Jump to content


Photo

I hate RED


  • Please log in to reply
75 replies to this topic

#1 Chris D Walker

Chris D Walker
  • Basic Members
  • PipPipPip
  • 108 posts
  • Other
  • Cornwall, United Kingdom

Posted 15 March 2010 - 09:37 PM

I Hate RED

I don't feel this way against any other digital camera. I have this hatred for the RED in it's mass hysteria of marketing, numbers and fandom. F23 and F35? Fine. Genesis? I can live with it. RED? I'm screaming in my head. This is wholly about what has been spoken about the camera, not the camera itself.

1) K's - Simply, a 4K bayer sensor isn't 4K. Marketing team, stop using this number. Also, as for the number of K's a camera has, how far can too far go?

2) Coming soon... - For obsoleting obsolescence, here's another build and after that an updated sensor followed by a new viewfinder.

3) Goodbye, intuition - Fashion and men's magazines have recently been using RED's and EPIC's to photograph their subjects, Esquire in particular. Said: "This makes it a lot easier on the photographer since he doesn't need to know, intuitively, when the best few seconds are to snap a stream of shots." Wait, the photographer doesn't have to click his shutter anymore? Wow, an artist is at work.

4) Wavelet goodbye - I'm not a master of software but a 10:1 compression ratio (Redcode 36) can't realistically carry uncompressed RAW data.

5) Indie Baby - "It's a professional camera for the prosumer market. Hooray!" I dispute. There are many great cameras that deliver an equally good-looking image for cheaper than a RED, both film and digital. Films get made on Super16 for under $30,000. Saying that, how much does the storage of compressed 'uncompressed', 4K, 12-bit data cost? Plus lens rental, DIT, colour grading, archiving etc.?

Everything written has been concerned with how the RED has been sold by the manufacturers. How the camera itself works in the field is not my point, I'll let others comment on that. I hate been given hype. Sony doesn't do this. Panavision doesn't do this. Arri doesn't do this. Conduct yourself in a gentlemanly fashion, RED marketing team.

Rant over. Has what I've said been fair?
  • 0

#2 Satsuki Murashige

Satsuki Murashige
  • Sustaining Members
  • 3510 posts
  • Cinematographer
  • San Francisco, CA

Posted 15 March 2010 - 11:41 PM

I Hate RED...

Rant over. Has what I've said been fair?

I have to question a few of your ideas here.

1, 4) "(Not) 4K/compressed Raw." I understand being upset about false advertising, but to me the most important thing is what the footage actually looks like. And in the right hands it can look very, very good. That is why everyone continues to use the camera.

2) "Obsolescence" etc. I think Red have kept their promise on this, for the most part. The constant flow of free firmware updates, software updates, and color science improvements have been very generous. Customer service is very responsive. And the new sensor is by all accounts like getting a new camera for $5k-ish. Can't wait to try one out myself.

3) Still camera replacement. I haven't heard about this, but if it is true then surely you can't blame how the camera is being used on the people who built it, right?

5) The indie market. In what world is a low-budget Super16 workflow cheaper than a low-budget Red workflow? Even if you get a free camera and discounted stock, you still have to pay for processing, an HD telecine (if you're gonna compare apples to apples), and digital storage. If you want to get footage that looks reasonably professional, you pretty much have to at least get an HD telecine on a Spirit or something better. I've learned to avoid the "el cheapo" HD telecine - you get what you pay for. The only way I see Super 16 being cheaper is if you get the stock, processing, and telecine for free, or next to. You can post Red yourself on a laptop at home if you really need to.
  • 0

#3 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 15 March 2010 - 11:50 PM

You're letting your dislike of being "sold to" get in the way of being objective. I know the feeling, a lot of us have a gut reaction against marketing.

What I try to do is separate the hype from the product -- they are two distinctly different things, and you shouldn't let the hype lead your decision-making, pro or con. I mean, look at some overly-hyped movies -- some are actually worth seeing.

Marketers are just doing their job. Sony, Panasonic, Canon, etc. all have their own marketing campaigns -- look at how the DVX100 and XL1 was sold to every aspiring filmmaker out there with a credit card. We just sort of got used to it after awhile and it became white noise. Red comes along as does it differently, almost like a viral internet-based campaign, got people talking, created a social network that did a lot of the marketing work for them by whipping up enthusiasm among enthusiastic types.

And secondly, I try and separate Red's own marketing tactics from the half-assed promotional tactics of some young people who are amateurs at marketing and have emotionally tied up the camera with their own hopes & dreams and now cannot be objective about it.

But honestly, I'm telling you that if you let your natural revulsion towards marketing get the better of you... you may be missing out on what has become a top-notch digital cinema camera at one-quarter the cost of its nearest competition. For all practical purposes, with the new M-X sensor, this camera is now one to two stops faster with almost two more stops of usable dynamic range. That aint chicken feed, it's significant.

But as to your points:

1) Red didn't create the mystique and confusion surrounding the vague phrase "4K", that started earlier. Remember Dalsa? It's become a shorthand and a vague standard rather than something specific. To make it specific, it need a descriptor, like "4K RAW" or "measurable 4K lines of resolution" or "4K RGB data", etc. And if Red is or isn't really "4K" then what is or isn't? The term by itself is pretty useless but it remains popular. If you think of "4K" as a vague goal or ideal standard -- a catch-all -- rather than something specific, you'll lose less hair or sleep. Red has never denied that 4K RAW measures something lower once converted to RGB, though 4.5K RAW, which the camera now is capable of recording, does almost hit that 4K benchmark. And 35mm film, that oft-named "true 4K standard", often doesn't hit that benchmark. So while I would love everyone to be more specific, it's a hopeless battle, 4K is here to stay as a vague term to be thrown around by everyone, not just Red.

2) The whole "obsolescence is obsolete" sales pitch, well, that sort of defies common sense so I never took it very seriously, but one could argue that Red has delayed obsolescence, even if they haven't killed it, by putting a new sensor in the camera. As for the constant upgrading, it's annoying but as long as it keeps making the camera better, it's an annoyance I can live with. I think we just aren't used to this method of releasing a camera and letting its users experience the "joy" of being involved in its evolution. For some people, it's just inconvenient, but for many private owner/operators, it has given them a sense of personal involvement in the camera's development, further reinforcing the emotional bond they have with the company. So it has its upside.

The downside for Red is that they get a bad reputation for an earlier build with a certain problem that is later fixed or improved, but people don't shoot often enough to see the benefit, so they carry this bad impression of a build that doesn't exist anymore. And with the big improvements lately with the new M-X sensor, it makes a lot of the earlier weaknesses ancient history, and negates most of the comparison tests that were shot last year. That's good and bad for Red -- good in the sense that their product is better, bad in the sense that impressions got formed based on earlier versions of the camera.

3) Bad / lazy habits caused by new digital technologies, negating the need to plan carefully, shoot carefully, etc. -- well, that's hardly Red's fault. Blame the filmmaker or photographer who takes less care in their work, gets sloppier. On the other hand, we may be seeing a fundamental shift in the way that images and movies are constructed, less formal, and some of us are just clinging to traditional ways of thinking about art-making.

4) Compression -- how good it is all depends on what you can do with it in post, the proof is in the pudding (the real phrase is "the proof is in the pudding's eating", which makes more sense...) If it works, it works, and Redcode, though not flawless, seems to work for most situations. It has made 4K RAW practical as a shooting format. But times change, processing speeds keep increasing, recording larger and larger amounts of data gets easier, so compression can probably become less strong over time. But compression isn't going anywhere soon and a lot of other camera's compression schemes are a lot worse than Red's. The h.264 compression used by the Canon DSLR's in video mode is pretty poor. Red will be offering less-heavy compression schemes in future cameras.

5) There are always cameras cheaper and more expensive than other cameras. The fact that Red is well-suited for indie production is proven by the fact that film festivals are now flooded by Red-originated features. Sure, it may not be suited for no-budget dirt-cheap features but it certainly competes well for projects that in the past would have opted for Super-16 or 24P HD instead of 35mm.

All the manufacturers have marketing campaigns, some in print, some passed on through spokespersons. Some are more low-key than others, that's all. Go to NAB and you'll pick up plenty of advertisements for Sony, Canon, Panasonic, etc. products. In fact, Red spends a lot less money on print advertisement than these guys and yet manages to have a larger presence through word-of-mouth. I think Red just suffers from not playing the same game as these companies that have been around for a long time; it's like the difference between heavy metal and classical music. Red is the rockstar that walks into a room full of classical musicians. It's a different culture -- look at Red's metal skull logo, it's something more in line with Harley-Davidson than ARRI. That's not a good or bad thing, it's just tonally different. Takes a bit of getting used to (rock-n-roll culture, motorbikes & cigars & supermodels, etc. -- not my cup of tea), but at least it's not overly corporate, it's not dull.

Red has been reaching out to traditional cinematographers lately, which is appreciated, so I don't really think it's useful to bring up the old problems from the early days. They are trying to find their niche among the other players without losing their unique identity. But they do generate passion among some users, which is a double-edged sword. Passion is an important thing in filmmaking but it also can lead to irrational decision-making if one does not balance it with some cold objective thinking now and then.
  • 0

#4 Brad Webb

Brad Webb
  • Basic Members
  • PipPip
  • 29 posts
  • Other

Posted 16 March 2010 - 12:53 AM

This is like saying you hate Pepsi because they use celebrity endorsements, you won't eat at McDonald's because you hate clowns, or you won't use by an iphone because of all the hype behind it. None of that matters.

If you like the taste of Pepsi, drink it. If you like the images Red produces, shoot it.



Red has done something that no high end motion picture camera company (film or digital) has ever done before, which is make a high quality camera that is relatively affordable.
  • 0

#5 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 16 March 2010 - 07:12 AM

We just sort of got used to it after awhile and it became white noise. Red comes along as does it differently, ....


Perhaps instead of white noise, we're getting red noise? ;-)

There's a lot more noise surrounding this camera than pretty much any other product in our industry. OK, we can live with that.

There are a lot more quirks and issues you have to know about and consider in order to use it than the more mature manufacturers have. But if you do your homework first, you can get pretty much what you need from it.

Red hits a new price/performance point all its own. There's nothing else nearby. For many shows, it's the right choice. For many others, it's not. I don't love it, I don't hate it. I just work with the files that come from it.




-- J.S.
  • 0

#6 Phil Rhodes

Phil Rhodes
  • Sustaining Members
  • 11934 posts
  • Other

Posted 16 March 2010 - 08:31 AM

I do hate Pepsi because it uses ludicrous injection-molded morons to promote itself.


Red didn't create the mystique and confusion surrounding the vague phrase "4K", that started earlier. Remember Dalsa?


Yes, but Dalsa were at least upfront enough to say, when challenged on this, "Well, yes, but we still think it's a great camera." Which it was - it was fantastic and wonderful. Red, on the other hand, get all wound up trying to prove the absolutely impossible then take their frustration out on me.

Not only is this impolite, it's also very scary from a company relations point of view. Can I trust these people to turn out a dependable piece of equipment? Well, funnily enough, it actually looks like I really can't, or at least couldn't for the first year or two.

To make it specific, it need a descriptor, like "4K RAW" or "measurable 4K lines of resolution" or "4K RGB data", etc.


12 Megapixel. DSLRs had the terminology for single chip imaging long ago. Even that would have been pretty impressive, but that wasn't good enough for someone's preconceived marketing scheme.

And if Red is or isn't really "4K" then what is or isn't?


My comments have always been based on comparisons with a hypothetical 4096-pixel cosited RGB array, which will always outresolve a 12MP bayer chip under the same circumstances.

This is the core of the argument. You can do what you like to make a 12MP bayer array look better on a test chart, but no matter what, if you did the same thing to the cosited RGB sensor, you'd get better results.

4K is here to stay as a vague term to be thrown around by everyone, not just Red.


Like who? I'd be the first person to point out that not much film gets that sort of effective resolution.

If it works, it works, and Redcode, though not flawless, seems to work for most situations. It has made 4K RAW practical as a shooting format


I don't think it has, though. It's made (extremely) heavily compressed 12-megapixel bayer data practical, but that was never really that difficult in any case - because there are major caveats with it.

The h.264 compression used by the Canon DSLR's in video mode is pretty poor.


Atrociously poor, yes. Almost unforgivably bad; it should be much better for the bitrate.

P
  • 0

#7 Chris D Walker

Chris D Walker
  • Basic Members
  • PipPipPip
  • 108 posts
  • Other
  • Cornwall, United Kingdom

Posted 16 March 2010 - 08:52 AM

Some great points here, guys.

I like Pepsi, more than Coca Cola, but I do hate their celebrity endorsements. I eat at McDonalds, but I do hate that Justin Timberlake sings "I'm lovin' it". I don't own an iPhone, but I hate that the small print in the ads read 'sequence shortened and stages removed', or something to that effect. I think it does matter because how it's sold is what first piques your interest.

If RED sold their cameras as professional HD with reasonably comparible compression to HDCAM-SR at a competitive price I would have fewer bones to pick. The fuss with many people isn't so much what it delivers, but what is promised.

About word-of-mouth, I think that is part of my issue. It has the same principle as a game of Chinese whispers; as it goes further down the line the message gets distorted and more simplified. There's a tendency to focus on the big numbers and fancy words.

I must stress this again, I have an issue with how the camera is being presented, not the camera itself. It would be a pleasant surprise to test a RED (currently out of the question due to a lack of funds) and find that I like the result. What unique advantages does a RED, ignoring cost for a moment, have over an F35 or a SI-2K?

I hate being sold 'untruth', that's all.

Thanks for very full and detailed responses.
  • 0

#8 Phil Rhodes

Phil Rhodes
  • Sustaining Members
  • 11934 posts
  • Other

Posted 16 March 2010 - 09:50 AM

Red and SI2K are pretty comparable in terms of technology: both use single bayer-filtered sensors and record to wavelet codecs. SI2K has considerably fewer pixels on its sensor.

The big advantage over Sony stuff is that you don't end up locked into a tape-based post path. This is the really critical thing about any of these data workflows: you can do on a thousand-unit-of-currency desktop PC what you would otherwise need a hundred-thousand tape deck to achieve. That's what's most significant about any of this stuff. Of course, you can put a file recorder onto a Viper or an F23 or an F35 or an F950 and end up somewhere really special.

Like you I didn't initially have a problem with the technology. However, it was fairly obvious in 2006 that humanity just didn't have the technology to make a really good 12MP CMOS movie camera. We all had a pretty good idea what Red were trying to do, and we all had a pretty good idea that it would create a fairly feeble camera, which is why nobody else was doing it. People weren't not doing it because it was somehow new or advanced or particularly clever. They weren't not doing it because they didn't know how. They weren't doing it because it simply wasn't a very good idea, and I think Red proved that more than adequately by producing something that was soft, insensitive, clippy and noisy. These are exactly the problems you would expect to get by taking 2006 CMOS stills technology and trying to make it refresh at 24Hz. It won't do it very well.

Now, that's a reason not to use it, but it isn't a reason to take any form of offence.

The main problem I have, like you, is that they claimed to be able to do a lot of things that weren't really possible, inevitably failed to achieve these things, and have since run around claiming to have achieved them all. They didn't make a 4K camera. They didn't make it in the timescale they claimed they would (based on the consideration that it was far from usable for over a year after "release"). It didn't have the accessories or abilities they claimed (dynamic range, uncompressed output, etc). And yet now we have these people running around pretending to be some sort of gestalt messiah bringing enlightenment to filmmaking.

I personally think that is quite offensive.

P
  • 0

#9 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 16 March 2010 - 12:36 PM

The past is the past -- the real question is what is the best tool for the job (and the budget). Anything else is a distraction ultimately. A lot of people have been carrying a chip on their shoulders about Red, including Red, and at some point, it seems adolescent (doesn't help that some people involved are actual adolescents though...) Time to move on.

I have been ambivalent about Red for awhile, even after using the camera on two features (or maybe, because I used the cameras on two features.) Most of my frustrations though were not on the set, where I only encountered a few annoyances and bugs, hardly more than what I typically run into with any camera technology including film. My biggest complaints were the relatively slow speed and the blue noise in 3200K lighting, and a somewhat mediocre dynamic range, OK but not great. And I suspected that the RedCode compression was affecting the noise in some way, sort of blurring it.

No, the real frustration came in post, from dailies that were slow to get made and off the mark in terms of look, from trying to figure out why the same files looked radically different at the efx facility compared to the D.I. facility, trying to figure out why my stills from the r3d files looked better than what I was getting during color-correction. And basically finding that a lot of the post people I was dealing with were as clueless as I was in many areas.

At that point, my only frame of reference was the F900, which I had shot eight features on. Obviously I knew the pitfalls of that camera, from HDCAM compression to Rec 709 gamma to 3:1:1 color-sampling, etc.

So after the two Red features, I did three projects on the Genesis, a Log camera. I dealt with LUT's and whatnot. There were some frustrations there as well, the camera really needs the option of creating an internal Rec 709 signal for monitoring that bypasses the recorder so you aren't always tied to a LUT box, nor so you don't always have to look at a Log image in the viewfinder and deal with zebras and whatnot that act differently for a Log signal, etc.

But I was happy with the speed and the noise in 3200K lighting, which is much closer to what is typical on a film shoot, i.e. 500 ASA tungsten-balanced stock.

But I wasn't so happy with the aliasing, which seemed more aggressive.

My general impression at the time was that the Genesis/F35 sensor used a weak OLPF to get maximum sharpness into a 1080P recording, whereas the Red used a heavy OLPF to reduce aliasing in a 4K RAW recording, figuring that they were benefitting from oversampling to compensate (though in terms of pixel count, the F35 and Red sensors are similar.) So the final effect seemed similar in sharpness -- your eyes did not go "wow, 4K is so much sharper than 1080P" -- but that the Genesis image was "crisp" and edgy (even without edge-enhancement) whereas the Red image was soft and smooth, maybe almost too "smoothed." Unscientific observations, I admit. But I do feel that this is one reason why the Red image feels less electronic and more film-like.

But the biggest issue I had with the Genesis was simply the size & weight (and power consumption.) It was like working with an old GII Panaflex without the advantages of working with film. I remember that Soderbergh had a similar reaction -- "what's the point of switching to digital if we have all the size & weight restrictions of 35mm?"

So when I went into job interviews for TV pilots in 2009, my general take was that Red = better size & weight, thus mobility for a location shoot, but weak in low-light and 3200K, Genesis/F35 = better in low-light and 3200K, but too heavy & large, better suited for a studio shoot. Also, I felt that the Genesis/F35 had better dynamic range than the Red.

This is ignoring the post issues.

But now with the faster M-X sensor in the Red, the better noise in 3200K, the wider dynamic range, etc. almost all the negatives have disappeared compared to the Genesis/F35 but you still have the advantages of the smaller body, etc.

So that leaves post, which was my major issue before.

I think in the past nearly two years since I shot my Red features, a lot has improved. More post houses now can do real-time transcoding of r3d files for dailies. With RedGamma, I think some of the color-correcting issues from working with PDLog have disappeared, the color-correction process is greatly simplified. I have yet to generate dailies from a Red shoot to see if that problem has changed, getting the process to be more accurate to how I set the cameras up in the field.

But I think this is still the major sticking point, the comfort most producers and post houses have with working with HD tape in Rec 709 or Log.

But from a camera person's standpoint, most of my reservations about Red versus other digital cameras have disappeared. It's a major turning point because things will only get better from here on out with everyone's new products.

As for compression, particularly RedCode, it's a mixed blessing -- I feel it has absolutely been the main reason for Red's success compared to Dalsa. It has made 4K RAW photography practical for the average shooter. And I haven't run into much problems with it. On the other hand, in the past at least, it has lead to some oddities in the recording, and perhaps is the reason why the blue noise in earlier builds was so blurred & chunky. Now Red has a less compressed format, RedCode 42, and has promised some even milder compression rates in later cameras. So I'm not too concerned about this aspect.

I'll be curious if uncompressed recording has become practical enough for ARRI and Aaton to proceed with their plans of not compressing their RAW output.
  • 0

#10 Dustin Lindgren

Dustin Lindgren
  • Basic Members
  • PipPip
  • 18 posts
  • Camera Operator

Posted 16 March 2010 - 01:12 PM

Hey Chris,

I'm selling a bunch of my old S16mm equipment right now if your interested. Ill make you a great deal. I need the cash to buy more stuff for my RED.

Cheers

Dustin
  • 0

#11 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 16 March 2010 - 03:39 PM

To make it specific, it need a descriptor, like "4K RAW" or "measurable 4K lines of resolution" or "4K RGB data", etc.


I'd suggest "4K Bayer". I think that conveys the most useful information in the fewest keystrokes.






-- J.S.
  • 0

#12 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 16 March 2010 - 03:43 PM

Of course, you can put a file recorder onto a Viper or an F23 or an F35 or an F950 and end up somewhere really special.


We have a pilot doing exactly that with a unit called a NanoFlash, on F-35's, and they're loving it. It's about the size of a cell phone, and getting used far more than was anticipated, even though it's limited to 8 bit 4:2:2. Clairmont has them.






-- J.S.
  • 0

#13 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 16 March 2010 - 03:52 PM

My general impression at the time was that the Genesis/F35 sensor used a weak OLPF to get maximum sharpness into a 1080P recording, whereas the Red used a heavy OLPF to reduce aliasing in a 4K RAW recording, figuring that they were benefitting from oversampling to compensate (though in terms of pixel count, the F35 and Red sensors are similar.)


The big difference is that one uses vertical stripes, with the same sampling structure and Nyquist limit for all three primaries, and the other's Bayer, with twice as many green samples as each of the other primaries. The Bayer cameras have to filter to near the red/blue Nyquist limit to control color aliasing to an acceptable level, which goes farther than need be for green. They all cheat on Nyquist to some extent, it's a judgment call as to how much aliasing to accept in trade for more sharpness. But stripes vs. Bayer is very apples and oranges where OLPF and Nyquist are concerned. So, go with your gut feeling on this one.




-- J.S.
  • 0

#14 Thomas James

Thomas James
  • Basic Members
  • PipPipPipPip
  • 844 posts
  • Camera Operator

Posted 16 March 2010 - 04:13 PM

I rented Che on Blu-Ray and I think the picture quality was horrible because the highlights would always clip. Now I realize Che was filmed with a prototype Red camera and I do not own a 4K television but I hope that the newer Red cameras are a lot better. Also I do not know why Che was shot at 24 frames per second. If the director does not like the film look and thinks that all movies should be shot with video cameras then why shoot at 24 frames per second?
  • 0

#15 Phil Rhodes

Phil Rhodes
  • Sustaining Members
  • 11934 posts
  • Other

Posted 16 March 2010 - 04:42 PM

NanoFlash, on F-35


I think NanoFlash is great if you want to get better results out of an EX3, or something. Then, it's very helpful. The problem with it is that it's only MPEG-2, presumably because that codec is now sufficiently mature that chips to do it in realtime, to HD, are reasonably affordable.

The problem is that MPEG-2 (as h.264) suffers horribly from diminishing returns on bitrate. It is easier to compress than h.264, but still not that easy, especially if you are trying to do it reasonably cheaply, in low power consumption, and a physically compact package, in realtime.

If you try to throw bitrate at the problem it becomes a vicious circle because you need data to fill that bitrate and that data has to be calculated; for this reason alone, quality is not linear with bitrate, even on a single device, which puts a question mark over NanoFlash's "Look how much bitrate!" claims.

Therefore although a NanoFlash will make an EX-3 look considerably better than its internal recording provisions and is therefore a worthy and useful thing, I think that the cinematography gods ought to cast lightning down upon anyone who puts one on an F-35.

What we need is one that does JPEG-2000 frames, or better yet Cineform, which has good cross-platform support. Disclaimer: Redcode is not JPEG-2000 with the serial numbers filed off, in any way whatsoever, perish the thought.

P
  • 0

#16 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 16 March 2010 - 06:14 PM

We're using the NanoFlash in its I-frame 280 Mbit/sec mode. By my math, the compression ratio is about 2.8:1, much like HDCam SR's 4:2:2 10 bit mode. We split screened it with SR, nobody could see a difference.




-- J.S.
  • 0

#17 Saul Rodgar

Saul Rodgar
  • Basic Members
  • PipPipPipPip
  • 1682 posts
  • Cinematographer

Posted 16 March 2010 - 08:55 PM

We split screened it with SR, nobody could see a difference.


Ultimately, that is the litmus test, if no one can see the difference, why argue tech specs until one is blue in the face?

Also, does every frigging project out there really benefit from RAW recording? Let's face it, for projects other than movies that will be historically archived or digitally or traditionally projected, the answer very likely will be no.

For bigger size recorders, but up to 10 bit 4:4:4 3D dual HD-SDI signal input capability, try some of these Wafian recorders.

http://www.wafian.com/Products.htm

Edited by Saul Rodgar, 16 March 2010 - 08:59 PM.

  • 0

#18 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19759 posts
  • Cinematographer
  • Los Angeles

Posted 16 March 2010 - 10:39 PM

Ultimately, that is the litmus test, if no one can see the difference, why argue tech specs until one is blue in the face?

Also, does every frigging project out there really benefit from RAW recording? Let's face it, for projects other than movies that will be historically archived or digitally or traditionally projected, the answer very likely will be no.


Whether it's RAW, Log, HyperGamma, whatever, yes, almost every project benefits from recording all the dynamic range of the sensor rather than record in Red 709 video gamma -- because almost every project is going to be color-corrected, and it's easier to achieve film-like highlight roll-off if you have more information to work with.

RAW is just something specific to single-sensor cameras where the signal has to be processed into color (RGB). Not recording RAW means that the camera has to do the debayering and conversion in real-time so it can be recorded as a color signal. And there are some advantages in leaving that conversion to post, mainly because the processors used in post are likely to be more powerful than the ones in the camera.

The other benefit to RAW recording is that it is a natural form of lossless data compression because converting it to RGB immediately triples the amount of data to record unless you then downscale and/or compress it.

Shooting Rec 709 gamma makes more sense for material that is not going to be color-corrected or doesn't have to have a film-like look, a video-ish look is OK.

As for not seeing the downside to a certain type of compression or color-subsampling -- for example, not being able to see the difference between 8-bit 4:2:2 versus 10-bit 4:4:4, or a more compressed signal versus a less compressed signal, sometimes the weakness of the more compressed or subsampled version pops up later in certain types of color-correction. This is one reason why so many people think that their consumer h.264 8-bit 4:2:0 HD recording seems great... until they try to do things in post to is, like pull chroma keys or do an aggressive color-correction to it.

I'm not doubting John that the Nanoflash recording looked similar to SR, but I'm sure he's aware of the potential downsides and probably would recommend that they shoot greenscreen shots, for example, on the SR deck.
  • 0

#19 John Sprung

John Sprung
  • Sustaining Members
  • 4635 posts
  • Other

Posted 17 March 2010 - 01:27 AM

RAW is just something specific to single-sensor cameras ....

I'm not doubting John that the Nanoflash recording looked similar to SR, but I'm sure he's aware of the potential downsides and probably would recommend that they shoot greenscreen shots, for example, on the SR deck.


Raw, so far, has only been implemented on single chip cameras. But there's no theoretical reason why it couldn't be done in the three chip world. Maybe it's on somebody's to do list....

Yes, green screen policy is to shoot 4:4:4 for the actual screen. The stuff you matte into the holes can be either way. This show won't have any composites, so we're good on that point.

To support more extreme color timing, the first thing I'd like to see from NanoFlash would be 10 bits instead of 8. That's where you slice you range into smaller steps, which lets you stretch things farther without seeing the steps -- like a contour map. The DP on this show is well aware of the limits of 8 bit, and given the style he wants, can shoot it close enough to where he wants to land. The convenience is well worth the trade-off of being in 8 bit.




-- J.S.
  • 0

#20 Saul Rodgar

Saul Rodgar
  • Basic Members
  • PipPipPipPip
  • 1682 posts
  • Cinematographer

Posted 17 March 2010 - 01:27 AM

Whether it's RAW, Log, HyperGamma, whatever, yes, almost every project benefits from recording all the dynamic range of the sensor rather than record in Red 709 video gamma -- because almost every project is going to be color-corrected, and it's easier to achieve film-like highlight roll-off if you have more information to work with.


Technically, I agree. Wholeheartedly. However, what I meant is that RAW may not be the best choice for every type of project out there. As you know, it is up to the director, DP and producers to choose the best format for the project at hand based on cost, conditions of filming, post-processes, audience delivery format, etc and not necessarily RAW for the sake of RAW.

Just like one would not necessarily film every webisode project on 70 mm film, RAW recording may not the best choice for them either. As an extreme oversimplified example, it's a bit like the joke of the webisode cameraman who is turns up to set armed with an iPhone for filming and when questioned by the stunned direcor, he says "Well, most people are going to watch this on an iPhone, so might as well film it on one!" ;)

Edited by Saul Rodgar, 17 March 2010 - 01:31 AM.

  • 0


Metropolis Post

The Slider

CineLab

Broadcast Solutions Inc

Glidecam

Rig Wheels Passport

Abel Cine

Wooden Camera

Opal

rebotnix Technologies

Aerial Filmworks

FJS International, LLC

Visual Products

Media Blackout - Custom Cables and AKS

Gamma Ray Digital Inc

CineTape

Tai Audio

Technodolly

Ritter Battery

Paralinx LLC

Willys Widgets

Rig Wheels Passport

Visual Products

Glidecam

rebotnix Technologies

Paralinx LLC

Gamma Ray Digital Inc

Willys Widgets

Abel Cine

CineTape

Technodolly

Metropolis Post

Broadcast Solutions Inc

CineLab

The Slider

FJS International, LLC

Ritter Battery

Tai Audio

Aerial Filmworks

Opal

Wooden Camera

Media Blackout - Custom Cables and AKS