+ Reply to Thread
Results 1 to 9 of 9

Thread: Asking for some suggestions about a little experiment

  1. #1
    Junior Member anormal is on a distinguished road
    Join Date
    Oct 2012
    Posts
    7
    Thanks
    2
    Thanked 1 Time in 1 Post

    Asking for some suggestions about a little experiment

    ed: please if an admin thinks this must be in general chat, plese move it, thanks!

    hi all,

    i am new in this forum, and reading posts here and there it seems this is full of pros :D
    maybe you can spare some words about a thing i want to test
    i am not a pro of printing or design, but i am fluent in windows design programs, read something about color theory, color profiles, etc... so i have a little base to understand

    well, what i want to test is the similarity of paints (enamel or acrilic) respect to some standards,

    i am interested in military scale models and as you probably know there are some color standards for military equipment (and industrial in general as RAL) and so exists exact definitions about some colors (for example one of the NATO camouflages are defined with Bronze green, RAL 6031, Leather brown, RAL 8027 and Nato black, RAL 9021)

    so?
    well, in the market there are a big list of brands which comercialize paints dedicated to this hobby, Tamiya, Vallejo, MrColor, Humbroll, Revell, etc, etc...

    i know i can paint some little patches of these paints in a piece of white plastic (or cardboard or whatever you think is ideal) then use a colorimeter or spectrometer to read the XYZ (Lab?, or other data) of these color patches, and the using some color distance measure (delta E?) to test which paint is more exact to the standars

    even i could use it to find substitute paints from other brands


    questions:

    i think i need some protocol to scan the colors maybe using some type of standard light?

    what i need a colorimeter or spectrometer? i am confused about this

    or maybe i can use a good and profiled flatbed scanner?

    also is there a recommended brand or model? money is of course important...

    what delta E, or formula do you recommend? i can test a lot of them (have some books about color theory)

    also what i need? XYZ?, L*a*b*, ???

    ...

    well, thanks for your time reading this, also excuse for not being printing related

    if you have some tips for me... i'll listen!
    thanks!

  2. #2
    Donor mcjam45 is on a distinguished road
    Join Date
    Aug 2012
    Posts
    76
    Thanks
    9
    Thanked 16 Times in 14 Posts
    Hi,
    What you need is a spectrophotometer. An inexpensive solution would be a ColorMunki Photo or a used i1pro(1). You don't need a standard light as a spectrophotometer has a built in illuminant and sufficiently shields the test sample from ambient light.

    Using the Colormunki software, you can measure your patches and calculate the color difference (dE) between the measured LAB values and the LAB value of your reference using [URL="http://www.brucelindbloom.com"]Bruce Lindbloom's Color Difference Calculator[/URL] for example.
    Last edited by mcjam45; 10-29-2012 at 01:17 PM.

  3. The Following User Says Thank You to mcjam45 For This Useful Post:

  4. #3
    Junior Member anormal is on a distinguished road
    Join Date
    Oct 2012
    Posts
    7
    Thanks
    2
    Thanked 1 Time in 1 Post
    thanks for your answer,

    i've read there are some differences in Colormunki versions, one is colorimeter and the others are spectrophotometer? so, the Photo version is the one i need! nice! i've seen it for aprox. 350€

    thanks also for the color calculator, i plan to code one myself

  5. #4
    Donor mcjam45 is on a distinguished road
    Join Date
    Aug 2012
    Posts
    76
    Thanks
    9
    Thanked 16 Times in 14 Posts
    Quote Originally Posted by anormal View Post
    i've read there are some differences in Colormunki versions, one is colorimeter and the others are spectrophotometer? so, the Photo version is the one i need! nice! i've seen it for aprox. 350€
    Yes, you need the photo version or the design version. All other versions are colorimeters and will not suit your purpose.

    Keep in mind that in contrast to the i1 pro, which is an industry standard, the colormunki is supported only by a few third party apps/profilers. But especially in combination with the freeware Argyll you still can do a lot with it.

  6. #5
    Donor Panos is on a distinguished road
    Join Date
    Sep 2012
    Posts
    22
    Thanks
    0
    Thanked 0 Times in 0 Posts
    I would go with the i1pro if I could spend the extra amount

  7. #6
    Junior Member anormal is on a distinguished road
    Join Date
    Oct 2012
    Posts
    7
    Thanks
    2
    Thanked 1 Time in 1 Post
    Panos: any feature you think it bill be useful in the i1 pro? better than the colormunki i mean?

    thanks!

  8. #7
    Moderator Aaron125 is on a distinguished road
    Join Date
    Sep 2012
    Posts
    140
    Thanks
    0
    Thanked 1 Time in 1 Post
    Quote Originally Posted by anormal View Post
    ed: please if an admin thinks this must be in general chat, plese move it, thanks!

    hi all,

    i am new in this forum, and reading posts here and there it seems this is full of pros :D
    maybe you can spare some words about a thing i want to test
    i am not a pro of printing or design, but i am fluent in windows design programs, read something about color theory, color profiles, etc... so i have a little base to understand
    Welcome mate,I hope you enjoy your time at these forums, always nice to have such polite new members join.
    Quote Originally Posted by anormal View Post
    well, what i want to test is the similarity of paints (enamel or acrilic) respect to some standards,

    i am interested in military scale models and as you probably know there are some color standards for military equipment (and industrial in general as RAL) and so exists exact definitions about some colors (for example one of the NATO camouflages are defined with Bronze green, RAL 6031, Leather brown, RAL 8027 and Nato black, RAL 9021)

    so?
    well, in the market there are a big list of brands which comercialize paints dedicated to this hobby, Tamiya, Vallejo, MrColor, Humbroll, Revell, etc, etc...

    i know i can paint some little patches of these paints in a piece of white plastic (or cardboard or whatever you think is ideal) then use a colorimeter or spectrometer to read the XYZ (Lab?, or other data) of these color patches, and the using some color distance measure (delta E?) to test which paint is more exact to the standars

    even i could use it to find substitute paints from other brands
    For both finding out which paint is closest to the standards you mention, you would first need to get some more, detailed info regarding the actual colour of your paints. By that, I mean you need to figure out some way to either convert from RAL numbers into either CIELab or CIEXYZ. Otherwise, how would you know or be able to work out how far off the paints are when no spectrophotometer or common colour management app would have any way to convert between RAL and anything else (at least I've never seen RAL mentioned in any colour management app I've ever come across).

    It's only once you have the Lab or XYZ measurements for both your original standards and your measurements of the actual paint you are thinking of using that you will be able to figure out any delta E distances.
    Quote Originally Posted by anormal View Post
    questions:

    i think i need some protocol to scan the colors maybe using some type of standard light?
    You are absolutely correct. All colour management is based around 2degree observer and D50 standard illuminant. There are ways to convert to 2/D50, say if your illuminant was perhaps D65 but that's getting into chromatic adaption and stuff you don't need to bother with. Any spectrophotometer you use will be measuring your samples as 2/D50 CIELAB or 2/D50 CIEXYZ. The spectro itself will likely use 45degree/0 for making its measurements but that has nothing at all to do with the 2/D50 mentioned above. The 45/0 or 0/45 is referring to the actual lights and measurement hardware inside the spectro. Don't get confused with that (at least try not too ;-)
    Quote Originally Posted by anormal View Post
    what i need a colorimeter or spectrometer? i am confused about this

    or maybe i can use a good and profiled flatbed scanner?
    Absolutely do NOT use anything other than a spectrophotometer. Colorimeters are mainly used for screen (emissive) measurements. And for the vast majority of colorimeters, there is no way to take a reflective measurement (measure the colour of an actual object, not something which is self-luminous, such as a computer screen/TV/etc.).

    Forget about even thinking of using a scanner - it can't be done for the requirements you have.
    Quote Originally Posted by anormal View Post
    also is there a recommended brand or model? money is of course important...

    what delta E, or formula do you recommend? i can test a lot of them (have some books about color theory)

    also what i need? XYZ?, L*a*b*, ???
    By far the most popular spectrophotometer on the market for at least the past 5-10 years is the X-Rite i1 Pro. This is the only spectro I would recommend for you to purchase and they can be found quite inexpensive on ebay, I've seen many for around US$200-$250. Just make sure you DO NOT get a UVcut version - this is quite important for your needs.
    Quote Originally Posted by anormal View Post
    ...

    well, thanks for your time reading this, also excuse for not being printing related

    if you have some tips for me... i'll listen!
    thanks!
    Quote Originally Posted by anormal View Post
    Panos: any feature you think it bill be useful in the i1 pro? better than the colormunki i mean?

    thanks!
    The ColorMunki is not really going to do what you need it to do. It uses an LED for its internal light source and are a UVcut device. (not 100% sure abou this but if I'm wrong, I'm sure someone will say so). Also, it is very much a pure beginners device, definitely not suited to your requirements as it interfaces with only a very few apps (there isn't all that much support for it really) and isn't quite as accurate as the i1Pro.

    The i1Pro on the other hand, interfaces with just about ANY colour management app out there and has SO much support from both manufacturers and users so if you run into any difficulties you will have a much easier time trying to sort out your problem.

    If you end up purchasing an i1Pro, I can give you a list of applications which I think will suit your needs or at least get you started.

    Happy hunting!

  9. #8
    Donor mcjam45 is on a distinguished road
    Join Date
    Aug 2012
    Posts
    76
    Thanks
    9
    Thanked 16 Times in 14 Posts
    Quote Originally Posted by Aaron125 View Post
    The ColorMunki is not really going to do what you need it to do. It uses an LED for its internal light source and are a UVcut device. (not 100% sure abou this but if I'm wrong, I'm sure someone will say so).
    The Colormunki is in fact UVcut (> not a filter, but a limitation of it's LED illuminant) and the measurable wavelength is a little bit limited on the short side (starting at 430nm), but apart from that it's a great device, perfectly suitable for what the thread opener's experiment in my opinion.
    It's not less precise than the i1pro (except for the wavelength limitation I mentioned)…I use both on a regular basis.
    The i1pro of course is supported by far more application and this is the main reason why i also recommend to get a used i1pro, but the Munki will do the job.

    btw. The x-rite Isis also uses LED as it's light source.
    Last edited by mcjam45; 11-01-2012 at 03:06 PM.

  10. #9
    Junior Member anormal is on a distinguished road
    Join Date
    Oct 2012
    Posts
    7
    Thanks
    2
    Thanked 1 Time in 1 Post
    Aaron125: i just want to say: WOW! you nearly solved all my questions in a pro way

    i guess i can get colour data from standards, (RAL,Federal Colors,etc) because, well, they are standards, for FC there are already published oficial data (example, here: http://www.colorserver.net/) so i think i'll have a base data for then comparing

    second, thanks for the tips about light source, this is a little complex for me but i understand, i'll have to read more about D50,D65, etc...

    also, thanks for all the info about colorimeter and spectrophotometers, the tip about UVcut, etc...

    mcjam: thanks for confirming the thing about uvcut in the colormunki

    you guys were so helpful!

    i've just now need to get a i1pro, get some standard color data, and start scanning paints

    i'll keep writing about my experimentations!

+ Reply to Thread

Similar Threads

  1. Any suggestions / help ..
    By sunnytimes in forum General Chat
    Replies: 0
    Last Post: 09-28-2012, 02:55 PM
  2. classic error suggestions "film stuck in conveyor" Dolev800
    By rodrigoinge in forum Computer To Plate
    Replies: 3
    Last Post: 09-09-2012, 12:59 PM
  3. Need suggestions to accomplish accurate deviclink
    By epoch in forum Color Proofing Management
    Replies: 5
    Last Post: 01-07-2012, 09:25 PM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts