Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

accuracy of colour temp readings

P

Peter Savage

Well-known member
Joined
Feb 29, 2008
Messages
165
Reaction score
0
Points
0
Location
UK
has anyone seen the following issue or is it just my camera?
during some testing we set up a single light source which we measured to be 3300 Kelvin. We tested with 2 other cameras (both Sony) and they both gave a reading of 3300 K. When I tried my Red it gave a reading of 3831 Kelvin. Further tests with 5600 K light source gave the expected readings on the other cameras but 6500 K on the Red. I have noticed that I am consistently 'correcting' my shots from the red by white balance shift in the same direction ie always towards a higher Kelvin figure. Can anyone (especially someone from Red) comment on this please.
thanks

peter
 
I did a shootout between RED and D21 a while ago.
One thing I noticed was that when both cams was set to 3200k the RED was colder on its output and the D21 was warmer.
If I dialed in 4100k on the RED they seemed to match.
 
Anders, thanks for the quick response.... maybe this is a bit of an issue with the Red. Of course its easy to make the adjustments but it would be good to know from Red (if only for peace of mind) if this is consistently the case. Your figures seem to concur with our tests.
p
 
it's disappointing that no one from Red responded to this.... does anyone from Red have anything to add?
 
Peter Savage said:
it's disappointing that no one from Red responded to this.... does anyone from Red have anything to add?
Click to expand...

Hey, give the fellas some time!

This thread has been up for only 12 hrs, and it`s almost impossible for them to keep track on all the threads!
 
ahem... i guess you're right with that to some extent ... being in a different time zone means for me it's two days since i posted this... i apologise.
peter
 
I shoot with gelled (CT Blue) tungsten which gives me a color temp of around 4700. Even manually balancing with a white card gives me a cool blue look almost the way the gelled light looks to the human eye. I always have to drop down the color temp before transcoding. I read time and again here that the camera needs gain for the blue channel but I'm thinking of taking those gels off the lights to get a warmer look to begin with.
 
ture...
 
Every color meter has small calibration differences, even if you use exactly same brand color meter, you can get slightly different results, I am talking about reflected-light and incident-light both color meter including industry's standard Minolta Color meter 2 or 3 and new Kenko's color meter also.
Best bet, you shouldn't depend on camera's reflected-light color meter; use incident-light color meter together, you'll get best result.
 
well... still no response from RED. If someone from RED can provide some sort of answer to this I'd appreciate it. The real question is whether RED One cameras are consistently giving a different colour temperature reading and if so why. As I've already mentioned it can always be 'corrected' in post. This came about because of tests for a presentation about white balancing cameras, having set up our control lighting arrangement it became very confusing (especially for the audience) bringing the RED into the equation and finding such a difference in colour temperature. So if someone from RED can shed some light on this (ha!) I for one would me most grateful. How much of a problem might it be if there were a few different makes of camera on a shoot each balanced to such a different degree Kelvin figure?
p
 
Zakaree Sandberg said:
i dont even pay attention to the color
(when im shooting)
color doesnt matter to me until i hit post
Click to expand...

But it still can affect how you light, no? I know the sensor is right around 5000º Kelvin, but there are lighting decisions you might make to influence things one way or another.
 
steve, this is exactly why I would like someone from RED to respond to this. I guess they have not seen it yet...
 
I'll bump and add some more info:

We are currently shooting a drama with two REDs. One of the cameras are in the early 2000s, the other mid 4000s (serial numbers).

I have reset and black shaded both cameras twice. Each time the camera in the 4000s are between 250 and 500 degrees Kelvin warmer than the other. And it gives different color temp readings when doing Auto white balance.
Anyone experienced anything like this?

EDIT: I guess I should mention, the raw files are actually different, with the newer camera being warmer than the older one. So it seems like the sensors have slightly different native color temps???

I would love to get some info from RED on this too.

EDIT #2: David Mullen experienced the same thing on Manure:

Then I noticed again something I noticed the day before, that by the end of the day, it seems that the image on C-camera (our main “B” camera for the shoot since Day One) is slightly yellower than A or now-returned B camera. We thought it was just a monitor/output/cabling issue but at the end of Friday, Conrad looked at the RAW files and we saw that the yellowish difference is in the original recordings. It’s a simple color-correction fix but the question is what is causing it. We’ve been doing a black shading every morning but we may try doing it twice a day to see if that helps. It’s like the old days when you used a lens that was warmer than another lens, annoying in color-correction later when you’ve got a lot of intercuts in a scene, but workable. Maybe there are slight differences between digital cameras of the same model that only become visible when doing RAW recording where you don’t have a lot of electronic adjustments happening to the recording to match signals between multiple cameras.
Click to expand...
 
I'm curious, I'm just starting out in motion having worked digital capture for the past 10 years on the still side.

Since Red takes it's origin from digital still cameras, and those in digital still balance with the use of a digital grey card, not white, would this make any improvements in the color outcome?

I would never start a still shoot without setting the color (or at least getting a grey reference card in the first frame).

I've owned two Phase One P30 digital backs, and although they were made doing the same production run, they did differ. However, once they were gray balanced prior to the start of shooting, they looked identical. It's a trait of all digital cameras I think, but you can dial them in.
 
Guys,
Hoping that two Red cameras will automatically match is a pipe dream. They're like snowflakes. Same with HD and SD video cameras. And motion picture film from batch to batch. Red does not have a magical elixir that makes their cameras match each other better than Sony, Panasonic, or Kodak. And as others have pointed out, even CT meters are all different.
In the world of video, doing a multi-camera setup, you need an engineer to match the cameras. When making a feature film with different batches of film stocks, you'll need the timer's help to smooth out the subtle differences in those batches. With Red it's even easier, and can be done on the desktop (although done best by a colorist with Scratch)
My big beef with Red and CT is that the Rec 709 preview often exaggerates the subtle variations in a person's skin tone, or makes a shadow or a hot white cyc go unnecessarily blue, all of which leads to confusion until we shoot a little bit and do a quick curve in RedAlert! The skin tone issue is particularly annoying...
I hope that the preview out on the next cameras are of slightly higher quality.
Cheers,
Harry
 
Harry,

thank you for your reply.

I must admit that I thought RED was different and more fine tuned than other video cams. For some reason I was under the impression that the native color temp of the sensor had to do with the design, and that it was a pretty consistent number.

I guess it's good to have confirmed that this is normal variation IMHO.

Regards,
Fredrik
 
Peter, I did the same test a few days ago with a tungsten source. The RED said it was 3800, my Minolta Colour Meter II said it was 2800.
 
You must log in or register to reply here.
Back
Top