Monitor Output Color Correction?

[Casey Green]

(Sorry for the duplicate post - but with all the activity lately, the first one seems to have been lost in the shuffle) :)

I thought I read about this here before - can't seem to find it though... but in theory, could you take a still or short "camera test" clip (RAW) while on set, off-load it into REDCINE make your adjustments, and then take that profile (LUT?) and load it back onto the Camera (by USB, etc.) for use only on your Monitor output?

This would allow the DP and Director to have an idea what a specific adjustment would look like without needing any filters and also without needing to make any counter adjustments on the Assist Monitor at Video Village. (and yet still Record the RAW unfiltered clip). I suppose this depends on the limitations of what can be processed in real-time on the monitor output signal (LCD, SDI, etc).

Is this possible? (or only possible for the actual Recorded Output)?

Many thanks to anyone who can answer - plenty of soup for you! :)

[David Battistella]

(Sorry for the duplicate post - but with all the activity lately, the first one seems to have been lost in the shuffle) :)

I thought I read about this here before - can't seem to find it though... but in theory, could you take a still or short "camera test" clip (RAW) while on set, off-load it into REDCINE make your adjustments, and then take that profile (LUT?) and load it back onto the Camera (by USB, etc.) for use only on your Monitor output?

This would allow the DP and Director to have an idea what a specific adjustment would look like without needing any filters and also without needing to make any counter adjustments on the Assist Monitor at Video Village. (and yet still Record the RAW unfiltered clip). I suppose this depends on the limitations of what can be processed in real-time on the monitor output signal (LCD, SDI, etc).

Is this possible? (or only possible for the actual Recorded Output)?



Here is the way I understand it. A DP can come up with several "looks" for a film. These looks are stored in the metadata of the RED Raw 4k file. When the file is played back, it looks to this metadata to display the file as you have asked it to be displayed, but it has not been recorded this way, it has only been recorded RAW.

This means that the original 4K raw file is never changed. You are only giving the file instructions to play back with certain color values. So any look you come up with is never baked into the file. Just the metadata. So if they want to get back to the RAW image, tweak or start from stractch they could later, but you can provide them with your LUT's as a reference point in post.

I think that this alone is going to change everything. Getting a great histogram for your file will be important as the more picture information ou can record, then the more you can manipulate that image in post.

It's a still photography concept applied to moving images.

David

[Mark L. Pederson]

Casey -

YES.

You can take footage into REDCINE on set, create a LUT - load that into the camera - and it will be used on the monitor outputs - EVF, LCD, HDMI, HD-SDI - in the REDCODE RAW file, the LUT will be added to the meta-data - so, when you bring the footage back into REDCINE - you will have the OPTION to use it as a starting point for transcoding, color grading, or you can disregard it and start from scratch.

You can make several LUTs, one for each set or location, for example, or complex lighting set-ups, etc. - and load them in - not sure how many the camera can hold, or if you just stack 'em on the SD card and then select one - we shall see.

I think folks will quickly realize that shooting with RED, because it's RAW and the sensor has an insane range, that "color" on the set is not anywhere near as important as ISO, exposure selection (and obviously focus).

[Stuart English]

Yes. Mark summarized it really well.

The only clarification is that the file you upload isn't a LUT - which is just a table where the combined result of all the "one light" color correction parameters are stored - instead its the discreet values used in the one light color correction itself; e.g. you could load one up, adjust any one of those parameters and then re-save it under a different name / number.

[Casey Green]

Casey -

YES.

You can take footage into REDCINE on set, create a LUT - load that into the camera - and it will be used on the monitor outputs - EVF, LCD, HDMI, HD-SDI - in the REDCODE RAW file, the LUT will be added to the meta-data - so, when you bring the footage back into REDCINE - you will have the OPTION to use it as a starting point for transcoding, color grading, or you can disregard it and start from scratch.

You can make several LUTs, one for each set or location, for example, or complex lighting set-ups, etc. - and load them in - not sure how many the camera can hold, or if you just stack 'em on the SD card and then select one - we shall see.

I think folks will quickly realize that shooting with RED, because it's RAW and the sensor has an insane range, that "color" on the set is not anywhere near as important as ISO, exposure selection (and obviously focus).

! This is incredible! Just what I was hoping - I don't know how to describe how revolutionary that alone is! From a Directors viewpoint, I think having that creative freedom from Pre through Post (including right there on set) is insanely exciting!

[Casey Green]

Yes. Mark summarized it really well.

The only clarification is that the file you upload isn't a LUT - which is just a table where the combined result of all the "one light" color correction parameters are stored - instead its the discreet values used in the one light color correction itself; e.g. you could load one up, adjust any one of those parameters and then re-save it under a different name / number.

Thanks Stuart (and Mark and David).

Amazing... (especially when realizing you can still record RAW).

I would be curious as to what the exact parameters will be.

Stuart, might you be able to elaborate here?

[Kip Hewitt]

So the LUT's are created in Photoshop/Lightroom/Aperture? What file type is used to transfer over to the RED?

[Alex Boothby]

I think Red should design a 'video Ipod-esque' device designed for testing and tweaking monitor LUTs (or REDCINE parameters). Easier and more portable than a laptop. Could be a hot seller!

[David Battistella]

So the LUT's are created in Photoshop/Lightroom/Aperture? What file type is used to transfer over to the RED?

I am pretty sure that the REDCINE software is where you will be doing all of this. You may or may not be able to import LUT's from third party programs, but I can not see why you would not be able to do that.

David

[M Most]

Thanks Stuart (and Mark and David).

Amazing... (especially when realizing you can still record RAW).



Useful, yes. Amazing - well, this has been a feature of Silicon Imaging's camera since it was introduced. And it was an announced feature of the Kinetta, which never quite made it to a final product. It is, in fact, the way all cameras that allow one to record the RAW output should behave. Of course, there are currently only three that do this (Silicon Imaging, Dalsa, and the Andromeda modification to the DVX100), with Red set to be the fourth.

In practice, one would not likely create LUTs on the day for every scene that one shoots. It is more likely that individual LUTs would be created for day exteriors, night exteriors, day interiors, and night interiors during preproduction. That would allow for a predictable conversion for set monitoring. Production is really not the place to be doing color correction, as your time is much better spent attending to production requirements, and the viewing conditions are not conducive to proper monitoring anyway. The whole idea of RAW capture is that such steps can (and should) be saved for post production, where more time, proper talent, and proper monitoring conditions can be applied to make the images sing. That's not to say it has to be done that way, it just seems to make more sense.

[Casey Green]

Useful, yes. Amazing - well, this has been a feature of Silicon Imaging's camera since it was introduced. And it was an announced feature of the Kinetta, which never quite made it to a final product. It is, in fact, the way all cameras that allow one to record the RAW output should behave. Of course, there are currently only three that do this (Silicon Imaging, Dalsa, and the Andromeda modification to the DVX100), with Red set to be the fourth.

In practice, one would not likely create LUTs on the day for every scene that one shoots. It is more likely that individual LUTs would be created for day exteriors, night exteriors, day interiors, and night interiors during preproduction. That would allow for a predictable conversion for set monitoring. Production is really not the place to be doing color correction, as your time is much better spent attending to production requirements, and the viewing conditions are not conducive to proper monitoring anyway. The whole idea of RAW capture is that such seps can (and should) be saved for post production, where more time, proper talent, and proper monitoring conditions can be applied to make the images sing. That's not to say it has to be done that way, it just seems to make more sense.

I see your point... (I'm not quite as familiar with Dalsa or SI), but I also see how this should be considered more than just "useful" on the day. I think it's great to be able to get an idea what your scene will look like (if it is drastically different than what the RAW data will appear to be on the monitor) , in a pre-viz way.

As mentioned in some other recent threads, these features should not only be thought of as being used in just one type of production scenario. For features with long pre-production schedules, yes, this might be unnecessary or just an interesting option, however, for other productions, such as lower budget Indys, Music Videos, Event Videography, Documentaries, etc... the ability to have uploadable color/value parameters instantly available will be considered extremely valuable.

The feature here that I believe separates the RED from other cameras is that you can monitor the settings while still recording the RAW data AND have the parameters embedded in the clip as metadata, all while monitoring the applied LUT's image. The idea here is that you may never actually use the LUT/Monitored version, but it helps in the creative process to see an approximate version of what the scene will look like (and then also have that as a reference point when you start grading in post).

Not sure if this is true for the other Cameras mentioned, but in 4K combined with the rest of the RED ONE's features, this qualifies as amazing to me. :)

[Jason Rodriguez]

The feature here that I believe separates the RED from other cameras is that you can monitor the settings while still recording the RAW data AND have the parameters embedded in the clip as metadata, all while monitoring the applied LUT's image.

Yes, that's a very useful feature, but there are other cameras out there with that same feature-set as well.

[Casey Green]

Yes, that's a very useful feature, but there are other cameras out there with that same feature-set as well.

perhaps, but not with the rest of RED's feature set / pricepoint. And that is what I am referring to. It's a total package deal.

[Poi Boy]

Yeah , those other cameras are in serious shit.
Aloha
-A

[M Most]

perhaps, but not with the rest of RED's feature set. And that is what I am referring to. It's a total package deal.

True. But Jim would be one of the first to tell you that a number of the features and operational parameters of the Red were "inspired" by other projects, notably Jeff Kreines' Kinetta project. Jeff was really the first to pioneer the concept of RAW capture for motion imaging, the concept of embedded metadata in that RAW stream for LUT's and other information, the concept of on-board RAIDed hard disk recording, and even the concept of real time debayer and playout of the recorded files. What Jim and his team have done is extend a number of these concepts, up the resolution considerably, and use compression as a way of shoehorning all that data into a recordable form using today's technology. They've clearly also done a number of other things that are very original, but it's sometimes interesting to give credit to the pioneers, both current and near-current, when that credit is due.

And I'm certainly not overlooking the fact that Kinetta never actually shipped any camera product.

[Casey Green]

yes, kudos to all that inspired the project, and all of the innovation they accomplished.

I still feel that we will continue to learn more about the inner workings of RED that will continue to separate them from the others that came before.

[Laco Zamba]

This SI-2k demo is very useful

http://www.siliconimaging.com/DigitalCinema/Tutorials/SpeedGradeOnSetIntegration_1.html

[Casey Green]

Thanks for the link, Laco Zamba. That is exactly what I am looking forward to with RED and REDCINE. I see they mention being able to apply one LUT to viewing and another to Recording. This would be ideal.