Jump to content

Digitizing 120: Opinions regarding DSLR vs flatbed vs film scanner?


lukpac

Recommended Posts

  • Replies 134
  • Created
  • Last Reply

Top Posters In This Topic

If a flatbed would give results about the same or perhaps worse than my DSLR, I'm not about to spend the money on one. However, if a flatbed is likely to produce better results, I may consider one.

 

Flatbeds do better with B&W than color, so that falls in your favor, while DSLR tends to do somewhat better "scanning" in color. So in some respects its a wash. What a (really good) flatbed might offer is a more predictable workflow for B&W MF negatives, but so much of that depends on how well you've set up your DSLR. From the looks of your samples, you seem to have the rigidity dialed in, perhaps (as rodeo_joe suggests) you just need to get the film positioned flatter and more parallel to the camera. Again, if judging only by your web-quality samples, I think you're already getting about as good as you're gonna get without spending for a fairly high-end scanner: finesse the film a bit flatter, and you've aced it.

 

If these images are important enough for you to engage in this thread to begin with, they're probably important enough for you to sink a *little* bit of money into so you can get a better idea of the options. While the Epson V700/750/800 is the "gold standard" in attainable flatbeds, their price is somewhat inflated vs comparable capabilities in mid-priced Epsons. You can pick up a brand new v600 for under $200, which (with meticulous use) would get you within 90% of a v700. Much of the extra cost on the v700 goes to 4x5 wet scan optimizations most buyers will never use, the 750 just adds the fancy SilverFast software. You could also look for a nice second-hand Epson 4990, which was the "sleeper" flatbed of the line: very very good for a "budget" scanner. You do need to play around with the film holders quite a bit until you nail the perfect setup, then the workflow becomes routine. Note some claim to get sharper results just laying the negs flat on the glass, emulsion side down, with the film holders on top of them to keep film flat. Scanners are voodoo toys.

 

Anyway I don't think you're gonna get a definitive answer you're certain of until you try a flatbed yourself. At most, you risk $100 loss: buy a new v600, use it for a couple months, and if it offers no advantages in quality or workflow over your DSLR method, just resell it for near what you paid.

Edited by orsetto
  • Like 1
Link to comment
Share on other sites

Note some claim to get sharper results just laying the negs flat on the glass, emulsion side down, with the film holders on top of them to keep film flat. Scanners are voodoo toys.

 

I think one of the biggest sources of potential frustration with the Epson flatbeds is that the focus point seems to vary, and there's also no way to change it(that I know of) like on the Nikons(which AF on every frame, and IIRC Vuescan at least even lets you specify where on the frame you want to focus).

 

The Epson holders do come with shims and you're supposed to experiment to determine the optimum holder height, although they're somewhat limited. The BetterScanning holders use relatively fine pitched screws, which are even more fiddly to set up but the effort is well worth it in terms of the ultimate quality of your scans.

 

Of course, the V700 I use at least IS designed for on-bed scanning. Officially, the only prescribed use is to place the 8x10 mask on the bed and then lay an 8x10 sheet within it. Unofficially, I've scanned 110, 127, and a few others directly on the bed under a sheet of AN glass. When you tell the scanner you're scanning on the bed, it apparently uses the optics designed for reflected scanning, on which the resolution is limited(IIRC) to 4000 dpi and ICE doesn't work. Of course, you could use the glass holder(at least on he V700/750/800/850) and still get the higher resolutions along with ICE.

 

I still think this is one of the best scans I've ever been able to turn out on my V700. This was a wet scan on the glass tray, using straight heptane as the scanning fluid and under a second sheet of glass. Unfortunately, I wasn't able to get all the bubbles out. This was done without ICE...wet scanning does minimize the appearance of film defects. The second attachment is the full res scan...bubbles and all. This was taken around 2008 on what was probably then expired Provia using either my Rolleicord Va or Rolleiflex Automat III. Processing would have been done either by Dwayne's or the local(since closed) pro lab.

 

1721690351_frame28copy.thumb.jpg.2284d32837f766d75ddf92e7d437cfb0.jpg

 

2096807210_frame28.thumb.jpg.84d8ab6b210db68aae84f5b620c50eb7.jpg

  • Like 2
Link to comment
Share on other sites

Why the chemophobia?

 

I picked heptane because it's a relatively non-volatile hydrocarbon(at least as compared to hexane-n-heptane boils at 98ºC) and I had plenty of it on hand in HPLC grade. I understand that the wet mounting fluid Aztek sells is a mostly heptane and IPA.

 

The old Kodak film cleaner is a mixture of heptane and a heavy chlorocarbon.

 

I'm not overly worried about the small amount needed to wet mount a transparency at home....and yes at work I have work areas rated for significantly more dangerous stuff(like the elemental bromine that I was handling yesterday...stuff like that makes you think of a hydrocarbon with a boiling point similar to that of water as being relatively benign).

Link to comment
Share on other sites

Yes, bromine isn't flammable, but it's still easily one of the nastiest of what I'd call "common" laboratory reagents(I had someone storing a couple of cylinders of 5% fluorine in helium with me for a while, and I regularly begged them to get the things out even though a sealed cylinder is pretty harmless).

 

Still, if it were so dangerous would Aztek sell a solution that was a mixture of it and something even more volatile(IPA) as a mounting fluid?

 

I wouldn't use hexane at home as a mounting fluid-although its volatility would be as much an inconvenience(keeping an adequate fluid film on intact while scanning) as its flammability.

 

Truth be told, there's a lot more fear of flammability with a lot of things than actually exists. I often have the discussion at work that when the current batch of folks doing a PhD on the main GC-MS graduate, I'm going to convert it to using hydrogen as a carrier gas. A lot of folks lose their minds over this and are terrified of the thought of hydrogen as a carrier gas. In reality, it's very difficult to get a flammable mixture even from a bad leak in poorly ventilated room-much less when flowing 1mL/min in a lab, and you get big benefits in terms of resolution, analysis time, and cost($50 for a cylinder vs. $250).

Link to comment
Share on other sites

Digital ICE is unfortunately not an option for B&W, which is 99.9% of what I'll be scanning.

 

For sharp and detailed B&W scans, a used inexpensive Polaroid Sprintscan 120 or Microtek 120tf with an older version of Silverfast would meet your needs if you chose to go the film scanner route and could sandwich the film between glass. The two scanners are fast and have auto focusing

Link to comment
Share on other sites

For sharp and detailed B&W scans, a used inexpensive Polaroid Sprintscan 120 or Microtek 120tf with an older version of Silverfast would meet your needs if you chose to go the film scanner route and could sandwich the film between glass. The two scanners are fast and have auto focusing

 

- A 16+ year old obsolete and unsupported scanner that, if you can find one, goes for over £500 ($700 US) on eBay? What a bargain!

 

Honestly, the OP could probably have copied a couple of hundred negatives with his digital camera, and processed them, in the time spent posting and reading this thread. And got results no worse than any dedicated scanner bought for under $500.

 

Just seen the bargain price of Canon's Canoscan 9000F. That would be my choice if I was in the market for such a thing.

Link to comment
Share on other sites

Given that many of the former photography companies gave up making scanners, I'm not certain the scanners from that time are obsolete since there were no real attempts to replace them with serious consumer/pro models.

 

- They're obsolete in the sense that they're no longer made or supported in any way by the manufacturer. And in some cases the manufacturer is no longer in business.

 

Then again, some use hardware interfaces like SCSI, that are definitely obsolete technology.

Link to comment
Share on other sites

Honestly, the OP could probably have copied a couple of hundred negatives with his digital camera, and processed them, in the time spent posting and reading this thread. And got results no worse than any dedicated scanner bought for under $500.

 

I honestly haven't had that much time. But I did make it through almost 40 photos last night, over the course of a few hours. They look pretty good, although shadow details are tough.

 

257564714_ActOne023_02.thumb.jpg.e8ad91e60abc7ec3a4f48f2a7c21361c.jpg

  • Like 1
Link to comment
Share on other sites

It would be a simple matter to take bracketed images with a digital camera, which should penetrate the darkest parts of normally-exposed film, whether B&W, Kodachrome or Velvia. A two-stop over/under bracket would increase the DMax from 4.0 to about 4.6 (approaching that of a solar eclipse filter), and possibly pull some detail out of the thinner areas too. That only adds about 5 seconds to the "scan" time, but about 5 minutes to the processing.
Link to comment
Share on other sites

It would be a simple matter to take bracketed images with a digital camera, which should penetrate the darkest parts of normally-exposed film, whether B&W,

 

- Errrm, in negative film the shadows are the most transparent part of the film and don't need any 'penetrating'. Plus the useable Dmax of a B&W neg should be no more than 2.2D, otherwise the film has been heavily over-developed.

 

What negatives generally need is to use an S-shaped tone curve in the positive, which tails gently away to pure black.

 

This can usually be done by setting the black point just to the left of the hump in the histogram and then setting a pull-point about 1/3rd up the slope. Pull the straight line down into a curve that retains the required shadow detail, and then work on the highlights.

 

This must be done using a 16 bit/channel image got from a camera RAW file. 8 bit JPEG output is useless for this purpose.

 

BTW, the shadow detail in the example above looks fine. What do you expect to see under the wheelarch of a car that's of any interest?

Edited by rodeo_joe|1
  • Like 3
Link to comment
Share on other sites

Errrm, in negative film the shadows are the most transparent part of the film and don't need any 'penetrating'.

I'm aware of that. That's why I specified the darkest part of the [original] image, which would apply to either negative or positive media. Bracketed sets are usually symmetrical around the calculated exposure, which covers either possibility.

 

Digital imaging is almost linear spanning thermal noise levels to saturation. Curves are applied when processing RAW images, including bracketed HDR images. That's how an 14-18 stop range can be presented on a 6 stop medium, like prints or a computer monitor. A similar process occurs when you capture a scene on film, sometimes in reverse. Velvia, for example, has about 6 stops of capture range, which it translates into a 10 stop range of density on the film.

 

Each integer increment of density is a factor of 10, equivalent to about 3 f/stops.

Edited by Ed_Ingold
Link to comment
Share on other sites

All the 'useful stuff' in a B&W or colour negative should be contained within a density range of 1.8 to 2.2D. That's only 7 to 8 stops maximum, which can be located almost anywhere in the 12 stop range that any decent modern digital camera can hold in a RAW file.

 

This makes the exposure latitude for copying negatives very wide, and as long as there are no 'blinkies' in the digital copy, you should be able to show all the detail that the negative contains. However, it may be necessary to deliberately lose some shadow or highlight detail in order to prevent the overall image looking flat or grey.

 

Pictorial effectiveness should dictate the tone curve used, rather than some technical ideal of showing every pebble in a shaded patch of tarmac to the shine on a chromed bumper.

 

And of course, if the shadow detail hasn't been captured on film to begin with, then it's impossible to show it in a digital copy.

Link to comment
Share on other sites

BTW, the shadow detail in the example above looks fine. What do you expect to see under the wheelarch of a car that's of any interest?

Picture looks great. WHo cares about the shadows?

 

I wasn't referring to that photo. More about photos like this:502379133_ActOne019_06.thumb.jpg.678186f679f976e1e514c023a6af49e0.jpg

Link to comment
Share on other sites

Would there be any software that could fix that one ? I doubt it. But since your camera is doing a good job of copying, perhaps you should just keep going and do the rest that way. The negs seem to be in good condition which means less work editing the images
Link to comment
Share on other sites

I wasn't referring to that photo. More about photos like this:[ATTACH=full]1274809[/ATTACH]

 

- It looks as if the negative was very underexposed. Scanning, or digital copying and software can fix a lot, but it can't work miracles.

 

How are you capturing and processing the files? Automatic processing usually gives acceptable, but not optimal results. And if you're not shooting RAW with the camera, then the amount of correction possible is going to be severely limited.

 

I suggest you get to know your way around the curves tool, and not rely on auto-everything post processing.

 

I took the liberty of tweaking the above picture. This is about the best I could do with only an 8 bit JPEG to work from.

IMG_20181211_135459.thumb.jpg.c865d4a70c26b281d8d4b99f5b85d747.jpg

However, the histogram shows a very narrow band of density, and even with more bit-depth to play with, this image might be a lost cause.

Link to comment
Share on other sites

How are you capturing and processing the files? Automatic processing usually gives acceptable, but not optimal results. And if you're not shooting RAW with the camera, then the amount of correction possible is going to be severely limited.

 

I suggest you get to know your way around the curves tool, and not rely on auto-everything post processing.

 

Generally I've been letting the camera handle the exposure automatically. I've toyed with compensation a few times, but I don't think I've tried it for the above photo. Yet.

 

I've never done much with curves, but all of these have had manual levels adjustments, first on the separate R, G, and B channels, and then on the overall levels after converting to grayscale. Nothing has been automatic in that regard.

 

As a tangent, I wish I knew what camera my dad had used to take these. Unfortunately, I think whatever it was he must have gotten rid of it after getting a Nikon F.

Link to comment
Share on other sites

All the 'useful stuff' in a B&W or colour negative should be contained within a density range of 1.8 to 2.2D. That's only 7 to 8 stops maximum, which can be located almost anywhere in the 12 stop range that any decent modern digital camera can hold in a RAW file.

I don't dispute your assessment of color negatives, but B&W negatives can have a very high density, as can certain reversal films like Kodachrome and Velvia. You can see in the product information sheet below that the DMax of Velvia 50 is approximately 3.6, with a density range of 3.3, or about 10 stops. This should be in range for a modern digital camera (or legacy scanner). However the task of extracting information at the extremes is easier and less noisy using a bracketed exposure and tone mapping.

 

http://www.fujifilm.com/products/professional_films/pdf/velvia_50_datasheet.pdf

 

In my experience, both with scanning and the darkroom, it is generally easier to extract. useful information from dense areas in a negative than underexposed portions. IMO, ISO values tend to be inflated for marketing reasons. In a few short years, Tri-X went from 160 to 400 without any discernible change in the film itself. I got better results overexposing both Tri-X and color negative film by about one stop.

 

I can't improve much on the B&W image above, but it might work better to take more pains scanning the original film than adjusting a JPEG image on the internet. That said, getting good results from a thin negative or the deepest shadows of Velvia is a fools errand. You can't process what isn't there.

  • Like 1
Link to comment
Share on other sites

...but B&W negatives can have a very high density..

 

- Yes they can, but they shouldn't.

A properly developed B&W film should top out at around 2.2D - give or take a couple of tenths to allow for base+fog.

 

There's a popular culture of going for a 'good thick' neg. It's mistaken and bad practise.

  • Like 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...