Help with Kodachrome!

[bvy]

I'm new to scanner calibaration, and I have to admit, reading about it gives me a headache.

I have Kodachrome (64) slides that I'd love to scan. But the colors are all over the place. I don't need clinical color accuracy, but I want something closer to real than what I'm getting.

B&H sells these:
Kodak #Q-60K3 Target- 35mm Kodachrome Slide 1575141 B&H Photo

I have an Epson V500 scanner. Using just this reflective IT8 target and the software I have (Epson Scan), can I calibrate my scanner to get mostly accurate color from these slides?

Thank you!

 

[jd callow]

If its critical this is what I would do. YMMV

You need a film scanner that has a real not theoretical ability to scan to a dmax of about 3.4. The emulsion is different and it may be helpful to have a scanner that has a built in capability for Kodachrome (Nikon 4000/5000/8000/9000), but I think this may be hype or at least I don't know it to be true. Finally you need to forget about film profiles and scan at highest bit depth and the widest gamut and do your colour, density and contrast corrections in photoshop on a calibrated monitor. Your scanner needs to be calibrated assuming it can be.

I do not know if a v500 has the chops to do the film justice, but i wouldn't trust epson scan or any scanner software to make the right decisions regarding accuracy.

 

[pschwart]

The version of Epson Scan that was supplied with my V700 does not support creating a profile from an IT8 target. Profiling the scanner won't save you from having to make corrections in Photoshop anyway If you are still interested in profiling, check out VueScan at hamrick.com.

I have an Epson V500 scanner. Using just this reflective IT8 target and the software I have (Epson Scan), can I calibrate my scanner to get mostly accurate color from these slides?

 

[jd callow]

Things have changed, but there used to be a difference between profiling film and calibrating the scanner. The latter was a way to ensure the hardware was seeing colors as it should and generally involved scanning a series of patches and targets to ensure colour fidelity and focus accuracy. It now seems that all or many scanners leave the factory fully calibrated (for life). At least many scanners still test their focus. The former (profiling film), in my experience, was/is the fruitless pursuit of the holy grail. If all colour negs of a kind printed with the same filter pack on paper from a single batch I'd think film profiles would have value. In my experience one frame from the next on any given roll is as likely to need a different filtration than not let alone one roll to the next. And I have no understanding how a given B/W film, which is not part of this conversation, can possibly be profiled.

Sorry for the rant.

 

[samcomet]

This is what I do: If you can, scan RAW to a .dng file for both the Q Target and the Kodachromes. Do a comparison by eye (assuming your display is calibrated) between the Q Target as observed in a slide viewer (I use a light source to the view the slide with the same colour temp as the scanner) and the Q Target on the monitor. Note what "adjustments" were necessary to make both 'look" the same. Save as a PS CS5 preset or note the adjustments and apply them to the Kodachromes. I know that this is a bit lo-tech but it works for me. Cheers good luck!
sam

 

[jd callow]

A slight variation and one that might be followed in a lab is to make the (raw) scan of the slide then put slide on a light box and adjust the scan until it matched.

 

[Tim Gray]

Ditto on most of the above. Profiling my scanner (Nikon Coolscan V) with one of the Lasersoft targets did help get me more accurate color. It still needs a bit of adjustments to match what I see in the slide, but it's much easier to get there after profiling. This is probably because I'm playing a bit fast and loose with my scanning/profiling. I tried Vuescan's profiling solution (very easy to use) but ended up liking the profile I built with ArgyllCMS more.

I scanned (single exposure) the target slide with Vuescan and saved a raw linear scan. I built a profile with the resulting file. For the profile to be valid, you are supposed to use the exact same exposure settings for all of your scans. Instead of doing that (me playing fast and loose as mentioned above), I let the scanner figure out exposure. I also use the multi-exposure option, where Vuescan does a high and a low exposure to get more out of the shadows. Save as a linear scan. I then bring it into Photoshop and assign the profile that I built with ArgyllCMS. Convert to AdobeRGB and tweak. The 'tweaking' in this case usually involves a curves adjustment layer to bring the black point of the blue curve up a little bit and moving the midpoint of the RGB curve up to lighten everything up a bit. This is saved as an action so I can run it automatically and optionally tweak my default adjustments.

It might all sound a bit complicated, but once the profile was built, I can do rough color corrections on Kodachrome slides pretty quickly - less than a minute per slide? Of course, finer adjustments and dust removal takes longer... For the finer adjustments, I pull the slide out and compare to what I have on the light table.

 

[bvy]

Thanks for the responses. Lots of different ideas and good feedback. None of this is “critical,” and I understand I’m not going to get the most professional results from my V500 scanner. I further understand that some further post-processing in Photoshop will be required.

That said, it sounds like, at a minimum, I will need VueScan and the target slide that I linked to from my original post. Hopefully this will be enough to get me started.