RED DSMC specs detailed, prices start at $28,000 for EPIC-X, lower-end Scarlet still

[fancourt]

Originally Posted by Athiril

The current Red One is 11 stops, and the new Mysterium Monstro which is supposed to go into the Epics and the Scarlet FF35 is allegedly 13 stops.

Maybe, have you done your own tests? I've done a lot of side by side exposure tests over the years. Most dp's rate the r1 at about 7-9 stops. The shadows are/were extremely noisy, and most excluded those exposures as unusable hence their lower rated latitude. 5219 has around 12-14 stops depending on who you talk to.

[Athiril]

Allegedly, like Reala allegedly has 15+ stops.

Plus there is a difference between DR and latitude, though related.


I wont be using 5219 for any of my own stuff period I think. I dont plan on doing studio shooting.

And such a ridiculous dynamic range... why would you need that in a studio? You're supposed to be controlling your contrast ration in a studio.

Location everything for me.


edit: and as for their rating, of course you dont want anything in the deep shadows mixed in with the noise floor and lack of bits used to represent that portion of the image, especially for people who can afford such a camera, and will be having serious post work.

[fancourt]

So then why quote it has having 11 stops?
Its not a real world number, only one that exists from regurgitation of numbers off a website. Using 5219 in a studio saves money, the majority of studio based drama is shot on 19, it enables you to move faster. It could be argued that red hasn't picked up in this market as its a little too slow.

[Athiril]

Because its probably been measured as having 11 stops of separation between the light falling on an object as what ends up as pure black and pure white on the sensor.

Objective numbers are real world numbers.

Quoting it as 7 stops for DP x and 9 stops for DP y would be entirely innaccurate.

7-9 stops is not a real world number, that is subjective, and the fact that it varies person to person shows that it is neither correct nor accurate, and is dependant upon application.

Maybe they like to place their shadows 2-4 stops above straight black, that's choice.

I'm sure you'd find something completely different for the few productions where the DP directs the scene to be lit and shot as an end results and used as is (ie: those who have chosen to get what they want in camera and not in post), not an imtermediate intended to rely upon post-production.

[Dark_Ernie]

This may not be the most appropriate comment in the world...but this thread is making me so hot.

[fancourt]

Measuring, cynically in a lab is not indicative of the camera's performance, and is generally undertaken by manufacture's and those who don't have the means to shoot anything worth watching. Measuring from white point down to the noise threshold will not reveal what a camera is capable of, or how good its sensor is, or how much latitude you have to sculpt your scene.
Shooting night exteriors by rating 50d like 500 is possible, but not a real world scenario, and if you did it would end up in tears. Rating the r1 at 11 stops is possible, in the same kind of silly and inexperienced way, but to this day I have not seen one camera test shot in a real world scenario that proves this. Shooting in daylight/tungsten will also impact on digital capture as well, especially with older non-beta build's.
As a side note I've read some people rating the f23 at 13 stops, which has actually been out of beta and in production for a while now.

[Squeezer]

Originally Posted by ][ XpLoiT ][

retails for $28,000 (ouch).

I sense a Group Buy coming on

[][ XpLoiT ][]

Originally Posted by Squeezer

I sense a Group Buy coming on

maybe a time-share buy

[Athiril]

Originally Posted by fancourt

Measuring, cynically in a lab is not indicative of the camera's performance, and is generally undertaken by manufacture's and those who don't have the means to shoot anything worth watching. Measuring from white point down to the noise threshold will not reveal what a camera is capable of, or how good its sensor is, or how much latitude you have to sculpt your scene.
Shooting night exteriors by rating 50d like 500 is possible, but not a real world scenario, and if you did it would end up in tears. Rating the r1 at 11 stops is possible, in the same kind of silly and inexperienced way, but to this day I have not seen one camera test shot in a real world scenario that proves this. Shooting in daylight/tungsten will also impact on digital capture as well, especially with older non-beta build's.
As a side note I've read some people rating the f23 at 13 stops, which has actually been out of beta and in production for a while now.

Scientifically, not cynically.

I whole heartedly disagree, its the only way to measure dynamic range, as thats what dynamic range is, so the current RED is 11 stops assuming their testing was on the level and not marketing scam.

And like I said before, its perfectly fine to plan to make use of the whole range, if you intend not to treat your footage to anything in post production, ie: you finish your footage in the cinematography stage, which very few do, though some still work that way.

Accurately, and scientifically measuring the dynamic range and noise floor is an excellent evaluation of a sensor's performance, and is the only way, giving a numerical guestimation is just completely useless, as you said yourself, it changes from person to person, but the performance of the sensor isn't changing, it's fixed.

There are reasons to shoot everything in a 3:1 or similar ratio, but this does not at all alter what the dynamic range is, edit: some of those reasons being bit allocation 'per stop' or every per x fraction of a stop from the shadows of the sensor, on top of the noise floor.

It is exactly the same as film, you do not use the entire dynamic range of the film, or your highlights become too dense, and your shadows too thin - as well as the noise on the film in those areas (yes noise, there is no grain left on colour film (typically) but an impression of the grain left in the dye clouds) will be compounded by a scanner's noise from being on the extreme ends of the scanner sensor - one of the reason's some people may pre-flash their film.

It's not really right to say latitude either and give a single number, thats not telling anything useful, a film may not have even underexposure and overexposure latitude at it's manufacturer rated speed.

And in comparison to film, digital has 0 stops of exposure latitude, you can say it has several stops of underexposure latitude from the analogue signal running through an electronic amplifier (raising ISO/sensitivity) but this has to be planned pre-exposure.

And it is possible to shoot 50D @ 500 without it being as bad as you say, though it needs to be processed in non-standardised development.

In non-standard development film can be pushed and pulled a -lot- more than what the traditional wisdom of what you would find floating around on the internet.

I've pulled colour negative about 14 stops below box speed in development (though my exposure was only +10 stops on top of box speed, the image was still kind of decent, but would have been a lot better with an extra couple of stops)

I've also pushed Reala (of all) over 4 stops in my own aminophenol developer trying to improve the weak colour developing function of aminophenol - something which I'll try to repeat and use a colour developer on afterwards, the developed silver images have excellent tonality from shadows to highlights.

But this could be done in the some formulation by developing with that as a first developer, fixing, using a rehalogenating bleach, re-exposure, and then putting it through colour developer (C41 developer or other) until completion - which will result in a negative, with the flexibility of being able to use a first developer.

Underexposing a colour negative film 3 and a quarter stops and just running through the standard process of course will result in a pretty dodgy image.

[FB008]

The only thing epic about it is the price.

[proffesso]

Originally Posted by FB008

The only thing epic about it is the price.

the cheapness? or am I missing somthing?

[SyN]

Originally Posted by FB008

The only thing epic about it is the price.

how about it's fail?

[FB008]

Originally Posted by proffesso

the cheapness? or am I missing somthing?

Originally Posted by SyN

how about it's fail?

Sarcasm, people

Epic as in huge price tag, which isn't a good thing. Or it will just keep the fanbois dreaming.

[SyN]

fanbois always need to look up to things they cant afford to have something to worship

epic is a bad a name as holden epica(r).

[FB008]

Originally Posted by SyN

fanbois always need to look up to things they cant afford to have something to worship

epic is a bad a name as holden epica(r).

more like epicfail