Peter Moretti

Wolfgang, et al, How important do you think is having HD native resolution? Eizo's CG222W has just about everything the CG241W has, except it uses a 1680 X 1050 resolution. Should that be a deal killer? I'm tempted to drop $1.4K on one of these and use it until things sort themselves out more. Realistically, I'm not going to be doing my final grade for quite a few months. It'll be a big step up from what I'm using now.

Daniel, If there is one thing I've learned, it's that it all totally depends on many factors. Even though HD TV's should decode with the Rec 709 gamma, there is no way to tell for sure w/o using a monitor probe. That's b/c what looks good and sells more TV's versus what is accurate can be at odds with each other. The internet uses sRGB, so grading for that should be much easier, b/c it's using the same color space as your computer. You should still calibrate your monitor and try to set the black point as low as you can get away with. There are just too many links in the chain to get a definitive answer. I bet the best answer is to go with something like an eCinema monitor which was made with broadcast in mind. (The Quato broadcast monitor is not available in the US.) But short of that, we are all just trying to save a few grand here and there until the picture becomes clearer, LOL.

Well, actually, my Spyder 2Pro does have a setting for Rec 709 calibration. How accurate it is, IDK. I also don't know how accurate Avid's composer window is. Does it add anything to the color that the ColorVision software does not account for? But it looks like it is possible to use a calibration probe for Rec 709. Again, how accurate the calibration will be is still a VERY open ?.

Wolgang, The problem I believe isn't so much that modern high-end LCD's like an Eizo CG or a Quato can't display the proper color, black and grey scale levels, it's unclear how to get them to do that. We say proper calibration, but what does that mean? I send RGB signals through the profiling software and its read by the probe. But aren't those RGB colors sRGB colorspace? That is different from Rec 709, which is used for HD. http://www.13thmonkey.org/~boris/gammacorrection/ So when you calibrate a monitor, how do you send Rec 709 color values to it instead of sRGB? I have Datavision's Spyder 2, and there is no Rec 709 option, AFAICT. Phil, Actually, this question goes for broadcast monitors as well. How do you properly calibrate a broadcast monitor if the calibration software is sending sRGB values to it? Thanks very much for all the help guys.

So if I understand correctly, the gamma correction is performed by the computer/video card/(possibly monitor) but not by the NLE. So when I choose the Rec 709 color space in Avid, Avid uses the color range values and luminence weigthing for 709 but not the proper gamma correction?

I've read that a lot people recommend grading with a broacast monitor. But it seems to me that a properly calibrated high quality computer monitor might be at least as good (if not better). I believe by using a computer monitor--instead of a broadcast monitor with HD-SDI--one cycle of RGB to YUY/4:2:2 to RGB conversion is eliminated. Here's a diagram of the path of the video stream assuming Avid. (I know Avid is hardly recommended for serious grading and CC mostly because of its long render times and somewhat crippled CC tools, but I think the comarison holds no matter what software is used.): Computer Monitor DNxHD/YUV/4:2:2 >> Computer Video Card/RGB >> DVI Out/RGB/Uncompressed >> Computer Monitor/RGB Notice only one YUV to RGB conversion. Broadcast Monitor w/ HD-SDI DNxHD/YUV/4:2:2 >> Computer Video Card/RGB >> HD-SDI Out/YUV/4:2:2 >> Broadcast Monitor/RGB Notice a two extra conversions; one RGB to YUV (to meet HD-SDI's specs) and one YUV back to RGB (by the broadcast monitor to convert the computer's RGB color space to YUV/4:2:2 to comply with HD-SDI's specs display the image). Does this makes sense? I'm still learning about all this, so I could very well be missing something. Thanks MUCH.

And 3-D... YUCK! :blink: :ph34r:

If you want a camera you can grow with and have tons of controls to learn about and master, the XH-A1 is a brilliant choice. HDV isn't supposed to handle movement well, but Canon has done a great job with HDV. And the XH-A1 has CCD instead of CMOS sensors, so you won't have the potential problem of fast moving objects changing shape due the "rolling shutter" effect. The XH-A1 is GREAT. I really don't see how you can do better... as long as that size camera is acceptable for what you are doing.

Just interested in hearing your thoughts on Avid DS as a finishing and color correction platform. I realize DS can cut, do titles, audio as well as effects and CC. But don't know how it stacks up against more specialized programs like Smoke, Lustre, SCRATCH and DaVinici. Thanks much!

Which makes me wonder what lenses will be used with Red's upcoming Epic, which will have a 35mm SLR full size sensor. I've always assumed that cine lenses could fill a 36 X 24 negative, but the fall off would be significant, so the smaller negative size uses the sweetest spot of the lens. But perhaps I'm mistaken.

That's funny John, I was thinking the exact same thing. Essentially use some thermal paste to glue heat fins (like those used by computer overclockers) to the Red body.

Hi Karl, I read the articles and didn't catch the part about delaying the focus pulls, but I could have easily missed it ;). I'm not saying it wasn't a stylistic choice, I just found it odd it happened in only two scenes--for no clear (pun maybe intended) reason, IMHO. ... Well, I guess you could say it helped to convey the chaotic and mecurial nature of the Joker. I can buy that.

I am almost afraid to say this, but I really believe the focus was significantly off in a few scenes. And I'm surprised this hasn't been brought up. The shooting of the Joker, both when he crashes the gala fundraiser and when he's being interrogated in the police station, goes out of focus often and considerably, IMHO. I don't see how this could have been a projector problem. It looked like the focus puller had a hard time keeping up with the movement; perhaps marks weren't really followed in the takes that worked best. I very much liked the movie and felt almost empty afterwards because of the sorrow in its theme and the passing of Heath Ledger.

Brian, I realize that DV and DVCAM are identical in terms of compression. But there are still big differences between a 570 and PD150 that go beyond chip size. Probably the most notable being the 570 will have a much nicer lens. And that, IMHO, is a lot of what you're seeing if you had to de-focus one of the projectors.

Rob, I'd assume you'd import QT files from Redcine into Vegas. Then create an EDL and use that in SCRATCH to capture .R3D files. But of course ask others.

Maybe so, but there is an "issue" with Red footage not looking like one would expect b/c of the philosphy to leave much of the image processing for outside of the camera. Looking at completely unprocessed Red footage can be a bit of a shock if not viewed under the proper circumstances and with certain expectations in mind.

John, I'm not really sure that this is true. By having the different color channels, in post you can differentiate different colors which happen to have the same intensity. For example a green and red striped shirt may not have enough differentiation between the stripes. But if you keep the color values, you can adjust each colors' B&W intenisity in post. Make the green a little darker and red a little lighter. Now the stripes will show up in B&W. Converting from color to B&W can be as easy as setting the saturation to 0, or it can be rather complicated, where you empahsize certain colors in the conversion while de-emphasizing others. The differences can be very significant. They can make a subject look warmer, harder, older, etc.. So when planning a B&W finished product, you should still shoot in color, IMHO. BTW, don't use colored filiters either, b/c they remove color values, which is information that you may need. Neutral density or graduated density filters are fine. And generally try not to clip the highlights, so use a histogram or zebras.

I've got a script for you: http://www.hollywoodbookcity.com/MovieScri...e_MV101099.html :P

Thank God for those people, they're good for the economy ;).

If you are saying that the camera, or whatever piece of equipment, doesn't give some talent; of course I agree. But the back and forth in this thread has largely been about are people are biting off more than they can chew when they plan on wearing a lot of hats? And does the Red play into that psychology by offering a 3+K image at cut rate price (relative to other cameras)? All I'm saying is some people may be very good at one thing and not at others. It takes special talent to be a producer, editor, director, dp and so on. But they are not really the same talent. I know a girl who is a wizard at editing, but try to get her to plan a shoot or talk to actors and she withers. But there is a place for her in filmmaking--and using your logic she should go into another field. I've also met directors who still don't know the difference between HDV and HD. You think such directors can edit in... anything? You'd have better luck having the chimp house in the San Diego Zoo pound out Shakespeare. Some people can wear a lot hats and succeed. But those who only excel at one thing can also do very well.

So you are saying a single person should be able to: write, cast, direct, shoot and cut a watchable short using a handycam with amature talent and iMovie/Windows Moive Maker? Every now and then, yes. But lets be honest, specialization exists for a reason. I'm sure there are many fine DPs out there who can't direct or use FCP/Avid or write a screenplay. Spielberg's DP Michael Kahn use a Moviola. If he can't use Windows Movie Maker, are his three Academy Awards are undeserved?

I think if someone set out to make video look like film on SD broadcast TV, they could come very close and fool a lot of people. But they'd have to have that goal in mind. I agree a lot of it has to do with color representation and how highlights are handled, but these can be fixed to a very large extent by a good colorist.

Come on Mitch, I gave prices for what I included. What I left out obviously isn't included. And monitors good enough for grading is a whole can of worms that there is no way I can put a price tag on. Some people say you get away with some used CRT's on eBay for a few $hundred. Others say eCinema's LCD's are acceptable for $6K. Others say you need something like a Sony broadcast monitor for around $10K. And I have no idea if he's going to try to truly finish and make a DI himself or just do a really well CC'd offline. And there is the whole audio side that we didn't even touch. I feel my prices were accurate for what I listed. If someone wants a detailed list for an entire post production facility geared towards Red footage, then I agree, there is a lot more to it than two computers and a copy of SCRATCH.

Assuming you own the Red: $16K for the Red version of SCRATCH $10K for the SCRATCH workstation $6K for the FCP workstation $32K, IMHO.

Red is definitely a work in progress esp when it comes to postproduction, and even more so when it comes to Avid. The Red camera records .R3D files, which are "RAW" files that no NLE can presently use, FWIU. .RD3 files look like crap to the average viewer, but to colorist they look LOVELY. They are very gradeable. A philosophy of Red's is to remove image processing from the camera and put it into the computer. You use another program, called Redcine, to convert .R3D files into ones that can be ingested into FCP or Avid. Redcine Quicktimes can be imported into FCP. Avid's MC can also import the Qicktimes, but the timecode will be gone. There is a program called Metacheater that fixes this problem, and Avid is supposed to be working on a fix. Avid's DS can import Redcine's DPX files, so that's another route to take if you work with DS. I believe Assimilate's SCRATCH can import .R3D files directly. This area is constantly moving, so I hope the info. I've given is accurate. I belive it is.