Shooting to Scan

[Alpha]

Many of us who shoot film and do not have easy access to professional quality scanning equipment often butt heads with a significant hurdle: our scanners. Nearly all flatbed scanners are notoriously bad at recording fine shadow, and worse highlight detail, especially in film scans (negative and positive). Even expensive CCD negative scanners with LED back-lighting can leave some to be desired. I have many a chrome from which even a Nikon SuperCoolScan 9000 on it's highest quality settings has difficulty recording fine highlight detail, yet print wonderfully to Ilfochrome. This represents a significant hurdle to producing high quality digitized work, particularly when one considers the expensive and cumbersome alternative: the drum-scan.

So what's a film-shooter to do? You could make a print, or an internegative or interpositive enlargement that underexposes the highlights and then scan that. But this is a tedious and expensive proposition.

I propose a radical alternative for shots that are specifically taken for digital presentation, when time and money are of the essence: shoot to scan with color negative film. Underexpose your highlights in order to compensate for the scanner's inability to record them well, and correct in post.

 

[Helen B]

Why do you think that this a radical idea? It is common practice to use negative film for scanning. You should find that there is no need for any underexposure.

Best,
Helen

 

[Alpha]

The crux of the idea has nothing to do with the film, but intentionally underexposing your shots to compensate for the scanner's poor ability to record highlight detail. It could just as easily be some other film.

 

[Battou]

Why do you think that this a radical idea? It is common practice to use negative film for scanning. You should find that there is no need for any underexposure.

Best,
Helen

This could apply to those who are scanning prints via standard flatbed scanners, they tend to exibit color loss due to the glass. This often times can not be fully corrected in post. A slight underexposure could be useful (pending proper lab work was done on the prints) in that regard for those unable to purchass professional equipment for this.

What makes this a radical concept is the ideaology of intentionally of breaking the norm of proper exposure for the sole purpose of compensating digitizing shortcommings.

 

[Alpha]

This could apply to those who are scanning prints via standard flatbed scanners, they tend to exibit color loss due to the glass. This often times can not be fully corrected in post. A slight underexposure could be useful (pending proper lab work was done on the prints) in that regard for those unable to purchass professional equipment for this.

This is true. As I mentioned, if you want to retain the highest quality neg, you can expose correctly in camera and burn like hell in printing, then scan the print.

 

[Battou]

From what I have seen a lot of the scanning issues are coming from people who are using sendout labs (Wich is not to say home printing does not experiance this but).

Often times I am seeing (and have experianced) prints of poor quality being scanned in this manor. This is due to the lab they sent it to not their own exposure error. These scans often exibit sevearly blown out skyes and grey (some times pale) shadows in the darks. Intentionally underexposing would be a some help to these people who lack their own processing ability more or less to compensate for idiot lab techs first.

An example I posted reasently shows this quite clearly.

Scanning a poor print is going to have poor results, no two ways around it. Even then if you have a quality print one is still faced with this issue of color and highlight loss in scanning.

 

[Rick Waldroup]

Max, what about scanning B&W negatives.

A few years ago, I borrowed a Minolta film scanner, but I really did not know what I was doing and some of the scans came out okay, but a lot of them were crap.

I have been seriously thinking about going back to film for my personal work and all I will be shooting is B&W film. Do the same problems persist in scanning B&W film? Thanks.

 

[usayit]

I would imagine the same would apply for B&W negatives. How about slides?

A lot of this has to do with equipment.. no? Or is this a generalization? I find some digital cameras (I know this was originally about scanners) do a better job capturing details in the shadows. Underexposing by a stop and correct post-pro works pretty well.

I personally have yet to find a local lab that does negative scans to my liking... most are geared to delivering photos to the regular consumers good enough for web posting and/or small prints. I usually ended up scanning at school or at home.

 

[Battou]

I would imagine the same would apply for B&W negatives.

Yes, it does, The B/Ws I posted in Off topic got sevearly blown out even with my dedicated, and required some processing, but not much, merely a color converson from CMYK to RGB.

 

[Alpha]

From what I have seen a lot of the scanning issues are coming from people who are using sendout labs (Wich is not to say home printing does not experiance this but).

Often times I am seeing (and have experianced) prints of poor quality being scanned in this manor. This is due to the lab they sent it to not their own exposure error. These scans often exibit sevearly blown out skyes and grey (some times pale) shadows in the darks. Intentionally underexposing would be a some help to these people who lack their own processing ability more or less to compensate for idiot lab techs first.

An example I posted reasently shows this quite clearly.

Scanning a poor print is going to have poor results, no two ways around it. Even then if you have a quality print one is still faced with this issue of color and highlight loss in scanning.

This problem is common with non-professional labs, however most of the people I am speaking to are those who scan with flatbeds or low quality neg scanners (like a SprintScan or similar).

Max, what about scanning B&W negatives.

A few years ago, I borrowed a Minolta film scanner, but I really did not know what I was doing and some of the scans came out okay, but a lot of them were crap.

I have been seriously thinking about going back to film for my personal work and all I will be shooting is B&W film. Do the same problems persist in scanning B&W film? Thanks.

Yes, the same problems most certainly do apply to black and white negatives. Scanning also gets a lot more interesting if you use a staining developer for your negatives. However, this technique is oriented toward people who shoot film for volume or high-throughput work, need very fast turn-around time, and/or are shooting specifically for the production of a digital file. Examples would be a client who needs an image for their website, a model who doesn't need prints, etc.

I would imagine the same would apply for B&W negatives. How about slides?

A lot of this has to do with equipment.. no? Or is this a generalization? I find some digital cameras (I know this was originally about scanners) do a better job capturing details in the shadows. Underexposing by a stop and correct post-pro works pretty well.

I personally have yet to find a local lab that does negative scans to my liking... most are geared to delivering photos to the regular consumers good enough for web posting and/or small prints. I usually ended up scanning at school or at home.

Yes, and yes. Slides represent a particularly difficult challenge. They already have a much reduced exposure range compared to negative film, and underexposing will further reduce that range. Additionally, underexposed slides aren't going to give you a 1:1 proof of your final shot, since you're going into it knowing that you're going to correct. However, when you're working with tightly-controlled lighting, say in a studio setting, this can be pulled off easily. I might not recommend it for a landscape. I don't have the file with me at the moment, but I have a headshot example shot indoors with Agfa RSX II 50 slide film that I intentionally underexposed and then corrected in post. It looks great.

As for equipment, there are no flatbeds to my knowledge that excel in recording fine highlight detail from film. As I mentioned, even very expensive dedicated neg scanners can fowl up. I have an Ilfochrome I just had printed for my portfolio, that I shot on Velvia. Looks great in the the print. The CoolScan 9000 took 45 minutes to scan the 6x9 chrome set to single laser, 16-pass and most auto-options turned off. The highlights are still blocked. As such, with difficult negatives/slides, even taking them to a pro lab with top non-drum-scan equipment can leave some to be desired. Also, most labs enlarge at the time of scanning, as opposed to scanning archivally (by my own definition). By my definition, an archival scan of a neg/pos is 1:1 in size and at as close to the film's native resolution as possible (we're talking thousands of dpi here). Yet what most labs consider a "high quality non-drum scan" is scanning directly to 8x10 at 300dpi.

But remember the idea is that this technique is not designed to be used for general shooting/portfolio work, but for times when you want/need to shoot film but all you want/need in the end is a digital image.

 

[Battou]

This problem is common with non-professional labs, however most of the people I am speaking to are those who scan with flatbeds or low quality neg scanners (like a SprintScan or similar).

Yes, I did understand that, hence my original post, but after some thought I felt I should include the processing issues as it can compound the issue on top of the scanning.

Your suggestion applies to a much broader range than you where originally intending and I figured I would try to make that connection, many people (including my self) who are using non-professional labs are also using the same scanning equipment to wich you refer, many of wich the prints they have recieved.

 

[usayit]

Yes, and yes. Slides represent a particularly difficult challenge.

But if you underexpose on a negative to compensate during a scan, shouldn't you over expose a slide (positive) that is also intended for scan?

 

[Helen B]

I wouldn't recommend underexposing negative film for scanning - if anything give generous exposure to keep the shadow detail away from the toe. Colour negative film will show higher graininess with less exposure, and there should be no need to underexpose to preserve highlight detail in all but the most extreme lighting conditions. Colour negative film has a large dynamic range but a low gamma, so D-max is not excessive.

I'm puzzled about why Max is having problems with highlights 'blocking up' in a slide scanned on a Coolscan 9000. Something's wrong there. What pixel value are you setting your white point at?

B&W negs.
I use a couple of Minolta Elite 5400 (not the 'II' version) scanners for my 35 mm B&W negs. These are the best low-cost scanners I have come across for silver-image film, and I have a few (Polaroid 4000, Nikon 4000, Nikon 5000, Nikon 8000, Nikon 9000) along with ten years of experience scanning film.

Negatives that have been exposed and processed for printing by enlargement onto plain old silver gelatin will usually scan very well. You should be able to get every last bit of detail out of them.

Best,
Helen

 

[bhop]

I dunno why everyone's always so down on flatbed scanners. I think in the past they might've sucked, but i've been using my Epson 4490 for all my film scans and I haven't had any problems at all with blowouts, so they must be improving. I really, really like the scanner. I don't shoot in any special way for scanning.

 

[Alpha]

I wouldn't recommend underexposing negative film for scanning - if anything give generous exposure to keep the shadow detail away from the toe. Colour negative film will show higher graininess with less exposure, and there should be no need to underexpose to preserve highlight detail in all but the most extreme lighting conditions. Colour negative film has a large dynamic range but a low gamma, so D-max is not excessive.

I'm puzzled about why Max is having problems with highlights 'blocking up' in a slide scanned on a Coolscan 9000. Something's wrong there. What pixel value are you setting your white point at?

B&W negs.
I use a couple of Minolta Elite 5400 (not the 'II' version) scanners for my 35 mm B&W negs. These are the best low-cost scanners I have come across for silver-image film, and I have a few (Polaroid 4000, Nikon 4000, Nikon 5000, Nikon 8000, Nikon 9000) along with ten years of experience scanning film.

Negatives that have been exposed and processed for printing by enlargement onto plain old silver gelatin will usually scan very well. You should be able to get every last bit of detail out of them.

Best,
Helen

Oh, Helen dear. You always seem to have these oh-so-easy be-all-end-all solutions to such common problems. How fortunate for you. You should write a book. I'm sure you'd make a killing solving everyone's problems with simple theoretical propositions. While I admire your scientifically-geared testing, most of your results are not replicable by mere-mortals. You must have a silver halide touch.

As for color negative film, I'm well aware of it's general capabilities, including its tolerance for overexposure. Grossly underexposing the film will introduce a lot of grain into the shadows just as you say...not so much with slight underexposure, though the effect of underexposure on grain will multiply in the faster films. Flatbeds especially have a very hard time dealing with fine highlight detail, whatever the film type. Color neg is no exception. I've devised this method precisely because this is a routine problem, not one that rears its head on rare occasions of "extreme lighting conditions."

The Nikon scans are uncorrected TIFF. But perhaps you're right. There must be something terribly wrong going on in the multi-million dollar multimedia lab I work in. The Nikon, for all its reliable and excellent general performance, still does have its shortcomings. One need only compare it against a Flextight or better yet a drum scanner to see that.

As for black and white, I can't speak for Minolta as I've never used it. But the SprintScan series are really terrible pieces of crap as far as I'm concerned. When I wrote earlier about these problems occurring with black and white negatives, I was referring mostly to flatbeds, though I have had the occasional problem on nicer equipment. Either way I think black and white is a moot point. I don't advocate anyone underexposing all of their shots unless they really know what they're doing, i.e. do it as part of a finely-tuned shooting workflow (Nick Brandt, for example, grossly underexposes his subjects in order to nail the background/sky and then sucks every bit of detail out in post that he can find). What I was speaking on was not whether you should underexpose b&w, but rather whether average quality scanners are capable of recovering fine detail from a normal exposure. Applying this method to black and white film, in many cases, would produce underwhelming results.

Please bear in mind the audience I am speaking to here. Mostly I am not speaking to those with access to excellent equipment as I stated in my original post, but to people with everyday equipment who shoot to digitize and have had chronic difficulties. The problems I've described are endemic to most consumer and "pro-sumer" grade scanners. This is simply a proposition of one way to cope with their shortcomings.

If you have some better alternative, please, by all means, speak up because I'd love to use it myself. But for most, feasible does not include spending large amounts of money on a scanner (and let me preempt you by noting that judging from your previous posts, your idea of a modest equipment price has a seriously inflated price tag).

Best,
Max