V600 versus lab scans

[Montanawildlives]

Please excuse the complete newbie question. I was looking at purchasing the Epson v600 to scan 35 mm color and black-and-white negatives and in the future possibly medium format like 645. The v600 obviously has more resolution in terms of DPI than even enhanced or super scans from most Labs but I realize that's not the end of the story. So I'm wondering about the quality of the scans made from the v600 versus Labs such as the darkroom or even a local lab. For example does the v600 offer convenience and cost savings over the long term but the scans are lower quality? I'm not ready for the v800 or the V850 at this point any comments on that would be appreciated too. Thanks!

 

[faberryman]

If you currently only shoot 35mm, you might want to consider a film scanner instead of a flatbed. You will get much more detailed scans.

 

[Lachlan Young]

Please excuse the complete newbie question. I was looking at purchasing the Epson v600 to scan 35 mm color and black-and-white negatives and in the future possibly medium format like 645. The v600 obviously has more resolution in terms of DPI than even enhanced or super scans from most Labs but I realize that's not the end of the story. So I'm wondering about the quality of the scans made from the v600 versus Labs such as the darkroom or even a local lab. For example does the v600 offer convenience and cost savings over the long term but the scans are lower quality? I'm not ready for the v800 or the V850 at this point any comments on that would be appreciated too. Thanks!

Pretty much any minilab scanner outresolves an Epson V-whatever where it actually matters. The Epson specs are largely wishful thinking rather than a realistic representation of what they actually resolve (not a lot & not very crisply).

 

[Montanawildlives]

Thanks. You mean like the Kodak Scanza or something?

 

[Montanawildlives]

And this would be as good as or better than, for example, the 2048×3072 "enhanced scans" (~18.1MB they say) from The Darkroom ($4 per roll).

 

[faberryman]

Thanks. You mean like the Kodak Scanza or something?

The Kodak Scanza is a piece of junk. Take a look at PlusTek and Pacific Image film scanners at B&H.

 

[Les Sarile]

Minilab scanners can offer more detail and sharpness - as long as the data was captured on the film accordingly, but unfortunately not much control in terms of crop, color and contrast as they run them in fully automatic unattended mode with "enhancement settings" on. These enhancement settings usually include excessive brightness, uncontrolled auto levels and over sharpening. Since most people don't really know what the color negative should look like, they usually think this is just fine anyway. With the proper use of an Epson V600, you can control these things but that obviously takes more work then paying the minilab to do it.

 

[BMbikerider]

A dedicated film scanner if you can find a good one with little use, included in the shopping list has to be a Nikon Coolscan V. The may be a little slow by a flatbed scan terms, but the ultimate quality has to be seen to be believed. There one drawback, using their original software they can only be used with a computer using XP, but Silverfast software can be used to programme their use on Windows 7 and higher.

However the D max the Nikon scanner can produce gives enhanced highlights and shadow detail is 4.2. An Epson V600 will only yield around 3.5. In the Epson range and indeed any other amateur flatbed scanner will not come close to it except the Epson V800/850 which will yield a D max of 4.1. However that is around 3 times the price of a V600.

The Nikon scanner has far a better Digital Ice capability than the Epson and will get rid of almost any unwanted artifacts on the film, whereas the Epson version is, to be kind, a bit 'hit and miss'.

 

[Alan Edward Klein]

Unless you pay a lot of money for better outside scans, I don;t think the results will be as good as you can get with a V600. Check my Fickr page to see results of my V600 scans of 35mm and 120 MF film. There is a learning curve and lots of time is spent scanning and editing. But that's what it is with film.

 

[MattKing]

Every time you use a flatbed scanner for film, you are trying to focus through a flat piece of glass that can get dirty and be scratched.

Scanning 35mm film well with a flatbed scanner isn't as easy as the advertisements indicate. And you will lose a substantial portion of the resolution in the film, which you will need to replace with a fair amount of digital manipulation, including artificial "sharpening". In general, flatbed scanners don't give you more than about 1800 ppi in real/native resolution, so that means decent quality for prints up to about 6"x9" unless and until you add those manipulations and that artificial sharpening.

It goes without saying that the necessary digital manipulation is rarely achieved "automatically".

If you are limiting the results to display on the internet, limiting resolution in that way isn't all that consequential. You have almost enough native resolution to fill a "4K" screen. But if you want to print as well.....

All that being said, for the relatively small proportion of my film that I need a digital file for, I almost always use a flatbed scanner. But if I want high quality results, I either print optically in my darkroom, or send the slide or negative out to be scanned professionally.

 

[Wallendo]

I have scanned 35mm negatives and slides with a V600 and feel my results are better than the scans I get from thedarkroom.com. It takes a lot time and effort but you can get decent scans from a V600.

I now use the V600 for medium format and prints. I use a Plustek film scanner for 35mm.

 

[hsandler]

I find flatbed scans with a V500 can jst barely give you 3000x2000 real resolution from a 35mm frame if you take care to scan at the optimal height for the particular scanner. I find typical lab scans (I think many use the Pakon scanner for 35mm) are about the same resolution but have sharpening turned up so they look crisper. They maybe even are sharper, but the problem I have with typical lab scans is they clip highlights and amp up the contrast and saturation too much. So I would prefer flatbed scans. But it does take a long time to scan a 36 exposure roll of 35mm film with care, like an entire evening. That’s one reason I generally prefer medium and large fomat these days; if I’m going to spend so much time scanning, I want a high res result.

 

[Lachlan Young]

Honestly, I'd avoid pretty much any desktop scanner - they rarely deliver anything close to 50% of the claimed resolution (and even then, it's laden with optical shortcomings that no amount of time spent on the computer will solve) & the handful that do (Nikon) seem to produce a fairly strong degree of grain aliasing.

While the Noritsu or Frontier minilab scanners do have significant limitations compared to high end CCD or PMT drum scanning, they represent a drastic quality step above the V600 etc - and if isn't, you need to have a chat with your lab about how well maintained their machines are or look closely at how sharp your negs really are. The minilab machines have quite a wide array of user controls, so an amenable lab may be able to deliver what you want with fairly minimal fuss.

If you want to do your own scanning & don't want (or need) to fall down the rabbit hole of high end CCD or PMT drum scanning, a full-frame DSLR, a macro lens, a light table & the means to keep them in a tight relationship & control flare (not too difficult) may actually deliver better results than a consumer scanner.

 

[shutterfinger]

I keep reading these posts but there is no qualification to the comments.
My eye is calibrated to Gallery Quality Print be it traditional it inkjet.
In post 9 of https://www.photrio.com/forum/threads/flatbed-film-scan-greater-resolution.162327/ is some useful information on scanner resolution.
3rd party testers use a USAF 1951 test target for the scanner resolution. Its fine for testing film and lens resolution as designed but does not provide a true picture of a scanners ability. The ISO 12233 https://www.graphics.cornell.edu/~westin/misc/ISO_12233-reschart.pdf is much better.
Due to the sites limitation a full res scan cannot be posted. Full resolution scans must be downsized to post inline. I have a V500 and Plustek 7600i. The only difference between the V500 and V600 is the transparency area.
Silverfast Ai6 will not run on Windows 10.

Top left: V500 @ 6400dpi, 48 bit RGB, Vuescan, color set for the film type.
Top right: V500 @ 6400 dpi, 48 bit RGB Epson Scan with No Color Correction selected.
Bottom Left Plustek 7600i @ 7200dpi, 48 bit RGB, Vuescan with color set to the film type.
Bottom Left: V500 vuescan scan color adjusted in PS.
Film: Kodak Pro 100 (PRN).
DPI set to manufacturer stated optical limit.
Gallery quality, maybe. Proof quality for a magazine editor/gallery curator, definitely.
Good scanning takes practice. Scan for the best detail from the negative, adjust the scan for best image in post software.

The negative on a Logan light box with a Nikon D300 , 60mm f2.8 Micro Nikkor hand held.

 

[Alan Edward Klein]

None of the 4 pictures looks normal. What's your point?

 

[Herminus]

According to scandig the V600 has real resolution of1560dpi, this is very low.

https://www.filmscanner.info/en/EpsonPerfectionV600Photo.html

 

[shutterfinger]

According to scandig the V600 has real resolution of1560dpi, this is very low.

Another lens resolution target used to evaluate a scanner sensor. Misleading.

 

[Herminus]

Here is intertesting video with V800 vs V600 :

 

[Les Sarile]

Personally, I don't find sharpness and resolution as important as color fidelity. Here are a couple of examples of color failures I've encountered that would make you think they were from different frames of film when in fact they are the same one.

 

[bernard_L]

Here are a couple of examples of color failures

Re: Coolscan 5000 versus Epson V700. You do not state what software was used respectively for these two scans. The color balance depends more on the software+procedure: auto-colors? auto-contrast? fix and freeze gray point on gray card frame? profiled?... than on the harware model that you use to label the images.
But I'm sure you know that quite well.

 

[Les Sarile]

But I'm sure you know that quite well.

Coolscan + Nikonscan, Epson + Epsonscan - both fully automatic with no pre or post anything.
Over 40K frames of various films scanned to date and I am still learning . . .

 

[bernard_L]

More interesting would be both scanners with, say... vuescan, and same settings. There is a difference, but much less. In my limited experience (slides) the colors are more saturated with coolscan, possibly because the RGB LED's (Coolscan) have narrower spectra than the response of the EPSON RGB matrix sensor, and presumably less sensitive to dye cross-talk.

 

[klownshed]

EpsonScan on automatic results in ugly files with no headroom for editing.

Turn everything off and you get tons of headroom and if you are any good at editing you will have much better files. Editing the resulting flat scans is easy for B&W, not so easy for colour until you’ve worked out the method. I use RGB curves to get the colour how I like it. When I can be bothered with colour film, which is increasingly rare. I don’t love colour.

I love black and white. And a V750 is pretty good for black and white in my experience.

 

[Les Sarile]

In my limited experience (slides) the colors are more saturated with coolscan, possibly because the RGB LED's (Coolscan) have narrower spectra than the response of the EPSON RGB matrix sensor, and presumably less sensitive to dye cross-talk.

Be interested to see examples.

I've tried Vuescan but the Coolscan+Nikonscan combination is simply unmatched IMO and collection of results.

 

[Anon Ymous]

Another lens resolution target used to evaluate a scanner sensor. Misleading.

I don't think this is valid criticism. For starters, the USAF 1951 test chart isn't specifically a lens resolution target, but can be used, among others, to test scanner resolution. They use a glass slide target, which has elements of known resolution. Film curvature isn't an issue in this case and it's a simple assessment of what is easily distinguishable in 100% zoom. I don't understand what's wrong with this method. From comparisons I've seen on-line, these flatbeds have seriously overrated specs. A scan of the same film with an Epson and a dedicated film scanner shows a night and day difference.