Color libraries for ETC S4 LED on the Element

[JChenault]

I recently had an opportunity to sit down in a dark demo room with the new ETC S4 LED and a conventional fixture. My intent was to put some colors in the conventional unit, dial up the S4 LED, and compare them with my eye on human skin. I wanted to see how good the matching was both from a 'How close is this on a white wall' and 'How good is the Selador multiple colors at mimicking gel"

I was using an element to control the LED unit. The profile was set to a D-60 luster. The fixture was set to 'Theatre mode' which put it in HSI ( hue saturation intensity).


The attempt was a abject failure as the color matching was worse than awful. I was getting visibly different colors ( R-80 was especially humorous. It was a fairly bright blue green). The color matching was so bad that I have to think we set something up wrong.

Anyone out there tried something like this. What results did you get? Any idea what we might have done wrong?

 

[MNicolai]

IIRC, the LED array in the S4 LED differs from the other Selador fixtures thus far in how many of each color of LED were used. I'm sure ETC will get a calibrated profile for the new fixtures into their consoles in a future software version.

 

[derekleffew]

John, your experience is not unique.

... I will say that the color picker via gel manufacturer is a bit off, but again that might be expected... Not sure if it was because we were using the D60 profile or not. ...

It seems to me that R80 should be R80 no matter what the color mixing mode, RGB, CMY, HSI, cat walking in paint.

Did we ever get a clarification on:

...I wonder if one could just copy the profiles for the Desire D40 and D60?
EDIT: Oh wait, that won't work, as I believe they've substituted a white circuit for the orange in the X7.
And why does no one ever mention Metamerism when talking about Selador fixtures?

You can run the Lusters with the D60 profile, same LED colors. ...

?

 

[MNicolai]

R80 is sort of R80. I could be mistaken since it's been awhile since I've had my Congo in front of a Selador rig, but I thought that the color picker only calculated color information but not intensity. So with Selador or Desire fixtures using a calibrated profile, if you had a fixture set to R80, that would not look at all like the R80 you'd be used to because it'd be a dozen times brighter than R80 out of a 1kW fresnel. You'd have to dim the fixture down a lot to get into the ballpark.

You may have to hit up Novella Smith to see exactly what benchmark (fixture/lamp color temp/lamp wattage) they've calibrated their fixture profiles against.

As for using the wrong profile for a fixture, it does matter. If the LED arrays differ between the fixture you're using and the profile you've selected, the calibrated profiles will mess up the final color that gets rendered out of the fixture. Also remember that the Desire series fixtures have brains that render the dimming curves and colors very differently than classic Selador. They've even optimized the use of the LED's in Desire fixtures to make the same LED's glow brighter on a Desire fixture than they would on a classic Selador fixture. Using the correct profile matters.

 

[xander]

As MNicolai said, I would start by using the correct fixture profile before judging the equipment.

-Tim

 

[ScottT]

In the 1.9.11 software version for the Eos line:

Support for the Source Four LED has been added.

Source

 

[sclausenETC]

Hi there -

I wasn't going to jump in on this thread, but the statement by Derek has motivated me. If you'd like a good explanation of why R80 is very rarely "always R80" then please take a look at this article by Mike Wood. There are so many variables involved when trying to match a perceived gel color on an incandescent light with any other color mixing system or lamp source that we as console manufacturers must go for the "best" match we can get - which may mean the mix for the "brightest possible" R80 or perhaps it's the mix for the "best color rendering" version of R80 or perhaps it's simply the "best R80 we can get given this set of LED emitters." This may be matched to a tungsten lamp of some fixed wattage and voltage that may or may not be the lamp you are using. Different types of color mixing fixtures bring an extreme amount of variables to the mix on the other side of the equation. At best, gel pickers will always be a best attempt and will probably always need some tweaking before you are satisfied that R80 is in fact R80.

There are a few of us over here at ETC who could talk for hours about this subject. This isn't meant as an excuse for missed color "matches" - just know that it's an area where we are working every day to make both the fixtures and the consoles better at generating quality color for your use.

I hope this helps -

Thanks -
Sarah

 

[starksk]

It seems to me that R80 should be R80 no matter what the color mixing mode, RGB, CMY, HSI, cat walking in paint.

Which R80? R80 in front of a Tungsten source is going to be different than an HID source. R80 in front of a 3250K HPL is going to be different than R80 in front of a 3050K HPL.

Mike Wood wrote a great article in the current Protocol issue titled: Gel color pickers—an exercise in futility? In it, he writes about the challenges of gel/color pickers and why they may be inaccurate.

John, the issue that you are experiencing is a known issue with how the Hue/Sat values native to the fixture compare to the Hue/Sat values native to the console. We discovered it shortly after 1.9.11 was released and are working to correct this issue. As a workaround, I would recommend using the General Quick Set (Direct Mode) as it has color calibrated data [EDIT2]in the console software[/EDIT2] that will give you a closer match.

Derek, I am not sure what clarification you are looking for with those quoted posts.

[EDIT: Apparently Sarah and I are on similar wavelengths this morning and she posted whilst I was typing]

 

[sclausenETC]

You owe me a Coke... ;-)

 

[derekleffew]

...Derek, I am not sure what clarification you are looking for with those quoted posts. ...

Does the Source Four LED Lustr+ use the same X7 colors as the Desire D60 Lustr®+? I can't find where I the notion that the S4 substituted white for orange, but it's not something I would have invented on my own.

 

[starksk]

Does the Source Four LED Lustr+ use the same X7 colors as the Desire D60 Lustr®+? I can't find where I the notion that the S4 substituted white for orange, but it's not something I would have invented on my own.

Yes, the Source Four LED Lustr+ and the Desire D60 Lustr+ both use the same 7 colors. In fact, all Lustr+ fixtures use the same colors. Those colors are Red, 3200K White, Amber, Green, Cyan, Blue, Indigo.

Red-Orange is available in the Selador Classic line in Vivid, Vivid-R, Vivid Fire, Lustr, and Paletta fixtures. In the Desire line, this color is available in D40/40XT/60 Vivid and Vivid Fire Fixtures.

 

[joshua]

We recently purchased 15 D60 Luster+'s and I can tell you that the color calibration is not good at all. We have an ETC Ion running the most current software. All of the colors both from the gel picker and the color wheel tend to be much yellower than they are in reality. Not only that, the D60 ships with a 25º "lens". It is very spotty and in fact is 1/2 the size on the 26º Source 4 hanging next to it. If ETC wants me to buy additional lens's for the fixture, I'm okay with that but they should be more honest about both the color rendering abilities and the photo-metrics of their products.

 

[rochem]

Nothing against ETC (or any other line of console manufacturers, for that matter), but as they've alluded to, getting the color picker to accurately match a specific gel color is more often than not an exercise in frustration. Not all, but most of the professional programmers I know put very little faith in the color picker, preferring instead to mix their own colors. I've been doing this recently as well. It's simple, really - when you have that show with R80 backlight and a few LED products, turn the backlight to full and manually mix the LED product until your eye says it matches. Save it as a by-type color palette, and when you go to start a new show, just import that palette and you'll always have that color available that you've matched with your own two eyes. It takes a while to build up an arsenal of palettes, but it's a huge convenience - now when a designer asks me to put the MLs in R80, I can put them in my own palette and know that it will be right instead of needing to waste time trying to match to a color from memory.

 

[joshua]

I've got no problem with mixing a custom R80 or any other color. My larger concern is 2 fold. 1. When ETC releases the next update for the Ion range, the profile may change and that could change all of the custom colors that I've set. 2. The photo-metric issue is really bad. I took this photo from my booth. The unit on SR is the D60 with a 25º Lens. The unit on SL is a 26º Source 4. See what I mean?

 

[Brandofhawk]

The color in both fixtures is created alittle different right? In the conventional, the gel is litterally removing the wavelengths from shooting to the stage floor, starting with "all" colors, and cutting out all but the blue. While the LED S4 is adding light to the mix to get it to the blue. Right?

Color picking wouldn't be an issue for me i think. While it would be nice, I'd probably just pick a close color, than dial it in to exactly what I want (Which may not be the gel color I would have been able to go with in the first place).