Chroma not working anymore

[borgqueenx]

i really want to get that app to blend laser colors working again.
i tried alot but i cant get that stopid app to work
Any solutions on how to get this running? Is there a alternative to the application?
I want to know now how 650mW of 532nm blends in with 1-1.2W of 638nm.
And how much blue i then need to finish the projector.

 

[ElektroFreak]

You don't really need Chroma for that, the ratios are pretty well established. If you're using 635nm red you can go with a ratio of 3:1:2, meaning 3 parts red, 1 part green, 2 parts blue. If you're using 650-660nm red, you want a red green blue ratio of 4:1:2. In my experience you can have more red than suggested and still be fine, but don't use less. Red is always the least visible. Once you've got the system built you can then tune your balance in software.

For your numbers you'd start with your green power, 650mW. To find the 635nm red power needed you multiply that by 3, so 1.95W. To find blue you multiply your green power by 2, so 1.3W. This will give you a very balanced starting point. From there tuning in software will be very easy and shouldn't require any big changes.

 

[borgqueenx]

You don't really need Chroma for that, the ratios are pretty well established. If you're using 635nm red you can go with a ratio of 3:1:2, meaning 3 parts red, 1 part green, 2 parts blue. If you're using 650-660nm red, you want a red green blue ratio of 4:1:2. In my experience you can have more red than suggested and still be fine, but don't use less. Red is always the least visible. Once you've got the system built you can then tune your balance in software.

For your numbers you'd start with your green power, 650mW. To find the 635nm red power needed you multiply that by 3, so 1.95W. To find blue you multiply your green power by 2, so 1.3W. This will give you a very balanced starting point. From there tuning in software will be very easy and shouldn't require any big changes.

ah okay thanks ^^
why do people use it then?

 

[ElektroFreak]

ah okay thanks ^^
why do people use it then?

Good question.

 

[LSRFAQ]

Good question.

EF, whats with the NEG attitude. Its freeware dude, and a good tool!

Because if you have more then just RGB, and your planning to spend thousands on lasers of your own money, it makes sense to check against the science. Vision science is imperfect, and if your adding say yellow OPSL or a old dye laser to the mix, you want to know what is up.

I've had to match a Pantone color in a large projected logo as a subcontractor. The huge bank likes their signature shade of blue. That was not easy to do, because the "Color Spider" measuring widget is for LCD screens, not 16x80 foot displays. Two projectors with different amounts of 445 had to match on different sides of the building. Chroma is useful for that, too.

One projector company wanted a laser that did certifiable D2 "Standard White", and one does not argue with BIG BOX MART when they want identical colors in the projectors in the stores.

Does the fact that is fun to learn and play with it, matter?

Borg, are you running Win 7? There are a lot of issues with light show software and freeware running under 7, especially with 64 bit.

I've hit that with Anarchy. Chroma, ILDA viewers, and electronic design freeware such as SABOR, LTSPICE, and PSST!.Tocket might like to know if your having issues, so he can fix it for others. Please PM him with the problem.

Have you tried XP compatability mode or downloaded the latest Matlab/Labview runtimes?

Steve

 

[rhd]

I've never understood why Chroma:
A) Is so slow to boot, even on a Quad Core system with 8 gigs of ram.
B) Why coding it required reliance on a 250mB Matlab runtime.

That second part really blows me away. I have trouble believing that the math involved isn't doable without Matlab. I often wonder if it just happened to be the case that the coder knew Matlab well, and thus relied on it, as opposed to doing it in something more widely supported without extra runtimes?

Anyone know what the math behind Chroma is?

If there weren't any objections (IE, if this wouldn't be considered disrespectful - I don't know, because I don't know anything about Chroma's history), I'd be happy to try recoding Chroma in something web-based.

 

[LSRFAQ]

If there weren't any objections (IE, if this wouldn't be considered disrespectful - I don't know, because I don't know anything about Chroma's history), I'd be happy to try recoding Chroma in something web-based.[/QUOTE]

The math is attrocious when you factor in the laser power and air scattering. The way he solved it was to store a massive series of eye reponse lookup tables/equations, then use Matlab functions to search it and do the seriously heavy math.

That was probably the easy way out, and its accurate.

He works on dies that glow for his doctoral research, so the Matlab was of course, around.

If you want to take a crack at it, you can PM him at PL. I have not seen him post in a while and he is in Europe.

Steve

 

[Bionic-Badger]

Give me a break. It's not complex math, and the data (simple CIE tables) and equations are freely available and not hard to apply. I've already coded up stuff in javascript to plot and manipulate those CIE diagrams and matrices, and most of the time was spent on actually plotting things, not using the data and equations.

Chroma is a terrible program, not because it doesn't do what it should, but because it's some crummy Matlab script that was compiled into binary code. Whoever wrote it should release the source code so we can just use Matlab directly without some shitload of libraries for probably 2kB worth of Matlab scripts. I couldn't even get the program to run with the runtime libraries I installed, which was a reason I had to code up my own shit.

 

[LSRFAQ]

Where are the corrections for Munsell type measured response and CIE space in that link?

Yeah, Chroma is slow and clunky, but its accurate.

Steve

 

[Bionic-Badger]

Those are simply links to the kind data and type of equations that go into such a program, showing that it is not secret knowledge, nor that complex to implement.

Also, why would you need or want Munsell type measured response when the CIE data supersedes it, and every other color system is a simple set of well-documented transforms on that same data to more correctly represent human eye response. Also all the color spaces are simply matrix transforms of the data, transforming coordinates from one color space (such as sRGB) to another (say one built of laser wavelengths) in the manner depicted in the second link.

Not to take away from Tocket's efforts, but there's nothing to indicate that Chroma is applying anything more sophisticated than the CIE XYZ colormetric data (probably using Matlab's functions directly) and some matrix transforms.