.LOOK files Scratch and eCinema PRO40-ADV

[Seth Larney]

Hey all,

I am putting together a DI pipeline for use with the RED, which will be based around Scratch, .LOOK files and the eCinema PRO40-ADV (or maybe PRO24-ADV which has the same funcionality but smaller). The eCinema display has the ability to automatically create 3D LUTS internally, as well as load LUTS for certain film stocks etc, once calibrated.

As I don't have my camera yet and build 15 just introduced the .LOOK file functionality, I an thinking about possible workflows that wil enable us to get in the ballpark on-set (using .LOOK files), and then move over the the eCinema display LUT's with scratch for post.. is this a sensible way to go ?

Or.. should we just stick with once or the other ? (.LOOK or the eCinema LUT's).

Am I complicating matters ?

Any suggestions on a suitable workflow with these technologies for feature film DI for posible filmout ?

Cheers

 

[Jeff Brue]

From my understanding the camera's onboard lut is going to be a floating point 1D incapable of film emulation. Ideally we'd just be able to charachterize the sensors log mode and do the full manipulation at the monitor level.

 

[Fredrik Harreschou]

Seth.

I'm too in the process of working out a high end RED workflow for a feature film.
A couple of points to keep the discussion going:
1. Any special reason not to use LUTs in SCRATCH? And calibrate with RSR Cinespace for instance?
2. Right now the look files are only between cameras. I guess you are looking for a way to do looks in preproduction and apply during production and post? This will come eventually but right now I can't see how we can include the LOOK files in the workflow.

Please share if you have more thoughts. I will reply.

Cheers,
Fredrik

 

[Seth Larney]

From my understanding the camera's onboard lut is going to be a floating point 1D incapable of film emulation. Ideally we'd just be able to charachterize the sensors log mode and do the full manipulation at the monitor level.

I think you are exactly right, Jeff. This is what I am thinking as well.

 

[Seth Larney]

Seth.

1. Any special reason not to use LUTs in SCRATCH? And calibrate with RSR Cinespace for instance?

Fredrik

Not that I can see, this is likely what we will do.

 

[Seth Larney]

Seth.

2. Right now the look files are only between cameras. I guess you are looking for a way to do looks in preproduction and apply during production and post? This will come eventually but right now I can't see how we can include the LOOK files in the workflow.

Please share if you have more thoughts. I will reply.

Cheers,
Fredrik

Exactly. I was looking at ways that we can maximize the usage of what we have been given by RED (eg, the .LOOK files), and how best to integrate them into a working workflow for filmout (eg using LUTS at monitor-level in Scratch).

 

[Dane Brehm]

The use of LUTs for filmouts is a no-brainer. Having little experience with DCIM specifications and a recent annoucement of a $1.1bn digital distribution agreement. What is the typical workflow for a pure 2k/4k Theater distribution using the DCI spec?

 

[Fredrik Harreschou]

The DCI spec is 12 bit JPEG2000 (in XYZ) at 2048x1080 (2K) or 4096x2160 (4K). I guess it is more of a specification than a workflow. But I'm not sure if I understand your question...?

The DCI Digital Cinema System Spec v 1.2 was released about a week ago:
http://www.dcimovies.com/DCIDigitalCinemaSystemSpecv1_2.pdf

 

[Dane Brehm]

I was actually curious if it was a RGB colorspace or wider?

 

[M Most]

I was actually curious if it was a RGB colorspace or wider?

It is XYZ color space, the widest gamut available. But the explanation is not that simple. If you are interested, you need to read the spec as a starting point, possibly followed by Glenn Kennel's book on the subject ("Color and Mastering for Digital Cinema").

 

[Dj Joofa]

Glenn Kennel's book on the subject ("Color and Mastering for Digital Cinema").

MMost, I got hold of a copy of the above mentioned book as I take your advice seriously. I must say it is a nice book. The author has left out many technical details and mathematical notions to perhaps reach a wider audience, and I felt that at some places one needs the background to fully appreciate what he is saying. Nevertheless, I learned many things from a cursory browsing.

Just goes to say that learning is a process that never stops and the more one learns the more one realizes how little he/she knows. Just like I have learned so many things after joining this forum a short time ago. Honestly, I was not even aware of many things which informative people here were saying. Every time I re-read David Mullen's or Finner's posts I discover something new.