Welcome to our community

Be a part of something great, join today!

  • 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 :)

Issue with secondaries on Red footage

Ryan McGregor

Active member
Joined
Jan 12, 2009
Messages
42
Reaction score
0
Points
0
I just started grading a feature shot on the MX today with Color and I am having issues qualifying secondaries via the HSL. Even when trying to pick out a very contrasted color such as very red lipstick out of a shot, the color picker has issues isolating it, even when playing with the sliders. I am using the new RedColor2 RedGamma2, but I have tested other color sciences and it doesn't seem to matter. The preview window in the secondaries shows me what looks like a default image, not even the metadata that we are grading in the viewer. Anyone faced this problem before?
 
This is how Color works. I dunno if it works otherwise with the Rocket (wouldn't be totally illogical). Kaku can probably answer that.

Working with 10-bit log DPX/AJA from well developed files is the best current option with Color as far as I can see.

Cheers

Gunleik
 
Your only workaround in Color is the FX Room right now, it takes the graded input from the Primary Room and not the source like the secondaries. You can build about anything Color can do there as well, once you got the hang of it. BTW, you might want to get Graeme's plug-ins for it to have additional possibilities.

Your other option would be moving to Resolve. The node concept makes these things much more flexible than the room concept.
 
There's no doubt it is kind of possible to work with RAW in color, but for long form (you said "feature") I wouldn't recomend it.

Uli is right:
You can build your secondaries in the FX room...

But:
The workarounds take some time, too, and limited access to all metadata gives you a not ideal startingpoint.

It all comes down to budget, I know. But the results from 10-bit log files both practically (as to how it is to work with them and non-buggyness) and resultwise (as to technichal quality of the image) has been better on the tests we've ran on it on 10-bit log, than from RAW.

Color is quite ok with 10-bit log RGB material, though. There are some snatches, but they are liveable...

There's one snag with my advice, though. You or someone else must be good at preparing the 10-bit log files... -:)
 
... There are some snatches, but they are liveable...


You've clearly met the women in my family.

(J/K, I love them all, but what you wrote was very funny to a set of American eyes.)
 
Yeah.

There are preffered methods but if you have your feature broken into reels, then it is more than managable in Color. It's generally one or maybe two layers of video.

Gun, I'm just curious why you would advocate DPX over r3d in Color when it works quite smashingly.

David
 
Lack of metadatacontroll in the redroom, where the lack of curves is the most important.

Last time I tried, it didn't play well with a Rocket and a Kona at the same time. That may be history. (Didn't try that since the initial beta-release...)

General stability, predictability and consistencey.

Secondaries work as expected... -:)

You don't lose the firstlight (metasettings) when you reset a grade.... -:) As they are no more metasettings.

We tested this before our last two series done on Color and we had fantastic results... on log DPX...

Color will still hang in there, while we make a transition to "something else".

Next up for testing is Resolve....

Those kind of things.

There is other weirdities going on in Color as well, but that's another story.
 
Color is good to grade DPX and prores444 exported from RCX.
The debayer capability of RCX with Rocket is far superior compared to Color.
 
Luca,

Have you done some direct comparisons between using r3d in color and the RED TAB vs. dpx files with the same sequence to the same sequence OP size.

I can not see how adding an intermediate transcode will yield better images than directly from the R3D.

I agree that with the Rocket this can be a bit aster than it used to be.

David
 
Luca,

Have you done some direct comparisons between using r3d in color and the RED TAB vs. dpx files with the same sequence to the same sequence OP size.

I can not see how adding an intermediate transcode will yield better images than directly from the R3D.

I agree that with the Rocket this can be a bit aster than it used to be.

David
Yes, I've done some comparison and I'm more pleased with the result from redlogfilm dpx than red raw files in Color (also because Color doesn't have redlogfilm..). Michael Cioni did some tests about the debayer capability of different softwares and Color was not so good about raw.
 
Also in AE DPX playback is better compared to native r3d, rocket included.
The movie "Social network" for instance it's been finished with DPX not Raw. I've found DPX is very flexible workflow, especially if you have a rocket for transcoding.
I understand other people want to stay raw but I'm faster with DPX and the final result is more cinematic IMO. The beauty of Red is there are several ways to process the files and it depends on what you prefer.
 
Thanks for sharing all of that Luca. It's true that it is very nice to have all of these creative choices for the final look of a film.

So much to consider.

David.
 
Agree with Luca on all points.

AFAIK RLF is part of the latest Color update (haven't done it yet, though)

The dpx workflow in color is just so much more consistent and effective than RAW in Color, and if you need LUTs, I have had weird results with RAW, which i suspect is due to Color seeing the RedRoom as a LUT and that LUT + LUT = strange, according to Color.

I would consider it for very short projects, though, if I didn't need the curves in the metaroom. Thing is I very often find them neccessary before debayering for a good result.

Cheers!

G
 
Back
Top