cabbiinc
Slightly Irregular
I'd start with a very basic thing and make sure that the glass in your flatbed is clean. Turn the scanner on with the reflective material light showing. Do a preview scan and watch the light pass under the glass. If you see a lot of fog or other dirt it's time for cleaning.
If you have multi-pass turned on then turn that off.
If you have IR dust removal turned on try it turned off.
If you have multi-pass turned on then turn that off.
If you have IR dust removal turned on try it turned off.
venchka
Veteran
1. As pointed out above, place the film in the holders correctly.
2. Clean the scanner glass. Both sides if you can get the glass out without breaking the scanner.
3. Epson V700 with EpsonScan software will produce better results than mere mortals need. 12x18 (35mm original) and 16x20 prints (medium format & 4x5) are very possible.
4. Learn this:
http://www.kenleegallery.com/html/tech/scanning.php
5. Scan at 2400 dpi MAXIMUM. Output size = original size. Example: Scan 35mm film at 2400 DPI with the output size at 24mm x 35mm. Set EpsonScan to match film type being scanned: 48 bit for color, 16 bit for B&W. Save to a TIFF file.
6. Attached samples were scanned with a circa 2000 Epson Expression 1680. Several models older than the current V700. Click each sample for large version.
7. Practice. Practice. Practice.
Good luck.
Wayne
2. Clean the scanner glass. Both sides if you can get the glass out without breaking the scanner.
3. Epson V700 with EpsonScan software will produce better results than mere mortals need. 12x18 (35mm original) and 16x20 prints (medium format & 4x5) are very possible.
4. Learn this:
http://www.kenleegallery.com/html/tech/scanning.php
5. Scan at 2400 dpi MAXIMUM. Output size = original size. Example: Scan 35mm film at 2400 DPI with the output size at 24mm x 35mm. Set EpsonScan to match film type being scanned: 48 bit for color, 16 bit for B&W. Save to a TIFF file.
6. Attached samples were scanned with a circa 2000 Epson Expression 1680. Several models older than the current V700. Click each sample for large version.
7. Practice. Practice. Practice.
Good luck.
Wayne
HCEarwicker
Member
I'm thinking that could account for the lack of sharpness I'm getting in my scans? When I first got my old 3170 a couple of weeks ago I did notice a fogginess on the underside of the glass...
In the unlikely event that anyone's taken notice of my "DO use USM to scan negatives" advice: forget it - at least until I've cleaned the glass
Thanks, cabbiinc.
cz23
-
Help a scanning novice out with this, if you would, please.
According to filmscanner.info the actual maximum resolution of the V700 is 2300 dpi.
And according to Epson's web site, the formula for resolution needed for printing is:
(short side of output / short side of original) X 300 (or 360)
At 300 ppi, a 12 x 18" print from a 35mm negative (0.95") would therefore require about 3800 dpi. Even if you printed at 240 ppi, you would still over 3000 dpi.
I'm not challenging you, but rather trying to think this through and learn.
John
Swift1
Veteran
I looked through your scans and would just like to point out that some of what your seeing is happening before scanning.
Many of the images have motion blur or the focus is slightly off. No scanner will solve this.
From my experience, Fuji Superia 200 can be capable of excellent results, but only if exposure is very accurate. The film can handle over/under exposure fairly well, but I've noticed a rapid decline in overall image quality with over/under exposure. I've found that scanners dislike dense negatives.
Try exposing 1/3 stop under.
Also, and this is somewhat related to exposure, while the film probably has good dynamic range, I've found that my V500 has a difficult time with color negative film that has a very wide rang of exposure, so lighting conditions can play a roll.
I don't know much about the V370, but I've been using a V500 for over 4 years and it is capable of quite decent results. As others have suggested, scanning has a bit of a learning curve, so it may take some time and practice to get best results.
jammcat
Lick My Lens Cap
Photo_Smith
Well-known
No!
The scanners sharpness is limited by it's native resolution and how flat the film is. Sharpening in the Epson module is an uncontrolled guess with all the artefacts and posterisation etc built in to the resulting file.
What you need is a flat dull looking scan you import into Photoshop like this:
http://photo-utopia.blogspot.co.uk/2010/11/scanning-with-epson-v500.html
From there you can work in photoshop to apply radius based USM to get the tonality you require and then do a normal workflow.
The Job of the Epson software is to get the information from the negative without changing it, the resulting scan should look dull, flat and unsharp.
Put the brilliance in in the post processing:
V500, Kodak 160VC
Chestnut leaves at dusk by Photo Utopia, on Flickr
HCEarwicker
Member
Thanks for the reply and for the link, Photo Smith. I understand the principle, really I do. But in my very limited experience, post-scan USM has been looking more artificial than pre-scan USM - I've conducted graduated compare and contrast experiments with both. Perhaps that's becasue I'm using Paint.net and/or Flickr's aviary rather than PS (which I don't have and can't afford)?* Whatever, I'll clean my slightly fogged glass, make sure I flatten my negatives as much as poss, and try again - the right way.
* maybe GIMP's USM tool is better than Paint.net's?
BTW - I'm reassured to see that I've discovered all your other recommendations work best for me too.
* maybe GIMP's USM tool is better than Paint.net's?
BTW - I'm reassured to see that I've discovered all your other recommendations work best for me too.
Photo_Smith
Well-known
Well, if it works for you I guess that is hard to argue with. Pre scan USM is destructive uncontrolled where post USM is up to your skill, very fine adjustment.
Have you tried pixlr the web based app?
http://pixlr.com/editor/
The USM has the same adjustments as PS; amount, radius and threshold. Where a high radius will control tonal values when couples with a low amount say 25, radius 10 gives a contrast sharpen-do this first in the workflow, low radius higher amount say 50, 2, will be the last sharpen after all other edits.
Try Pixlr not as fluid as real software but quite powerful and the price is right.
Have you tried pixlr the web based app?
http://pixlr.com/editor/
The USM has the same adjustments as PS; amount, radius and threshold. Where a high radius will control tonal values when couples with a low amount say 25, radius 10 gives a contrast sharpen-do this first in the workflow, low radius higher amount say 50, 2, will be the last sharpen after all other edits.
Try Pixlr not as fluid as real software but quite powerful and the price is right.
Athiril
Established
There's one technique you can use if you're desperate for more detail out of flatbed scans
http://www.rangefinderforum.com/forums/showthread.php?t=130731
http://www.rangefinderforum.com/forums/showthread.php?t=130731
Ranchu
Veteran
Ranchu
Veteran
This has been my experience as well. I also find post scan sharpening somewhat weird looking more often than not. My sharpness needs are few, if I can see the grain, it's OK.
http://www.rangefinderforum.com/forums/showpost.php?p=1984519&postcount=49
maddoc
... likes film again.
I would have get one film (with good exposures, reasonable high dynamic range, and some frames with fine details) scanned by a good scanning service to have a reference and then start getting the film-holder of your Epson into focus.
For this you would use thin scotch-tape to increase the height of the plastic film-holder in small increments (layer-by-layer) until you achieve a degree of resolution as close as possible to the professionally scanned film. For this you would switch of any sharpening inside the Epson software.
Most important, the film has to be 100% flat, scanning a film with the slightest tendency to curl in either direction will lead to out-of-focus scans.
For this you would use thin scotch-tape to increase the height of the plastic film-holder in small increments (layer-by-layer) until you achieve a degree of resolution as close as possible to the professionally scanned film. For this you would switch of any sharpening inside the Epson software.
Most important, the film has to be 100% flat, scanning a film with the slightest tendency to curl in either direction will lead to out-of-focus scans.
kuzano
Veteran
Flat bed scanning just a PITA
Flat bed scanning just a PITA
That's my conclusion after a fair try. I tried with an Epson 4990. Then, I actually purchased and sold three Epson V500's. I used Epson Scan, VueScan, and Silverfast.
I gave each scanner and each software package a fair trial. I will say the EpsonScan software gave me the best result in all cases. I concluded trying to post process in the software was wasted time. Get the best scan possible and work in editing software.
It's not like I am a complete idiot on the computer end. I have been consulting and repairs on PC's, Windows, and networking for 25 years and selling banking software ten years prior to that.
I came to the conclusion that flat bed scanning just doesn't cut it. Or, perhaps my expectations were just to high. Likely the latter.
I do agree with the poster who said that resolution for scanners is HIGHLY overstated under the best circumstance, and am frustrated by that obvious lie in marketing the hardware.
I know a fellow who has a Nikon 9000, and when I want good scans I usually pay him to do them for me. I also send them out to a lab in Portland (140 miles away).
I sold all the scanners and got rid of my desire to home scan my images.
Funny, I actually expected my scans to look as good as my negatives and transparencies. Never again. I let people who know what they are doing with a scanner do it, and pay them for the few I really need to scan.
That's my plan. No more time spent wasted on scanning when there are professional services who can do it so well. It's not that expensive if you don't want every image you shoot, or in your archive, scanned
Flat bed scanning just a PITA
That's my conclusion after a fair try. I tried with an Epson 4990. Then, I actually purchased and sold three Epson V500's. I used Epson Scan, VueScan, and Silverfast.
I gave each scanner and each software package a fair trial. I will say the EpsonScan software gave me the best result in all cases. I concluded trying to post process in the software was wasted time. Get the best scan possible and work in editing software.
It's not like I am a complete idiot on the computer end. I have been consulting and repairs on PC's, Windows, and networking for 25 years and selling banking software ten years prior to that.
I came to the conclusion that flat bed scanning just doesn't cut it. Or, perhaps my expectations were just to high. Likely the latter.
I do agree with the poster who said that resolution for scanners is HIGHLY overstated under the best circumstance, and am frustrated by that obvious lie in marketing the hardware.
I know a fellow who has a Nikon 9000, and when I want good scans I usually pay him to do them for me. I also send them out to a lab in Portland (140 miles away).
I sold all the scanners and got rid of my desire to home scan my images.
Funny, I actually expected my scans to look as good as my negatives and transparencies. Never again. I let people who know what they are doing with a scanner do it, and pay them for the few I really need to scan.
That's my plan. No more time spent wasted on scanning when there are professional services who can do it so well. It's not that expensive if you don't want every image you shoot, or in your archive, scanned
Ranchu
Veteran
Not sure I've ever seen a how to scan thread here where you haven't said that, kuzano. People are asking for help scanning. Got any tips?
HCEarwicker
Member
Thanks for the links and tips, Ranchu, everybody. (Dear OP - my apologies if I've hijacked the thread. Hopefully this stuff is useful to you, too).
It's been said that pre-scan USM is destructive, but is it any more destructive than any other aspect of recombobulating film to digital? Sure, something is being 'done' to the original analogue negative, but the same goes for leaving USM off. Only difference I can see is that USM ON appears to render more detail; USM off doesn't. What makes the un-USM'd digital reading more 'true'? True to what?
What is true that I have little control over how pre-scan USM chooses to sharpen the negative (I have some: low/medium/high), but if I don't like what it does I can turn it off.
...Just trying to get my head round this stuff. I reserve the right to backtrack on everything I've said if/when I discover I've been doing something wrong, or my equipment's buggered.
PS - Currently, I'm only scanning a bunch of 40-year old B&W 35mm negatives - the most recently-uploaded 30 or so of these. Things may change when I get the first roll of colour film taken on my new (old) Oly RC developed:
http://www.flickr.com/photos/105343797@N08/
(the scans uploaded before the panda on the beach shot were all done by a friend on an Epson v750 at 900dpi, with slight editing by me, mainly in Flickr/aviary. Those negs are also 40 years old - check the hair and outfits! I'll have to ask him if he used Epson pre-scan USM. I've got a feeling he did - and auto-exposure. He got through 400+ negs in a few days.)
It's been said that pre-scan USM is destructive, but is it any more destructive than any other aspect of recombobulating film to digital? Sure, something is being 'done' to the original analogue negative, but the same goes for leaving USM off. Only difference I can see is that USM ON appears to render more detail; USM off doesn't. What makes the un-USM'd digital reading more 'true'? True to what?
What is true that I have little control over how pre-scan USM chooses to sharpen the negative (I have some: low/medium/high), but if I don't like what it does I can turn it off.
...Just trying to get my head round this stuff. I reserve the right to backtrack on everything I've said if/when I discover I've been doing something wrong, or my equipment's buggered.
PS - Currently, I'm only scanning a bunch of 40-year old B&W 35mm negatives - the most recently-uploaded 30 or so of these. Things may change when I get the first roll of colour film taken on my new (old) Oly RC developed:
http://www.flickr.com/photos/105343797@N08/
(the scans uploaded before the panda on the beach shot were all done by a friend on an Epson v750 at 900dpi, with slight editing by me, mainly in Flickr/aviary. Those negs are also 40 years old - check the hair and outfits! I'll have to ask him if he used Epson pre-scan USM. I've got a feeling he did - and auto-exposure. He got through 400+ negs in a few days.)
Photo_Smith
Well-known
The reason USM is destructive is you have little control, so the best thing is to turn it off and use a method where you have complete control.
The artefacts produced by a poor scanning methodology are passed on down the chain so it you create halo's, noise, aliasing (false grain) or other issues you're stuck with them.
The correct method is to supply your image 'digital negative' with as much information from each RGB channel as possible, no sharpening auto exposure etc.
The putting on USM doesn't give you more detail in your scan it just gives you more noise.
To answer why switching of USM would that be a truer rendition is simple; less noise you add to the original signal the better.
True to what? The original negative/slide, print conventionally to understand what you are being true to.
I was told by a scan expert that most flatbeds scanning 35mm can't resolve grain on medium speed films, so if you see grain you have to ask yourself 'what is that' its noise called aliasing.
It should be easy to get natural looking scans if you give your photo editor the right basic file, of course digitising film changes it, but with the right workflow it should look very natural-and sharp!
Barney by Photo Utopia, on Flickr
But hey, if you use USM in the scan module and are happy who am I to argue?
The artefacts produced by a poor scanning methodology are passed on down the chain so it you create halo's, noise, aliasing (false grain) or other issues you're stuck with them.
The correct method is to supply your image 'digital negative' with as much information from each RGB channel as possible, no sharpening auto exposure etc.
The putting on USM doesn't give you more detail in your scan it just gives you more noise.
To answer why switching of USM would that be a truer rendition is simple; less noise you add to the original signal the better.
True to what? The original negative/slide, print conventionally to understand what you are being true to.
I was told by a scan expert that most flatbeds scanning 35mm can't resolve grain on medium speed films, so if you see grain you have to ask yourself 'what is that' its noise called aliasing.
It should be easy to get natural looking scans if you give your photo editor the right basic file, of course digitising film changes it, but with the right workflow it should look very natural-and sharp!
Barney by Photo Utopia, on Flickr
But hey, if you use USM in the scan module and are happy who am I to argue?
HCEarwicker
Member
Thanks, PS.
Putting the control issue to one side - does post-scan sharpening also add noise (of one sort or another) to achieve the illusion of detail?
Edit - I can see you're getting better results than I am, so maybe I should just hush my mouth. ...Before I do - what scanner do you use? What did you use for 'Barney' and 'Chestnut Leaves'?
Edit #2 - I used to have my own darkroom, so I have printed conventionally. When I compare my scans to my original prints, I can see very clearly what I'm not getting: sharpness and tonal depth/richness. So far, I've accepted that as the price paid for digital conversion with an inexpensive scanner.
Another question: If, by not using pre-scan USM, I'm truly recording what's on the negative, how come I only get to see what appears to my eyes to be much more like what's truly on the negative - ie, detail - once I've sharpened (whether pre- or post-scan)? How come 'what's truly on the negative' looks so soft to the scanner? Where and why is the detail hiding?
Putting the control issue to one side - does post-scan sharpening also add noise (of one sort or another) to achieve the illusion of detail?
Edit - I can see you're getting better results than I am, so maybe I should just hush my mouth. ...Before I do - what scanner do you use? What did you use for 'Barney' and 'Chestnut Leaves'?
Edit #2 - I used to have my own darkroom, so I have printed conventionally. When I compare my scans to my original prints, I can see very clearly what I'm not getting: sharpness and tonal depth/richness. So far, I've accepted that as the price paid for digital conversion with an inexpensive scanner.
Another question: If, by not using pre-scan USM, I'm truly recording what's on the negative, how come I only get to see what appears to my eyes to be much more like what's truly on the negative - ie, detail - once I've sharpened (whether pre- or post-scan)? How come 'what's truly on the negative' looks so soft to the scanner? Where and why is the detail hiding?
Photo_Smith
Well-known
Most likely a case of PEBCAK being good with computers won't give you a free pass at judging colour, or having a good workflow.
I have no idea of how high your expectations are but for web based images and prints up to 12" wide a flatbed and medium format film is a perfectly capable device.
English Apples by Photo Utopia, on Flickr
Reasonably natural, good enough for web sharing and small prints.
Max by Photo Utopia, on Flickr
mdarnton
Well-known
I pulled one of your Flicker pix and gave it some sharpening, and that really helped. Sure, there's a bit more noise, but when a photo is reduced to a real size for printing or screen use, sharpness trumps noise.
I've been messing with my old 3170 scanner for film recently because I'm going to start stitching some 4x5 and I've learned a couple of things.
First, it's worth messing with finding the right focus, which can be as much as 2.5mm or as little as .75mm over the glass and still be in spec for Epson. Second, for me, scanning between glass added quite a bit. Including Newton rings, so I was considering getting some anti-newton glass, because I thought it would be worth the difference from no glass at all. Then I discovered wet scanning. It's a little more bother, but there's another nice bump in performance to be had there, again.
Summary: if you're scanning using Epson's neg carrier, there's still a lot of room to make things better.
I've been messing with my old 3170 scanner for film recently because I'm going to start stitching some 4x5 and I've learned a couple of things.
First, it's worth messing with finding the right focus, which can be as much as 2.5mm or as little as .75mm over the glass and still be in spec for Epson. Second, for me, scanning between glass added quite a bit. Including Newton rings, so I was considering getting some anti-newton glass, because I thought it would be worth the difference from no glass at all. Then I discovered wet scanning. It's a little more bother, but there's another nice bump in performance to be had there, again.
Summary: if you're scanning using Epson's neg carrier, there's still a lot of room to make things better.