- This topic is empty.
ACR colour calibration
-
..anyone had a play with this? Matching your camera to a gretag card for better colour accuracy?
I’ve played with it on and off for a while now and don’t get consistent results. (It will vary from lens to lens and under extreme light colour differences, it also seems to vay on focus distance!)
http://www.rags-int-inc.com/PhotoTechStuff/ColorCalibration/ for more info.
Pleasing colour and accurate colour aren’t necessarily the same, but I’d like to start from a position of accurate :)
Si
Simon, after seeing some of our prints and web shots at Nigel’s studio, I reckon whatever method you’ve been using so far seems to be working out just fine!:D
I’ve seen the results of that script on a DP Review ACR calibration script thread and if I was to base a judgement solely on that test, I’d give it the thumbs down. :( The ACR defaults are very generic and in no way camera specific, so hitting the right tones and contrast seems very hit-and-miss, but at least they don’t wash the colours out as much as this example.
I did another search for alternative ACR calibration scripts today and came up with one byThomas Fors which I believe has been recommended by Martin Evening and Bruce Fraser (rip) and one by Scott Tindemans which has yielded some good results with CRW files. I’ll play with these and see how they work out.
I use the Thomas Fors one myself, I gave the other link as it gives more explanation of what it’s all about. You have to edit the script a bit if you’re on the latest version of ACR.
What I’ve found is….
The light source does seem to make a difference.
The Nikons seem to persistantly make the reds a little orange (we’re talking a very small about, you wouldn’t notice unless you were shooting a beer bottle with a green and orange label for instance :( )…. Yellow skin tones
The Canons seem to have an oversaturated red but probably closer to the right hue. Pink skin tones
The Fuji S3 (only one I looked at), seems pretty good, perhaps a bit vibrant on the oranges. good skin tones.These are all very subtle and unless you have a wellcalibrated screen and printer, you’d probably never see it, apart from the Canon’s saturation on the reds.
Si
Cheers for the info Simon.
What I’ve found is….
The light source does seem to make a difference.
Yeah, from what I’ve gathered so far:-
the camera should be set to take a close-on perfect exposure of the colour card as under or over exposure may make significant differences;
the colour card should be placed in the most significant part of the scene when setting up a test shot;
at least 2 seperate calibrations are required for daylight and tungsten (mixed lighting could well be a problem);
seperate calibrations are also required if lenses with different colour reproduction characteristics are used.
There may well be other little foibles, but I haven’t got round to testing these scripts yet.Simon, would you recommend shooting the colour card at the start of every shoot or is it sufficient to have a small set of saved calibrations (for various lighting conditions) you can refer back to as necessary?
FYI: I’m using a Gretag Macbeth Eye-One Photo to generate profiles for the NEC 1980SXi (a very SXi monitor IMHO :lol: ), both of which I’m very happy with. I had used a Pantone Colorvision Spyder II previously, but never got a profile from that I was happy with.
Peter
You should be able to use the EYE 1 to make a specific camera profile! Somethying I haven’t got around to, I did my screen and printers, work brilliantly.
Best to shoot the Gretag card under even and controlled lighting from identical lights, a classic 45 degree one each side. There is a page on the Thomas Fors site that gives an example. I’ve tried it under window light, clouds, direct sun, camera flash and studio flash, get different calibrations for al of them, some are more consistent than others, what I do is apply a “Half-way” version of these. SO it’s nearer most of the time. :)
Si
Simon wrote:
You should be able to use the EYE 1 to make a specific camera profile!
Yeah, true enough, I should try it out alright. I had thought that moving the lights around after shooting the colour card may make the calibration less valid, but if the colour temperature of the light sources (irrespective of light direction) doesn’t change, then the calibration should be right on the money, (I guess).
It could well be the solution I was looking for; here’s fingers and toes crossed. :) Thanks for the top tips!
Peter
You must be logged in to reply to this topic.
