Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

DRAGON 6K SENSOR

Nathaniel Kramer

Nathaniel Kramer

Well-known member
Joined
Dec 13, 2011
Messages
387
Reaction score
0
Points
16
Location
New York City
So here is my question, but first let me preface that i have never shot my Epic on anything less than 5k. From what i understand if i down res to 4 k or 3k or less, then the sensor crops my image because it uses less of the sensor- it does not down res the sensor. Which imho is ridiculous because I want to shoot my lenses full frame, not cropped. So, this begs the question - how will dragon achieve 6k? If its a larger sensor, then we are all screwed ( dragons frame size will be larger than s35) unless we want to buy Imax lenses - 65mm i believe? And if the dragon actually and hopefully just squeezes more pixels into the S35 frame, then why the hell does the mx sensor not do the reverse when shooting resolutions less than 5k??
 
Dragon's sensor design will feature smaller pixels and a slightly larger sensor size. Most lenses that cover full 5k today will likely work with the Dragon sensor. These are things we know at the moment.

Keep in mind that 5k on the Epic at the moment is larger than S35. S35 is closer to 4k on Epic and Scarlet. When Dragon arrives S35 will be 5k (approx).
 
Sensor will only be minutely larger... smaller photosites as I understand it. FF lenses will still cover prety much in the same way they will cover the MX sensor. I agree with shooting the max output and downrezzing in post. The only time I have shot less than 5 k is when I'm shooting higher framerates and on a couple of occasions, when I had a certain lens mounted and wanted to suddenly get a close-up. When you go from 5k down to 2k, you are suddenly zoomed in and only shooting the sweet spot of the lens (cropped sensor.)
 
Elsie, thanks for that info but wouldn't we just be better off shooting 5k and cropping in post for that closeup? How much is the crop zoom factor going from 5 to 4 k? 5 to 3k?
 
This is all covered in this forum so you could just such the threads.

but it's quite simple actually. Since downsample quickly is quite hard to do.. it simply takes a lot of computing power... So if you think twice about it. The epic or no other camera that I know of today is able to take a 5k image stream with lets say 400fps per second and down scale it in real time to 1k. For that simple math/ data reason RED do what is possible, they capture less pixels at higher frame rates. Capturing huge amount of data is one trick, processing it is another trick... Leaving us with red rockets and larger computing bays.

The fact that the Epic actually has a video out that looks as good as it does from the 5k source is quite amazing even though it's only in 24p-60p. Getting the onboard playback from 5k to HD is even more amazing.

Regarding the sensor of th dragon.. it will be slightly larger than the 5k MX sensor. Still small enough to fit within the projection circle of S35 lenses. Also the 6k dragon sensor has higher "dpi" / more photo sites per mm then the mx sensor. So a 5k frame from the dragon will be a slightly smaller in sensor size than from the MX. So this will give you non vingetting 5k for the lenses that just does not cover the MX sensor.
 
Phil Holland said:
Dragon's sensor design will feature smaller pixels and a slightly larger sensor size. Most lenses that cover full 5k today will likely work with the Dragon sensor. These are things we know at the moment.

Keep in mind that 5k on the Epic at the moment is larger than S35. S35 is closer to 4k on Epic and Scarlet. When Dragon arrives S35 will be 5k (approx).
Click to expand...

I missed that, you are right S35 is less in horizontal width.

S35= 25mmx19mm for 4:3.

But as the dragon and MX is 16:9 they still do no go further out in the projection circle than S35 correct?
 
MX I believe is a 1.9:1-ish sized sensor.

S35 is typically a 4:3 aspect ratio. Any sensor that is 16x9 but claims a S35 equivalent fits within that "S35 box". You loose vertical resolution, but 4:3 is really not much of a desired delivery format or even capture format these days. However, if you prefer to shoot anamorphically it sure is.

Nathaniel, you may find this useful for some of your questions. Here's a link to the Epic Data Sheet:
http://www.reduser.net/forum/showthread.php?77349-Epic-M-and-X-Data-Sheet
 
Nathaniel Kramer said:
Elsie, thanks for that info but wouldn't we just be better off shooting 5k and cropping in post for that closeup? How much is the crop zoom factor going from 5 to 4 k? 5 to 3k?
Click to expand...

Good point Nathaniel. Another point I should mention though, is that you use a lot less of your SSD storage when cropping the sensor to 2k, so that's to be considered. But like you, I prefer to shoot FF and crop in post, if need be. Can't wait for the 6k Dragon.
 
For me some Zoom lens can't use 5k ,like Optimo 17-80,24-290, so i always using 4k for that kind of Zoom
 
Phil Holland said:
S35 is typically a 4:3 aspect ratio. Any sensor that is 16x9 but claims a S35 equivalent fits within that "S35 box". You loose vertical resolution, but 4:3 is really not much of a desired delivery format or even capture format these days. However, if you prefer to shoot anamorphically it sure is.

]
Click to expand...

S35 film was never a projection format, so there was never really a tyical aspect ratio.
It was used in 4x3 for TV, even as an extraction from 3 perf!
S35 was historically used for effects work where optical printing would be used, the smaller grain offset the extra generation loss.

Once a DI became affordable then many films were shot either 3 or 4 perf with a 1: 1-185 or 1:240.

An anamorphic frame is smaller than S35 as it's an original projection format, space is needed for the optical soundtrack.

Different manufactured will quote slightly different sizes for S35 Film , there were at least 2 standards!
 
Phil Holland said:
MX I believe is a 1.9:1-ish sized sensor.

S35 is typically a 4:3 aspect ratio. Any sensor that is 16x9 but claims a S35 equivalent fits within that "S35 box". You loose vertical resolution, but 4:3 is really not much of a desired delivery format or even capture format these days. However, if you prefer to shoot anamorphically it sure is.
Click to expand...

Well, to be accurate, Super-35 really just means using the Full Aperture (silent) width of the 35mm negative instead of Academy/1.85/Anamorphic (sound).

In the 1950's there was a format called SuperScope which involved shooting in Academy (1.37 4-perf) and framing for a 2:1 extraction, which was then printed to 2.35 CinemaScope with a 2:1 hard matte. But once matted widescreen projection to 1.85 became more commonplace, SuperScope seemed like a waste of effort just to get to a 2:1 image. So the inventors of SuperScope experimented with using the old Silent Era Full Aperture (1.33 4-perf) and framing for a 2.35 extraction and blowing that up to CinemaScope. That process was called SuperScope-235 but never caught on, I'm not sure any features were even shot in the process.

In the early 1980's, John Alcott was about to shoot "Greystoke", which has a 70mm release planned, and he didn't want to shoot in anamorphic 35, partly because of the low light levels he liked to use. So he thought about the old 2-perf Techniscope format and decided to revive it but use a 4-perf 35mm camera shooting Full Aperture and framing for 2.40 extraction. He put a hard matte in the camera gate so that something like a 1.50 or 1.66 area was exposed because he didn't want to protect all the way to 1.33 for a 2.40 image. He called it Super Techniscope. "Top Gun" and "Silverado" used the format next, and later, James Cameron on "The Abyss" and "Terminator 2". Some of these projects used a hard matte in the 4-perf camera to expose Full Aperture width but not the full height. The format was called "Super-35" eventually because "Super Techniscope" had associations with Technicolor Labs and Super-35 was more generic.

Eventually, people started also shooting in 3-perf Full Aperture, which sometimes is called Super-35 since the 2.40 extraction area is the same size whether you shoot 4-perf Full Aperture or 3-perf Full Aperture. But generally it's just called 3-perf.

But "Super-35" basically means using the Full Aperture width of the 35mm negative to extract a widescreen image, so it has nothing really to do with 4x3 other than the fact that the 4-perf 35mm negative is that shape. Super-35 in itself does not have an "official" aspect ratio other than what you compose it for, whereas 4-perf 35mm Full Aperture does have a 1.33 : 1 aspect ratio.
 
Super-35 wasn't really used for effects work for non-Super-35 movies too often, other than perhaps rear-projection plates and other background plates for compositing, for no other reason than why not use the full area of the negative for that stuff.

But the difference in size between the 1.85 area of a Super-35 negative versus an Academy negative wasn't going to give you much grain reduction, and the 2.40 area of Super-35 is smaller than in anamorphic 35mm so there is even less reason to use Super-35 for effects work for an anamorphic production (though there are some pre-D.I. examples, such as the "Lost in Space" movie.) Normally an effect person shooting plates for optical printer compositing would try to use VistaVision or 65mm if they had the budget for it, if they wanted to minimize grain.

Nowadays with digital compositing and digital grain management, all sorts of formats are shot for effects work even for an anamorphic 35 production. Elements might be shot in Super-35 instead of VistaVision because maybe they are going to be shrunk down in the composited frame, for example, or maybe the grain isn't so noticeable. These days you don't have the extra worry about grain build-up from working with dupe elements in an optical printer, just the grain in the original negative, so the need to shoot larger negative formats for effects work has declined, the main advantage now is just the ability to reposition elements in the frame.
 
David Mullen ASC said:
Super-35 wasn't really used for effects work for non-Super-35 movies too often, other than perhaps rear-projection plates and other background plates for compositing, for no other reason than why not use the full area of the negative for that stuff.

But the difference in size between the 1.85 area of a Super-35 negative versus an Academy negative wasn't going to give you much grain reduction, and the 2.40 area of Super-35 is smaller than in anamorphic 35mm so there is even less reason to use Super-35 for effects work for an anamorphic production (though there are some pre-D.I. examples, such as the "Lost in Space" movie.) Normally an effect person shooting plates for optical printer compositing would try to use VistaVision or 65mm if they had the budget for it, if they wanted to minimize grain.

Nowadays with digital compositing and digital grain management, all sorts of formats are shot for effects work even for an anamorphic 35 production. Elements might be shot in Super-35 instead of VistaVision because maybe they are going to be shrunk down in the composited frame, for example, or maybe the grain isn't so noticeable. These days you don't have the extra worry about grain build-up from working with dupe elements in an optical printer, just the grain in the original negative, so the need to shoot larger negative formats for effects work has declined, the main advantage now is just the ability to reposition elements in the frame.
Click to expand...

Excellent Mr. Mullen, what kind of extractions do you believe we will be able to achieve with the new sensor?

- Paul Ellington
 
Phil Holland said:
MX I believe is a 1.9:1-ish sized sensor.

S35 is typically a 4:3 aspect ratio. Any sensor that is 16x9 but claims a S35 equivalent fits within that "S35 box". You loose vertical resolution, but 4:3 is really not much of a desired delivery format or even capture format these days. However, if you prefer to shoot anamorphically it sure is.

Nathaniel, you may find this useful for some of your questions. Here's a link to the Epic Data Sheet:
http://www.reduser.net/forum/showthread.php?77349-Epic-M-and-X-Data-Sheet
Click to expand...

We are getting our Hawks so then I would love 4:3 full sensor.
And yes MX it's not precisly 16:9 but "a more letter boxed" format than FF. thats what I meant.
 
David, at what point in time, when Super-35 became popular, did camera manufacturers and larger rental houses start to "center" the lens port on the full aperture gate (away from the off set Academy gates)?
And the reverse of this: when did camera manufacturers (like Mitchell & B&H) change their camera designs to off set the lens ports from centered on the aperture (silent era) to centered on Academy ("talkies" era)?
I own several Moviecams that have dual PL mounts (S-35 & standard...built in the late 80's & mid 90's), yet my mid 80's 35-3 has it's PL lens port Academy offset.
I also have a few Mitchell's with full aperture gates, but I don't know if they are Academy or full aperture centered.
 
You must log in or register to reply here.
Back
Top