Ektar 100 versus Alpha 900, Velvia 100, Portra 160VC and TMax 100

Discussion in 'Film and Processing' started by beamoflight, Dec 2, 2008.

  1. Hi all,

    I thought I would drop here my contribution in the continuing evaluation of the new Ektar 100 film. Please note I
    am not a professional tester, but I have done my utter best to be consistent and thorough. Do note also that some
    of the scanned images are completely unsharpened, so essentially "RAW" type, so don't bother me too much with
    comments about softness! I have made clear remarks whenever an image is sharpened or not.

    I made scans on an Imacon 646 and put the Ektar 100 film up against Velvia 100, Portra 160VC, TMax 100 and
    last-but-not-least against the pixel monster Alpha 900!

    Have fun: (follow line Ektar 100):

    http://www.boeringa.demon.nl/menu_technic.htm

    Regards,

    Marco
     
  2. Thanks for all your work, Marco. Amazing how Sony has relatively quickly moved into the photography business in a big way. Who would have thought 10 years ago that they would be competing against Nikon and Canon. I just sent out some exposed Ektar to be developed. I don't have access to an Imacon 646, but I look forward to scanning it on my Nikon 9000. Looks like a great color reversal film.
     
  3. Thanks Marco! I love these tests you do! <p/>It looks like the Ektar has a redder cast than the Velvia 100. I would have thought the other way around.<p/>I wonder how the Ektar would compare to Provia 100F? Just kidding!
     
  4. Excellent work Marco, thank you. Now all you need to do is compare the Alpha to MF :).
     
  5. A test like this was done at DPReview….maybe it was you Marco….I can’t remember. It was not well received by some as showing that humble 35mm was so, so close to a 24mp DSLR gored the sacred cow of some. After all, the 3mp D30 was proclaimed by some to beat 35mm film at even 10x15……something I noticed was blindly incorrect when testing my D30 with my old Imacon 343.

    I prefer 4x5 to 35mm for reasons obvious. But your test was well conducted and interesting to read. Amazing what results can be obtained by someone who knows how to scan film.

    Best regards.
     
  6. Wonderful test, Marco (I hope you'll run one involving 100UC and perhaps Velvia 50, at some point, as well). I'll
    look over it with a more keen eye later. Thank you.
     
  7. "Amazing how Sony has relatively quickly moved into the photography business in a big way. Who would have thought 10 years ago that they would be competing against Nikon and Canon."

    Sony introduced the first CCD camera in 1981 (Mavica). They made the first commercially available solid state image sensor...they were the first company in the business of electronic cameras. They've been in the "photography business" from the very beginning. They've been making solid state video cameras from about 1982. They just never addressed the dSLR market. They've probably made far more solid state imaging devices than Canon and Nikon combined if you count the millions of video cameras they've made.
     
  8. Steve....thanks for the info about Sony. I know they were in video cameras, etc. But still, I wouldn't have thought 10 years ago that they would be competing against such well established, old-line camera makers such as Nikon and Canon. But...it all makes sense when one considers that much of the "value" in todays DSLR is the electronics, an area where Sony has deep expertise.
     
  9. Sony is obviously doing well with their camera line. The samples I've worked on from the A900 are better than what I've seen from the 1Ds Mk3.
     
  10. To one degree, it made sense for Sony to start making DSLRs, since some of their loyal sensor customers like Nikon started making their own. So to keep volume on large sensors, becoming their own customer was a way to maintain economies of scale.
     
  11. Thank you Marco, I just downloaded the images and it looks that you set up the chart to close to the camera for a
    usable resolution comparison (all the shots seem to outresolve the chart). It maybe late and I may be missing a
    detail you explained in your setup.

    Anyhow, in my test of 35mm Ektar a and Tmax (Xtol 60/40) I get the following lines per picture height:

    Ektar: 3700 lpph (using coolscan)

    TMX: 5300 lpph (under microscope, coolscan showed 3900 which is its max true resolution)


    Can you please tell us what is your assessment of the A900 true resolution in lpph?
     
  12. Also

    Velvia 50: ~4100 lpph (slightly outresolving the Coolscan).
     
  13. Just as a side observation based on your scans, D76 seems to produce drastically softer, less resolving and grainier results than Xtol 60/40.
     
  14. Marco, is it me, or does the Ektar 100 (in your test) look to have less sharpness and resolving power than the 160VC?
     
  15. We are very interested to see what the true resolution of the A900 is horizontally and vertically.

    Could you please post a shot of a resolution chart (farther away from the camera so the A900 cannot resolve it)?
     
  16. Arjun,

    I noticed the same thing. It looks like the focus was off a bit on the Ektar sample.
     
  17. I wonder about the half-tone printing of the chart. Is it possible that some of the "grain" is really actual detail of the dots on
    the chart?
     
  18. Same here, also if the resolution shown for number 20 is just 2000 lines per picture height, there may be a problem with the test scan.
     
  19. All your film scans show visible grains while Alpha900 at the max resolution has none. I am no expert but my immediate impression from looking at your comparison scans is no matter how fine grain films are they are no match to digital cameras that have no grains. However, I just printed a couple of 6x7 400VC images in 8x10 size. I see no grain on these prints. 400VC is known to be not a fine grain film. But practically it is grainless in moderate enlargements. I can imagine it will remain fairly fine grain in 13x15 or even at 16x20 blowing up from a 6x7 neg. Of course blowing up from a 35 mm neg is a different story.

    My point is the grains in your comparison scans caught my eyes immediately. It immediately caused my mind to drift away from focusing on the resolution comparison. In my opinion these films and the Alpha900 are all very high resolution so the comparison of resolution among them are not very meaningful. Rather, dynamic range, colors, highlight and shadow performances are where the comparison will be more meaningful. Just my opinion. Correct me if I am wrong. Thank you for your effort.
     
  20. Tin, there is no right or wrong and every time someone undertakes the effort to post a test it benefits everyone.

    That said, I agree with you, digital cameras in the 25 megapixel range will have comparable detail to sharp 100 iso film. Both comfortably producing fine detailed 16x20 prints. Then dynamic range, color relationship (not color balance that can be corrected), etc, become more ever so the distinguishing factor.
     
  21. Nice work, Mauro.

    I think folks tend to forget how good film really is. I shoot both, and I do get tired of battles over which is better. The reality is that both film and digital are excellent.
     
  22. Prints larger than 16x20 would still be better detailed with film larger than 35mm.
     
  23. Agreed. That's why I shoot medium and large format. Besides, I really enjoy darkroom printing.
     
  24. Hi all, As the OP, I though it time to chime in again. However, with one big warning: I am not going to fulfil any or all of your wishes and answer to all of your questions. Why? Well, for one reason, and that is: Whatever I would do / say / comment, there will ALWAYS be someone disagreeing / saying "you should have done this or that". It will NEVER satisfy everyone. However, to give a more brought idea of how I see this test: First of, I am not a professional tester. This is the first time ever I tested films and a digital camera in such a manner and went so deep. Am I satisfied with how it went? Well, partly. Did things go wrong: no, not dramatically, but I have had issues like the unfortunate coincidence of when you most need it, the Imacon 848 not being available and having to make do with a lesser problematic scanner. Would I do things differently a second time: Yes, probably, the test setup would require some adjustment / additions. But all in all, I think it went pretty sound. Do I think that any of the issues I encountered render the test results completely useless? No, not at all, and that's the reason I posted here and on APUG to share it with all of you, as I still think many things are to be learned from it. In the end, well, I think I have given enough information and background for you yourself to evaluate it's value... I have given more background and insight than most professional testers / websites / photo magazines ever do, for example giving you excess to the full high resolutions scans directly. To respond to some remarks: - QUOTE1: "I am no expert but my immediate impression from looking at your comparison scans is no matter how fine grain films are they are no match to digital cameras that have no grains." You missed one crucial point about film and it's usage: film users generally WANT a film to have grain. We don't mind seeing it, nor scanning it. I never intended to obliterate grain from the scans (as is more or less possible in drum scanning or using smoothing functions in PS). As one former active, but now unfortunately deceased member (Flotsam) of the APUG (The Analog Photography Users Group) so simply and eloquently said it: "That is called grain. It is supposed to be there." - QOUTE2: "Thank you Marco, I just downloaded the images and it looks that you set up the chart to close to the camera for a usable resolution comparison (all the shots seem to outresolve the chart)." I only partly really agree with this remark. Yes, in hindsight it would have been better to use a test chart with higher resolution test patterns. However, as I noticed after doing all the scans and processing the files, that would also require a higher resolution offset printing process, as the test chart's dot screen pattern already is on the brink of being revealed. HOWEVER: that last fact in itself is a HUGE!!!! COMPLIMENT to all of the films and the Alpha 900 in particular in their high capacity for fine detail / high resolution capture. Just imagine, this chart is 60x90(!) cm (2x3 feet for you Americans and Brits, well, disregarding the differences in shoe sizes ;-) ). If you don't have any feeling of how big that really is, draw it out on your desktop, and amaze yourself of how much detail must be captured to get to that point! - QUOUTE3: "We are very interested to see what the true resolution of the A900 is horizontally and vertically." Well, as I said in the test, the theoretical max resolution is about 84 lp/mm based on the sensors size and pixel count. Now I have proven that the Alpha 900 and also Velvia 100 resolve the 40 lp/mm line pattern with ease, with enough contrast left for the Alpha 900 shot to resolve considerably more. So if you want my "BEST GUESS", than I would say, as I already stated in the test, that the Alpha 900 in reality comes pretty darn close to reaching it's full 84 lp/mm, at least with high contrast black / white transitions. It's not a huge jump from the 40 to 84 lp/mm, but it DOES require even two pixels directly next to each other to be able to distinguish themselves. Looking at the absolute cleanness of this test image line pattern in the Alpha 900, I am personally convinced the Alpha 900 will reach it's theoretical maximum with an appropriate line source in a test chart. Just look at this image, also on page 2 of the test results, once again (Alpha 900, 40 lp/mm line pattern at 800% viewing in Photoshop): Marco
    00RgjG-94715584.jpg
     
  25. "It's not a huge jump from the 40 to 84 lp/mm, but it DOES require even two pixels directly next to each other to be able to distinguish themselves."

    No offense Marco, I really appreciate your responses, 40lp/mm is just 5 megapixels. Big difference between 5 and 24 megapixels.

    Your test is good but can't be used for resolution comparison. You must be measuring/doing something incorrectly.

    The resolution of the film in your test is not an issue, we understand what the films can do, I posted the results I got myself for people to have as reference.

    But***

    1) Could you just shoot a picture of the chart with the A900 farther away and post it - It is a simple digital snapshot. We can do the true resolution calculation ourselves.

    2) Could you also set up a subject with lighting measuring f64 with the meter with the sphere out and directed to the camera, and shoot it at f8? This is to compare to the Ektar's dynamic range test I posted.

    Thank you again for all the contributions.
     
  26. Marco,

    This is the resolution test for Ektar:
    http://shutterclick.smugmug.com/gallery/6616619_YJEwK#429860538_sAEAm-O-LB

    This is the dynamic range test for Ektar:
    http://shutterclick.smugmug.com/gallery/6616619_YJEwK#424020444_n2LsD-O-LB
     
  27. Very nice and detailed review. I also agree with a previous poster, that I'm not willing to get caught up in the debate which format is better. Rather, I am happy that digital is now resolving in the film range and has given me a choice to consider more looks for my shooting, (I have a 1Ds and 40D). With digital I always get the shot first, and follow up with film if desired. I also shoot at medium format film for those special scenes that I feel would benefit.

    The fact that Kodak is releasing a new emulsion is comforting in itself in terms of film being around for a while longer. I may just try a few rolls to see how it works for me in lieu of 160VC. I don't have a dedicated 35mm scanner, (I actually shoot my negs against light table on stand with a digital (fix in lightroom), and use an Epson Expression for medium format. Most of my prints are up to 30", and it suits me well.

    Fantastic times were are living in, more choices... for now.
     
  28. Well, for those like Mauro concerned with the max. resolution of the Alpha 900, it turns out DPReview has done a good job and answered all of your questions. They photographed a test chart with double the resolution of mine, so going all the way up to the Alpha 900 theoretical max. resolution of about 84 lp/mm (4000 line pairs per picture height). See here for the result page with a comparison to Canon's top model 20MP+ beast:
    http://www.dpreview.com/reviews/sonydslra900/page35.asp

    To download the actual test chart photo, it's here:
    http://a.img-dpreview.com/reviews/NikonD700/samples/comparedto/res/A900_Res_f7.1.JPG

    As you can see, my prediction of the Alpha 900 fulfilling almost it's full 84 lp/mm max resolution, was justified and confirmed by DPReview. Scroll down on the main page to the resolution table as well.

    Cheers,

    Marco
     
  29. Not a biggie, like I said, resolution is getting to where it can be compared to film. It is just I always like first hand test/discussions with the person actually running the test.

    It looks like the A900 resolves similarly to Ektar100 (from DPreview).
     
  30. Per the chart on DPreview, it can resolve 2700 lines per picture height. That is very good. They mentions slightly better than the 1DsIII.

    For nature, Ektar may have the edge on resolution and dynamic range, but for general purposes it looks like the A900 is comparable.
     
  31. Marco, I think I did not make it too clear to you and I think you have misunderstood my response. My fault. I love films. What I was trying to point out is the grainy scans of films could easily mislead one's mind to jump on a negative perception toward films. Some would probably look at the grains and say: look, films are clearly inferior. I understood that wasn't your intention. That should not be a significant factor in comparing those films vs. Alpha 900. When printing at moderate enlargements these films are grainless. At max enlargement those grains could easily carry people's minds away. I actually do not like grains. But at the sizes that I print I do not have a problem with them.

    On the other hand, if the test focused a little more on comparing dynamic range, color characteristics, highlight and shadow performances then the differences among these films and Alpha 900 will be more objective. Maybe you did in your test. Maybe I missed it. Any way, knowing the differences is a tremendous benefit for us photographers. So, again thank you for your effort. I hope I got it right this time.
     
  32. So would everybody be happy if I put 154 lp/mm as the resolution for Ektar 100 in my film comparison table? I got this number by dividing 2700 LPPH by 24mm, the frame height for 35mm film. That number seems close, because datasheets say TMX resolves 200 lp/mm, and by the same calculation above, Marco's test concluded 220 lp/mm.
     
  33. The travesty is that the Ektar is not available in 120! what a crock. I find it hard to believe that from a production standpoint it would cost that much more to manufacture.

    I would think demand in 120 would be quite high for this emulsion.
     
  34. Bill, my Ektar test had 3700 LPPH. (not 2700). Yes, it is 154 lines per mm.

    This is the resolution test for Ektar: http://shutterclick.smugmug.com/gallery/6616619_YJEwK#429860538_sAEAm-O-LB

    My test of TMX measured 220 lines per mm (slightly higher than the datasheet).
     
  35. Velvia 50 measured 160 l p mm in one roll and 170 l p mm in the second roll rested.
     
  36. Mauro, in another thread you say: "the new Ektar 100 has very fine grain, wide latitude and it outresolves the new 24MP Sony A900."
    I'm sorry, but where's your direct evidence for this? I must have missed it. Marco's data clearly shows otherwise.
    Also, I'm really curious about your numbers of 3700 lpph for Ektar 100 and 4100 lpph for Velvia... and how that relates to how much of the detail the CoolScan can really extract out of the film compared to what's on the film. Do you have an extended thread with these results? I'm curious about 4000ppi of LS-9000 vs 8000ppi of Imacon for scanning 35mm film. I have some results from real world pictures I haven't posted yet, but haven't shot resolution test charts yet to compare scanners... though I fully intend to. Have you done this (shot test charts to compare scanners)?
    Thanks,
    Rishi
     
  37. And Mauro, the chart I have (Stephen Westin's) doesn't allow me to test higher than 2000 lpph, correct? Do you use a different chart?
    And, sorry, naive question -- do you just set it up so that it is normal to the lens axis, and fill the entire 35mm frame with the chart?
    I'd really like to try this all myself to convince myself.
    Thanks,
    Rishi
     
  38. Rishi,
    "Mauro, in another thread you say: "the new Ektar 100 has very fine grain, wide latitude and it outresolves the new 24MP Sony A900."
    I'm sorry, but where's your direct evidence for this? I must have missed it. Marco's data clearly shows otherwise."
    Yes. I posted the result above in this thread.
    "
    And Mauro, the chart I have (Stephen Westin's) doesn't allow me to test higher than 2000 lpph, correct? Do you use a different chart?
    And, sorry, naive question -- do you just set it up so that it is normal to the lens axis, and fill the entire 35mm frame with the chart?"
    Just shoot it from farther away so you can't resolve the number 20 detail but close enough so you can resolve at least the number 1 detail.
     
  39. Any distance you pick is fine as long as you resolve between 1 and 20.
    The OP was resolving over 20 so his test is inadecuate to measure resolution.
     
  40. Mauro, perhaps I'm not understanding something correctly.
    The 4.2 you point to in your Ektar resolution chart image: If the test chart is photographed such that it fills the full frame of the camera, DPReview says that if the value '20' is sufficiently resolved, then the sensor resolves 2000 lpph. Meaning their 'chart factor' = 100 lpph.
    Now, I know that your resolution chart doesn't take up your full frame, but how did you determine your 'chart factor' to be 890 lpph, therefore resulting in something as low as resolving a value of '4.2' equating to a resolution of 3700 lpph?
    How do you determine this 'chart factor'?
    Thanks in advance for your help,
    Rishi
     
  41. Open the picture in photoshop, note the total height in pixels and divide it by the total height of the chart in pixels. The height of the chart is measured where the black and white arrows converge.
     
  42. Multiply by 100 of course.
     
  43. Basically, in my test, the chart fit 8.93 times in the picture vertically. So every step in the chart equals 893 lines per picture heaight.
     
  44. Oh, wow, I didn't realize that was a 1:1 crop representing only 1/9 the total image height!

    Is it? If so, you really zoomed out from the resolution chart!

    Still, I think 4.2 is a bit of an exaggeration. If that's a 1:1 crop, I'd probably cap it off somewhere between 3.2 and 3.5, not 4.2. Which would then yield: 3.2 x 8.93 x 100 = 2858 lpph.

    Don't you think?
    Rishi
     
  45. No. The criteria for the chart is "the highest number where you can still count the lines. Not ambiguos.
    Same criteria is applied to all films and DSRs tested.
    Here you can see the full image shot:
    http://shutterclick.smugmug.com/gallery/6560295_hraSq#418129352_UCwSt-XL-LB
     
  46. Also Rishi, remember Ektar is cutting it close to the Coolscan's resolution.
     
  47. Scanning Ektar (same as Velvia and TMX) at 8000dpi would allow to a more precise evaluation of the resolution.
     
  48. Marco, I hope you don't mind, I put your 800% Ektar sample next to mine for the 2000lpph mark. Using nearest neighbor.
    http://shutterclick.smugmug.com/gallery/6616619_YJEwK#432647170_Kn5Kr-O-LB
    The difference tell me that something may have gone wrong in your test or scan.
     
  49. The films look like trash compared to the A900, like 35mm vs. 4x5. Seeing and seeing clearly are two different things, a point too often missed in discussions of "resolution."

    It would be nice to see a real scene photographed on all of the above. The films are all at their best with a high contrast, black and white resolution chart. But it's down hill from there for film as film's resolving power is strongly related to detail contrast. The A900 should show similar resolving power even with mid and low contrast details. In the real world the gap will be wider.

    Pretty much any full frame body will produce higher IQ on real photographs than 35mm film. This A900 comparison, on a target which lends itself to film, just shows how big the gap is with current 20+ MP bodies.
     
  50. The scanned images are not contrasty even for the low resolution chart. Based on my experience with velvia 50 and 100 and the USAF test chart, the line pairs on the slides are very contrasty when viewed under a 30X microscope with a proper light source. The detection limit of the 30X microscope is 80-90 lps/mm.

    Richard
     
  51. Daniel,

    We know resolving power goes down with contrast on films.

    "Pretty much any full frame body will produce higher IQ on real photographs than 35mm film."
    We are trying to discuss based on actual results. Do you have any side by side tests to support your claim?
    Say like a tree shot with 35mm film and any full frame body?
    "This A900 comparison, on a target which lends itself to film, just shows how big the gap is with current 20+ MP bodies."
    If it makes you happy to believe that no one will stop you. Not sure what was wrong with the OP's test, my resolution results are drastically better than his.
     
  52. Daniel, (from above)
    I put the OP's 800% Ektar sample next to mine for the 2000lpph mark. Using nearest neighbor.
    http://shutterclick.smugmug.com/gallery/6616619_YJEwK#432647170_Kn5Kr-O-LB
    The difference tell me that something may have gone wrong in your test or scan.
     
  53. I suppose the question is whether the Sony Alpha 900 can replace 35mm or 6x6cm fine grain film in making photos that might be enlarged to 16X20 or 20X24?
     
  54. Newton,

    Yes, resolution wise the A900 is very good and can make prints comparable to 35mm film.
    No, it won't be comparable to medium format film - especially detail wise for landscape prints 16x20 and 20x24.

    I shoot 6x7 and get around 98 true megapixels of data with velvia (high contrast). The A900 gets about only 12 true megapixels of data (high contrast). Also my system (RZ67II) just captures more information all together through the lens onto the film plane than possible on a FF(35mm) system.
     
  55. The A900 has 30% more linear true resolution than the 40D (2700 vs 2100 lpph).
    This is what medium format scans (coolscan) look vs the 40D. All test done in studio control.
    http://www.shutterclick.smugmug.com/gallery/6302153_PLzKe#407096809_FeZJ4-O-LB
     
  56. Mauro:
    The A900 gets about only 12 true megapixels of data (high contrast)​
    Says who? Any data to back this up?
    As for your 800% comparison of Marco's results vs. yours: you're right, there's something terribly wrong with Marco's results. I'm inclined to believe your results, Mauro, as your explanation of 8.93x100 to calculate LPPH all seems reasonable & correct to me.
    As for an 8000ppi scan -- do you have an extra frame of film to send me? I can scan on an Imacon 848. Or, I could just repeat the test myself :)
    What size would you suggest I print Stephen Westin's resolution chart at so I can repeat the test myself, then compare a LS-9000 vs. Imacon 848 scan?
    Thanks,
    Rishi
     
  57. Mauro, I have to say going through your smugmug stuff: nice methodology.
    Couple of questions for you:
    1. In your 'Color Comp' image comparing TMX to Provia to Velvia -- did you use individual color profiles for each film when scanning, to make sure you're truly representing what's on the film?
    2. You list the 40D's 'True Resolution' as 5.6 MP. How did you arrive at this?
    Thanks in advance,
    Rishi
     
  58. Rishi,
    "The A900 gets about only 12 true megapixels of data (high contrast)"
    - at 2700 lpph --> Resolution = 2700 * 2700 / 24 *36 = 11 megapixels
    "As for an 8000ppi scan -- do you have an extra frame of film to send me? I can scan on an Imacon 848. Or, I could just repeat the test myself :)"
    - I have lent the shot I scanned to someone else but I have the one next to it which is one stop under exposed. It should work about the same. Can I cut just that frame for you or you need a strip to feed in the scanner?
    "What size would you suggest I print Stephen Westin's resolution chart at so I can repeat the test myself, then compare a LS-9000 vs. Imacon 848 scan?"
    - It doesn't matter. The larger and highest resolution the better of course.
    "In your 'Color Comp' image comparing TMX to Provia to Velvia -- did you use individual color profiles for each film when scanning, to make sure you're truly representing what's on the film?"
    - No same standard setting for all.
    "You list the 40D's 'True Resolution' as 5.6 MP. How did you arrive at this?"
    - From my own test. Roughly at 2000 lpph --> Resolution = 2000 * 2000 / 24 *36 = 6 megapixels
     
  59. I scan Ektar on a Nikon Coolscan V, which has a true dpi of 4,000. If the Ektar is underexposed then a small amount of grain shows up. With normal exposure there is no grain at all. It's the only color reversal film for which I don't have to use the scanner's digital GEM grain removal. I am MUCH more impressed by it than I expected to be.
    Is it totally sharp at 4,000 dpi? No. I have to reduce it to 3200 dpi for that. That's using a Nikkor 35/1.4 AIS lens, one of the sharpest lenses they make. Is it very sharp? Yes!
    Also, I wasn't doing wet scanning. It's possible that a wet scan would sharpen it up more. Of course, it's also possible that a wet scan would reveal grain.
    My two cents.
     
  60. Ektar is not color reversal film.
     
  61. Mauro, I don't understand your calculation.
    Let's assume that for each line pair, you need 2 pixels to represent the line pair (one black, one white). If you can resolve 2700 line pairs per height, then essentially you need 5400 (2700 x 2) pixels along the y-axis (height) to resolve those 2700 line pairs.
    For the horizontal axis, if we assume equal resolving ability, we'd be able to resolve 2700 x (36/24) = 4050 line pairs per width. Again, if we assume you need 2 pixels to represent each line pair, that means we have 8100 pixels across the width of the sensor.
    Now, if we multiply 5400 x 8100, we get 43.74 megapixels.
    Which also doesn't seem right, but for the opposite problem of yours (i.e. it seems too high).
    Help me out here,
    Rishi
     
  62. We are trying to discuss based on actual results. Do you have any side by side tests to support your claim?
    Say like a tree shot with 35mm film and any full frame body?


    Google is your friend ;-)

    http://www.sphoto.com/techinfo/dslrvsfilm.htm

    http://www.ales.litomisky.com/projects/Analog%20versus%20Digital%20Shootout%20(Hasselblad,%2035mm,%20Canon%205D).htm

    http://www.luminous-landscape.com/reviews/shootout.shtml

    http://www.photographical.net/canon_1ds_35mm.html

    You can argue line pairs all day long. As I said, seeing and seeing clearly are two different things. Even if the A900 ultimately resolved fewer line pairs on some chart than a particular film (I'm sure you could soup some B&W emulsions to beat it here), it will produce the superior print with the higher overall IQ.

    People have been producing fantastic prints with lesser DSLRs for a long time now, regardless of line pairs on a chart. Image quality is not just about high contrast line pair resolution.
     
  63. Mauro,
    Sorry but I'm going to have to slightly counter your conclusion, but let me start off by saying that, to me, your results show that the Ektar 100/Coolscan 9000 combo show that, for black & white resolution test charts, the resolving power of Ektar100/LS-9000 seem to be on par with that of the Sony A900. This says nothing, of course, about lower contrast scenes and/or 'cleanliness' or 'perceived sharpness' of the two formats.
    Now, the folks over at dpreview.com gave the Sony A900 a conservative rating of 2700 LPPH. I'm going to compare your image & their image blown up to 800%, to compare what their line pattern looks like @2700 LPPH and what your line pattern looks like at ~2700 LPPH. Here we are:
    [​IMG]
    Link to Full-Size Image
    I'd say, in all fairness, your pattern at 2670 LPPH looks like their Sony A900 representation of the pattern at 2700 LPPH.
    If we look at your Ektar 100 image at the region pertaining to 3700 LPPH, which you claim to be the resolution limit, I'd have to say that the analogous region on the Sony A900 pattern is somewhere around 3800 LPPH. Actually, that's me being very generous to the film scan. More like the region in your image pertaining to 3560 LPPH (4, on your chart) corresponds to the area pertaining to 3800 LPPH on the Sony image. I still think that's me being overly generous. Evidence presented below:
    [​IMG]
    Link to Full-Size Image
    So... I don't think you can say that the Ektar 100/Coolscan 9000 combo outresolves the Sony A900. They're about on par with another, comparing these two tests, for black & white (high contrast) regions. I think resolution will fall off faster for film at lower and lower contrast due to a formidable portion of film being needed to represent many tones... but I could be wrong. Let's save that discussion for somewhere else and focus on what's at hand :)
    Rishi
     
  64. That being said, Mauro, I would like to make an 8000ppi scan to make sure we extract all the detail out of the Ektar 100. I'll also place it under my 'scope to see what I can resolve by eye.
    I'll PM you my address.
    Cheers,
    Rishi
     
  65. Daniel, sadly you are quoting flawed or ill itendened marketing oriented examples. If you have a true interest I'd advice you to test yourdself.
     
  66. Rishi, "Mauro, I don't understand your calculation."
    The chart is LINES per picture height not LINE PAIRS per picture height.
     
  67. Rishi, "So... I don't think you can say that the Ektar 100/Coolscan 9000 combo outresolves the Sony A900. They're about on par with another, comparing these two tests, for black & white (high contrast) regions."
    The criteria for the chart is "Can you count the lines?". So this is your comparison:
    http://shutterclick.smugmug.com/gallery/6616619_YJEwK#433420448_yQERk-X2-LB
    Give it to an unbias person and ask him/her to count the lines on both charts. Then tell us the results (answer is 5 and 9).
     
  68. Rishi, I put the Ektar film in the mail for you. As a bonus I included the same shot with TMX and the same shot with Velvia 50.
    Putting them through an 8000dpi scan should give us a better assesment on the resolution.
     
  69. Mauro, if you look at the images I posted above, at 2800 LPPH for the Sony, I count 9 lines. For 2670 LPPH in your example, I count 5 lines.
    Any further down the chart, I can't say for certain that there are 5 lines. For the Sony, a little further down the chart I can still guess 9 lines, but certainly by 3800 LPPH, I cannot discern 9 lines. Nor can I discern 5 lines at your 3700 LPPH mark. I can discern like 2 or 3, but to be fair to the Sony, I can also discern 5 or 6 lines at 3800 LPPH.
    So I still think it's fair to say that the Ektar/Coolscan 9000 combo did not outresolve the Sony. Feel free to challenge.
    Great, I'm even more curious to check out the Velvia 50 results, as that's what I shoot. I'll scan them this weekend if I receive 'em by then!
    Cheers,
    Rishi
     
  70. No need to challenge, both Ektar and the A900 are very high resolving.
    I heard of people having problems scanning Ektar with the Imacons. I'm looking forward to the results.
    FYI, you'll flip over when you scan the Tmax...
     
  71. Mauro,
    Scanning any negatives on the Imacons is an exercise in futility. At least, prior to the 949, which I hear has a softer light source. Any little defects picked up by the Imacon are, of course, magnified upon processing the negative (the 'stretching of the rubber band' explanation that Marco provided on his website), which is NOT the case with slide film.
    Therefore, scans just look like they have holes all over them. I never scan negatives on the Imacon because of this... IR dust removal and/or the softer light source of the LS-9000 are essential to me when scanning negatives.
    Additionally, you can't do the cool 'super advanced erik krause workflow' with the Imacons to subtract film base color :)
    -Rishi
     
  72. ha ha, well don't worry about cleaning the films or exagerated grain. We are just benchmarking resolution not framing them for posterity.
    To clear grain quicky and sharpen for resolution evaluation try this -> 1- Sharpen (150, 1.5,0) 2- Neat Image 3- Sharpen (50,1,0)
    As you will see, my development process for TMX takes resolution pass the 200 lines/mm from the TMX specs.
    It'd be nice if you could add a crop of each film next to your collage above.
     
  73. Of course, I'll do a nice comparison of the Coolscan vs. the Imacon scan vs. the Sony.
    Also, I find it interesting that you sharpen, then apply NR, then sharpen again... typically I apply NR and then sharpen... why do you add in the first sharpening step? Additionally, my final round of sharpening (no not output sharpening) is typically done with a mask, a la Bruce Fraser, but thank heavens for the kind souls over on the Lightroom dev team that built this functionality into LR.
     
  74. A presharpen (with no other edits to avoid modifying the noise pattern) often/always increases N.I. efficiency.
    The last sharpen I do after all edits are done and it is sized for print.
    Try it both ways sharpen-NE and NE-sharpen, and let me know your observations.
     
  75. Sorry, too many abbreviations being thrown around. What's N.I. and NE?
     
  76. NI = Neat Image
    NE = NI misspelled
     
  77. Also, Mauro, I'm having some trouble understanding this LPPH vs. lpmm business.

    Marco claims that 2000 LPPH where the resolution test chart takes up the full frame translates to 40 lpmm. Since a frame has 24mm height, that's 2000 lines per 24mm, or ~83 lines per mm. Which he then translates to 40 line pairs per mm.
    So then that essentially means that when they talk of LPPH, they're actually talking about 2000 successive lines with no blank spaces in between? Kind of like saying 2000 rows of pixels per picture height. OK, I guess that makes sense. Let me know if I'm wrong.
    In which case, your calculation now makes sense to me. Because if you're seeing 2700 LPPH (well, that's what I see anyway for your Ektar example), then that's only 1350 line pairs per picture height, which is 56.25 lpmm. If you then assume that 2 pixels are needed to represent each line pair, that gives you 112.5 pixels per mm. Or 2700 pixels per 24mm and 4050 pixels per 36mm. Multiply those together and you get 11 megapixels. Which is what you're assigning the Sony & what I'm assigning the Ektar scan since I think they're on par with one another.
    So now it'll be interesting to see what the Imacon tells us about the Ektar film.
    But now I'm really curious... If at 2700 LPPH for your Ektar scan (my subjective resolution limit estimation), it's resolving 56.25 lpmm, and under your estimation of 3700 LPPH it's resolving 77 lpmm (3700/2/24), then how the heck are you saying that you've seen 160 lpmm on Velvia?
    That seems like insanity, given that I already think your 3700 LPPH is an overestimation, no offense :)
    Confused,
    Rishi
     
  78. "So then that essentially means that when they talk of LPPH, they're actually talking about 2000 successive lines with no blank spaces in between? Kind of like saying 2000 rows of pixels per picture height. OK, I guess that makes sense. Let me know if I'm wrong."
    You are correct. Just like true-pixels with actual information (not DSLR fake-pikes).
    3700 lines per picture height = 154 lines per mm. (not line pairs).
    Velvia has 160-170 lines per mm. (not line pairs).
     
  79. Haha, man I am thick in the head today.
    So, out of curiosity, what's the correct abbreviation for 'line pairs per millimeter' and 'lines per millimeter'?
     
  80. A great piece of work. Well done and thank you Marco. It is amazing to see the differences between film and digital, but even more striking is how all the pictures look good in their own way. That's why so many people shoot both formats I guess.

    Like some of the other readers, I'd love to see a test of similar quality using MF.
     
  81. Marco -- how come you haven't commented on Mauro's findings? It really does seem to me like maybe your Ektar example suffered from focusing issues? Especially in light of the fact that the Velvia seemed to have resolved all the lines... I doubt Ektar is that *that* much worse as your results show...
    Still, isn't it interesting to see how one person's interpretation of the results leads to a resolution calculation of 11 megapixels (mine, since I put the Ektar/LS-9000 combo at 2670 LPPH), whereas someone else's interpretation of the same results leads to a resolution calculation of 20 megapixels (Mauro's, since he puts the Ektar/LS-9000 combo at 3700 LPPH). No wonder we can't reach a consensus!
    Also, Mauro, I do think you need to concede that if you think the Sony A900's 'real resolution', according to your own interpretation of the data over at dpreview.com, is only 11 megapixels, then that's also the resolution of your Ektar/LS-9000 combo. Or, conversely, if you think that your Ektar/LS-9000 combo resolution is 20 megapixels or more, then the Sony really is close to its rates resolution of >20 megapixels. Because when putting those line chart at 800% next to one another, the Ektar certainly does not exceed the Sony in resolving ability for the same LPPH. Wouldn't you agree?
    Cheers,
    Rishi
     
  82. "So, out of curiosity, what's the correct abbreviation for 'line pairs per millimeter' and 'lines per millimeter'?"
    I don't know. It is probably best to clarify on every post...
    "Or, conversely, if you think that your Ektar/LS-9000 combo resolution is 20 megapixels or more, then the Sony really is close to its rates resolution of >20 megapixels."

    Not true. DSLRs are advertised knowingly that only 2/3rd of the pixels have actulal luminosity/color data. The only true statement from DSLRs is when they state for e.g. 24megapixels (true), but it becomes a lie when they add resolution to it (although they ussually avoid it to prevent lawsuits). Still confuses the buyers ingeniously.

    On the other hand, a Coolscan 9000 provides very close to true 20mp in 35mm and 90+mp in 6x7. The rest is out for the film's resolving power.
     
  83. If it weren't because of the above, it wouldn't make sense to test the resolution of a 24 megapixels camera, would it? That is why DPreview, etc test every camera to provide buyers with a value for resolution.
     
  84. e.g: the A900 ha a 24 megapixel output and has aproximate 12 megapixels of RESOLUTION.
    You can double that with a 35mm Coolscan, if you have the film providing the data.
     
  85. Also the 12 mp resolution of the A900 is a soft number because DPreview's test was not optimal. It could be higher... say 16 megapixels..?
    A correct way to evaluate resolution is by placing the chart so it extiguishes resolution in the lower numbers. Say 1-5 not 10 to 20. The two main reasons for this are:

    1) the jump in resolution at the lower numbers of the chart is much higher so it makes it possible to evaluate resolution better. For e.g. If a camera resolves step 2 of the chart, it comfortably resolves step 1 (half the linear resolution), and will clearly show it doesn't resolve step 3 (150% of the linear resolution).
    When you are trying to evaluate resolution between the mark 20 of the chart and the mark 21, the jump in resolutions are just marginal. It becomes very subjective.
    2) the second reason is that the cross patterns at the corners of the charts are useless otherwise since the camera will outresolve them.
     
  86. Interestingly enough, Popular Photography back in 2004 rated the LS-9000 at 3414 ppi (67.2 line pairs per millimeter), which would give you 15.6 megapixels scans at most.
    Which places your >20 MP resolution of the Ektar/LS-9000 combo a bit high -- but now I'm curious to see what the Imacon will resolve.
    Mauro, do you, or anyone else, have a link to a LS-9000 review that places its true optical resolution near 4000 ppi? I remember reading this somewhere, but can't find the reference.
    Rishi
     
  87. The Coolscan is not true 4000. My particular scanner resolves 3900 in one direction (the one I use for film testing) and 3650 in the other direction. These are what I use to calculate true resolution.
    35mm=> 20.2MP
    and 6x7=> 91.7MP
    This is a link to an extensive review with similar results:
    http://www.filmscanner.info/en/NikonSuperCoolscan9000ED.html
    Also remember Rishi that each pixel of a Dslr captures only one color per pixel whereas the coolscan captures every color on every pixel. Massive difference.
     
  88. True Mauro... wait a minute. So when Canon advertises 21.1 megapixels... are they saying that the effective resolution is 21.1 megapixels, or that there are exactly 21,100,000 pixels on the CMOS sensor, or that are actually many more pixels but that after interpolation & Bayer pattern processing and whatnot, you end up with an effective 21.1 megapixels?
    Thanks,
    Rishi
     
  89. Les, thank you for the links.
    Rishi, I hope the question is sarcastic... obviously you have only 21 million pixel but the effective resolution is much lower. This has always been the case - not a new development.
     
  90. Les, can you explain the 3650v 4100h resolution to me. That's better than mine grrrr!
     
  91. Daniel, sadly you are quoting flawed or ill itendened marketing oriented examples.
    Mauro, not one URL I pasted has anything to do with marketing. They are all real comparison images performed by people just like you and I.
    If you have a true interest I'd advice you to test yourdself.
    Been there, done that, have the prints to show for it. The thing is though, I test real scenes printed to real paper. I'm not all that interested in high contrast test chart resolution because I don't shoot high contrast test charts.
    If all you're looking for is a number on a chart, then I'm sure you can soup a B&W emulsion to beat the A900. If you're interested in overall image quality, then just look at Marco's samples again. The A900 looks like a larger format compared to the 35mm, with a target that favors the film.
    Not to throw a wrench into a healthy discussion about line pairs, just pointing out a flaw with the idea that line pair numbers tell the whole story about image quality.
     
  92. Daniel, yes it is all about the print.
    A couple questions regarding how you have formed your experience:
    1) To form your experience with digital; What digital camera have you shot more often and what printer and size you print?
    2) To contrast your digital experience; What film format, camera and films you use for printing on the the same printer and size that led you to your conclusions?
    3) What scanner do you have?
     
  93. If possible, after you tell us about your equipment, please be kind and post a "real life" example comparison of the same subject you shot with both your digital and your film equipment.
     
  94. Two things come to mind:
    First, the Imacon is good, but it isn't a great scanner and far from being the best drum scanner using fluid mounting; Second, scanning negatives requires LOTS of interpretation and judgment.
    This means that in the first case, the smoothness as well as the sharpness of the scans from film are compromised and in the second case, the color, saturation and contrast of all films and negatives in particular are all subjective.
    You've done a wonderful looking report, but it means very little. To confirm your results, I'd like to see high quality, totally RAW, 16 bit drum scans of each film where there's no interpolation whatsoever. The conclusions might be the same, but at least the test would be more accurate and fair.
     
  95. Mauro,
    I received the film(s) -- thanks!
    Unfortunately, I have some extremely disappointing news. I went all the way over to our scanning facility yesterday, with high hopes, only to find that they're closed until January :(
    So I'm going to have to hold your frames of film hostage for a bit longer than originally planned -- I hope that's ok?
    In the meantime, though, I've been doing some LS-4000 scans at home, as well as dropping them under the microscope. You're right, I flipped when I placed both the TMX and the Velvia under the 'scope. Not so much the Ektar, which paled in comparison to the previous two.
    Anyway, I'm putting together some LS-4000 scans as well as my own calculations from under the microscope. But before I post back with that info, let me make sure I'm making my calculations properly.
    Note the line on the right edge of the chart and the two crosshair points -- is that the 'chart height' I use to calculate the Line Pairs Per Height? I'm using the measurement tool in Photoshop to get an exact measurement. If so, I'm getting 8.6 chart heights per full height of the frame. If I measure from literally the top of the chart to the bottom I get 7.2 chart heights per full height of the frame, but I'm pretty sure that's the wrong way to measure it right?

    Please confirm. Thanks!
    Rishi
     
  96. Thanks for the links to those scanner tests, Les. Those were exactly what I was looking for.
    I'm going to have to say that I feel the resolution for the LS-9000 is overstated, whereas the resolution estimation for the LS-5000 seems spot on.
    Take a look here:
    [​IMG]
    For each scan, the upper right red circle indicates what I'll call the '6.2 horizontal' (vertical lines indicate horizontal resolution) element, and then lower left red circle indicates what I'll call the '6.3 vertical' (horizontal lines indicate vertical resolution) element. For the LS-5000, it's clear that 6.2 is resolved horizontally, and 6.3 is resolved vertically, which is indicative of 4095 dpi. 6.3 is resolved vertically, indicative of 3650 dpi. Which gives ~20MP worth of information.
    For the LS-9000, though, while I agree that it can resolve '6.3 vertical', giving it 4095 dpi in the vertical direction, I feel that '6.2 horizontal' is an overestimation. I can hardly make out those vertical lines. And this is at 300% magnification.

    So does the LS-9000 really perform worse just because of the optical diffuser? In my own scans, the LS-5000 appears sharper than the LS-9000 (but to me, offers too much offensive pepper grain with certain emulsions).
    Mauro, I'll try and scan your resolution test chart frames with and without the Scanhancer optical diffuser, to see if I see any differences in resolving power.
    Also, does 'vertical resolution' refer to the resolving capability of the actual row of pixels on the CCD, whereas 'horizontal resolution' refers to that resolution achievable by the stepping motor? Think about it for a while, orient yourself and the film and the scanner, etc., then get back to me :)
    Cheers,
    Rishi
     
  97. Also, doesn't Nyquist sampling theory hold that in order to truly discern 4095ppi of data (that is, 4095 lines per inch, we'd need to sample at twice that or, namely, 8190ppi?
    Let's think in line pairs:
    Say the black and white lines don't perfectly match up with the pixels. Then instead of one pixel getting the black line and one pixel getting the white line, one pixel will get, say, 75% of the black line while the next pixel gets 25% of the black line and 75% of the white line. This'll give 75%:25% for the two adjacent pixels (instead of the ideal 100%:0%). In other words, loss of contrast.
    I guess that's why the lines in the image above are not black/white at 6.3 and 6.2... but rather 'darker gray/lighter gray'... if we sampled higher (8000ppi), perhaps the contrast would be better?
    Am I understanding this correctly?
    And if so, doesn't that mean that 8000ppi really will extract more USABLE (not dye clouds, but actual resolution test chart lines) information out of film, given that my own (and Mauro's own) assessment of true optical resolution of film under a light microscope indicates ~26MP for TMX and 21MP for Velvia (and 12.5MP for Ektar 100)... sorry I haven't posted my observations yet, but I will in my next post...
    Rishi
     
  98. Let me state that last question a little more simply:
    Mauro and I have clearly seen 20 megapixels worth of information on resolution test chart shots on Velvia under a light microscope (no scanner involved!). In other words, 3665 lines per picture height or 3880 lines per inch. Which is ~1940 line pairs per inch, which would need 3880 pixels per inch to resolve. But then according to sampling theory, wouldn't that mean that we'd actually need to sample at 7760ppi to really get all that data?
    That is, an Imacon :)
    -Rishi
     
  99. This thread is really getting down to the discussion of the ultimate resolution (contrast 1000:1). More dpi (1.5 to 2X) for a scanner is needed to resolve the same amount of dpi (line pairs). Otherwise, fuzzy line pairs will appear.
    Richard
     
  100. Richard, thanks. Yeah I think I pretty much answered my own questions.
    So, basically, a 'true' 8000ppi scanner would have resolved 6.2 & 6.3 as black & white lines (higher contrast) than the darker gray and lighter gray lines that the Nikon LS-5000/LS-9000 resolved them as.
    Yes?
    Not that this same effect couldn't be simulated by adding significant sharpening to the Nikon scans...
     
  101. At least 2x dpi are needed to resolve the same dpi (line pairs) because the lines (black) are getting close to each other at the utimate film resolving power. The space between 2 black lines must be resolved by 2 pixels. Otherwise, fuzzy scanning results will be obtained that do not reflect the true film resolving power. Too bad, the bottleneck is the resolving power of scanners (from film to digitalization).
    Richard
     
  102. I guess I see this as an alignment problem (which Scott Turner has confirmed in personal communication)... basically if the pixels on a 4000ppi scanner were perfectly aligned with the line pairs of a resolution test chart, such that the black line fell perfectly on one pixel width, and the white space next to the black line perfectly projected into the immediately nearby pixel, then one could get, theoretically, nearly 100% response instead of darker gray/lighter gray lines (lowered contrast) for the lines on a chart corresponding to 4000ppi resolution.
    But what are the chances of that always happening? Slim. Half the black line may fall on one pixel, while half the black line falls on the next pixel, creating 50% | 50% for two pixels that should've been 100% | 0%. Hence the drop in contrast for details on par with the resolution of the scanner.
    But for 'real world' detail, which Fuji Velvia data sheets itself claims the film only records 5.5 MP of information (or 2000ppi, roughly), this probably really isn't an issue. Which is probably why 'in the real world', the Imacon scan doesn't make a whole lot of a difference compared to a properly done Nikon LS-9000 scan.
    Still, it'll be very interesting when my scanning facility reopens and I plop Mauro's film shots of resolution test charts on the Imacon vs. the Nikon.
    Cheers,
    Rishi
     
  103. Rishi, regarding the Coolscans resolution comparisons:
    1) The difusser has certainly and impact and at least you should sharpen before assessing resolution.
    2) The Coolscans have better resolution across the steps of the motor than across the CCD arrangement.
    In evaluating the resolution of my charts, my lines pph factor measured 890, If your calculations are close (+/- a few percentage points) to that number then you are measuring the correct distance.
     
  104. I still haven't posted my assessment, sorry.
    But I need to correct a previous statement... I looked at the films on the microscope again using a different diffuser, and my estimation of TMX or Velvia didn't change, but I was able to see the Ektar more clearly (color negative film is hard to judge under the microscope given the orange mask and the inherently low contrast).
    But the Ektar held up well against the Velvia on my second look...
    My estimations are:
    TMX: 26MP
    Velvia: 21MP
    Ektar: 21MP
    This goes well with another site that claimed they could resolve 154 lines/mm on Ektar 100, right up against 160 lines/mm for Velvia.
    I'd really like to test 'real world' contrast.
    In the meantime, an interesting article by Norman Koren on 4000ppi vs 8000ppi scans:
    http://www.normankoren.com/Tutorials/Scan8000.html
     
  105. Rishi, I can't believe it, for once, our results are similar for the first time.
     
  106. Rishi, what is you true MP assesment of the A900, based on the charts from DP review, to compare to the films above?
     
  107. Haha, yes Mauro... but don't get too excited... it was based off of the exact same test, literally, given you mailed me the results :)
    But yes, in general I agree with your results.
    HOWEVER, I realized you applied a, well, what I would call, ridiculous amount of sharpening to the film scan (unless your LS-9000 is magically 10 times superior to mine... and I checked focus many times plus held the film flat using AN glass).
    So, I also applied what I felt would be a similar amount of sharpening to the Sony A900 image. In which case, barring aliasing, I think the Sony can resolve up to 3700 LPPH, which makes it a 20.5MP camera.
    I'm really curious to see the Imacon scan because it's clear my LS-4000 can't resolve all the detail on the film. Perhaps one of the LS-5000 or LS-9000s in the lab can do better... I will post back results in January.
    Regardless, though, let's keep in mind that digital's resolution (as far as I know) doesn't drop off as fast as film (20MP --> 5.5 MP), or does it?
    Mauro, or anyone, can you comment? I haven't found digital SLR tests of 1.6:1 charts.
    Cheers,
    Rishi
     
  108. "Regardless, though, let's keep in mind that digital's resolution (as far as I know) doesn't drop off as fast as film (20MP --> 5.5 MP), or does it?
    Mauro, or anyone, can you comment? I haven't found digital SLR tests of 1.6:1 charts."
    Partially true, but it doesn't work that way.
    First, it is true resolution in film goes down as contrast goes down, same as the human eye. Digital drops off as a cliff. But scenes in real life are a blended result of low-high contrast and low-high frequency detail. Where high contrast-high-frequency intertwain low contrast-low-frequency.
    Second, resolution itself in film is obviously a random array - same as nature - not a discrete array as in digital.
    Third, a black and white chart in my opinion, favors digital - not film, as it hides the terrible color resolution of digital.
    "So, I also applied what I felt would be a similar amount of sharpening to the Sony A900 image. In which case, barring aliasing, I think the Sony can resolve up to 3700 LPPH, which makes it a 20.5MP camera."
    Sharpening does not invent detail. Can you post a side by side of the point where you consider the a900 still gives you individual lines at 3700?
     
  109. Can you post a side by side of the point where you consider the a900 still gives you individual lines at 3700?​
    I already did, Mauro :) Look above. I do a comparison of a 3700 LPPH section of the Ektar scan next to a 3800 LPPH section of the Sony A900... and I assert, once again, that there's no way one can reasonably say the Ektar resolves the lines better than the Sony around this region of 3700 - 3800 LPPH.
    Second, resolution itself in film is obviously a random array - same as nature - not a discrete array as in digital.​
    Very true. Which is why buildings and lines look so much better in digital (to me), whereas random detail (trees), to me, look better in my film shots [in side-by-side comparisons].
    First, it is true resolution in film goes down as contrast goes down, same as the human eye. Digital drops off as a cliff. But scenes in real life are a blended result of low-high contrast and low-high frequency detail. Where high contrast-high-frequency intertwain low contrast-low-frequency.​
    Digital dropping off as a cliff is desirable to film dropping off gradually (assuming that the gradual drop-off begins at a higher contrast than the cliff drop-off, of course). In order to accurately represent low to high contrast (or any contrast range), you want the recording medium to have the its maximal response throughout the contrast range. This is the best way to accurately represent a signal: have a 100% response to it no matter the signal. This is very basic theory... So I'm not sure what you're trying to say. The lens already decreases the contrast its recording, the film decreases it more, and so what you get in the end is not the best representation of what was originally there. If you could up the response to 100% for either the lens or the recording medium, you'd do a better job for most of the contrast range... up until, of course, where the digital 'drops off like a cliff'. But this'd be much better than dropping off gradually.
    Rishi
     
  110. And Mauro... why doesn't a film lover such as yourself do some 1.6:1 contrast resolution chart tests as well as red, green, & blue color resolution chart tests with digital SLRs??
    :)
    Has anyone performed these tests? I can't find them on Google.
    Thanks,
    Rishi
     
  111. It's fair to say Sony Alpha 900 resolves some lines at 3700 not all of them. It happened when the lines were aligned with the pixels. Otherwise, a blurred gap occurred. Mauro in his test includes some real life subjects to go beyond the contrast of 1000:1. If I understand correctly, it's easy for the software in the DSLR to guess straight lines, simple shapes, etc.
    Richard
     
  112. Agreed Richard. I'd also say that it's fair to say in the Ektar scan that 4/5 lines were resolved at 3700 LPPH... for the Sony, 7/9. That's 80% vs. 77.8%. They're on par with each other.
    Under the microscope, though, all the lines can be resolved (in my observation)... so perhaps the film slightly outresolves the Sony A900.

    To me, that doesn't matter though. As I only deal with scanned film. Again, the Imacon results will be very interesting. Trying to enjoy the holidays despite the anticipation of being able to back to the scanning lab :)
    -Rishi
     
  113. Wishes for Xmas:
    1) I wish the Coolscan 9000 were 8000 dpi to honor my RZ67 and TMX combo.
    2) I wish I had a 60" printer.
     
  114. Mauro, you do agree that maintaining 100% response (or as close as you can get to it) and then sharply dropping off is better than the gradual drop-off of film's response, no?
    In fact, Velvia boosts the response to >100% for lower frequency detail, which really makes it pop. And that's the best part of Velvia.
    Also, Mauro, I forget -- do you fluid mount, or use glass carriers?
    Rishi
     
  115. "Mauro, you do agree that maintaining 100% response (or as close as you can get to it) and then sharply dropping off is better than the gradual drop-off of film's response, no?"
    Like you said, it depends. If you are shooting a building the former is better. If you are shooting nature, the latter. In every portionf of the picture you will have amix of finer detail of high contrast and lower detail on low contrast - same as the natural eye response.
    "Also, Mauro, I forget -- do you fluid mount, or use glass carriers?"
    Glass for 6x7 and bare for 35mm.
     

Share This Page