Jump to content


Photo

4k numbers


  • Please log in to reply
82 replies to this topic

#1 Mike Brennan

Mike Brennan
  • Basic Members
  • PipPipPipPip
  • 581 posts
  • Cinematographer
  • London

Posted 20 September 2006 - 07:55 AM

Comparing numbers for uncompressed 4K output

RED states uncompressed RAW is 323MB/s @24fps 12bit linear.
Dalsa states reconstructed 4k is 1.2Gigbytes 12bit per channel.


Do I understand that the RED RAW at 323MB/s is bayer pattern uncompressed.(but not reconstructed)


Will Red have a reconstruced 4k output available on camera? (in the hope of a 4kfield monitor you understand!)



Mike Brennan
  • 0

#2 Max Jacoby

Max Jacoby
  • Sustaining Members
  • 2955 posts
  • Other

Posted 20 September 2006 - 09:01 AM

The Dalsa does output raw as well, doesn't it? I had a look at the camera and the Codex system at the Cinec and one of the guys from Codex said that reconstruction is done once the film is edited by Dalsa themselves at about 5 seconds per frame. The thing I don't understand is the size difference between the Dalsa and the Red, presuming the Origin outputs raw as well.
  • 0

#3 Graeme Nattress

Graeme Nattress
  • Basic Members
  • PipPipPip
  • 145 posts
  • Other

Posted 20 September 2006 - 09:29 AM

Comparing numbers for uncompressed 4K output

RED states uncompressed RAW is 323MB/s @24fps 12bit linear.
Dalsa states reconstructed 4k is 1.2Gigbytes 12bit per channel.
Do I understand that the RED RAW at 323MB/s is bayer pattern uncompressed.(but not reconstructed)
Will Red have a reconstruced 4k output available on camera? (in the hope of a 4kfield monitor you understand!)
Mike Brennan


4096x2304x(12/8)x24/1024/1024 = 324MB/s - so that's for 12bit packed linear sensor raw data for 4k.

There's not much point outputtting reconstructed data as you're just inflating the file size by a factor of 3 for no good reason. Best to reconstruct at the point of viewing.

Graeme
  • 0

#4 Mike Brennan

Mike Brennan
  • Basic Members
  • PipPipPipPip
  • 581 posts
  • Cinematographer
  • London

Posted 20 September 2006 - 11:13 AM

4096x2304x(12/8)x24/1024/1024 = 324MB/s - so that's for 12bit packed linear sensor raw data for 4k.

There's not much point outputtting reconstructed data as you're just inflating the file size by a factor of 3 for no good reason. Best to reconstruct at the point of viewing.

Graeme



Thats the idea, to see it in the field.

Is a overscanned 4k on a 1920x1080 display an option? i guess it is asking too much from a small camera body, is there a device that can do it live?


Mike Brennan
  • 0

#5 Sam Wells

Sam Wells
  • Basic Members
  • PipPipPipPip
  • 1751 posts
  • Cinematographer

Posted 20 September 2006 - 11:21 AM

4096x2304x(12/8)x24/1024/1024 = 324MB/s - so that's for 12bit packed linear sensor raw data for 4k.

There's not much point outputtting reconstructed data as you're just inflating the file size by a factor of 3 for no good reason. Best to reconstruct at the point of viewing.

Graeme


Well that's what I was first thinking when I read the above. I mean it'll end up as JPEG2000, or whatever display format - right ?

That is, IF the final image is digital.

If it's a 35mm film out then it's not about the point of viewing, it's about the transfer. Is the full reconstruction neccesary ? One might prefer to work with 4K as opposed to 2K here.

?

-Sam
  • 0

#6 Alan Lasky

Alan Lasky
  • Basic Members
  • PipPip
  • 55 posts
  • Other

Posted 20 September 2006 - 01:07 PM

Just for the record, the DALSA Origin outputs:

4096 x 2048 (50 pixels are 'black reference' so the true imaged resolution is 4046 x 2048)

16 bits per pixel LINEAR.

We most definitely record in RAW mode. We have on-set software for Bayer-to-RGB reconstruction called "Visualize" that allows the crew to see full resolution 4K frames for exposure analysis and color evaluation. CODEX has a quite brilliant 'quick de-Bayer' solution built into their box that allows for video resolution and 1K playback utilizing the CODEX virtual file system. It is not the full implementation of the high quality de-bayering algorithm but it works quite well.

As the representative from RED quite correctly pointed out, there is not much use carrying full RGB on the set; in our case it is the difference between 16 megs per frame versus 48 megs per frame. We use no compression of any sort on the material, no wavelets, no DCTs, no FFTs, simply the RAW Bayer output + metadata recorded as a single channel DPX file.

As for questions of the physical size of the Origin vs. the RED I have no direct experience recording 4K RAW imagery with the RED camera, so I am afraid I can offer no insight into how they provide that functionality. I do have a great deal of expereince recording 4K with the Origin and I know what sort of support infrastructure is required for our system.

Alan Lasky
DALSA Digital Cinema
  • 0

#7 Phil Rhodes

Phil Rhodes
  • Sustaining Members
  • 11943 posts
  • Other

Posted 20 September 2006 - 02:33 PM

Hi,

I'd be highly suspicious of any system that didn't have an HD-SDI output providing at least a monitoring-quality reconstructed RGB image.

Phil
  • 0

#8 Alan Lasky

Alan Lasky
  • Basic Members
  • PipPip
  • 55 posts
  • Other

Posted 20 September 2006 - 03:31 PM

Hi,

I'd be highly suspicious of any system that didn't have an HD-SDI output providing at least a monitoring-quality reconstructed RGB image.

Phil


Phil,
I assume your quote was directed at DALSA?

What we have for real time 'monitoring quality' output is a 1024x768 8 bit DVI output for viewing the image in real time. We do not use HD-SDI because we are not a 'video' camera and using a video based monitoring solution would be pointless, as well as cumbersome.

None of the professionals who have used our camera from the ASC, BSC, or DGA were the least bit 'suspicious' (let alone "highly suspicious') of our system's monitoring functionality. Once you understand that we are shooting like a digital still camera in RAW mode the use of RGB monitors for viewing makes perfect sense.

In fact many of our users complain that going back to regular video assist, either standard def or HD, is a let-down after using our RGB based DVI system. Of course the DVI is only 8 bit, so one must rely on our VISUALIZE tool in order to see the full captured dynamic range of the image. Much like standard video assist one can see the imagery in real time but you would not want to light to it. That is what our exposure tool is created to accomplish.



Alan Lasky
DALSA Digital Cinema
  • 0

#9 Robert Sanders

Robert Sanders
  • Basic Members
  • PipPip
  • 79 posts
  • Director
  • Studio City, CA

Posted 22 September 2006 - 03:38 PM

The RED camera has Dual HD-SDI output. So you should be able to view 444 1080P on set while recording 4K internally.
  • 0

#10 Alan Lasky

Alan Lasky
  • Basic Members
  • PipPip
  • 55 posts
  • Other

Posted 22 September 2006 - 04:25 PM

The RED camera has Dual HD-SDI output. So you should be able to view 444 1080P on set while recording 4K internally.



Uhhhh, ok. Great.

If you are recording 4K as RAW Bayer pattern imagery and "Dual HD-SDI" output MAY be one of your chosen output formats why should a system carry all the overhead of formatting an HD signal for viewing?

We use a pure 8 bit RGB signal for viewing because it gives us the most 'neutral' viewing environment for our 4K Bayer images. We record in RAW to get away from all the gamma, knee, and other circuitry/menus needed to bit pack a signal for HD viewing and recording. We should put it back in for recording 4K?

I would use the VIPER in FilmStream mode with the Luther box or a Cinetal monitor with a custom LUT for Dual-HD-SDI.

I have no experience with RED's dual-link HD-SDI output, and when I saw it at IBC it did not have any outputs at all. Not to say that it will not support that functionality but please tell me "the RED camera has 'X' features" when it ships.

Alan Lasky
DALSA Digital Cinema
  • 0

#11 Graeme Nattress

Graeme Nattress
  • Basic Members
  • PipPipPip
  • 145 posts
  • Other

Posted 22 September 2006 - 06:23 PM

Yes, we do have HD-SDI for monitoring or recording purposes, and yes, a full image processing pipeline to drive that. We also have a DVI port for cheap, plug it into an LCD monitoring. Why? You've got to see what you're shooting. We need the image processing pipeline for the EVF, so again, you can accurately see exactly what you're shooting. Meanwhile you can be recording RAW, or, if you don't want that, RGB. I don't see why you should buy an expensive LUT box when it's cheaper / easier to do it in camera, and the metadata from those settings can travel along with the RAW data.

Graeme
  • 0

#12 Mike Brennan

Mike Brennan
  • Basic Members
  • PipPipPipPip
  • 581 posts
  • Cinematographer
  • London

Posted 23 September 2006 - 02:42 AM

Phil,

None of the professionals who have used our camera from the ASC, BSC, or DGA were the least bit 'suspicious' (let alone "highly suspicious') of our system's monitoring functionality. Once you understand that we are shooting like a digital still camera in RAW mode the use of RGB monitors for viewing makes perfect sense. ......
edit
.....In fact many of our users complain that going back to regular video assist, either standard def or HD, is a let-down after using our RGB based DVI system. Alan Lasky
DALSA Digital Cinema


I know these guys are future users of the Dalsa, but the ASC BSC DGA (in general) decided not to use a digital camera in their digital cinema standards testing!

They are of the ilk that have relied on meters not screens to judge exposure and contrast and have a bee in their bonnet about anything with the word video attached, call it snobbery.

They will say the Genesis is a Digital cinema camera not a HD camera for instance.

The subject of monitoring brings a lot of video baggage that is irksome and worrying to those DPs with little video experience.
Digital Cinema, LUTs RGB is more Hollywood than "HD"

But worth bearing in mind when referring to quotes from DPs as there are two camps, those that have used light meters and those who never will.

Give me as many monitoring options as possible.


Mike Brennan
  • 0

#13 Deanan DaSilva

Deanan DaSilva
  • Basic Members
  • PipPip
  • 27 posts
  • Digital Image Technician

Posted 23 September 2006 - 11:26 AM

Give me as many monitoring options as possible.
Mike Brennan


I don't know if this is what Alan is trying to say but one very important thing about
shooting raw is that you need evaluate the raw signal independantly of the monitoring
signal. It's very easy to be misled by a signal that has been converted into a completely
constrictive color space. Doing so, you add additional layers of interpretation to the raw
data that could lead you to easily make the wrong decisions.

Even things like waveform monitors are risky because they were not designed to be
use with linear (to light) or log images and you have add a layer or interpretation to
put into a video space so the tool works as expected.

Of course, monitoring for talent and everyone else is wholly different issue.

Deanan
  • 0

#14 Graeme Nattress

Graeme Nattress
  • Basic Members
  • PipPipPip
  • 145 posts
  • Other

Posted 23 September 2006 - 11:35 AM

Although, at some point, you do have to convert RAW into something more viewable, as you point out, monitoring RAW directly is going to be very misleading. But again, as you point out, converting RAW to REC709 is also misleading as there's so much more data in the RAW than 709 can allow for. At some point, the skill of the cinematographer, is to "see through" these constraints to "see" what the finished result will be. We can all supply tools to do this, but in the end, that's the skill.

Graeme
  • 0

#15 Deanan DaSilva

Deanan DaSilva
  • Basic Members
  • PipPip
  • 27 posts
  • Digital Image Technician

Posted 23 September 2006 - 11:57 AM

Although, at some point, you do have to convert RAW into something more viewable, as you point out, monitoring RAW directly is going to be very misleading. But again, as you point out, converting RAW to REC709 is also misleading as there's so much more data in the RAW than 709 can allow for. At some point, the skill of the cinematographer, is to "see through" these constraints to "see" what the finished result will be. We can all supply tools to do this, but in the end, that's the skill.

Graeme


From my experience, I would prefer not to look at mapped images to judge exposure. If you do map to rec709 then you are going to be deliberately compressing your highlights and you're doing the cinematographer a disservice by making it harder to judge his exposure. It is the equivalent of trying to evaluate your negative by looking at the print. Now, if you provide mappings that let you see a very flat image or map the shadows and highlights separately, the dp can then make a more informed decision.

Ultimately though, once you get used to shooting RAW on a particular camera, you really only need a light meter. Only in more complicated setups or when you're starting out with a camera does having good exposure tools really help alot. Shooting raw is alot more like exposing film than it is shooting video and the tendancy to make it work like video isn't necessarily adventageous (unless you just want things to work the way you know them to, which alot of people do.

Deanan
  • 0

#16 Phil Rhodes

Phil Rhodes
  • Sustaining Members
  • 11943 posts
  • Other

Posted 23 September 2006 - 02:54 PM

Hi,

> We do not use HD-SDI because we are not a 'video' camera

Yes you are.

No matter how impolitic it is.

Phil
  • 0

#17 Alan Lasky

Alan Lasky
  • Basic Members
  • PipPip
  • 55 posts
  • Other

Posted 23 September 2006 - 04:58 PM

Hi,

> We do not use HD-SDI because we are not a 'video' camera

Yes you are.

No matter how impolitic it is.

Phil



No Phil, we are not. I would not call a Nikon Digtal still camera a "video camera," nor would I call a film scanner a "video camera," hence I would not call a frame transfer CCD based system that shoots in RAW mode to a hard disk a "video camera" either.

"Impolitic" is not the "I" word I would use in this situation.

Alan Lasky
  • 0

#18 Graeme Nattress

Graeme Nattress
  • Basic Members
  • PipPipPip
  • 145 posts
  • Other

Posted 23 September 2006 - 05:25 PM

Got to agree with Alan here. The Dalsa is most certainly a "Digital Cinema" camera of very high quality. Video, to me, implies many things that are not "Digital Cinema" - interlace, "sharpness", blown highlights, over-processed images, over-compressed tape format etc. etc.

Digital Cinema is about getting the best quality images you can via a digital / electronic capture. It's a quality aesthetic that seperates it from video, it's a way of thinking that differs from video.

A Digital Cinema camera is much closer to a high end D-SLR than any video camera as Alan points out. Even if the Dalsa had HD-SDI on it, it wouldn't make it video camera, any more than a D-SLR becomes a point and shoot because it can make JPEGs. It's more of a philosophy thing than a technology thing.

Graeme
  • 0

#19 Jim Jannard

Jim Jannard
  • Basic Members
  • PipPipPipPip
  • 361 posts
  • Cinematographer

Posted 23 September 2006 - 05:33 PM

I echo Graeme's comments. Dalsa has truly set the mark for Digital Cinema (certainly the one we are shooting at) :-) and cannot be described as a "video" camera.

Jim

Edited by Jim Jannard, 23 September 2006 - 05:34 PM.

  • 0

#20 Phil Rhodes

Phil Rhodes
  • Sustaining Members
  • 11943 posts
  • Other

Posted 23 September 2006 - 06:01 PM

Hi,

As an engineer, I have an almost pathological disregard for this kind of chicanery.

You see, while you think you're trying to make this non-film technology seem friendly to film-oriented producers and DPs, what you're actually doing is preparing the ground for big problems later on. My position is to call a spade a spade, then later on, if there's problems, there won't be any room for your clients to say "Hang on, this appears to be a spade, whereas you sold me a Super Spade-O-Matic."

Yes, I know it's time-consuming and problematic to make nontechnical people understand that yours is a particularly high-standard, specialised video camera with certain attractive features; but it is, fundamentally and in all operating parameters, a video camera.

Selling people a fluffed-up, glossed-over version of the truth because you think they're too stupid to understand is talking down to them and it is asking for trouble. I have been forced to do this by various people in companies you'd have heard of on several occasions, and it never ends well.

And the reason you use HD-SDI over DVI is that the DVI connector will break the first time you look at it wrong.

Phil
  • 0


Visual Products

Technodolly

rebotnix Technologies

The Slider

FJS International, LLC

CineTape

Tai Audio

Media Blackout - Custom Cables and AKS

Ritter Battery

Wooden Camera

Metropolis Post

Broadcast Solutions Inc

Rig Wheels Passport

Glidecam

Paralinx LLC

Opal

Aerial Filmworks

Abel Cine

Willys Widgets

CineLab

Gamma Ray Digital Inc

Aerial Filmworks

CineLab

CineTape

Visual Products

Tai Audio

Willys Widgets

FJS International, LLC

Broadcast Solutions Inc

Media Blackout - Custom Cables and AKS

The Slider

Technodolly

rebotnix Technologies

Paralinx LLC

Wooden Camera

Metropolis Post

Ritter Battery

Abel Cine

Rig Wheels Passport

Glidecam

Gamma Ray Digital Inc

Opal