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 :)

New RED Color Science - Test Pilot Feedback Thread

Yes please, some one...

The test is very simple, set up Red with Build 18, point it as a room with only a single 3200k light on that's underexposed, then upgrade to the alpha and shoot, post both pics.

This takes under an hour, including setup, bathroom and whisky breaks
 
blue chanel x blue chanel with 80b

blue chanel x blue chanel with 80b

even better

A001_C005_0605G5_0000.jpg no filter

A001_C007_0605MV_0000.jpg filter 80B
 
I did a pilot test and here we go with my report>>>


What I use in the test:


Lens: Leica Summilux 50mm f/1.4 @ f/1.4 (wide open),

Tungsen light: (a big light bulb E27 75W) coming from right side and it is positioned at approximately the same distance (about 1m) between object and camera.

Color light meter: Gossen Colormaster 3F.

Color Charts: Color checker by GretaMacBeth, Gamma-3cP Chart by Gamma & Density Co, Grey Card by FotoWand (Neutral Grey DN=0.75 log D, 17.68 % Reflection).


Red Team test pilot questions (Q) and my answers (A):


- Q>>RED Team question: Improved noise in blue channel under Tungsten light?

- A>> Sanjin's answer: It's true less noise almost not visible, the file is usable even at 800 ISO.

- Q: Improved skin tone color response on video monitor outputs?

-A: Yes, you can see a skin color reference part at Gamma & Density Cinematographer's Chart and also Gretamacbeth Color Chart.

Light temperature data:

- Q/A: camera's AUTO white balance (?) = 2605K

- Q/A: white balance tool in RED Alert (?) = 2498K

- Q/A: light (color) meter (?) = 2450K

Displays:

- Q: Visual appearance of image in RED-EVF and RED-LCD, including comments on noise under tungsten light?

- A: Improved, almost no visible noise at 320 ISO.

- Q: Visual appearance on external monitor - such as HD-SDI broadcast or cinema monitor?

- A: Improved, no visible noise at Apple Cinema Display 20" driven by HD-SDI preview out to BM HD-SDI converter to DVI then to monitor.

Applications:

- Q: Visual appearance in RED Alert! ( skin tones and colorimetry in general, and noise under tungsten light )?

- A: All improved, now color is pretty nice, also skin tones at the charts are much better then before.


See more about other application reports later... below... scroll down...



Download test shot 4K TIFF file 48MB>>>



REDAlert_test_pilot1.jpg

Test shot in RED Alert v3.72 Beta

REDAlert_test_pilot2.JPG

Test shot scaled from 4K to 1280x640

Tungsten_bulb1.JPG

A big tungsten light bulb 75W involved in the test by Convexa Seletti, Made in Italy.

TungstenBulb_Gossen2.JPG

Gossen Colormaster 3F (13 cm hight) that measured color temperature compared with a big tungsten light bulb 75W (24 cm hight) by Convexa Seletti, Made in Italy.
 
Now the test shot processed at 3 different LUTs in Shake:

Shake_RedLog.jpg

Base: Camera RGB and RED Log file imported in Shake.



Shake_Truelight.jpg

Truelight LUT in Shake, see more about set up parameters at the picture.



Shake_PanaLog2Lin.jpg

Panalog to lin in Shake, see more about set up parameters at the picture.



Shake_Sapphire_FilmEffects.jpg

Sapphire Film Effects with a grain amp set to 0, see more about set up parameters at the picture.


--------------------------------------------
 
Can anyone comment if the new colorimetry helps with the old problem of red lips shifting towards purple/magenta on a rec.709 monitor?
 
Searching for more solutions before sent a file to final CC grade: Now with Gamma & Density 3cP on set look app>>>

Imported tiff file as Camera RGB +RED Log from RA output ISO 320, 2605K,...

3cP_test_pilot_report1.JPG

Gamma & Density 3cP Test Pilot file report 1

3cP_Test_Pilot_report2.JPG

Gamma & Density 3cP Test Pilot file report 2: result at right and source at left side also down look at luminance + RGB 3cP graph analysis... blue channel is under control and reduced...


All those reports show a huge improvement in blue channel, less noise and more pleasant skin tones.
 
Wow... amazing....
 
Skin tones

Skin tones

SKIN TONES>>>Again one more shot at tungsten light bulb 75W:

Test_pilot_skintoneISO_320.JPG

Shot handheld on R1 at mirror with Canon 50mm FD L f/1.2 @ f1.2 (wide open) ISO 320, 2618K RA WB,
all in RED Space and because wide open f/1.2 is super fast for 75W tungsten light bulb then clipping occurred at the arm's skin tone.
Maybe if I would step down to f/2.8 that could get the best but I didn't because it was a test partly also for wide open lens performance.
Also don't forget that tungsten light bulb has a distance about 50-60 cm to the arm, actually it was too close.
Dont't forget that the speed of f/1.2 could refer to the speed of T1.3 that is offered by the fastest and sharpest S35mm PL lenses like a Carl Zeiss/Arri Master Primes set...
One more thing is the fact that fast Canon FDs or Leica Noctilux M @ f/0.95 (that is even fastest FF35 lens available) are all pretty good sharp at wide open...


Test_pilot_Rec709_1.jpg

Shot handheld on R1 at a bath mirror with Canon 50mm FD L f/1.2 @ f1.2 (wide open) ISO 320, 2618K RA WB, Rec 709 in RA, tungsten light 75W, almost no clipping at the arm's skin tone.
Also don't forget that tungsten light bulb had distance about 50-60 cm to the arm, actually it was maybe too close.

Test_pilot_skintoneISO_160.JPG

Shot handheld on R1 at mirror with Canon 50mm FD L f/1.2 @ f1.2 (wide open) and ISO 160 was "corrected" in RA, 2618K RA WB, then it got almost no clipping.


Now I would say that "RA Colour Science" is one of the biggest improvements (build upgrade, new RA, etc,...) in a last period,

no any noise at tungsten light shooting, skin tones are as they are (accurate), nice or not nice, as in a real world/nature.

Amazing!!!

It was enough tests from my side at the moment and I'm waiting here for reports of other RED Users/Test pilots...
 
alex andreermel said:
even better

A001_C005_0605G5_0000.jpg no filter

A001_C007_0605MV_0000.jpg filter 80B
Click to expand...

Thanks- but I was referring to before and afters with the current build 18. So could you shoot the same thing with build 18 with no filter, and then with 80B. Then shoot the same subject with the Color Science with no filter and with the 80B filter. This comparison shows the benefits of the 80B filter but does not show the benefits of the new build. (Other then just taking your word that it is better.)
 
Underexposed tests?

Underexposed tests?

So anyone have some underexposed tests that they can share?

Underexposed images in tungsten light will really show how much better this Color Science is over the current build. As Shawn has already posted- can someone shoot an underexposed image in tungsten light using the current build 18, and then the same shot underexposed with Color Science? (And provide the settings)

Thanks again to all of you who are a part of this testing program! :)
 
Hi Ryan,

Just finished...

Yes, we've done tests indoors at 3200K at normal exposition and both underexpose and overexpose (not with the "old" Build though...). I also did the same with outdoors, cloudy but clear day.
The results are GREAT! I really love it!

I'm getting all the results together, and first send them to Graeme and Stuart, but i'll try to post something (pictures and some comments) for tomorrow morning.

Cheers!

Antoine
 
Antoine,

sounds great,

looking forward to see your test results tomorrow (Sunday) morning!!!
 
Love to see some real world before/after examples beyond the land of Macbeth :-)

let me tell you build 16 made this camera a lot easier to sell into agencies and clients but the last remaining stalwarts for me are food clients who still hold onto their 'No digital' policies are yet to be re-convinced because of this exact issue..

You can live and die by the last job SOMEONE ELSE did for your client, I can show them all the Red NAB reels and Knowing trailers i can get my hands on but many clients are either converted OR turned away by whoever shot their last job on RED and its going to take another leap like build 16 to get some of them back onside and undo someone elses bad karma

Crossing everything that this next leap will be, as Jim says, 'as important as build 16'
in helping me bridge the gap with these people as while I still do love film, Im sick of spending a Red One's worth of cash on every 35mm job, just to buy the stock and process it! ... go Jim, Jarred, Ted and Red

Jas
 
We had Amy Vincent, ASC in on Friday doing some tests for an upcoming RED feature under all tungsten lighting, and they gave this new alpha build (B18 3.7.2) a go, but didn't see enough of a difference between it and B17 3.4.1 in a quick REDAlert! session or live on the monitor (Panasonic 1760). Hoping I can take some more time early this next week to put this build through its paces with what we're learning from the rest of you on here.
 
Antoine Fabi said:
Hi Ryan,

Just finished...

Yes, we've done tests indoors at 3200K at normal exposition and both underexpose and overexpose (not with the "old" Build though...). I also did the same with outdoors, cloudy but clear day.
The results are GREAT! I really love it!

I'm getting all the results together, and first send them to Graeme and Stuart, but i'll try to post something (pictures and some comments) for tomorrow morning.

Cheers!

Antoine
Click to expand...

Brilliant, thanks! I'm really looking forward to seeing this! :)
 
You must log in or register to reply here.
Back
Top