Scanned film vs. traditional darkroom

[John51]

How much difference is there between an optical RA4 print and scan+laser RA4 print?

 

[Bob Carnie]

How much difference is there between an optical RA4 print and scan+laser RA4 print?

None if both are done correctly

 

[peoplemerge]

I thought laser was limited to 8 bit color depth... is that only in the case of the machine used by my local service provider? I think it's a lightjet, but it may be something different.

 

[MattKing]

I thought laser was limited to 8 bit color depth... is that only in the case of the machine used by my local service provider? I think it's a lightjet, but it may be something different.

When you get to the printing stage, 8 bit is fine.
It is the editing steps before that that need more.

 

[alanrockwood]

When you get to the printing stage, 8 bit is fine.
It is the editing steps before that that need more.

I think that at the scanning stage 8 bits is also OK. Possible exceptions might be for scanning something like Velvia. It's in the intermediate photo processing steps where more bits might be needed.

The thing is, if the variability in the scanned film (through a combination of film grain and scanner sensor noise) amounts to at least roughly a half of a bit increment then the signal is effectively dithered, which swamps the finite step size of the digital signal. In those cases you won't see banding or related artifacts. However, to be sure that banding, etc., do not occur in the processing of the image subsequent to digitization it may sometimes be necessary to convert to a higher number of bits before doing the processing.

I should point out one thing. If the dither in the signal is near the threshold of where dithering becomes effective to prevent banding then there will be a little more average noise in the digitized signal. This is due to the coarseness of the digitization step size. However, if the dither is much above that limit then the added noise due to the digitization process is basically negligible in comparison to the dither itself. Also, the effective dithering will be most effective if you scan at the highest resolution setting of the scanner. (But not at settings that result in interpolation.)

One way to test this on your scanner is to set the scanner to the highest resolution setting. Then scan a slide or negative where there is no gradient in the image. Then zoom in on the histogram for that region and see if there is just one or several different peaks in the histogram for that region. If there are several peaks then the signal is effectively dithered and there is virtually nothing to be gained by going to a higher bit depth. This should be done at both maximum density and minimum density. I have done some limited testing like this on my scanner when scanning negatives and have so far found that there were several peaks present in each of my tests, so for my scanner when scanning those films there is nothing to be gained by going to higher bit depth during the scanning process.

 

[peoplemerge]

@alanrockwood - interesting. It always seemed like the place where you'd want particularly high bit depth is where you _do_ have a gradient, such as the sky. I'll try to digest what you're saying about peaks, is there some research on the web that elaborate that technique? Sorry to go on a tangent from the OP's topic.

 

[alanrockwood]

@alanrockwood - interesting. It always seemed like the place where you'd want particularly high bit depth is where you _do_ have a gradient, such as the sky. I'll try to digest what you're saying about peaks, is there some research on the web that elaborate that technique? Sorry to go on a tangent from the OP's topic.

There is a thread at photrio that (among other things) discusses the issue of banding, grain, sensor noise, etc. The best post to look at is probably #38. It shows an example of how noise in the image can eliminate banding. It comes from a For a more theoretical post look at #27.

Here is the link:

https://www.photrio.com/forum/threads/sensor-noise-in-film-scanner.162143/page-2

The images in post #38 in the link just referenced came from a thesis. Here is a link to the pdf of the thesis.

https://uwspace.uwaterloo.ca/bitstr...d=1830F1F5528E43CDB940023C1BD664E2?sequence=1

The thesis shows black and white and some color examples. There's a lot of math in the thesis. Feel free to skip over the math.

To clarify a little bit about my use of the word "peaks" it refers to peaks in the histogram. And if you zoom in over a very limited range in the x-direction of the histogram you will eventually come to the point where there are a series of sharp peaks. The separation of the peaks in the x-direction corresponds to one-bit increments of the digitizer. If I get around to it I will try to find some images to post illustrating this.

 

[dmr]

How much difference is there between an optical RA4 print and scan+laser RA4 print?

LOL, I daresay that most of the users of this system could not consistently pass a side-by-side comparison test of the two. Ditto even for things such as a typical lab wet print and a properly-printed inkjet (excuse me, giclee', to use the buzzword-compliant term) print.

 

[DREW WILEY]

Depends on what your skill set is, along with your equipment and personal preferences, and in fact on how critically trained your sense of color is. Don't laugh; 50% of color differentiation is psychological rather than physiological. I've trained professional color matchers. Now when someone like Bob Carnie chimes in, you have to realize he's a switch hitter with quite a bit of experience in both realms. The rest of us are lucky to get proficient in just one or the other. I dislike computer work and enjoy darkroom work, so stand firmly in the camp that true optical prints are "better", because that's what I do best. And I'm very well equipped for that. But it involves stinky chemicals; and not many people these days can afford space for a serious color darkroom. The RA4 process per se is fairly easy; but everything leading up to that can be either as simple or as complicated as you wish. I'm one of the few people that does advanced unsharp masking techniques with color negative film. Scanning and laser printing does not eliminate that need, but just does it differently via PS controls. I like that tactility of real film; others like to
fiddle with keyboards and a mouse (is the plural "mouses" rather than mice?).

 

[mshchem]

Depends on what your skill set is, along with your equipment and personal preferences, and in fact on how critically trained your sense of color is. Don't laugh; 50% of color differentiation is psychological rather than physiological. I've trained professional color matchers. Now when someone like Bob Carnie chimes in, you have to realize he's a switch hitter with quite a bit of experience in both realms. The rest of us are lucky to get proficient in just one or the other. I dislike computer work and enjoy darkroom work, so stand firmly in the camp that true optical prints are "better", because that's what I do best. And I'm very well equipped for that. But it involves stinky chemicals; and not many people these days can afford space for a serious color darkroom. The RA4 process per se is fairly easy; but everything leading up to that can be either as simple or as complicated as you wish. I'm one of the few people that does advanced unsharp masking techniques with color negative film. Scanning and laser printing does not eliminate that need, but just does it differently via PS controls. I like that tactility of real film; others like to
fiddle with keyboards and a mouse (is the plural "mouses" rather than mice?).

Well said. I'm not proficient at scanning color negative film. I do great with my Dad's Kodachrome slides, print with inkjet, get great images. Medium format optical prints from Portra on Fuji CA paper are fabulous.

 

[jtk]

OK, this is not quite your standard "film vs. digital" question because film is involved in both branches of the comparison.

The question is, how does a print that is generated by scanning film (along with doing the digital processing of the film, such as converting the negative to a positive image, retouching, etc.) and printing the scan with a high quality desktop printer compare to a print generated entirely in the darkroom?

A related question: How does a digital image generated by scanning film (along with the software processing required) compare to making a wet print in the darkroom and then scanning the darkroom print?

Obviously, I am assuming that the same negative is used in both branches of a comparison.

IMO an optical print from a negative can only hint at whats normal for a reasonably skilled person (i.e. most minimal Photoshop skill) with a reasonably good inkjet printer (e.g. Canon Pro). Enlarging optics always soften detail and "grain" and inkjet pigments produce better blacks.

 

[JWMster]

jtk: I thought you were ONLY digital these days. Do you print both ways, or is it prior vs. current experience? Do you use or have you tried Piezography? Either way... I appreciate your comments all the more 'cause so few will actually suggest similar. As a late comer, I learned digital printing and pushing it hard there, I'm not inclined at the moment to go down the path of learning to wet print for its own sake. Steve Shaub and others have demonstrated a lot of wonderful digitally printed images with good papers (expansive in some cases) and solid equipment (less expensive) that I'm satisfied that images that don't meet my aesthetics from digital printing probably are my responsibility and not a shortcoming of the medium. And trying to get to the same level with wet prints - especially with color - looks like a pretty steep climb for guy of my vintage. I watched a Dodge and Burn video of a Large Format landscape / nature photographer out in Seattle somewhere working with old Cibachrome stock, and the guy was amazing what he could tease out of his images. But also.... it just seemed like.... "really.... are you kidding me?" and not likely to happen in my lifetime.

 

[jtk]

jtk: I thought you were ONLY digital these days. Do you print both ways, or is it prior vs. current experience? Do you use or have you tried Piezography? Either way... I appreciate your comments all the more 'cause so few will actually suggest similar. As a late comer, I learned digital printing and pushing it hard there, I'm not inclined at the moment to go down the path of learning to wet print for its own sake. Steve Shaub and others have demonstrated a lot of wonderful digitally printed images with good papers (expansive in some cases) and solid equipment (less expensive) that I'm satisfied that images that don't meet my aesthetics from digital printing probably are my responsibility and not a shortcoming of the medium. And trying to get to the same level with wet prints - especially with color - looks like a pretty steep climb for guy of my vintage. I watched a Dodge and Burn video of a Large Format landscape / nature photographer out in Seattle somewhere working with old Cibachrome stock, and the guy was amazing what he could tease out of his images. But also.... it just seemed like.... "really.... are you kidding me?" and not likely to happen in my lifetime.

I ONLY print inkjet (Canon pigment) but I shoot film and digital AND I'm turning into some kind of archaeologist...because I enjoy printing from old flatbed scanned prints and weirdly-sized negs and, recently, ancient Kodachromes.

I've never tried Piezography but thanks to Print Exchanges I have seen many fine examples of it...and am not convinced that it is better than OEM pigments from Epson or Canon.

Way back, I was a reasonably good semi-professional wet darkroom color printer. That's where I learned to specify color differences in terms of Color Compensating units (now a lost art)...which I wish was directly applicable to Photoshop and inkjet printers...but doesn't seem to be.

Life changed, I quit pro photography after 10 years, quit wet darkroom printing entirely (and almost quit photo) taking a few years to discover flatbed scanning and inkjet printing. Got a Nikon V, got excited about inkjet B&W, then got a crappy little digicam, then got a Pentax DSLR etc etc etc. You can see that mix in my Photrio Media.

Not gotten far into PS, but I'm increasingly using it for dodging and burning.

I never pursued Ciba printing more than superficially but it did make better blacks than Ektacolor. I found Ciba transparencies more useful professionally. Recently re-discovered Duratrans, which would be more tempting than Ciba if I still had the opportunity to make exhibits.

 

[wilson109]

My general rule is color digital, B&W film. Nothing compares to printing from a B&W neg and watching an image appear by magic. I currently use 35mm but I am looking into a good used medium format film camera. However when I travel however I only take my DSLM because of size and weight with a small P&S as backup. There is only so much you can get in a carry on.

 

[markjwyatt]

Not sure if this was brought up (saw references to color), but how about scanned B&W negatives sent to a lab who then make wet prints? I know this type of enlarger (laser scanner) is not affordable for home use, but I wonder if there are any advantages/disadvantages to this process? I think being able to digitally preprocess the image could be an advantage. You are more at the mercy of the skill of the operator, but there are custom services, and this seems to be a very common way for commercial labs to make B&W silver prints.

 

[peoplemerge]

It seems the best way to get a fine print out of a simply horrible neg that won't print out of an enlarger. Or possibly if you need to make a fairly large run of truly identical silver prints.

 

[EdSawyer]

optical RA4 beats any scanned/printed variety, hands-down. Even if for nothing else than an entire generation of interpolation (scanning, RIP, inkjet printing) not happening in the workflow. Using top-of-the-line enlarging lenses (apo el nikkor) extracts the most possible quality from the optical workflow.

 

[John51]

It's the top of the line part that I'm not keen on.

If reasonably priced kit, eg. Mid range Durst + non apo 6 element lens will get within a size or two of lab prints, I'd take that as good enough to consider having a go at RA4. iow, if the lab can make say a 20x24 print from a scanned neg that will take close scrutiny, home processing a 16x20 RA4 (or even 11x14) that will take close scrutiny (same neg) will do for me.

 

[alanrockwood]

Not sure if this was brought up (saw references to color), but how about scanned B&W negatives sent to a lab who then make wet prints? I know this type of enlarger (laser scanner) is not affordable for home use, but I wonder if there are any advantages/disadvantages to this process? I think being able to digitally preprocess the image could be an advantage. You are more at the mercy of the skill of the operator, but there are custom services, and this seems to be a very common way for commercial labs to make B&W silver prints.

I did this with one negative and it turned out well.