Newbie question

[Brian Langeman]

I have yet to see a single frame of real-world RED footage. At least, that I know about. (Note: Downloads are NOT real-world footage.)

 

So far there has been sizzle overkill and very little steak.

 

"How does the RED compare with the D20 or Genesis?"

Good question. It probably gives a sharper picture, but for all practical purposes it will probably look very similar, since there is currently little call for >2K resolution. It is a much cheaper camera to buy, but it doesn't fit into existing tape-based workflows, which is a lot more of a hurdle than many people realize.

 

Eventually all this will become clearer.

 

Hello everyone. This is my first post, mostly cause I usually just read everything and make my own decisions, or learn info that way.

 

What do you consider real-world RED footage? I'd consider it anything shot for the purpose of being used that isn't a test chart or something. Why are downloads not real-world? You can download clips off of REDRelay.net that are good examples of what's possible on a RED. I'd consider two very good reels of what you can get out of a RED here http://www.mammothhd.com/MHD_RED5.html Watch those and decide if it's a capable camera. But in the end, you obviously need to test it yourself to decide whether it will work with the project that you have in mind for it.

 

One thing that's almost always overlooked with the high resolution is the ability to crop. You can punch into a shot by quite a bit if your final delivery is HD or even more if it's SD and still have FULL resolution on your final medium. And it gives more possibilities for post work like stabilizing shots cause you can afford to through away lots of edge pixels even if finishing in 2K. If you really want to get technical it's been said that the true full resolution it'll give because of it's bayer sensor is 3.2K which is still a lot more than the Genisis will give. Clearly resolution isn't everything and you've got to look at dynamic range and other things too, but the RED clearly has the Genesis beat on resolution.

 

As for workflows, that greatly depends on the type of project you are working on. I don't think the RED is intended to be a camera for news broadcast where companies have invested huge amounts on tape workflows. It really is a cinema camera first. That being said, take a look at this clip http://www.red.com/red_mythbusters If you're working with HDV at the moment, I don't see how it's ANY harder to work with the quicktime proxies which will give you plenty good quality footage if outputting to HD or SD.

[Brad Grimmett]

In Australia, US "football" does not rank that highly as something to be watched to be honest.

I believe it's called "Gridiron" in Australia. I hung out with a bunch of Aussie's when I was in Thailand last year and they were all quite fascinated by "Gridiron" and never stopped asking me questions about it. I've watched quite a bit of footie in the past and think it's great. But, like "gridiron" in Aus, footie is hardly ever on tv here, so it's hard to follow it if you're a fan.

 

Regarding Indy 4....it's very soft because of the amount of diffusion used.

Regarding the Red....it's getting used more and more, and over the next couple of years we'll find out for sure if it will be around for the long haul.

[Keith Walters]

What do you consider real-world RED footage? I'd consider it anything shot for the purpose of being used that isn't a test chart or something. Why are downloads not real-world?

Well, in my experience anyway, people who actually make their living from producing or shooting motion pictures, tend to want to see how their end product (or more to the point, somebody else's, using some new technology they are investigating) is going to look to the average end user. That means seeing it actually broadcast on PAL, NTSC, SD and HD Digital, digitally projected in a classy cinema, screwed up by inept projectionists in suburban fleapits, taped on VHS and so on.

 

Apart from that, advanced operators will need to be aware of what happens to commercial messages if the viewer fast-forwards, mutes the sound and so on. You may notice that some well-made ads still manage to convey their message under those conditions, whereas with others the message is completely lost.

 

Downloads tend to be "set pieces", often squeaky-clean and only showing you what the originator wants you to see. Real world footage is more likely to be what a client has told someone that wanted shot, not what the shooter thinks would look good on their showreel!

 

A second issue is that strangely enough, a lot working professionals (again, at least in my experience), do not seem terribly interested in providing themselves with gigantic and expensive Internet download plans. A common complaint is "download bloat" where users are forced to download hundreds of megabytes to get information that would have been conveyed just as well with something of YouTube quality.

 

The third issue is that, just as not everybody who talks about heaven is actually going there, not everybody who has to make decisions about shooting formats is a fanboy technogeek who spends 12 hours a day surfing the Net.

 

But all that to the side, none of this explains all the Nostradamus-type heming and hawing? If there are RED-shot ads currently being broadcast, why can't we simply have a list of them? How hard is that?

[Keith Walters]

To provide you with Oscar Wilde's original:

"To loose one parent may be regarded as a misfortune; to loose both looks like carelessness."

"Lose" not "Loose" :lol:

 

And while we are at the importance of being earnest on matters discussing digital video, the quest for actual footage, and the resulting opinions and posts formed and expressed nowadays thanks to marketing efforts that unsubstantiately inflate those cam's contribution to cinema and innovation for cinematography rather than practical experience on set or location, reminds me of another Wilde:

 

"Ignorance is a delicate exotic fruit; touch it and the bloom is gone."

 

And so we will think of current video gear in 5 years time, as we now ridicule U-matic ENG, which was once regarded as far superior to 16mm & 35mm VNF cine-films... oh, the blunder...

There is nothing particularly wrong with the quality of U-matic itself. Most of the problems can be put down to the primitive cameras they were attached to, and sloppy practice in TV stations.

 

There are still large numbers of 8-10 year old Sony HDCAMS earning a nice bit of change out there, so I think the days of "2 years and it's landfill" are well and truly over. Sony et al no longer have the money to play that game any more.

[Keith Walters]

here's part of the matlab code to pull in any dpx file...

 

% DPX image file reader (SMPTE 268M-2003 Reference)

%

% Originaly from:

% © 2006 Jeff Mather, Mathworks

% Extended by:

% © 2007 Amilcar Lucas, IDA TU-Braunschweig

% Chopped and hacked by:

% © 2008 Glen Alexander, Gestalt Prod

%

% Convert the buffer to an array of output pixels.

%

% DPX files can also contain bit-depths that cause pixel samples not to end

% on byte boundaries (e.g., 10-bit and 12-bit images) plus 14bit either LI or BI

switch bitDepth

case 8

pixels = buffer((startOfPixels + 1):endOfPixels);

case 10

% SMPTE 268M-2003 Reference

% Fig. C.3 10-bit components filled to 32-bit boundary

pixels = typecast(buffer((startOfPixels + 1):endOfPixels), 'uint32');

pixels = swapFcn(pixels);

if (numChannels == 1) % grayscale

% do some bit splicing to extract the gray information

pixel0 = uint16(bitshift(pixels, -02, 10));

pixel1 = uint16(bitshift(pixels, -12, 10));

pixel2 = uint16(bitshift(pixels, -22, 10));

% Rearrange the data to follow MATLAB's conventions.

pixels = [pixel0 pixel1 pixel2];

pixels = reshape(pixels', [columns, rows])';

else % RGB

% Rearrange the data to follow MATLAB's conventions.

pixels = reshape(pixels, [columns, rows])';

% do some bit splicing to extract the color information

pixelsb = uint16(bitshift(pixels, -02, 10));

pixelsg = uint16(bitshift(pixels, -12, 10));

pixelsr = uint16(bitshift(pixels, -22, 10));

% pack it back to a matlab friendly array

clear pixels;

pixels(:,:,3) = pixelsb;

pixels(:,:,2) = pixelsg;

pixels(:,:,1) = pixelsr;

end

% return because we do not need the normal function flow in this case

return

case {12, 14}

% SMPTE 268M-2003 Reference

% Fig. C.5 12-bit components filled to 16-bit boundary

% 14-bit components filled to 16-bit boundary

pixels = typecast(buffer((startOfPixels + 1):endOfPixels), 'uint16');

pixels = swapFcn(pixels);

% do some bit splicing to extract the information

pixels = uint16(bitshift(pixels, int8(bitDepth)-16, bitDepth));

if (numChannels == 1) % grayscale

% Rearrange the data to follow MATLAB's conventions.

pixels=reshape(pixels, [columns, rows]);

else % RGB

% Rearrange the data to follow MATLAB's conventions.

pixels=reshape(pixels', [3, columns, rows]);

pixels=permute(pixels,[3,2,1]);

end

% return because we do not need the normal function flow in this case

return

case 16

pixels = typecast(buffer((startOfPixels + 1):endOfPixels), 'uint16');

pixels = swapFcn(pixels);

otherwise

error(['Unsupported bit-depth: ' num2str(bitDepth) '-bits per component']);

end

% Rearrange the data to follow MATLAB's conventions.

pixels = reshape(pixels, [numChannels, columns, rows]);

pixels = permute(pixels, [3 2 1]);

Great. Why didn't you mention this earlier; you could have saved us all a lot of trouble :P

[George Ebersole]

"Alta Vista"... oops, I won't make that mistake again, heh. I guess I'm showing my age :)

 

A cameraman I used to work for years back (Jim Iacona? Can't recall) used to say that chip camera's couldn't capture skin tones like tube cameras. I was just a stage manager and special effects assistant at the time, and not a cameraman, much less a die-hard DP, but, he his point about new technology was that it always claimed to do more than deliver. I guess there's always a balance between those cheering for new tech, and those who are skeptical.

 

Keith; thanks for the feedback. I wont' fret so much. And yeah, DLing the user manual would be a step in the right direction. I'll have a look at your comments.

 

Scott; I guess it was all the CGI that threw me. It was a real distraction to my eyes. Sequences just seemed to be tossed together with digital inserts. It looked convincing in terms of story, but not in terms of the reality the film was trying to portray. That's kind of why I thought the whole thing might've been shot digitally. Digital stunts, digital extras, digital everything-else... I didn't feel like I was watching a "real movie".

Edited June 7, 2008 by George Ebersole

[Keith Walters]

"Alta Vista"... oops, I won't make that mistake again, heh. I guess I'm showing my age :)

 

A cameraman I used to work for years back (Jim Iacona? Can't recall) used to say that chip camera's couldn't capture skin tones like tube cameras. I was just a stage manager and special effects assistant at the time, and not a cameraman, much less a die-hard DP, but, he his point about new technology was that it always claimed to do more than deliver. I guess there's always a balance between those cheering for new tech, and those who are skeptical.

 

Keith; thanks for the feedback. I wont' fret so much. And yeah, DLing the user manual would be a step in the right direction. I'll have a look at your comments.

The first couple of generations of CCD cameras had trouble with accurate skin tones. The problem is simply that we have to make cameras using the physical properties of the 92 elements nature gave us. Ultra-pure Silicon is the only thing that can be used to build camera sensors, and its spectral response is far from ideal.

 

With tube cameras and colour film, complex mixtures of materials can be used to precisely "tune" the spectral response.

 

There is no question that the last generation of Saticon and Plumbicon-based tube cameras gave far better rendition of the red part of the spectrum than the first CCD models. In fact, a lot of earlier Betacam tube cameras were "recycled" for lower-end studio use.

 

On the subjec of "pre-heating" yourself for your first RED experience:

Now here's a strange thing. The Australian Film and video rental house LEMAC (www.lemac.com.au) has (or had) a series of 15 Megabyte or so quicktime downloads on the progress of the all RED feature: "The Nothing Men" (which has a good writeup on the www.RED.com site).

 

If you click on this link: http://www.lemac.com.au/sales/RED/REDmain.htm you will get to that page. BUT, if you go through the "Front Door" ie www.lemac.com.au there's no longer any mention of RED, although it was fair plastered with RED-related notices a few months back. The only way I could locate their RED page was through a Google search; none of their menus carries the link.

 

The first couple of quicktimes were probably exactly what you are looking for, where the operators give a bit of a rundown on the operation, give you a bit of a look at the camera and so on. I made a DVD from two of the files, and they come up roughly VHS quality, which is more than good enough.

 

But when I tried downloading them just now, neither file would play. They appeared in the temporary Internet Files folder right enough, but I couldn't access them.

 

WTF bigtime????

[George Ebersole]

Thanks again, Keith.

 

Watching now.

[George Ebersole]

I'm scoured RED's website, and haven't found the manual.

 

http://www.red.com/support

 

*edit*

 

Woops, nevermind. I think I found it.

Edited June 9, 2008 by George Ebersole

[Brian Langeman]

Maybe I'm a bit confused here, or don't know what you're trying to see with real-world footage. But I thought you would want to see what the camera can do, and not what the editors can do, or the people compressing it into various output formats. If you're looking for what the camera can do, isn't judging it by the final output kind of backwards? After all of the grading and compositing it can easily be made to have more dynamic range, for instance compositing in other elements that couldn't have been shot with it. And the colour reproduction is clearly not the same after grading, and watching it on a non calibrated monitor. And then there's the issue of it being encoded into whatever format your watching it on, either broadcast or DVD, again which changes it from it's original form, and is also different for each different output format you would see it on. Watching the finished work often doesn't show anything about what the camera can do, and I think you would want vanilla files that haven't been touched to judge it by.

[Keith Walters]

Maybe I'm a bit confused here, or don't know what you're trying to see with real-world footage....

I cannot comment on your level of industry experience, only my own.

Take your average feature film. The list of credits at the end will often run on for five minutes. Every one of those people (with a few well-known exceptions :lol: ) does something.

But what percentage of them would have any idea what you are talking about? One percent if you are lucky?

This is something that frighteningly few posters on this (and other forums) seem to understand: The vast majority of people who work in the film and TV industries know little or nothing about how images are captured.

But most of them will recognize a good picture when they see one.

And most of them will also know that when they see a really high quality commercial or TV drama series, what they are seeing is the very best than can be achieved with current imaging technology.

35mm film origination consistently produces stunning images.

Video origination can also produce excellent results, if the subject and available shooting conditions are favourable. And you have a top-notch post production team. A lot of "ifs".

If at the end of the chain, some new system consistently fails to deliver the goods, or it is a very hit or miss affair, they are not going to be interested in exactly WHY the system fails, and they are not going to listen to excuses.

All they want to see is an easily accessible selection of end results. That is what I would like to be able to give people, but so far I have had no luck at all compiling such a list.

We wait.

As somebody else pointed out, another frightening thing is the amount of "Awesome" sample footage people have made available for download, which is utter garbage.

[Brian Langeman]

OK. So you want to show this list of commercials or finished products to the 95% of people who don't understand all of this, and present it in an easy way to view it? That makes sense then. Although I don't know how much say those 95% of people have in selecting which camera to use for a project. I would hope that the people in charge of selecting the cameras have some of the understanding that we have, and at least know that the finished product is far from what the camera delivers. But the world isn't perfect, so maybe they don't even understand that basic concept.

[Patrizio De Sica]

Alta Vista? You mean CineAlta?

OK, first off, the CineAlta is a TV camera, despite what George Lucas and a few other chuckleheads tried to tell us. Star Wars II was captured on 1440 x 800 pixel videotape. Better than film? Super-8 maybe.

The CineAlta/F900/F950 etc are excellent for making TV programs, mainly because they were designed from the ground up to fit the workflow of TV studios. Movie cameras they are not.

Amazing... The more glue you add the more silly putty you'll get on your stupid falsehood. Typical KW BS :P

[Keith Walters]

Amazing... The more glue you add the more silly putty you'll get on your stupid falsehood. Typical KW BS :P

Ah well, it just goes to show, it DOES result in blindness AND insanity :lol: