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 :)

Two Dragons with Same R3D Won't Match in Post...?

Jason Honeycutt

Well-known member
Joined
Sep 19, 2012
Messages
283
Reaction score
1
Points
0
Location
Los Angeles, CA
Hi everyone. I shot a feature with two Epic Dragons and a Scarlet. I'm trying to mainly match the two Dragons. Our director was rapid fire where we didn't have any DIT time load in the same presents for the upcoming scenes so I figured, fine, I'll just load in the R3D file I create in RedCine then import that into the Source Settings via Avid. However... when I tried that on camera B, in Avid, one comes out all super blue and green. To test variables, I also tried it in RedCine - same thing.

My confusion is this... is it not supposed to be pure raw with zero metadata applied until we plug-in a look or LUT? Maybe I've been understanding RED's 'raw' for a while and maybe there is a little 'something' baked in.

Also, I did shoot charts but is there anyway to balance via charts in Red Cine? Taking a sample of, say, blue to blue, green to green, red to red, etc. I'm doing it manually and it's 'meh', not spot on so I was hoping there was more of an eyedropper/color match method.

But the main question is about why the same R3D file applied to two of the same cameras shooting side by side comes out with two entirely different looks. (Also, no filters, NDs, etc. were being used.)

Thanks!
 
Different OLPF's between the dragons, i.e. one had lowlight and the other highlight?

Craig Lees
ED & SMX UK
FD SSC & Lomo Anamorphic Sets
 
My advice would be to not try to use RedCine-X Pro for anything except dailies and viewing. If you need to do actual color correction (particularly matching between cameras), you need a calibrated monitor and decent color-correction software like Baselight, Lustre, Mistika, Nucoda, Pablo, or Resolve. Different cameras absolutely do look different, even with identical setups. Having color charts can help, particularly with scopes and a good monitor.
 
Different OLPF's between the dragons, i.e. one had lowlight and the other highlight?

Craig Lees
ED & SMX UK
FD SSC & Lomo Anamorphic Sets

Yeah, both Skin Tone Highlight ones :/ The weird thing is that it's not even close. Like, the Scarlet MX by default is closer... the one Dragon comes out super blue/green with the same look.
 
Yeah but maybe the OLPF setting in-camera was different - settings - maintenance - calibration - OLPF? I've done that' myself too many times! It will make one image mega green that you can just about correct with crazy OTT WB & tint values. I.e. the cams are set differently to one another. The result of which, one will go crazy green.

Craig Lees
ED & SMX UK
FD SSC & Lomo Anamorphic Sets
 
Have you reset the white balance and tint levels? If the lighting conditions are the same, it's the same OLPF with the correct settings in camera, same color science, all settings neutral and the white balance / tint levels set to the same levels they should match. If they don't then there's something else.
 
I agree with Christoffer - before going for any CC or LUTs etc - reset all the metadata and then see what you get.
 
need to see some frames to really help you out on this.

Option 2.

Create a look in RCX and apply that look to all the footage prior to importing it into the AVID.
What makes the most sense is that the two dragons had mismatched OLPF's and that one of them was not set to the proper OLPF.

David
 
did you use any ND Filters?
I also had optics before that had a different color. (and they where Leica Summilux-C)
 
there is a way to force redcine x to use a different olpf setting but i cannot remember how, someone from red might chime in
 
I think the problem is how metadata is used. Metadata is never baked in but can come along the file up to the Avid import or DNx file if transcoded.
But perhaps there is another problem (lens, olpf, different angle...) wich outcomes should be smaller but if you send us a screen grab with both images (ref/green) we can answer your question with more accuracy.

Pat
 
Every single variable affecting the light's path must match to get the same exact color.
The lighting, lens filter, lens, lens APERTURE!, OLPF, shutter, sensor.

The less light that hits Dragon, the more green the color will be.
For example, shoot at F/1.4 ISO 200 and F/2.8 ISO 800... F/2.8 will look more green because less light is hitting the sensor.
 
The less light that hits Dragon, the more green the color will be.
For example, shoot at F/1.4 ISO 200 and F/2.8 ISO 800... F/2.8 will look more green because less light is hitting the sensor.

So a gray chart will go from neutral gray to green just by underexposing?
 
So a gray chart will go from neutral gray to green just by underexposing?
Not Incredible Hulk green, but a more green tint than it would otherwise have with more light. It's more noticeable with larger differences in exposure. I've done many ISO tests that show this.
 
I will assume you had the same look metadata, same lenses set, same settings, same OLPF.
That leaves the black shading.
If a Calibration map is done with a cold temperature (55°), and the other one with a high temperature (70°), it can lead to some kind of shift in how the sensors capture the image. Shift will be higher with older cameras who ran too hot for all their lives. The sensor themselves are not equal depending on their total usage.
 
need to see some frames to really help you out on this.

Option 2.

Create a look in RCX and apply that look to all the footage prior to importing it into the AVID.
What makes the most sense is that the two dragons had mismatched OLPF's and that one of them was not set to the proper OLPF.

David

That's exactly what I did :) That was giving me the blue/green look, which is what I found so odd. Like, of the OLPF is slightly off in the camera, it shouldn't be 'that' off, it would seem.
 
Okay everyone... really great tips and great discussion. I tried different looks via Red Cine, eyeballed them per camera to get them close. (I wish I could 'eyedrop' a chart in RCX, would be sooo easy.) I'm really trying not to go in and out of DaVinci, since I use the same .cube files in Avid and I create tracking power windows, layers, etc. Plus, I don't have the 4K version of it :)

Also, for the variables... both cameras were using the same lens set, same shutter, no ND (it was a night shoot the entire time), etc. Everything was the same.

If it helps, this is what I did in Avid...

- First, put your main camera's color chart in a sequence. Next clip, put the color chart of the camera to be matched.
- Under color correction, I went to the charts and did the "Match Color" which kinda flies under the radar a lot in Avid's Color Correction.
- (To the right of the sliders) I used the "H + L" dropdown... the two windows above that tell you the exact numerical color values.
- Hover the mouse to over the "Select a Reference Color" window tab.
- I eyedropped, say, the "red" the Camera A reference chart which you can see on in the "Previous" window. The numeric values will pop up in the right little window.
- Using the little window to the left, hover the mouse and it turns to an eyedropper to select to color to change. Drop that on the same cart color.
- Press "Match Color." The numbers should be almost identical.
- Do that for the other colors... and it all matches :)

I totally forgot that Avid can do this but it's a lifesaver.
 
If all the settings were the same and there still is a difference between the cameras you might A) have had wrong OLPF settings in one of the cameras or B) they weren't black calibrated correctly.
Have you checked the cameras again? I would do a checkup of the cameras and their settings to see if there's something wrong. Also do a test shoot with the same settings on some charts to see if they match or differ.
 
If all the settings were the same and there still is a difference between the cameras you might A) have had wrong OLPF settings in one of the cameras or B) they weren't black calibrated correctly.
Have you checked the cameras again? I would do a checkup of the cameras and their settings to see if there's something wrong. Also do a test shoot with the same settings on some charts to see if they match or differ.

Yeah, it was a 'difficult' shoot... normally I'd do all of that, lens map, etc. I got a chance to calibrate the blacks at least. The only thing that might have been off was the OLPF, didn't get a chance to get into the settings. Being a public forum... I'll say it like this, to put it into perspective... was a deferred payment feature (if that puts it into perspective) but a good credit. We weren't supposed to shoot that first night, that was supposed to be the 'show up and build' night and begin production the following night -- but, surprise, we're shooting. I drove out to the location and literally put my luggage down and started rapidly building the three cameras while the director had the actors and everyone on stand-by. Not to mention that I haven't even lit yet and 10 minutes into the build, I get the AD poking his head in, "So how're we looking? 15?" I'm like, "I JUST got here... these are rental cameras. And then, as a heads up, I still have to light after that." Then he's like, "So I'll tell the director... 20... 25?" I remind him that they literally gave me ONE grip and nobody else to help with lighting. Got those three together, raced out to light... then shot :)

Definitely a good lesson... but thank God I told everyone to chill out, we HAVE TO CHART... because it's saving our asses right now. I'm just glad I remembered how to match in Avid. But yeah, I wish we could do that in RedCineX, like do a screengrab of charts, store it to the side as a reference, then eyedrop the colors.
 
Back
Top