Epson V700 out of focus?

Discussion in 'Film and Processing' started by andyorr1982, Feb 14, 2009.

  1. Hello Everyone,
    I just bought an Epson V700 to scan Medium format 120 film. I have done quite a bit of reading and almost everyone says this scanner has good quality for a flatbed scanning negatives. I'm not happy with my results at all so far and I wanted to get some feedback to see if I should send the unit back. I've tried scanning 120 film on the MF holder and on the glass surface itself. Both times I haven't had good results. I've also tried the height adjustment spacers for the film holder and they don't seem to make much difference.
    I have uploaded a sharpness comparison between the local lab's scan and my scan.
    http://www.aretephotograph.com/ComparisonSharpness.jpg
    I don't know what scanner they're using but they refuse to scan it more than 2078x2078 pixels and for a roll of 120 film they charge 15 USD! That's the main reason I bought the scanner...to obtain larger res. scans.

    To me, this lack of sharpness in the Epson is not tolerable. In this example the Unharpen Mask is turned to High and ICE and dust removal are turned off. Grain Reduction is also off.
    Should I worry about sending the unit back or do you think this is as good as I'll get out of the Epson V700 in general?
    Thanks much,
    -Andy
     
  2. I just started playing with a V750, basically the same scanner. And I've found out 2 things so far:
    1. Scanning directly on the glass is a big no-no. You should see a *huge* difference between scanning on the glass and in the holder.
    2. The height adjustment spacers do make a difference, albeit a minor one. Also, make sure you switch the one in the middle of the holder, too. I don't have the MF holder in front of me, but I know the 35mm slide holder has 5 feet. For my scanner, the + setting worked best.
    2078 x 2078 is approx. 900-1000 dpi depending on crop for a 6x6. Despite marketing claims, the V700's real resolution is said to be between 2000-2400 dpi, and that agrees with what I've seen. Without looking over your shoulder, I can't say what you might be doing differently than me, but you should be able to get some very good results from your scanner.
    --Greg
     
  3. First of all you need to make an estimate of how much resolution you are actually getting (in dpi). Scan the same negative over and over at higher and higher dpi until you stop seeing any more detail in the files. If the number you get is about 1800 dpi then sorry, that is all this scanner is capable of (same for other commercial flatbeds) - you need a Coolscan 9000. If you get significantly less then 1800 dpi then yes its broken.
     
  4. Andy, I feel your pain. I got the same results with my V750. Got one of Doug Fisher's custom film holders. BINGO! sharp scans. They have far more height adjustability and hold the film flatter. It turns out my focus point was above the highest setting I could obtain with the Epson film holder. The difference was amazing. Also those lab scans are sharpened, almost too much. An unsharpened Epson scan will always come in on the soft side but sharpen up nicley in post.
     
  5. Thanks for all your answers/opinions. It seems one of the Doug Fisher film holders is worth the extra xx dollars.
    Oddly, I made a custom foamboard holder for pressing the negative against the glass and it worked better than the film holder. This leads me to believe my focus point might be between the glass and 2mm. Any thoughts there? Does the Doug Fisher holder get closer to the glass than the Epson holder? I've also thought that film curl may be the culprit. I like the idea of having the braces between frames that the betterscanner holder has.
    Is the ANR glass worth the investment for Doug's holders? I'm thinking it might be worthwhile. On closer reading his site, I see that the ANR glass is a replacement of the "T" holders. Neat. Which is the best way to go?
    Thanks for all your help/opinions!
     
  6. Doug can answer the height question better. I think it does go lower. Your right, these scanners are all over the map. I have the glass, too and it works well, especially for older negs with a curl. I find if I cut and page my negs right after processing, and place them under some big books over night, they are plenty flat just to use the bars without glass. depending on your film and lab YMMV. You can see some of my scans in this threaD HERE: http://www.photo.net/digital-darkroom-forum/00Pfev
    Good luck...............Lou
     
  7. Hello Louis,
    Thanks for the post. Those are coming out much cleaner than mine have been so far. I'd be very happy if I was seeing those results. I'll place an order for the holder right away. :)
    Take care,
    -Andy
     
  8. Results vary with flatbeds.
    Folks always want the flatbed to be like a high resolution dedicated film scanner.
    Back when a pro flatbed as just 1200 dpi; they cost as much as a Hasselblad kit; ie back; body and lens; amateurs had 300 dpi units.
    After amatuer units breached about 1600 dpi; marketing kicked in; flatbeds are hawked like 5 horsepower vacuums.
    A 12000 dpi flatbed could be marketed tommmrow; many folks would flock to them like flies on dung.
    Scans from flatbeds; even low end ones have been used for a 1 1/2 decades for many types of pro work; whether it is acceptable depends on the application.
    Its really nothing new that best focus varies with a flatbeds Z height; thats why pro units were adjustable 14 years ago. Its also not new that a flatbed can appear less sharp than a high end drum scan; this is a couple decades old now.
    Go scan an rigid object with micro detail; place it at a slant so the objects surface goes from zero to say one or two millimeters. One can "see" the best focus point; this was shown at a trade show meeting in Reno back about 1992 that caters to service bureaus and print shops.------
     
  9. "scan an rigid object with micro detail; place it at a slant"
    Good tip Kelly. A great way to find your starting point and fine tune from there. In the "old" days we just would make shims for the holders out of cardboard to raise and lower the film holder. Andy, be aware my scans were sharpened for screen output. The scans come in somewhat soft looking.
     
  10. ... make shims for the holders out of cardboard to raise and lower the film holder.​
    Use a couple of the larger 3M yellow sticky notes. These are great because the whole pad starts out at about half an inch thick, but the individual sheets are rather thin.
    Do a binary search for the sharpest offset. Start with the whole pad, then compare the scan with one with half of the sheets removed. If the half stack result is sharper, do the next scan by removing half of the half. If the whole stack is sharper, do the next scan by adding back half of the half.
     
  11. OMG! Thanks to Andy for the topic.
    I always wanted to start this conversation, but I never had enough material to present.
    Firstly, Louis, thanks for the suggestion. That's something worth of looking at.
    I got the scanner (V700) a couple of weeks back and was playing with it since then. So far, I'm more or less Ok with 120 negs scans (never tried on (+) settings though), 120 slides - even with (+) settings on the height adjustment - out of focus. 35 mm - tried on the default settings, results more or less Ok but still prefer my "old pal" Plustek 7200i :eek:) for scanning 35mm.
    Robert, thanks for the idea - I had the same thoughs on "what I can do to try different height settings for the film holder".
    -Andrey
     
  12. Thanks for the idea on finding the focal point for the scanner.
    I just had a thought...
    I put a series of four or five plastic wallet-sized cards on the scanner glass. I staggered them so I could see the edge of each and some text. From here I selected the "reflected" option for scanning in the epson software so I could have the lower light turn on. I scanned this and tried to see which of these plastic card edges was in focus. There's a slight problem though. They get more blurry as they progress away from the glass. Only the one on glass is in focus. Does the Epson use a different lens when in "Reflected" light mode so that it can focus on the pane of glass?
    I've figured I can probably determine when I've found the proper focal distance when you can see grains in the film. does this sound about right? This scanner should be able to scan well enough to pick out some grains, right?
    Thanks for the help...In the mean time I'm waiting for my Doug Fisher tray. :)
    -Andy
     
  13. ... found the proper focal distance when you can see grains in the film.​
    I use a V500. The optics are so bad that it can't sharply resolve film grain on Tri-X. Pretty crappy in absolute terms, but I can't complain about a $100 scanner.
    The V700 and better has "high pass" optics for film scanning. Supposedly this increases scanner resolution by some real margin. Could you post some crops when you get a chance? I can't justify $2500 for a Nikon 9000, but a few hundred for better flatbed scans of my 6x7 would be worthwhile.
     
  14. Robert, noise, dynamic range, resolution and sharpness. All will come ruin your work if you scan with a flatbed.
    I'd advice you to get a used Coolscan 9000 - same price as a punny 40D.
     
  15. The Coolscan 9000 resolves 4000dpi in one direction and 3600dpi in the other. The Epson 2000dpi and 2100dpi.
    The Coolscan dynamic range is from 1-10,000. The epson 1-1,000.
    The Coolscan has almost zero noise. The Epson is totally intrusive in the shadows.
    The Coolscan is grain sharp. The Epson is like looking through the bottom of a bottle of wine (after you drank it).
     
  16. ... All will come ruin your work if you scan with a flatbed ... I'd advice you to get a used Coolscan 9000 - same price as a punny 40D.​
    Wow, that's serious.
    Here's how the cookies crumble (for me.) The Nikon scanners has twice the linear resolution of a low end Epson. However, 6x7 MF has four times the film area of 135. To a very rough first order, this means that 135 on my CS5000 and 6x7 on the V500 both yields the same ~20MP of usable image information. This is enough to generate technically flawless 8x10 prints, and very good 11x14's or a bit larger. Practically, these print sizes are just about right for manageable portfolio and albums.
    The key for getting tonality right with the Epson, or any scanner really, is to shoot and develop with the scanner in mind. Run some test exposures to understand the density range that the machine can comfortably digitize. When taking an exposure, and in the subsequent development, the goal is to map the desired scene dynamic range out to the appropriate negative density excursions. Fundamentally, this is nothing new - it's what fine art B&W photographers have been doing for decades. It's a quite practical workflowl for MF systems with removable film backs (I use primarily the RB-67.)
     
  17. "When taking an exposure, and in the subsequent development, the goal is to map the desired scene dynamic range out to the appropriate negative density excursions."
    Hmmm... how do you propose to do that? Say capture 14 stops with TMAX but only using half the density range?
    Robert, you need to step back and rethink what you believe you are accomplishing. Exposure to density has very little room for expansion. You can contract a bit with additional development. But you are tricking yourself otherwise.
     
  18. ... how do you propose to do that? Say capture 14 stops with TMAX ...​
    So, a few points here, the main being to explore what the media, the tools, and the specific workflow is capable of. It actually doesn't take a whole lot to get very useful information.
    1. Get a gray card. Mounted it nicely for some picture taking. Use a room with a constant light source.
    2. Take a reflected light reading from the card, and then an incident reading. That both shows the same EV gives comfort that nothing is drastically out of whack.
    3. Find some matte material that's a stop brighter and a stop darker than the gray card. Mount these adjacent to the gray reference. Having done step two, it's probably a good bet that the light meter can be trusted in this search.
    4. Load up the camera with film. Adjust shutter and aperture so that the first exposure is 5 stops darker than initially indicated. Shoot so that each subsequent frame gets one stop more light. A 6x7 camera gives exactly 10 frames - for a nice ~12 stop test strip.
    5. Oh, might as well blow another roll of film for this test. Have a frame totally exposed, and leave one frame totally unexposed. This gives a good reference for what max density, and clear film base looks like.
    The last time I've done this was about a year ago. For 120 Acros, a normal development in HC-110 dilution H gets about 10 stops. This is from direct visual inspection of the negative, and relative to the max density and clear film base reference frames.
    The "about 10 stops" comes about because the shadows gets very compressed. Practically, there's not enough usable tonal separation here even if technically there might be.
    Importantly, the cheap 4490 scanner was able to digitize what was directly apparent on the film. This means the scanner is good enough to pickup the bulk of the density excursions on Acros, independent of the source scene dynamic range. Good'nuf for govmn't work.
    As for 14 stops from TMAX, well I dunno, maybe with a super aggressive compensating development regime; it's certainly not particularly consistent with my recollection of the TMY characterstic curves. Even then it'd be interesting to see whether the compression leaves enough real highlight tonal separation for anything useful. Practically, keep in mind that a white card is just about three stops above gray, and a 3% black card about two stops below gray.
     
  19. Robert,
    I find my Epson V700 tops out around the 2200ppi mark in horizontal....when perfect focus is achieved. I use it primarily for 4x5. With that said, what I see is this: When using the V700 for 4x5, I only get a slight bit more resolution than I would with the Nikon 9000 and 6x7.....but I get that rez with 1/2 the grain. What it basically means is that you need to use 4x5 to achieve 6x7 quality on a good scanner. Of course, there's nothing stopping you from obtaining pro level scans from your best work at a lab, and using the flatbed for everything else. I find with the V700, with 4x5, I get superb results to 24x30....with 6x7, the results are good (not superb) to 16x20.
     
  20. I too found that on the scanner glass to be the sharpest. What a pain. Can't use frames. (Epson750)
     

Share This Page