Film-Tech Forum ARCHIVE: Barco DP2K-12C : Color Measurment procedure

This is topic Barco DP2K-12C : Color Measurment procedure in forum Digital Cinema Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=16;t=001024

Posted by Benoit Delval (Member # 6521) on 02-10-2012, 12:58 PM:

Hello,

We are a postproduction house @ Brussels-Belgium.

Can anyone help me on this issue ?

For a Barco DP2K-12c projector :
In the procedure "Color Measurement", which of these targets should we use for a DCI calibration to setup a grading room?

DC28_DCI_Xenon.TCGD
DC28_DCI_XYZE_314_351.TCGD
P7v2 telecine.TCGD
P7v2 theatre.TCGD

And what are the difference between?

Thank you so much !

Ben

Posted by Steve Guttag (Member # 268) on 02-10-2012, 06:17 PM:

When calibrating the DP2K-12C (or any Barco projector), if you use the Communicator Software (which you should...it is free!), it will ensure you have it right.

For the Color Measurement Procedure, it will load uncorrected colors and apply zero correction to it. After you are done, it will give you the option of naming/saving your MCGD (what you measured) to be used with any suitable TCGD (what you want).

To prove to yourself that you did a good job...after saving/loading your MCGD, verify with the Verify Colors TCGD and measure color by color (it will tell you what they should be) and it should be to within DCI (or REC 709 or whatever color space you want) given the tolerance of the specification.

The key is you MUST measure uncorrected colors

If you are doing this through the TI ICP program one can also select the uncorrected colors.

When you are done, for DCP/JPEG programs, you want the DCI_XYZ_314_351.TCGD as your target colors.

That said, Barco uses .PCF files (Projector Configuration File) that takes care of getting the TCGD selected for you. Again, DCI_XYZ_314_351 is the PCF you are looking for on 2D JPEG sources.

For REC709 (HDTV) they have a PCF and for 4:2:2 DCPs they have other PCF files (Depending on your player)

The PCF is a collection of parameters, including gamma, TCGD and such to ensure all of the bits and pieces are there for a particular format.

Posted by Benoit Delval (Member # 6521) on 02-11-2012, 11:19 AM:

Thank you so much for the detailed feedback Steve [Smile]

)

We will use DC28_DCI_XYZE_314_351.TCGD as Target.

Do you know the difference between these 4 TCGD ?

DC28_DCI_Xenon.TCGD
DC28_DCI_XYZE_314_351.TCGD
P7v2 telecine.TCGD
P7v2 theatre.TCGD

Thanks.

Ben Delval
Puzzle Film & Video
Brussels-Belgium

Posted by Steve Guttag (Member # 268) on 02-11-2012, 11:47 AM:

DCI_Xenon is a "perfect" version of the typical xenon color space. It has the same values as the "Nominal" MCGD...thus when those two are paired, the projector will not color correct at all.

The P7 color files, if I recall correctly, relate to the 4:2:2 color spaces and MPEG and date back all of the way to series 0 projectors (Barco DP40, 50, Christie DCP-I and H).

If you want to see what their coordinates are, connect to the projector via the TI ICP program and then load each one and see how the various boxes fill themselves.

If you have the Barco PostProduction version of Communicator, you too can see what is going on and create PCF files that have the parameters you want.

With Christie, you can also go to the TCGD screen and load the various TCGD systems and see what gets loaded.

I don't believe NEC gives one a means to see the TCGD within the DCC program...they too load a PCF to take care of it. But, with the TI ICP program one can see ANY of the DLP projector's TCGD files (and create ones of your own).

-Steve

Posted by Dave Macaulay (Member # 813) on 02-11-2012, 12:34 PM:

This is what I do,:

turn on the lamp and let it heat up for 5 or 10 minutes.
Select a test pattern. This means that in the next step the active MCGD will be <TEST PATTERN> - I don't know if a different one means different results but that's my procedure. I aim the meter at the screen centre, using the projector crosshair pattern. I like to adjust the lamp power for around 16 ft-L with the white pattern, if the light is really dim my color meter takes longer to stabilize.
log in as service tech
go to "installation", "color calibration", "measure native colors"
read color x/y values for the 3 colors plus white and enter the color meter readings in the boxes.
"save to file" to a name that makes sense, ie "onsite_2D" for 2D.
then "apply and exit"
go to the next choice, "select target colors" and pick "color verification". This makes you check a couple of "yes" boxes before you get back to the main menu.
next "verify corrected colors". If you're anal you can check all 6 colors and white, but if white comes in close to 314x/351y then the other colors are going to be pretty good too.

The color measured on a white test pattern is never 314/351 on a Barco in my experience, you only get it through this color calibration page by selecting "color verification" and then verifying the white in the next stage.

Communicator Post needs a key from Barco to run. The latest version gives you 30 days play time before it shuts off if you don't get a key. It doesn't have any features I need enough to bother asking for a key so far.

Posted by Steve Guttag (Member # 268) on 02-11-2012, 03:46 PM:

If you are not creating new PCF files, then you don't need the Post Prod Version. I do and have said key (it is keyed to my laptop only). Having PCF creation capability has the advantage of also being able to change Gamma as well as TCGD (had a customer that wanted Adobe 1998 color space for graphics and another that wanted me to match for Tungsten slide projector...that sort of stuff). This is totally unnecessary for 99% of the DCinema users/installers out there. Again, if you are handy with the TI ICP program, you can do much of this stuff but not as graphically.

-Steve

Posted by Benoit Delval (Member # 6521) on 02-12-2012, 02:58 AM:

Thank you Steve and Dave,
I also already played around with the Post version of the Communicator, and it is actually very interesting for a post house.

About the contrast, have you ever measured it actually with the Barco set to 14ftl? What is your result ?
Personally I use an X-Rite Hubble and I cannot get the 2000:1 announced.
For the calculation, I divide the value measured in ftl for white with the value for black. Is this correct ?

Thank you

Ben Delval
Puzzle Film & Video
Brussels-Belgium

Posted by Marco Giustini (Member # 4544) on 02-12-2012, 06:38 AM:

Benoit,

Did you measure ANSI contrast or ON/OFF contrast? in the former, the room is playing a huge role since reflections from the walls - if not pitch black - will raise your black level.
In the latter you should be able to measure something good, but again the room must be pitch black, watch for any stray light.

2000:1 is ON/OFF contrast. DCI spec for ANSI contrast - or intra frame contrast - is 100:1.

Posted by Benoit Delval (Member # 6521) on 02-12-2012, 12:04 PM:

Hi Marco,
My current measures are in contrast ON / OFF mode.
This is a grading room, all in black absorbing surfaces.
My measures do not exceed a contrast of 1200:1 @ 14 ftl.
Has anyone taken measures for other Barco Series 2?
I'm curious to compare results....
Thank you for any feedback.

Ben Delval
Puzzle Film & Video
Brussels-Belgium

Posted by Marco Giustini (Member # 4544) on 02-12-2012, 12:33 PM:

I did with series 1, but I don't have the results handy.
What white and black source did you use?

Posted by Benoit Delval (Member # 6521) on 02-13-2012, 03:46 AM:

I used a DCP I created with full white 1023 and full black 0 frames,read by our Doremi 2K4 server. The Barco set to 14ftl with a PCF targeting the DC28_DCI_XYZE_314_351.TCGD.

Ben

Posted by Benoit Delval (Member # 6521) on 02-25-2012, 06:12 AM:

Hi,

Some feedback about this Contrast issue.
Due to to high ftl level for the screen size (3m base) and the 1200W lamp, the Barco PCF has been modified with contrast set to 75% in PCF editor.
This decreased the level of white @ 14ftl, but did not decrease the black's level, as would a decrease of the lamp's current.
Which ultimately ruined the contrast.
Cheers,

Ben

Posted by Marco Giustini (Member # 4544) on 02-25-2012, 11:25 AM:

Sure, if you reduce the contrast electronically you reduce the contrast on screen.
I know that Christie has aperture plates to be put in front of the lenses to increase contrast for post production, maybe Barco has something similar?
If not, you could try defocussing the lamp to reduce the light output - keeping an eye on light uniformity.