My film scanner died!

I've seen a lot of scans from the Plusteks. They just plain don't have the real resolution that the Nikon has. They won't work for me. The problem is that they have fixed focus lenses, and the lenses just aren't that good in them. Epson flatbeds have the same issues, but worse.


The Plusteks have a claimed resolution of 7200 dpi. That's because they use a 7200 pixel sensor, but the real resolution is only about 3200 dpi according to every test I have seen from them. Its the lack of focusing and the low quality lens that causes that resolution loss.



The lens I shot my tests with costs more than a Plustek scanner!

Here's a 3600dpi full res scan with only minimal export sharpening - 3600dpi scan link

Thoughts on that 3600dpi scan mate?
I've always been satisfied with the output, but then I've never owned a Coolscan so I probably don't have the best comparison point.
 
The Nikon ES-2 works pretty well for camera scanning. I purchased some 62mm tubes to make it work with my 90mm macro lens (it's designed for a 50mm macro). The tubes were just a few bucks from camerafilters.com - I can get the full list of parts I used to make the ES-2 work with my Sony 90mm macro.

Also, if you are feeling adventurous, the lens in your scanner makes a GREAT lens for scanning 35mm film - who would have thought, right? You will have to rig it up in some type of mount, but it is supposedly one of the best lenses out there for 1:1 macro/scanning.

https://www.closeuphotography.com/scanner-nikkor-ed-lens
 
Thoughts on that 3600dpi scan mate?
I've always been satisfied with the output, but then I've never owned a Coolscan so I probably don't have the best comparison point.




Its not bad at all, but the fine detail resolution isn't quite as good as my Nikon. I've got a copystand on the way. If I ultimately find camera scanning to be greatly inferior to the Nikon, I may buy one of the Plusteks to try.
 
The Nikon ES-2 works pretty well for camera scanning. I purchased some 62mm tubes to make it work with my 90mm macro lens (it's designed for a 50mm macro). The tubes were just a few bucks from camerafilters.com - I can get the full list of parts I used to make the ES-2 work with my Sony 90mm macro.

Also, if you are feeling adventurous, the lens in your scanner makes a GREAT lens for scanning 35mm film - who would have thought, right? You will have to rig it up in some type of mount, but it is supposedly one of the best lenses out there for 1:1 macro/scanning.

https://www.closeuphotography.com/scanner-nikkor-ed-lens




The Nikon ES-2 looks nice but I don't have a Nikon camera, and I don't think it would work well with the 60mm Macro I have for my Olympus system. I just bought a nice copystand, it'll be here Wednesday so I'll be able to try 'scanning' my film using it with my Olympus gear.
 
The Nikon ES-2 looks nice but I don't have a Nikon camera, and I don't think it would work well with the 60mm Macro I have for my Olympus system. I just bought a nice copystand, it'll be here Wednesday so I'll be able to try 'scanning' my film using it with my Olympus gear.

Forgot to write originally: Definitely stinks about your scanner :(

But yeah, I'm using the Nikon part on a non-Nikon camera with a non-Nikon lens that isn't even the correct focal length. With some adapters, it can work quite well. The copy stand sounds like a good solution as well, but for others that are reading this, the ES-2 is a good solution if one wants one.
 
Chris I am thinking along the same lines (i.e. using a digital camera to scan) as I have a significant number of slides and negatives that have never been scanned and I think it is time. I have several suitable macro lenses that no doubt would work (just a matter of choosing which one) and I am considering using my Sony NEX 7 because of its 24 megapixel sensor which I felt might help maximize the resolution available to me. This camera also has the advantage of having a movable rear screen so I can view it from above which is ideal for this application.

In terms of light source I recently changed over 25 old style quartz halogen lights in my home's ceiling to the new style LED lights as the transformers on some of the old lights had begun to fail due to age. It is very easy these days - the LED lights for this purpose can be bought cheaply (mine cost about $8 AUD each - maybe $5 US). They come ready made with a standard wall plug (as did the old lights I found once I got up into the ceiling) so swapping them involved no rewiring - making it ideal for a slide copier too. I had a couple spare and it occurred to me that one would be perfect for the scanning project. There is no (or minimal) heat and best of all the lights each have a switch to allow you to change the color from warm to cool / neutral to another setting perhaps a bit blueish. Set on neutral should work fine. That makes using one of these an ultra cheap way of producing lighting though I may have to fritz around a little with diffusing the light to get the best light dispersion and brightness. As these lights have built in diffusers I suspect that irregular brightness should not be much of an issue though.

It occurred to me that I could easily make a wooden base from a short plank, maybe 600mm long with the light at one end, a slide holder in the middle somewhere (exact position to be determined depending on the lens I use and how bright the light is) and a camera tripod quick mount attachment at the other end to mount and hold the camera that does the actual work.

To mount the LED light scavenged from my ceiling light project I would cut a circular hole (90mm in my case) in a piece of medium density fibre board (MDF) to match the diameter of the light fitting. The MDF would be mounted vertically to the base. The light would fit easily into the hole and be held in place pointing at the slide holder by the two integrated spring clamps (already fitted into these lights to hold them into the ceiling). The slide holder could probably also be fabricated by cutting a rectangular hole of the correct size (36mmx24mm) in another piece of MDF. A short rail along the bottom of the hole made from another piece of wood or maybe extruded aluminium or something of that sort, should work to complete the slide holder. If I am clever enough I think I should be able to make the slide holder long enough to hold several slides to speed up the digitization process such the the holder works kind of like a magazine. BTW the vertical MDF pieces used to mount the light and the slide holder could each be mounted by routing a channel in the base if you want to be fancy and have a router or simply can be affixed to a small block of wood which in turn is screwed / glued to the wooden base. The important thing is that the MDF pieces are absolutely vertical and absolutely parallel to each other and to the plane of the cameras sensor when it is mounted.

If needed to prevent extraneous light confusing the camera's meter or producing flare it would also be a simple task to mount a tube (eg made from plastic rainwater pipe also of 90mm - a standard size) from the light to the back of the slide holder. This could be held in place by a couple more short pieces of wood attached to the wooden base.

All I need then is a remote trigger for my camera or failing that I could use a 2 second shutter delay setting to avoid shake.

You can see that I have the plan worked out in my head and I will start as soon as I finish another project I am presently embroiled in. A few adjustments may be required as I undertake the construction and work details out in practice. As I already have all the parts needed except for maybe some wood for the base I think I can do this for no more than a few pennies. Even if I needed to buy everything (excepting the camera) I could not imagine anything being cheaper. And so long as the bits that need to be vertical are vertical everything should align nicely for maximum resolution.

I am simply not inclined to spend a few hundred dollars on a scanner that I will most likely only use to scan maybe a couple of hundred photos if that.
 
Thoughts on that 3600dpi scan mate?
I've always been satisfied with the output, but then I've never owned a Coolscan so I probably don't have the best comparison point.

Try a good scanner. You'll likely see the difference. Epson has "real" resolution of around 2000 DPI at best, when optimally setup.

You can see this for yourself. Take your image file, and resize it to 50% (1800 DPI). Then size it back up to 3600. Look closely while flipping between the original 3600 DPI and re-resized 3600 DPI version and you'll see almost no difference. The reason is because there's hardly any information lost when resized to 1800 DPI, because it doesn't exist. If it had actual 3600 DPI it would be very obviously lower resolution when sized back up. Make sure to use the appropriate resizing algorithms in Photoshop.
 
Try a good scanner. You'll likely see the difference. Epson has "real" resolution of around 2000 DPI at best, when optimally setup.

You can see this for yourself. Take your image file, and resize it to 50% (1800 DPI). Then size it back up to 3600. Look closely while flipping between the original 3600 DPI and re-resized 3600 DPI version and you'll see almost no difference. The reason is because there's hardly any information lost when resized to 1800 DPI, because it doesn't exist. If it had actual 3600 DPI it would be very obviously lower resolution when sized back up. Make sure to use the appropriate resizing algorithms in Photoshop.




gavinlg used a Plustek scanner, not an Epson for the 3600 DPI sample he uploaded.
 
Guess I mistook that. Well, the test I mentioned still shows the limitation in actual usable resolution. I see only the slightest difference. Of course there's plenty of other factors - film, developer, aperture, editing, etc. etc.
 
I hope Olympus 60/2.8 macro is much better that the 30/3.5. I tried 30/3.5 macro with E-M5 mkII (16MP) and compared to a real scanner the corners are horrible (Epson flatbed horrible). I bought that combo for cheap to have something to play with during the corona lockdown and after reading through some 60/2.8 macro reviews and tests I wasn't convinced that the 60/2.8 macro would be much better in the corners than 30/3.5.

In the center pixel-shift does seem to get pretty much all the detail that a good (4000+ dpi) scanner can pull from a negative, but I don't like the "texture" of the pixel-shift files. Hard to describe what it is and my light source (I use colour enlarger head for light source) or raw processor might have something with it...
 
Its not bad at all, but the fine detail resolution isn't quite as good as my Nikon. I've got a copystand on the way. If I ultimately find camera scanning to be greatly inferior to the Nikon, I may buy one of the Plusteks to try.

Fair enough, I'd love to have a crack at some sharp negs with a Coolscan sometime and see the difference. You don't have any full res scans lying around do you?

Does your Olympus have the pixel shift mode? Might be worth experimenting with. I'm interested in how it goes for you.

Try a good scanner. You'll likely see the difference. Epson has "real" resolution of around 2000 DPI at best, when optimally setup.

You can see this for yourself. Take your image file, and resize it to 50% (1800 DPI). Then size it back up to 3600. Look closely while flipping between the original 3600 DPI and re-resized 3600 DPI version and you'll see almost no difference. The reason is because there's hardly any information lost when resized to 1800 DPI, because it doesn't exist. If it had actual 3600 DPI it would be very obviously lower resolution when sized back up. Make sure to use the appropriate resizing algorithms in Photoshop.

Erhmm..

I actually tried this and regardless of resizing algorithm used there's significantly more detail with the original higher res file. It makes the difference between being able to read the numberplate of some of the cars and not being able to.

Like I said, my testing indicates around 4200dpi is where the resolution tops out on the plustek - in fact out of the scanners I've owned (epson v550, v600, Minolta dual scan IV x2, Plustek 8100 x2) the Plustek has the highest resolution for sure. The epson's are frankly garbage. I have one for my 120 scanning, I hate using it and honestly the plustek gets more resolution out of a 35mm file than the epson does out of a 6x7 negative.

FYI - when I got my second plustek a few years ago I sent some negatives off around the world to Richard Photo Lab in CA to have maximum resolution scans done on their Noritsu, and compared them with the same negs (which they sent back to me) scanned on the plustek. The Plustek resolved more detail in my testing.
 
I hope Olympus 60/2.8 macro is much better that the 30/3.5. I tried 30/3.5 macro with E-M5 mkII (16MP) and compared to a real scanner the corners are horrible (Epson flatbed horrible). I bought that combo for cheap to have something to play with during the corona lockdown and after reading through some 60/2.8 macro reviews and tests I wasn't convinced that the 60/2.8 macro would be much better in the corners than 30/3.5.

In the center pixel-shift does seem to get pretty much all the detail that a good (4000+ dpi) scanner can pull from a negative, but I don't like the "texture" of the pixel-shift files. Hard to describe what it is and my light source (I use colour enlarger head for light source) or raw processor might have something with it...




My test image is sharp all the way to the corners.


Something to consider is that I have found that there is a LOT of sample variation in Olympus's micro four thirds lenses. I don't even bother buying them online because I have had to return so many that were decentered (sharp on one side of the frame, soft on the other).


I buy them in person at a store that lets me test the lenses before I buy. I have tried as many as six copies of a lens before finding one that was good. I have never gotten a good one the first time. Some are so bad that the image looks like it was shot with a holga on the unsharp side.


Once you find a good example, the Olympus lenses are incredible. The quality control just sucks, though. It isn't just Olympus, either. I have a Panasonic 45-150mm lens that was the 5th copy I tried. I really wanted the Panasonic Leica 8-18mm lens, but after trying all eight of them they had in stock and finding that ALL were severely decentered, I gave up and tried the Olympus 7-14mm. Got a good one the second try on that lens.


Never had issues like this when I was shooting Canon or Nikon in the past. The old Olympus OM-System lenses for 35mm film were all great, too. No QC issues. I think with everything being plastic junk now, no one cares about quality control. They just hope we won't notice if the lens is not great. Many people don't. I do because I regularly sell large prints and they simply cannot be soft on one side, not for the prices I charge my patrons for them!
 
The Epsons are horrid. I need scans that can produce large exhibition quality prints; selling my prints is my main source of income, so I can't compromise on quality.


Chris, so you need the best possible quality. But only for your best selected images. Those who are worth to be presented for sale.
Then I would recommend the following workflow:
Cooperate with a drum scan service. Drum scanners offer by far the best quality of scans. Not only concerning resolution, but especially also concerning recording details in the highlights and shadows (dynamic range), and the recording of the film grain (most natural and no grain enhancement by scanner noise like with all other scanner types).
Drum scanners are much closer to the real film look and film qualities then all other scanners. I am going this way for years now for all the shots were optical enlargement is not an option (like prints from colour slides). Never regretted it.
The only other way which is surpassing drumscanners in quality are real optical prints made with APO enlarging lenses. They offer the best detail rendition, surpassing even drum scanners in resolution by a big margin.

As you are in the US, you may contact Martin Strickland for drum scans. He has a very good reputation.
 
My test image is sharp all the way to the corners.


Something to consider is that I have found that there is a LOT of sample variation in Olympus's micro four thirds lenses. I don't even bother buying them online because I have had to return so many that were decentered (sharp on one side of the frame, soft on the other).


I buy them in person at a store that lets me test the lenses before I buy. I have tried as many as six copies of a lens before finding one that was good. I have never gotten a good one the first time. Some are so bad that the image looks like it was shot with a holga on the unsharp side.


Once you find a good example, the Olympus lenses are incredible. The quality control just sucks, though. It isn't just Olympus, either. I have a Panasonic 45-150mm lens that was the 5th copy I tried. I really wanted the Panasonic Leica 8-18mm lens, but after trying all eight of them they had in stock and finding that ALL were severely decentered, I gave up and tried the Olympus 7-14mm. Got a good one the second try on that lens.


Never had issues like this when I was shooting Canon or Nikon in the past. The old Olympus OM-System lenses for 35mm film were all great, too. No QC issues. I think with everything being plastic junk now, no one cares about quality control. They just hope we won't notice if the lens is not great. Many people don't. I do because I regularly sell large prints and they simply cannot be soft on one side, not for the prices I charge my patrons for them!

Chris, so what’s your method for testing lenses in the field, say in a camera store?
 
My test image is sharp all the way to the corners.


Something to consider is that I have found that there is a LOT of sample variation in Olympus's micro four thirds lenses. I don't even bother buying them online because I have had to return so many that were decentered (sharp on one side of the frame, soft on the other).


I buy them in person at a store that lets me test the lenses before I buy. I have tried as many as six copies of a lens before finding one that was good. I have never gotten a good one the first time. Some are so bad that the image looks like it was shot with a holga on the unsharp side.


Once you find a good example, the Olympus lenses are incredible. The quality control just sucks, though. It isn't just Olympus, either. I have a Panasonic 45-150mm lens that was the 5th copy I tried. I really wanted the Panasonic Leica 8-18mm lens, but after trying all eight of them they had in stock and finding that ALL were severely decentered, I gave up and tried the Olympus 7-14mm. Got a good one the second try on that lens.


Never had issues like this when I was shooting Canon or Nikon in the past. The old Olympus OM-System lenses for 35mm film were all great, too. No QC issues. I think with everything being plastic junk now, no one cares about quality control. They just hope we won't notice if the lens is not great. Many people don't. I do because I regularly sell large prints and they simply cannot be soft on one side, not for the prices I charge my patrons for them!


Never heard of this issue, never had it on any of my Olympus/Panasonic lenses. Are you sure it's not your camera's bayonet/sensor that is misaligned ?
 
Never heard of this issue, never had it on any of my Olympus/Panasonic lenses. Are you sure it's not your camera's bayonet/sensor that is misaligned ?




It showed the same on both my Pen-F and my OM-D E-M1 mark II. Also, it was not always the same side of the image that was unsharp. One sample would be soft on the right side, another unsharp on the left, and another unsharp on the bottom. If the camera was misaligned, the same side of the image would be unsharp with every lens, and no lens would be good on that camera.
 
You do not need a copy stand. You do need to make a fixture to hold the camera in normal way for landscape photos and then a vertical wall to mount something to hold the film like neg carrier for enlarger.

I use a Lime cube which is 5000K and bright enough for short exposures 1/100 sec F8 With D800 iso100. use a diffuser between light and film.

I use Nikons ES2, but requires a really good fixture to work well. Properly set up is is as good as my Minolta 5400 which does not work with current computers.

If you do color, WB on a blank frame and invert to get proper colors.
 
Erhmm..

I actually tried this and regardless of resizing algorithm used there's significantly more detail with the original higher res file.

You're right, there's a bit more, but I wouldn't say significant. That said, I tried it again at 2400 DPI and there was absolutely no difference. Indistinguishable. Actually the resized one might even be a tad sharper perceptually because it sharpened up some of the granularity. So I would say the scan has roughly 2400 DPI of resolution, and the extra stuff at 3600 is not doing anything for you.

If you see improvement up to a higher scan resolution, that's because likely you have to scan at a higher rez to get max resolution, when downsized. In other words, scan this again at 2400 and it might have less detail than the 3600 one downrezzed to 2400. This is fairly common - I think the Epsons are supposedly best at 4800 downrezzed to 2400.

I agree with you on the Epson regarding 120. I've always said that a decent scanner with 35mm film will give better results than an Epson with MF. A lot of folks don't like to hear it though.
 
Lens sample variation is real: thirty plus years ago, when we were picking lenses for ground truth photography at NASA, we'd get Nikon to send us a dozen lenses and typically found two-three of them matched the required specification, based on the published specs for these lenses. (For scientific data, the requirements are MUCH more stringent than for general purpose photographic use ... Typically, 11 out of 12 would be just fine for normal use.)

I've been scanning film since 1984, for work, and then for personal/exhibition uses since 1995. I have one dedicated film scanner left—Nikon Coolscan V ED— and will not replace it when it dies. A high resolution copy camera approach is the right way to go now.

For production work, don't even consider the notion of doing this without a copystand or other dedicated camera stand setup. It's a waste of time. You need a GOOD copystand, a GOOD light box, and a GOOD fixture to hold your film flat and registered accurately to do high quality captures. I use a Novoflex Magic Stand copystand setup, with a Novoflex focusing rail. For print captures, I have Color-balanced LED lights that I set up with light stands and use a Sekonic L358 meter to balance the lighting evenly and eliminate shadows. For negative captures, I've been using a home-made film carrier/guide until now. I just received the Negative Supply 120 Carrier device and ProMount ... expensive, but superbly made and extremely robust, precise negative holders, essentially. When the Nikon dies finally, I'll buy their 135 Carrier too.

For a copy camera, a good quality FourThirds or APS-C format camera with 16 Mpixels or greater is fine for most of what I do with Minox through 6x6 format negatives. I've used my Olympus E-M1 quite a lot, both with their ZD 35mm f/3.5 Macro lens (with and without the ED-14 teleconverter) and with my favorite old 1969 Micro-Nikkor 55mm f/3.5 lens. Both return very high quality results, although the square format captures get down to about 9 Mpixel with the E-M1.

Switching to the Leica CL (24 Mpixel camera) nets a newer sensor with more dynamic range and a 16 Mpixel capture of 6x6 format negs. That's a healthy improvement. I use it with the superb Macro-Elmarit-R 60mm or Summicron-R 50mm lenses, both with superb results. (Turns out the Summicron-R 50 actually outperforms the ME60 when I go to greater than 1:1 magnification, for example, when capturing Minox negatives at 1.7:1.)

Looking at the larger format negatives primarily, I've now set up the Hasselblad 907x/CFVII 50c with the V-system Makro-Planar 120mm or Planar 80mm lenses and tested copy work with them. The results are just phenomenal, better than I need: for square format capture, resolution is up to 39 Mpixels and detailing is just plain mind blowing, plus the net when shooting raw with this is 16 bits-per-pixel depth ... That's about 14 stops tonal range effectively (the best I can get out of the Leica is about 12-13 stops). It's a super high end setup, but eh? I have it, and I might as well use it for as many things as I can dream up to get my value out of it. :D

Good luck with your endeavors, Chris!

G
 
Interestingly, I kind of traveled down this road less than a year ago. I have a Plustek 7600i, and wanted more resolution for larger prints (on the rare occasion to go big). So, I invested some time and money in a solution that let me scan 35mm film with my A7RII. I don't have a real macro lens, I used a close-focus modern 35mm lens (Tamron 35/1.8...super sharp).

I found that although I had more resolution from the A7RII, the Plustek was just as sharp, easier to set up and use (although limited to about 16-18MP), but the Sony mirrorless was much faster for multiple scans.

So I now chose my workflow...old Epson flatbed for contact sheet scan (at max resolution, just for seeing on a large screen), then scanning selects at 3200-3600dpi on the Plustek.

I'm not sure it'll get gallery-sized prints, but it's plenty for 20" prints.

Chris, I'd say the BEST way to get what you're after is a Hassleblad scanner like a Flextight X1 or X5. Expensive, yes, but no better solution short of a drum scan, and you can deduct it on your taxes.
 
Back
Top