This content has been marked as final. Show 157 replies
I'm certainly interested! It seems a natural for a scripted solution in that you typically have to do everything, then do everything again, to converge on the ideal solution. Feel free to contact me offline if you think I can be helpful.
Thanks Bruce. I'll get in contact with you after the holiday.
Warning: Your digital photos may be fading. | http://rawformat.com
I would love to have a script for calibration. I've had success with several cameras following Bruce's techniques.
I've posted the first public beta version of the ACR-Calibrator script here:
I've only been able to test this with three different RAW files, so I'm sure there are bugs. If it fails or results in a bad calibration, I'd appreciate hearing about it.
Read the documentation on the page for info on how to use it.
unsupported? obsolete? incompatible? | http://rawformat.com
Looks promising. I'll give it a try.
Great work, Thanks.
Cheers, Rags :-)
I just posted version Beta 2 of the script
This fixes a few of the bugs reported since yesterday:
- Problems leading to desaturated results
- Text alignment in the status window
- Incorrect patch reading caused by ruler units other than pixels.
See the documentation on the web site for more details. The documentation has also been revised with additional details to answer some common questions.
In order to conserve bandwidth, this will be the last beta announcement I make in this forum. Keep checking the web site for further updates or, if you would like an email announcement when a new beta version is posted, send me an email. Otherwise, the next announcement I'll make here will be when the script reaches 1.0 Release status. And, of course, to answer any questions I can.
8.2 million pixels held captive. | http://rawformat.com
Just wanted to say thanks for the ACR Calibration Script. I ran it for my Canon 20D and it worked great. I was having difficulty doing it manually, and your script worked very well. I created one calibration for Daylight conditions and a separate one for Tungsten lighting conditions.
As I understand it, (and I may be mistaken), the ACR Temperature and Tint sliders interpolate between natural light and tungsten light settings built for each generic camera model. I was wondering if there is a way to use the custom settings generated by your script to do the same interpolation when using the sliders? Probably not. I was hoping I could use my custom calibrations for daylight and tungsten, then use the sliders to interpolate between the two.
Would you use the daylight calibration settings for all but tungsten lighting conditions? What about mixed lighting conditions, coll white florescents, etc.
Thanks again for sharing your work.
You're welcome. I just posted Beta version 3 of the script this afternoon. It fixes a few bugs and tweaks the optimization a bit.
Bruce Fraser certainly has more experience than I have with the calibration procedure. In fact, I use his manual settings in many cases as a standard to measure the script's performance against. I've seen Bruce mention that the calibrations generally hold under various lighting conditions except for tungsten. Tungsten usually requires a separate calibration.
As far as mixed lighting, that's always a fun one. My approach would be to use the daylight calibration to do the raw conversion and then do any necessary cleanup work on the image using local color corrections in photoshop.
Used your Beta 2 and works great with 1Ds, 1D Mkii and Oly 8080 (I used daylight for the calibration).
Thank you for sharing the script ... certainly has helped me out get the best out of ACR and my RAW files.
beta 3 is looking very good. It' s probably true that working by hand could produce a result faster than the script, but I really doubt that anyone making manual adjustments could beat the script for accuracy and, most of all, consistency. I sure can't.
I think you've done the ACR community a huge service!
I'd like to add my thank you for a great script! I agree with Bruce's comments: it is slower but more accurate than I could do manually. The slowness is quickly forgotten, but the accuracy lives on.
Does this script run on the Macintosh?
Yes, it does run on Mac.
It's great, it works. Beta 3 is noticeably better than Beta 2.
Be sure to follow the directions so that it runs as efficiently as possible.
It can also run in the background which lets you do other things; it makes it run a bit more efficiently since it does not have to literally redraw each opening of the raw file onto the screen as a full image.
As Bruce said - Tom has done ACR users a huge service.
Thanks. The reason I was asking is because, instead of downloading, the link opens as a text file in a browser page (Firefox). :(
If you have a two button mouse then right click the link and select "download linked file as", if you don't have a two button mouse then hit "control" and click the link - you'll get the same menu.
You can save it as a java script to your desktop and then drag it into the appropriate Photoshop folder.
Yes, I had already downloaded the file, I was just explaining why I asked in the first place. Thanks for responding.
I am in the midst of doing a Tungsten Cal, I did strobe last night and what little I checked on it looked good.
I see lightning, and when I do, I typically unplug my Mac. Does the script happen to remember it's position, or do I have to start fresh?
Thanks for your service and fine product,
Any settings you make on the Adjust tab of ACR when you open the file to start the script will be read and used as a starting point. So, if you have to abort the script, it's worth recording any reported results as well as the 'Best Result' for the currently running optimization and using those settings when you open the file and restart the script.
It's not exactly a resume feature, but it's close.
Based on the calibration method outlined in Bruce Fraser's ACR book: After manually calibrating ACR to my test image, I then need to save the settings to an XMP file (Calibration Only) for later recall. When I look at that XMP file it does not include Color Temp, Tint, Exposure, Shadows, Brightness, Contrast or Saturation. Using Tom's script, I get values for each of these.
How important are these values and should they be included in the XMP File (Calibration) or not?
Thanks again to Tom and Bruce for their tremendous effort!
See here http://www.creativepro.com/story/feature/21351-3.html "Saving your Settings".
See you Saturday! http://www.johnmaclean.com/asmp_oct9.pdf (shameless plug!)
Keep your last dollar!
You only want to save the settings from the Calibrate tab. The purpose of doing all those adjustments in the Adjust tab on the target capture is simply to massage it to the same (very limited) tonal range as the Macbeth chart so that it's easier to match the color patches in the capture to those in the reference target.
The resulting calibrate settings should be valid across a wide range of exposures and color balancesthey're simply tweaking the R, G, and B primaries of the built-in profiles for the camera model to better match the behavior of your specific unit.
I may not be one of the "experts" but.... it's only the calibration settings you want to save as a subset (not the settings for the other tabs).
The other settings (Color Temp, Tint, Exposure, Shadows, Brightness, Contrast or Saturation, etc.) are specific to the image and are used to bring the photo of the colorchecker to a neutral point so that the calibration can take place accurately.
If you have to run Tom's script again, you may want to hold onto a copy of all of the settings, plugging them in will get the script to the point of calibration faster and thus help the script run more efficiently. A case in point is if Tom releases another beta or final version of the script.
@John - Thanks for the links, I have re-printed and re-visited the info in Bruces article. Saturday!! Wuhoo, yes!!
@Bruce - Thanks for the added instruction, BTW, love the ACR book, looking forward to the addendum on DNG ;-)
@Jon - Thanks for the clarification on the on the "other" settings. I guess the bottom line is, that the other settings will help me if I need to re-calibrate, otherwise they won't be used when correcting images.
I ran the script about four times on one image and have gotten two results that were almost identical and two that were very different. I guess logic would dictate that I use the settings that are closest to what my eyes say are correct.
If I go to my OS X finder for any reason, and back into PS, I get a dialog with something about the Channel Mixer being unavailable. I just click ok. Am I interfering with the operation?
You caught my "last dollar bet" post before I edited it! ;-)
I am learning about color management and desire to use your script with Adobe RAW. I am going to purchase GretagMacbeth 24-patch ColorChecker, however, there are many different ones to buy. Could you be specific on which one, and a web site link would be great also. Really enjoying adobe raw, and looking forward to learning more through your help. Also, just ordered Bruce's adobe raw book.
Very impressive script Thomas , thank you !
I´m trying to evaluate my ACR calibration .
See before/after conversion for a daylight image
I kept only calibration "subset " values from my script results ,
other figures are set using " normal " approach ( using my eyes and Bruce techniques ).
It looks promising , but I´m still not quite sure what I have.
calibration figures looks quite high, compared what I'm used to ( zeros )
There is a huge difference in "contrast " setting , calibration gave me -20 for contrast , normally I would never go under 70 ( i guess I prefer "snappy " apearance) .
Are these numbers normal ( see values against gray in the right corner ) ?
( macbeth chart shot in the rainy Stockholm
daylight , I would expect 5500K )
Macbeth chart is 2.5 year old , so it could become a bit yellow.
What I can see is that colors seems to be more vibrant , cleaner .
I´m not sure if this makes slightly "over saturated look " ...
Everything looks promising anyway .
I'd like to thank Thomas Fors for his ACR calibration script (Beta 3) and Jeff Schewe, who reminded me to to Control click the link in order to download Thomas' script.
I ran the script the other day and the result is a daylight setting far superior to the camera default for my Nikon Coolpix 5700. Amazing. Truly amazing.
Thanks big time, Thomas! I've had some major problems with ACR and Canon DR(300D) RAW files. Installing ACR v2.3 helped a great deal but the script has added the finishing touches. In fact, I would say the results are even better than using the otherwise clunky Canon software (:
I just replaced ACR 2.3 with ACR 2.4 and verified that the calibration settings I added while using 2.3 carried over to the new release. I checked a few images and they look OK, perhaps just as before.
Still, I wonder if changes made to ACR 2.4 warrant calibration all over again. Is a make-over indicated by changes in 2,4?
According to his site, it just adds compatibility for ACR 2.4.
Tom has posted a new beta version of the ACR-Calibrator script.
Beta 3.2 (January 31, 2005)
- Added support for official released version of ACR 2.4.
Thanks again Tom!!
Thanks for letting us know about the script update. I've downloaded the new version and will use it to replace Beta 3.
Thanks for keeping up with ACR.
Thank you for the wonderful (and exhaustive) work you have done with this script. It truly unlocks the potential of digital capture.
The calibrations I have created with your script are far superior to what I can produce manually. The manual calibrations (following Bruce's guidelines) are close, but the shadows always had an uncorrectable cast. (Getting the red hue and red sat right seemed impossible.)
These script-based calibrations are right on.
Thanks again. Excellent work.
I have to add my thanks. Your script is the best Color Management tool I've found since diving into digital photography. MCG
I need to calibrate ACR but I don't have a GretagMacbeth 24-patch ColorChecker.. Could someone PLEASE post me a raw capture of it in a daylight conditions. I'm gonna use a flashbulbs mostly and it's temp is the same as a daylight.
That wouldn't work. The purpose of calibrating is to adjust ACR to your specific camera's response. An image taken by someone else's camera is meaningless to yours.
You need to take the photograph with your camera. Buy or borrow a ColorChecker chart.
Great, big thank you for such a quick respond! Of course It wouldn't, I don't know what was it wrong with me with posting such a stupid message:)
Will try my best to be smarter in future;)
Regards to all!
I am having trouble getting this to work. I have acr 3.2 and the latest version of the calibrator script downloaded from http://fors.net/scripts/ACR-Calibrator/
But when I try to run the script I get the follwing error:
"Your installed version of Adobe Camera Raw (ACR) is not supported by ACR Calibrator.
If you are running an older version of ACR, please upgrade:
If you are running the latest version of ACR, check for updates to the ACR Calibrator script:
But I have done both these things. Anyone else had this problem and have a way to fix it?