Please Hell with Scanner Colors

[bvy]

I have a Canoscan 8600f scanner, and it scans film quite nicely. With one problem. I can scan the same film strip several times, and each time the colors are slightly different. I execute calibration and zero out all the bells and whistles in the software, but the colors are still a moving target.

I've been reading about IT8 targets and think that's what I need. And I think I need a reflective target (for film). What about software? Do I need special software, or is this something that I can do with the software that came with the scanner? This is the part I'm not understanding (not for lack of surfing and reading).

Am I even on the right track with the IT8 target?

This is really frustrating. Maybe I'm being fussy about the colors, but that's just me. Can anyone help? Thanks.

[Marco B]

I have a Canoscan 8600f scanner, and it scans film quite nicely. With one problem. I can scan the same film strip several times, and each time the colors are slightly different. I execute calibration and zero out all the bells and whistles in the software, but the colors are still a moving target.

I have the 9950F, but must admit I do not do much color scanning in the classic sense. Most is sepia or selenium toned analog B&W prints scanned in 48 bit color mode.

I do find the colors consistent though if I use the "execute calibration" option in the scan software, even the sometimes very subtle color casts of these partially toned prints.

Are you sure you give all of your machinery enough "warm-up" time? I would recommend switching on all equipment, especially monitor and scanner, at least 1/2 hour before starting to work.

The shifting colors might otherwise be caused by the warming up.

IT-8 calibration might be a good thing, but even in that case, the above warning should be taken in consideration.

Marco

[Walt Calahan]

Yes, an IT8 calibration can help, but even then it isn't perfect. Color can still be a moving target, especially if there are voltage fluctuations in your power line, etc.

I use SiverFast scanning software with their IT8 calibration system. Check to see if you scanner driver can use an IT8 calibration target before buying a 3rd party solution.

Good luck.

[pschwart]

I have a Canoscan 8600f scanner, and it scans film quite nicely. With one problem. I can scan the same film strip several times, and each time the colors are slightly different. I execute calibration and zero out all the bells and whistles in the software, but the colors are still a moving target.

I've been reading about IT8 targets and think that's what I need. And I think I need a reflective target (for film). What about software? Do I need special software, or is this something that I can do with the software that came with the scanner? This is the part I'm not understanding (not for lack of surfing and reading).

Am I even on the right track with the IT8 target?

This is really frustrating. Maybe I'm being fussy about the colors, but that's just me. Can anyone help? Thanks.

Calibrating your scanner will get you in the ballpark, but you will still need to make adjustments in Phototshop. A reflective target is for prints; you will need a transmission target for film.

[Denverdad]

Profiling your scanner with IT8 targets is generally the way to go to get control of color when scanning film. You would need the transparency type target, not the reflective kind. But note that the whole IT8/profiling thing pretty much only works for slide film, not negatives.

Having said all that though, if you can't rely on your scanner to make a consistent scan in the first place then I'm not sure the IT8 profiling is going to help. You really need to get a handle on the changing output colors problem first. I like the suggestion to make sure you have allowed sufficient warm up time before scanning.

[bvy]

These are good suggestions. Thank you. A few comments.

Yes, I typed "reflective" but was thinking "transparency." Thanks for setting me straight.

Warm up time is a good tip. And you mean warm up time above and beyond what the scanner suggests, it sounds like.

So, an IT8 target won't work with negatives? I hadn't considered that.

I'll keep trying. There are obviously a lot of variables when it comes to scanning film.

Thanks everyone for your hell -- er, help!

[pellicle]

Negatives

I thought you were talking negatives, but saw your IT8 and assumed I was wrong.

some reading on my blog should get you going in the right direction;

* using the Epson for bulk scanning of 35mm colour neg (and getting your desired settings applied across all evenly)
* my colour negative workflow (which is similar)
* registration problems on Epson scanners and why you shouldn't go too far in trying to adjust this or that because when you improve focus you may disturb registration
* using your scanner to understand film density.
* driving your scanner software differently to effect some changes in scanner side exposure on Epsons (meaning better noise characteristics in the dense areas of your negs).
* don't forget colour management on your Epson (its not where you might think it is)

[rnwhalley]

If you are scanning slide film/transparencies then yes creating a profile for your scanner will improve results. To do this properly you will need an IT8 target, preferrably for the film type you intend to scan + you also need the data files to match the target. To use these you will need software that can generate the ICC profile.

I calibrate both my scanners from time to time with this method using Vuescan scanner software and also IT8 calibration targets. Have a look in the downloads section of my web site as there is a free tutorial explaining how to do all this (www.lenscraft.co.uk).

If you are scanning negatives then the IT8 targets won't help. Some software such as Vuescan includes profiles for negative scanning that remove the colour mask some films use.

One final tip if you are using a flatbed is to ensure your scanner is warmed up. This can take up to 30 mins for some scanners but most will be fully warmed up after around 5 mins of use. If you are scanning with a cold lamp this might account for some colour shift.