Red camera, Obsolescence Obsolete, or Obsolescence overboard?

[androbot2084]

What I would like to see is pricing which is not based on dollars but rather based on hours. For example let us say that the price of a Red One will cost you 1000 hours based on your hourly wage which could be anywhere from $10,000 to $50,000 depending on how much money you make. Of course this would mean that the rich would be subsidizing the poor but I think in the long run this would free up a lot of money for research and development so that everyone would benefit and technology would no longer be crippled because everyone would have equal acess to high end technology. The only thing that would hold you back would be your commitment of how many hours you would dedicate to financing your Red.

 

[Stephen Williams]

What I would like to see is pricing which is not based on dollars but rather based on hours. For example let us say that the price of a Red One will cost you 1000 hours based on your hourly wage which could be anywhere from $10,000 to $50,000 depending on how much money you make. Of course this would mean that the rich would be subsidizing the poor but I think in the long run this would free up a lot of money for research and development so that everyone would benefit and technology would no longer be crippled because everyone would have equal acess to high end technology. The only thing that would hold you back would be your commitment of how many hours you would dedicate to financing your Red.

Hi,

It would be way cheaper for the high earners to shoot on film! or rent cheaply from a low earner.

Stephen

 

[James Gardiner]

James, you still seem to be missing the point. This is a RAW camera - you need to oversample, it's not a luxury. If you want good 1080P RGB, you need to shoot with a Bayer-filtered sensor that has 70% or so more pixels in order to generate a good 1080P image with a minimal amount of artifacts.

1920 x 1080 is not "all you need" in order to get 1920 x 1080, not with a RAW camera. You need more than 1920 x 1080 if all you want is 1920 x 1080 -- do you get it? If you don't oversample, you can't use as heavy an OLPF to get rid of aliasing because now you are trying to compensate for resolution loss from debayering, so you end up with some ugly artifacts in the final color image.

The only question is whether you would prefer that a 4K or 3K RAW capture be processed into 1080P RGB inside the camera and recorded as 1080P in real time... or done in post later using a possibly better debayering algorithm.

You still seem to be questioning the need to shoot 4K RAW when all you want is an HD image -- if you want a good HD image, then you should at least be shooting 3K RAW. Not to say that you can't get away with 2K or less, but it is not optimal due to debayering artifacts and line resolution.

I'd say that modern 3-CCD 1080P cameras would deliver similar quality to a 3K RAW camera.

Again, your arguments against 4K have to be clarified because you are jumbling up the issues of acquisition resolution, recording resolution, and mastering & release resolution. My suggestion is that you go back through your article and everytime you use the vague term "4K" you change it to be specifically "4K RAW Bayer" to describe camera capture or "4K RGB" to describe some final mastering format, and see if your arguments against the RED shooting 4K RAW mode if you want 1080P still make sense.

Firstly, I agree with your NEED to quote 4K as RAW or RGB. As most people who talk to me about RED cameras don;t know what a bayer filter is.. As such, by default, in general I find people expect it to mean 4K-RGB.

In general, however, when some one comes up to me and says, I shot on a RED1 in 4K, it incorrectly implies (To them) that 4K luminance and a 4K workflow. Usually always they go down to 1920.

What I am trying to point out here, and of which you did gloss over, is that, RED1, 5D, and other sensors that sample at very high mega pixel but interpret in real time down sample output to 1920x1080 4:4:4 12bit should look just as good as 4K RAW files. This leads to a codec to store it but that is a matter of preference. However, you have less processing to do and less data to deal with. As such, saves time, and money.

(Note: a comment on this list that says the 5D does not down sample, but drop lines to archive the 1920x1080. Where can I get canon material to back this clame? Please post link.)

 

[Tim Wildenhain]

Note: a comment on this list that says the 5D does not down sample, but drop lines to archive the 1920x1080. Where can I get canon material to back this clame? Please post link

Read the following thread: http://reduser.net/forum/showthread.php?t=23082


It's incredibly doubtful that you'll get Canon to announce this on any publicly available format - it's not common for companies to announce "Yeah we added this feature; but the way we got it to work is a cheap parlor trick, good luck getting it to work right for you"

The above thread has some detailed tech info from Graeme and other members on the board that took apart the zone plate shot on a 5D.

Dont get me wrong; The 5D is definately on my wish list; but not for video purposes.

 

[Jason Sinclair]

Im one of those people that doesn't get into the finer technicalities but i am finding in real world situations that 4k first color correct into 2k color correct is a very simple, highly versatile system that beats any workflow i have previously worked with, in consideration of the final outcome. There's simply a lot more information to work with and that translates to better renders in color correcting. The tests i have seen with the 5d so far have been less than appealing and i think that is because they were color corrected and the information loss was too great for it to be balanced as much as say a red cam. (see the zacuto tests) I'd love to see more tests of course and will keep an open mind. I think some of the newer hd cams are better used for your argument. I think 4k is perfect for the current situation in my opinion from what i am seeing around me... There's a number of other factors that makes red very appealing most notably it's non proprietary lens use and its "open source" approach to a design process that allows for third parties to come on board. Of course with the close collaboration of it's customers in the design process sets it apart and it's hard not to become a fanboy when the Ceo of a company is asking for your opinion and implements your ideas if they merit well. There is the democratization of ideas and this translates through the hive mind, in sometimes heated debates (as seen here), to take into account many different viewpoints (from projectionists to dop's to post specialists etc.. ) and attains a very nice balance between competing factors. It permits the inclusion that no one person has the total answers to everything, although we all like to think we do...

 

[Albert Armes]

It makes sense that skipping lines on the Canon sensor readout is what they're doing. Their fastest camera with a mirror shutter does 10fps. There's no need for them to design a sensor that has a faster fps than 10fps, and with a mirror shutter, skew is not an issue. Designing that sensor to read out at 20fps or 30fps would just be wasteful. Line skipping turns a 10fps sensor into a 30fps sensor for video. That zone plate did look like there was something very much different going on in the vertical direction than the horizontal direction. It all makes sense.

 

[Chris Kenny]

What I am trying to point out here, and of which you did gloss over, is that, RED1, 5D, and other sensors that sample at very high mega pixel but interpret in real time down sample output to 1920x1080 4:4:4 12bit should look just as good as 4K RAW files. This leads to a codec to store it but that is a matter of preference. However, you have less processing to do and less data to deal with. As such, saves time, and money.

Sure. If the real-time debayer and downsample performed by a chip using a few watts of power in-camera is as good as the the results of letting an 8-core Mac Pro tower spend 12 seconds processing each second of footage.

Mind you, this isn't totally impossible, because specialized processors can be a lot better at specific tasks than general-purpose ones. But color me skeptical.

Anyway, it looks like Red is planning to offer this option with its new cameras, and maybe even with a special firmware build for the Red One.

 

[Vincent Rice]

Digital cinema exhibition will certainly not end up at 2K. One of the reasons that exhibitors are baulking at DCI rollouts is that it is already clear that 2K will not be enough. IF they are going to enter into the various funding schemes and partnerships (or heaven forbid, pay for it themselves) they want the gear to provide a superior-to-home experience for as long as possible. However the present 4K options are a little too costly. A non-Sony-priced 4K projector that could slot into the DCI infrastructure would be of considerable interest... Jim?

5K is therefore likely to be the sweet-spot acquisition resolution for DC using Epic/Scarlet. 5K RAW should produce excellent 4K output.

Of course nothing further is going to happen on the DCI front for a year at least due to the prevailing economics

 

[Vincent Rice]

What I am trying to point out here, and of which you did gloss over, is that, RED1, 5D, and other sensors that sample at very high mega pixel but interpret in real time down sample output to 1920x1080 4:4:4 12bit should look just as good as 4K RAW files. This leads to a codec to store it but that is a matter of preference. However, you have less processing to do and less data to deal with. As such, saves time, and money.

... and completely negates one of the great advantages of a RAW workflow... being able to drastically alter exposure and colour balance in post without destroying the image

(Note: a comment on this list that says the 5D does not down sample, but drop lines to archive the 1920x1080. Where can I get canon material to back this clame? Please post link.)

You must realise that Canon are not going make this explicit in any literature.

 

[George Wilcox]

2K is not as good as 35mm film, producing pixels on a 50 ft. wide screen that are .3 inches square, 4K is in the zone at .15 inch/pixel, 6K is a little better than the finest resolution 35mm film at .1 inch per pixel. 9K is on par with IMax's 70mm format. While today's digital projectors run $100K, I suspect Jim will change that too, just like he did with the cameras. This could spawn a renaisance of bigger screens, because more resolution on big screens makes for a richer experience (both visually and financially). Our brains can process approx. 1.2 GB/sec. of visual information, so I'm sure your brain will appreciate the better resolution, color depth, etc. more, even if you don't think it will, at least up to 6K. But since you don't process the whole frame when watching a film on a big screen - you process the center of your attention, so 9K and 28K would no doubt be amazing. I'm looking foward to it sooner rather than later for the reasons enumerated by others on this thread.

 

[Bing Bailey]

I don't think people should be trading on their ability to afford a certain technology it should be on their talent which will show through whether they use an epic or a scarlet or anything else. people hire james t mathers because he's a great dp not because he has huge pockets and can afford the best gear.

while I'm not targetting you james there are an elite bunch on these boards that want to deny the rest of us access to this type of quality/technology. if they are so damn elite let them compete on their skills not their wallets. I'm glad jim isn't listening to these people. if he did epic would be 300k and he'd only sell 50 of them. if you're the elites of cinema what are you so afraid of. after all the rest of us are just digital cinema peasants.

since the technology in scarlet and epic is going to be very similar RED will amortize the costs in two ways, in the higher prices on epic and with volume on scarlet so he is putting the technology out at an appropriate price on both and people who buy either are contributing. I don't see a difference.

 

[J. Bernard Vallon]

Valid point.

I say in the next decade 'who has what gear' won't matter spit. Talent will be king, as it is now. Once canon, nikon, panasonic, red, sony, and whoever else all produce great 4k+ cameras with 13 stops DR with s35 size sensors or bigger, all for under 10 grand, everyone on the street will have great gear, but talent will have a huge premium.

I hope our audiences can tell the difference.

 

[James T Mather]

I don't think people should be trading on their ability to afford a certain technology it should be on their talent which will show through whether they use an epic or a scarlet or anything else. people hire james t mathers because he's a great dp not because he has huge pockets and can afford the best gear.

while I'm not targetting you james there are an elite bunch on these boards that want to deny the rest of us access to this type of quality/technology. if they are so damn elite let them compete on their skills not their wallets. I'm glad jim isn't listening to these people. if he did epic would be 300k and he'd only sell 50 of them. if you're the elites of cinema what are you so afraid of. after all the rest of us are just digital cinema peasants.

I agree to some extent - but I have found that there are actually very few obstacles to people working in the film business. Anyone with dedication, common sense, reliability and some degree of ability can work in the field IMVHO. Red is the first pro camera I have owned. Prior to owning a Red camera, I was hired purely as a cameraman and the gear separately rented from Panavision. The only kit I brought to a shoot was a light meter. In short - Owning gear hasn't had much to do with the success of any cameraman I know. I know plenty of people with gear which sits on shelves mostly gathering dust.

In my opinion people who were going to make it would make it anyway. Right now anyone can buy a Red package and get some work on novelty value but as word spreads of their work, only people whose work is up to scratch will make it - or be doomed to shooting super-low/no budget work - in which case you will always be busy.

The truth is that very few people think they have no ability/talent. The argument that an elite have been employing "exclusionist" tactics based on equipment is IMVHO not true. Rather some people who haven't earned their stripes over years of shooting and carving out a career path feel that owning the camera is a shortcut into the game.

And who knows - maybe they're right.

 

[James T Mather]

Addendum to the "Crippling" conversation.

My concern is that I have invested a tidy sum in Red products - and they have done me well - its a great system. I plan on upgrading both my Red cameras. However, I now have a choice - upgrade to the EPIC X with a s35 sensor but less DR for 28 G's (minus my 17.5k investment) - or get a scarlet with a FF35 extended range sensor for nine grand plus.

I am somewhat (unreasonably, I grant you) perturbed that I have spent quite a lot of money with RED and now face having an inferior sensor as my "reward" package (which I fully realize is very kind of Red and above the call of duty). It has presented us Red One owners with something of a Hobson's choice. Avail of the offer (Lesser sensor) or buy the FF35 Epic and lose out on the extras.

All that concerns me is my reward package is inferior sensor technology to their entry level system. I am not concerned with a different body finish (ie:titanium colour) but the chip itself. My other option is to get the FF35 Epic sensor (with my generous trade in) but then I am committed to buying a few grands worth of extras (I estimate with the modules) in addition to the extra chip cost. Multiply that by 2 cameras and it mounts up a lot.

I am not trying to crush anyone's dreams or "cripple" cameras (to use the emotive language of this debate) - I have invested a lot of money in this, is it not unreasonable to want the best sensor I can get considering I am spending a lot more than nine grand with Red? I don't want my 28k camera having lesser specs than their entry level camera - why is this unreasonable?

If I might cheekily suggest two possible solutions?

SOLUTION A
​

Allow red one early adopters to get the additional bits 'n pieces with whatever sensor they opt to upgrade to (ie: the Epic x package with whatever brain you opt to get (within reason)- Naturally the costs of each package would be in line with the sensor - ie : the Epic X package, The FF35 X package (at higher trade in cost, naturally) or the 645 X package (at duly increased package cost).
​

or B
​

Put a Monstro in the Epic X (all I am referring to is the extended DR - yes I realize this will up the price) - most will only ever use cine lenses so are not looking for FF capability - just the best DR/ASA available at the time.
​

For all of this griping, I commend Red and Jim Jannard for remaining true to his/their word. Unfortunately for them, it seems you can't please everyone all of the time and we all have our agendas. They make great cameras and I imagine I shall be utilizing their products for a long time to come. The Red One has been, quite simply, a quantum leap in digital photography. Long may they last.

 

[Meryem Ersoz]

This is the part of the new specs that are causing the most angst among the existing users...that a $9K camera will have a better sensor, with greater latitude, than a $28K camera is somewhat mystifying.

If you are going to have distinct product lines, with an accompanying price hierarchy, then the price/feature hierarchy should make sense and be easily explained (especially to our clients, to whom we will have to justify the added expense of EPIC, in order to sell EPIC...).

Clearly, from the amount of debate and discussion, the price/feature hierarchy has a lot of people concerned and uncertain.

These are real concerns of real users with real clients. Re-framing the discussion to some BS about crippling cameras is just inflammatory and unconstructive. We're all acknowledging that there is a price/feature hierarchy, but there is some difference of opinion regarding what that should look like.

It doesn't make sense to me to try to sell them on a $28K camera when the $9K sensor is superior - and selling them on frame rates, well, that is simply not a selling point that equates to better quality. It's more like a nice perk...

 

[DRappazzo]

I agree to some extent - but I have found that there are actually very few obstacles to people working in the film business. Anyone with dedication, common sense, reliability and some degree of ability can work in the field IMVHO. Red is the first pro camera I have owned. Prior to owning a Red camera, I was hired purely as a cameraman and the gear separately rented from Panavision. The only kit I brought to a shoot was a light meter. In short - Owning gear hasn't had much to do with the success of any cameraman I know. I know plenty of people with gear which sits on shelves mostly gathering dust.

In my opinion people who were going to make it would make it anyway. Right now anyone can buy a Red package and get some work on novelty value but as word spreads of their work, only people whose work is up to scratch will make it - or be doomed to shooting super-low/no budget work - in which case you will always be busy.

The truth is that very few people think they have no ability/talent. The argument that an elite have been employing "exclusionist" tactics based on equipment is IMVHO not true. Rather some people who haven't earned their stripes over years of shooting and carving out a career path feel that owning the camera is a shortcut into the game.

And who knows - maybe they're right.

Great posts, but in my limited experience I have found that not having the ability (i.e. money) to afford to buy your own gear is a huge barrier to breaking into the field. It might not be a shortcut, but it is an advantage for sure. But with Red and others making giant leaps in camera technology, not only in technical specs but in price point, in the future everyone will be shooting high res footage and that is when the truly talented will be separated from the pack.

Until that day I will continue trying to figure out how to make it on based on knowledge, passion and little bit of luck.

 

[Tim Wildenhain]

I am somewhat (unreasonably, I grant you) perturbed that I have spent quite a lot of money with RED and now face having an inferior sensor as my "reward" package (which I fully realize is very kind of Red and above the call of duty). It has presented us Red One owners with something of a Hobson's choice. Avail of the offer (Lesser sensor) or buy the FF35 Epic and lose out on the extras.

This is the part of the new specs that are causing the most angst among the existing users...that a $9K camera will have a better sensor, with greater latitude, than a $28K camera is somewhat mystifying.

Sorry guys I have to disagree with you. They're introducing 2 new sensor lines. The entry level to the bigger Monstro is the Scarlet. Which may have the same sensor, but will not have the same usability and flexibility that you have come to expect and count on with the R1. You will cripple yourselves with if you choose to "upgrade" from R1 to Scarlet FF35 by going the cheap route. You have options, they've been laid out on the table.

There's 2 steps from R1 to Monstro - with being around in the semiconductor industry long enough, I know that Red in thier engineering wizardry, have decided to give the best compromise which is offering you Epic X which is a 1.5 step upgrade, as opposed to a 1 step upgrade. I wont say anything further at fear of starting yet another fierce argument, but suffice it to say that AMD went this route and NVIDIA went this route as well, both were insanely successful at it. Only difference, is the Epic X wont be available to anyone outside the R1 owners list.

 

[Meryem Ersoz]

suffice it to say that AMD went this route and NVIDIA went this route as well, both were insanely successful at it. Only difference, is the Epic X wont be available to anyone outside the R1 owners list.

Since I'm a shooter, not an engineer, I'm not sure that I understand this post at all. In what context was AMD insanely successful? That company appears to be on life support, at the moment. As is Nvidia, although their pipeline has many innovative products that I believe will rescue them over time.

I don't really understand the engineering terminology of 1 step, 2 step, or 1.5 step...

what I do understand is
Monstro = latitude is 13
Mysterium X = latitude is 11

If image quality is your priority, then sensor quality is a priority, and things like more frame rates and other features may be desirable but not essential. So you may be getting more of something you don't really care about (additional frame rates), while receiving less of something you do care about (latitude).

I'm not complaining here. Images from Mysterium are amazing, I'm sure that I will be ridiculously delighted with Mysterium X.

 

[Tim Wildenhain]

Since I'm a shooter, not an engineer, I'm not sure that I understand this post at all. In what context was AMD insanely successful? That company appears to be on life support, at the moment. As is Nvidia, although their pipeline has many innovative products that I believe will rescue them over time.

I don't really understand the engineering terminology of 1 step, 2 step, or 1.5 step...

what I do understand is
Monstro = latitude is 13
Mysterium X = latitude is 11

If image quality is your priority, then sensor quality is a priority, and things like more frame rates and other features may be desirable but not essential. So you may be getting more of something you don't really care about (additional frame rates), while receiving less of something you do care about (latitude).

I'm not complaining here. Images from Mysterium are amazing, I'm sure that I will be ridiculously delighted with Mysterium X.

AMD back in 99 came out of bankruptcy with the Athlon and dominated the processor market (became much more popular than Intel) for a long while. I'm not talking current trends in the business world, I'm talking about successful engineering advances. NVIDIA as well, think GeForce 3 era cards. Again, engineering advances.

Epic X is not a Mysterium X sensor, it's a Mysterium X Select sensor, which will be better than Mysterium X. Only the final release will show the difference of how that will appear on screen. I think that the reason the full specs of the Mysterium X sensor haven't been released as of yet, is because I'm not 100% sure that even Red knows exactly what those will be.

Could you enlighten me as to where your latitude specs came from? If they were before the "engineering breakthroughs" you may be getting upset over nothing. I saw nothing since Dec 2nd stating what the DR was of any sensor.

As for the steps. That's not an engineering term, that's just an explaination of the different lines of sensors. To go from R1, one step, would be the S35Epic (comparable size sensor with upgrade) The next step is the FF35 Epic, which is a step above the previous. The Epic X with it's Mysterium X Select sensor is actually a better sensor and will be a middle ground between the Mysterium X and the Monstro lines.

Sorry if I'm being vague.

 

[Irmeli R]

***