative approaches are on the way to solve this, but they aren't here yet.   I love shooting and working in digital. The flexibility it gives me is way beyond anything film could ever provide me. I know my medium very well and it will not, ever, look the same as film does and nor should it ever try to. It will replace film eventually, but it won't ever look, or respond, the same way that film can. Go talk to a good DP Aaron who works in both mediums. Shooting in digital requires a different way of lighting, a different way of thinking, and a different set of technical limits to film. These differences will actually be more easily understood, and adapted to, by someone from a video background. In that fact you'll get a hint as to how wholly different the two mediums actually are and why digital ultimately has its own look and will never accurately emulate film.   From: Mark Longolucco <> > (not to mention a somewhat long and steep learning curve with film)   There is an equal, if probably not longer and steeper, learning curve shooting and working in digital Mark. Unfortunately most digital filmmakers can't be bothered to actually educate themselves fully.   - Stray.

FrameWorks mailing list
Received on Sun Oct 09 2011 - 01:40:43 CDT