Bulletproof scanning?

[markbarendt]

Okay,

I've been considering scanning slides and negs but everywhere I look it seems that it's a real struggle to get it right. It looks to me, from the outside, that almost every image will need it's own work.

I'm not into that.

Even my local film scanning guy cringes when I (or one of my slide shooting buddies) say match the print to the slide, yes I know the reflective/transmitted issue changes the look, but overall brightness and color balance should not be an issue in my head.

So, before I head down this road at home I want to know if scanning can be made bulletproof:

First my assumptions.

1- A good quality scanner should be able to provide predictable, consistent results. After all it is a perfectly controlled lighting situation in the box.

Right?

2- The film base and color pallet of well cared for, in date, high quality films like Fuji Provia 400x or pro 160s , should be exceptionally consistent and should not require any ongoing scanner adjustment to get good results.

Right?

3- The scanner shouldn't need to "fix" anything if the scanner has a proper icc profile for the film in question.

Right?

4- If the film wasn't exposed well that's my problem not the scanner's. All I want from the scanner is an accurate import.

So, are my assumptions fair?

Is bullet proof scanning possible?

What does it take to get there?

Thanks

Mark

 

[MikeSeb]

No, scanning cannot be made bulletproof, as neither can any other technical aspect of photography. You can, though, set things up so that variables are controlled and deviations minimized; but you'll always be fine-tuning if you want the highest quality results. As to your assumptions:

1. Yes. You'll need to profile it, however, for best results. You can profile for slide film or for reflective materials; most scanners come with their own profiles that install automatically to cover color reversal, B&W, or color negative film. SilverFast comes with the largest variety of profiles via its NegaFix function. You can also make your own profiles with either VueScan or SilverFast, using ICC targets for slide films and reflective materials. These are not available---AFAIK---for color negative films since the orange mask screws things up.

2. Not in my experience. Different emulsion batches and variations in processing runs means you'll always be tweaking, even though you may start out in the ballpark if you've exposed, developed, and profiled accurately.

3. What do you mean by "fix"? You'll still need to set black and white points, and adjust contrast. You'll also need to tweak color balance. If you have an entire series of images on the same roll or from the same emulsion batch, shot in the same lighting conditions with similar exposure, you will find that any needed corrections for one image should be similar for all.

4. Not sure what you mean here. You are absolutely right to strive for accurate exposure and development, since everything after that will be much easier with a good original.

Once you've nailed your exposure and development methods, and obtained accurate profiles for your scanner, and learned to do the basic image adjustments you need in the scanner software (my practice is to set black and white points only, then do everything else in Lightroom or Photoshop), you will have done about all you can. You will still be in for some grunt work, however.

 

[markbarendt]

Thanks Mike,

Happy Thanks giving all!

 

[lenny]

Okay,

I've been considering scanning slides and negs but everywhere I look it seems that it's a real struggle to get it right. It looks to me, from the outside, that almost every image will need it's own work.
So, are my assumptions fair?

Is bullet proof scanning possible?

Mark

I'm with Mike on this one. I'm sorry, but your first statement was more correct. Every image needs its own work. Scanning is a real art, just like printing.

No such thing as bulletproofing. A lot depends on what you want your prints to look like... is there such a thing as bulletproof printing? I think not.

Lenny
EigerStudios

 

[wogster]

Okay,

I've been considering scanning slides and negs but everywhere I look it seems that it's a real struggle to get it right. It looks to me, from the outside, that almost every image will need it's own work.

I'm not into that.

Even my local film scanning guy cringes when I (or one of my slide shooting buddies) say match the print to the slide, yes I know the reflective/transmitted issue changes the look, but overall brightness and color balance should not be an issue in my head.

So, before I head down this road at home I want to know if scanning can be made bulletproof:

First my assumptions.

1- A good quality scanner should be able to provide predictable, consistent results. After all it is a perfectly controlled lighting situation in the box.

Right?

2- The film base and color pallet of well cared for, in date, high quality films like Fuji Provia 400x or pro 160s , should be exceptionally consistent and should not require any ongoing scanner adjustment to get good results.

Right?

3- The scanner shouldn't need to "fix" anything if the scanner has a proper icc profile for the film in question.

Right?

4- If the film wasn't exposed well that's my problem not the scanner's. All I want from the scanner is an accurate import.

So, are my assumptions fair?

Is bullet proof scanning possible?

What does it take to get there?

Thanks

Mark

The problem is that every negative or slide is different, even if shot on the same film at the same time of day, there are subtle differences. Much of the problem is that when you view a slide on a light table and on a computer screen, there are already differences. Computer screens are heavily influenced by ambient light, even though they are effectively lit from behind, this is different between an LCD and CRT based screen.

Another problem is that a specific shade on your scanner is different from the same shade on your monitor and that is different again from your printer. You need a colour profiling device that will use a colour checking chart that you can scan, so that the scanner and monitor are talking the same colour, this should also be applied to your printer, once all three are talking the same colour, it gets easier. This profiling needs to be repeated every few months, as colour can drift with the age of the light source. Scanner bulbs, monitor back lights or electron guns, and print heads all age, and can drift in different directions by different amounts.

I find it easier to work with different media as stand alone images, if I scan a slide and it looks good on the monitor, it doesn't matter if it matches the slide 100%?

 

[markbarendt]

Mike, Lenny, and Wogster,

You guys are no help.

Have you (or anybody else lurking out there) tried using targets like those offered here. http://www.targets.coloraid.de/

and

Software like offered here. http://www.littlecms.com/

to create your scanner/film icc profiles?

 

[keffs]

Whilst it is true that each image requires it's own processing, not all of this processing has to take place at scan time. I think a lot of people just concentrate on captuing everything they can during the scan, and do the artistic bit later in photoshop. There is no technical advantage in doing all tweaks at scan time.

Steve

 

[markbarendt]

Whilst it is true that each image requires it's own processing, not all of this processing has to take place at scan time. I think a lot of people just concentrate on captuing everything they can during the scan, and do the artistic bit later in photoshop. There is no technical advantage in doing all tweaks at scan time.

Steve

What I'm hoping for Steve is to be able to calibrate the scanner so that when I stick Provia in the scanner a very close copy is created.

I've been using a color managed workflow for digital cameras for several years. I can go all the way from the camera to the paper printed at either of two different pro-labs without any significant shift in brightness or color.

This works because everything in the system is calibrated, camera, monitor, the pro-labs workflow, the paper; everything.

What I'm trying to figure out now is if the scanner can be reliably calibrated.

 

[Donsta]

I actually use the Coloraid targets and they work fine. They are transparency only - not for negative films. While I have a few different targets, you really should be a lot closer using just one transparency profile, even if it is not an exact match to transparency film you are scanning. You need to turn all the auto stuff off when you scan the target and use the same parameters for scanning once you've made an icc profile from the target. I open it PS, attach the scanner profile (suddenly your scan then resembles the transparency in the scanner) and then convert to my working color space. Once you have a made an icc profile for your film and scanner, you probably won't need to do it for some time. I last made profiles for my Minolta Scan Multi Pro about 4 years ago and the profiles still give me excellent results. It really does make a huge difference to your workflow if you have have gone through the calibration process for film scanners. There's a nice little walk though what you're doing here: http://www.photographical.net/scanner_profiling1_2.html

 

[markbarendt]

I actually use the Coloraid targets and they work fine.
... I open it PS, attach the scanner profile (suddenly your scan then resembles the transparency in the scanner) and then convert to my working color space.

Thank you Donsta, that's exactly what I needed to know.

I kinda figured it worked fine but all I ever heard were complaints about scanners and about all the work it took to fix.

It's nice to know I can get reasonable scans without doing 3 backflips.