Fuji GW690III = How many megapixels ?

Discussion in 'Medium Format' started by john_dowle|1, Dec 8, 2011.

  1. I've been wondering for a while now just how many mega-pixels in a modern DSLR do you think would give the equivalent quality that you get from a Fuji GW690 ?
    Any feedback much appreciated as usual.
    John.
     
  2. Not really relevant! Sure you can scan in whatever number of pixels you want but the limit resolution wise will be that of the film/lens combination. Gradation and the ability to cope with a wide dynamic range are something else again! You also have the question of whether you prefer the much cleaner look of digital or not. A couple of things do however need pointing out:
    Medium format digital can deliver the goods but so can medium format film.
    Getting ultimate quality in either medium depends on your skills as a photographer.
    You will do your best work with the camera that you work with best- so use that and worry less about whether to use film or digital capture.
    End of sermon!
     
  3. John,
    I have a 6x9 view camera (Horseman VHR) and a Nikon D300. I recently looked at the same composition made from both of these cameras. I used Velvia in the film camera. I found that looking at 100%, the digital file looked sharper. However, the Velvia image had quite a different pallet to work from. I'm now actually thinking of selling my horseman vhr. However, I haven't yet because of lens movements and the fact that sometimes I miss shooting velvia.
    Hugh
    Signature URL removed. Not allowed per photo.net guidelines.
     
  4. ...I found that looking at 100%.....
    What did you look at? The slide with a micro?
     
  5. Mag,
    I compared a 3200 dpi scan made on a Minolta Multi Pro. This was by no means a perfect test. Also differences could be due to focusing of either the camera lens or scanner????
     
  6. You're going to get all sorts of answers to this, because "quality" can be defined in terms of so many things: extinction resolution, MTF50 resolution, signal-to-noise, dynamic range...
    Out of these, I presume you meant resolution. You can get massive megapixel equivalents for film if you consider extinction resolution with high contrast black & white target patterns, on slow film. It's much more practical to consider "viewer impressions" of resolution - the holistic impressions that you get when you compare a large print from film to a large print from a digital sensor. There is a good correlation between perceived sharpness in this sense and measures of resolution at normal contrast, like MTF50. Film has a much lower megapixel equivalent when the comparison is done this way.
    So I rate the perceived sharpness of my 16.7 MP square digital back (without an anti-aliasing filter) as similar to what I've obtained with 6x6 film. People who use the 22 MP, 4:3 aspect-ratio backs would probably agree that they're comparable to 6x7 or 6x8 film. If there were a 25MP, 3:2 aspect-ratio back without AA filter, that should be comparable to 6x9 film. (Pixel size = 9 microns in all these examples, so just as 6x9 film is the same film as 6x6 but just more of it, that 25MP back would be the same underlying quality as my 16.7 MP back but just more of it).
    Now it's a little different if you're comparing 6x9 film to a 35mm-format DSLR, as they all use anti-aliasing filters, which reduce resolution somewhat. That pushes up the megapixel equivalent from 25MP to probably something more like 40 MP (ballpark guesstimate - based on assuming that the AA filter decreases linear resolution by about 25%). The 35mm lenses will also have to have inherently better MTF than the medium format lenses, as the pixels are smaller.
    I have also observed that my 21 MP Canon 5DII results are subjectively comparable to my 645 film results. Doubling 645 (41.5 x 56 mm) gives you 6x9 film (84 x 56mm), so doubling the 5DII gives you 42MP - again ending up in the 40MP ballpark for a DSLR with an AA filter, but via a different route.
     
  7. No one's going to agree on this question. That said, with current technology, I don't think you could get enough megapixels on, say, a full-frame, 35mm-film-sized chip to equal a 6x9cm piece of film. I believe you'd need to jump up to a medium format digital back.
    While you could use any 6x9cm camera's film, I happen to have some scans from a GW690II camera. Using a Nikon LS 8000 scanner at full 4000 d.p.i. resolution and full bit depth, from a 6x9 cm piece of film I shot at ISO 200 with a GW690II, I get a 637 megabyte scan. I can comfortably make 16-inch prints from these scans that have bottomless detail from any viewing distance. Press your nose up against these prints and the image shows no sign of breaking up or digital artifact.
    Opening a RAW image from a Nikon D700 shot at ISO 200, full bit depth, I get a 69 magabyte file. In a 12x18-inch print from a D700 image, I can see digital artifact.
    Long story short, if I know I'm going to make an 11-inch or larger print from an image, I'll shoot medium format film and scan it. For making big prints, there's just no way a FF-35mm-sized digital camera is going to compete with a well-executed scan from a piece of medium format film.
     
  8. how many mega-pixels in a modern DSLR do you think would give the equivalent quality that you get from a Fuji GW690 ?​
    It depends on final use. Anyone can make scans and claims that one is better than the other but in reality, if they both give you the results you want in the application you want, then they are equivalent (which is what you asked for).
     
  9. There are all kinds of theoretical responses to this question. But there is one practical response. How good a scan can you get from the Fuji's 6x9 negative? And do you want to pay for it?
    From my GW690 I can get 6mp images cheaply, in an hour. My local lab uses a Noritsu and produces very nice scans at 3000x2000 pixels. Compared to the output of a 6mp DSLR, I prefer the film. I like the colour and dynamic range, but I also find that 6mp originated from film can be blown up slightly larger than 6mp from digital. Cost = $11 per roll, processing and scanning included.
    Some labs are offering even bigger minilab scans -- about 16mp size. I hear they are fantastic, and offer all the same advantages I just listed. I expect the cost to be incrementally greater and you have to find a lab that does this. (I hear that North Coast and Richard Photo Lab do these in the U.S.)
    If I'm willing to spend more per frame, I can walk to a lab that offers Flextight scanning. Not fast, and not cheap. But with an output of about 60mp, these scans are pretty jaw dropping.
    The point of this is to say, the limits of your Fuji GW690 are not the camera or film. The Fuji lens is fantastically sharp, the camera uses a leaf shutter which eliminates the resolution-robbing mirror slap and focal plane shutter bounce. A 6x9 piece of film is huge and inherently carries massive amounts of resolution - spatial, light and colour resolution.
    But there are real limits to a how good a film workflow can be, once it passes through a scanner. Objectively that means I can make my GW690 a really good 6mp, 16mp, or 60mp camera. Until scanners improve.
     
  10. But there are real limits to a how good a film workflow can be, once it passes through a scanner.

    Take the scanner out of the equation and it's even better.
     
  11. Hear! Hear!<br><br>But even when considering the scan results as well... I'm with Eric on this.
     
  12. I own the Fuji, a 5dmk1, and a nikon scanner.

    Last year I happened to shoot a photo whtih each, tripod mounted and compared them.

    Keep in mind that I shot iso 160 color neg film.

    And the winner?...almost a tie. The film was a hair more detailed, but fell apart due to grain structure. The 5d was super clean. A
    higher res scan would not have extracted any more usable detail from the film.

    A 5d II would proabably give a slight edge to the digital file, albeit with a little bit less dynamic range. The Fuji costs between 1 and 2
    dollars per frame to shoot, so factor that in. The scanner, would be about $1500 used today, I think.

    Of course the film purists will disagree...by I stand by my observations:)
     
  13. I recall reading that the FBI a few years ago figured out they would have to have somewhere around 14 to 16MP or so to equal a 35mm film camera. So, since the 6 x 9 has--just a quick estimate--maybe 5 time the area of 35mm, then that would put us in the ballpark of around 75MP. I'll assume we are comparing with a film of fairly high resolving power, such as TMax 100 or Delta 100 Pro.
    Here's another idea: Film has better dynamic range than digital. Digital camera with the best resolution may not have the highest pixel count, since larger pixel sensors have better dynamic range than smaller ones. So what if we wanted a digital camera with the dynamic range of Tri-X? Our pixel guess must then be lower. To be comparable to a 6 X 9, I'll guess we might be looking at something like--what? Maybe 14MP to 20MP? Just as the Epson R-D1 with "only" 6MP outperforms the DR of the Leica M8 with twice the pixels.
    We all agree: there's no exact answer. No one right answer, either. But it's interesting to think about what a reasonable ballpark figure might be. And even that depends on what aspect of the problem we think is important.
     
  14. I'm not a "film purist." A completely digital workflow is much less time-intensive, and allows opportunities like HDR and for color correction that are very difficult or impossible to execute with film and scanners.
    Certainly, if I had an extra $75K lying around, I'd buy a high-end Hasselblad digital kit with four or five lenses, and I'd sell my MF film gear. Also, if the largest print I ever made was 8x12 inches, I'd work exclusively with my Nikon D700s.
    Unfortunately, I like biggish prints and I'm not independently wealthy. As such, I still shoot and scan medium format film. With all due respect, if you can't make a more detailed large print from a 6x9 negative than any 35mm-SLR-styled-FF-digital camera, something is wrong with your film-shooting or scanning technique. A competent 600-plus megabyte scan from a well-shot 54-sq-cm piece of film is so much more detailed than a sub-100 megabyte capture from a six-times-smaller digital sensor, it isn't close when looking at jumbo prints.
     
  15. B G, I respectfully disagree when you say "A higher res scan would not have extracted any more usable detail from the film." I have made this mistake in the past.
    It is easy to look at a scan of any resolution (including low-res scans from poor scanners) and see grain. The natural assumption is that it would be a waste of resolution to increase the quality/resolution of the scan because you have already extracted all of the detail "down to the grain."
    But I have never seen an example where scanning a piece of film with a better scanner at a higher resolution does not extract more detail and a better image. The Nikon scanner you are using is pretty good, the best of the consumer desktop scanners. But seek out comparisons to an Imacon scan, or a drum scan, and you will find that scanning "down to the grain" means something different to each scanner.
    I suspect that part of the reason film offers up better and better image quality from better scanners is because of the fractal nature of the silver or dye clouds. You can quickly resolve the largest individual grains/clouds, but with increasing magnification you reveal progressively smaller grains/clouds. Each magnitude of grain carries some infitessimal amount of real image data - after all, grain is not something overlaid on top of the "real image." When you are seeing grain, you are seeing the building blocks of how the image is formed itself.
    To take an analogy, imagine the whole photography is a cathedral. Magnify your view of the catherdral and you quickly can resolve large stone blocks. Magnify more and you see small blocks and bricks. Magnify enough and you can see the grains of sand that make up the structure of the blocks themselves. Each of which are a valid part of the structure of the cathedral.
     
  16. I run a lab and exhibition printing service. I use film, 35mm digital and 37.5mp digital medium format. I use a Hasselblad X5 scanner and Epson 9900 printer. I only say this to indicate that I see a lot of different types of files, and work on all sorts of film and digital originals. Fundamentally, you are comparing apples and oranges to compare digital and film. They behave completely differently with respect to their resolution. For example, film has lots of resolution, but it is shrouded in a lot of noise (the grain). Digital often does not have as much resolution, but it has a much higher APPARENT resolution up to its 100% detail. In this sense, digital often looks sharper in normal sized prints, but if you push it much past about double its standard resolution, it tends to look rather poor. Film on the other hand will look soft, but the way the detail softens tends to give the appearance of more resolution, so you can push the enlargement a bit further than with digital of the same "resolution".
    So, when it comes to 6x9? If I were to scan it at 3200dpi on the X5, you could print it to 1mX1.5m or so and it would still look very good. In order to take a digital file and make the same sized print, I would want about 25-30mp of non AA filtered digital. Medium format digital would do the job no problem. It would likely do better too. The M9 at 18mp with no AA filter looks very good at 1mX80cm, but perhaps not quite as good as 6x7 or 6x9 film. It has a bit sharper resolution of fine detail, the but the film picks up a few tinier details that makes it look a little better. As you see, it is not an exact science. But I would say that the top of the line digital cameras today and medium format film both output so well as to make these sort of questions more the realm of pedantry than utility. Use what you like, it is all good enough these days!
     
  17. Pixel counts are not an accurate measure of quality so it's hard to know how to answer the question. Just look at modern digital cameras and you will see rather quickly that more pixels does not equal better image quality. Each sensor has different capabilities and each software app varies too. I just bought a 24mp digital camera and the image quality is clearly not twice as good as its predecessor, which was 12mp.
     
  18. ...how many mega-pixels in a modern DSLR do you think would give the equivalent quality that you get from a Fuji GW690 ?​
    Probably a lot less than you think!
    There've been some really unfounded assertions made here. Notably by David and Rob. Film is not some magical and totally analogue system that can be infinitely magnified to get more and more detail. The dye clouds of colour film have a finite size of around 3 microns diameter, and an average clumping or cluster size of around 9 microns. This roughly equates to 3000 pixels/inch and there really truly is no more useable detail to be got below that "resolution" level. Furthermore those dye clouds are fairly uniform in colour density, making the colour depth per equivalent unit area of film far lower than the 16.7 million colours per pixel of digital capture. Black and White film fares hardly any better in its resolving power, and can be a lot worse in its ability to render tone smoothly. And if you're going to champion film, forget the scanning process altogether and compare wet-printed prints with digital. At least that gives film a fighting chance.
    Dynamic range is also ridiculously overestimated by film enthusiasts. A good DSLR can currently encompass a dynamic range of over 12 stops, and lens and camera flare will conspire against anything much above that being meaningfully translated onto either film or a digital sensor.
    I'm not anti-film myself, but every time there's one of these threads the pro-film lobby trots out some totally usupportable nonsense as "evidence". If film really is that much better, show us. Post some comparison pictures to prove your point instead of all the hyperbole.
    My own comparisons are NOT favourable to film, as shown below.
    00ZiFt-422857584.jpg
     
  19. I'm with Steve Smith on this. If you can, just print directly through the negative onto photographic paper.
    I'm limited by my darkroom to an 11" by 14" print, which looks splendid when mounted and framed.
     
  20. I'm with Rodeo Joe on this :) Here's my examples below. This is a 4000dpi scan vs. 13mp digital capture. I think a few more megapixels for the 5d and it just might come out ahead. Notice that while the film image is more detailed, the crack in the wall gets lost in the film grain, but not in the digital version.
    [​IMG]
    fuji 6x9 full image
    [​IMG]
    5d full image
    [​IMG]
    fuji 6x9 detail
    [​IMG]
    5d detail
     
  21. Hugh, when you scanned the 6x9 frame at 3200 PPI you made an 85 mp image. If you compare that to a 12 mp
    image, both at 100%, the 85 mp image can look a lot less sharp but actually contain more detail. The real test is what
    the images look like printed at some size that's large enough that pixels in excess of 12 mp are relevant.

    B G, same comment, but you have a 133 mp image from your film scan. Even at this size you can see that your film
    scan has way more detail than your DSLR image.
     
  22. Yes Andy, more detail. But to see that detail, you also need to see all that noise/grain. Some detail gets hidden in the grain, so it's a judgement call for the photographer which he/she prefers I think.
    The color rendition from the digital camera is superior in this instance, in my opinion though. Of course choice of film will make a bit of difference here, as might processing.
    Reversal film will look less grainy, but it's just too contrasty for my style.
     
  23. You can't judge by scanned film. The real test would be in a final traditional print from Velvia (or at least a drumscan) versus print from DSLR . My guesstimate, from experience with 120 format Velvia, 4x5 Velvia, 35mm Velvia and 42 MP stitched images from a 5D II and Canon 17mm f4 L TS-E would be at least 80 MP and more likely 120 MP.
    I have not yet taken the time to directly compare my stitched 5D II images to my Velvia 120 format enlargements and transparencies but my impression so far is that this is very close to 35mm and 645 format Velvia. So, above I have simply interpolated from 36mm x 48mm (stitched image size from 5D II) to 56mm x 84mm, to get to 114 MP and I threw in a few more MP for good measure.
    Yes, this could come down without an AA filter, but I used a Kodak SLRn which was driven from the marketplace due to the lack of an AA filter and I can't understand how medium format is getting away with it.
     
  24. Yes, this could come down without an AA filter, but I used a Kodak SLRn which was driven from the marketplace due to the lack of an AA filter and I can't understand how medium format is getting away with it.​
    From what I've read, the SLRn & SLRc were driven from the market place primarily because of issues with their Fill Factory CMOS sensors: their very poor performance above base ISO (high readout noise), and very limited long exposure capability (high dark noise). Their predecessor (Kodak 14n) had even worse sensor issues, and that sort of reputational damage was hard to shift; and by the time the improved SLRn & SLRc came along, the full-frame Canon 1ds had already set new benchmarks for low-noise high-ISO performance, and long exposure performance, and the Kodaks just could not come close. Nikon did not have a full-frame camera to compete with the Nikon-mount SLR/n, but the overall package they offered with their contemporary APS-C cameras (D100 and D2H) was still winning more customers.
    I have a thing for the older Kodak DSLRs; I love the concept of the F5-based ones with the interchangeable viewfinders and user-changeable IR/AA filters, so I picked up a DCS720x not so long ago. The RGB filters on Kodak CCDs deliver beautiful colour (I can attest with my DCS645M digital back; and there are all those other Kodak-chipped medium format digital backs and cameras, and the various digital Leicas), and each generation has significantly improved on long exposure noise.
    Just imagine if Kodak hadn't shut their DCS division in 2004: we could now have a "Kodak DCS1800" with the same Kodak 18MP full-frame CCD as the Leica M9, in a fully modular F5 (or better) body, with modern batteries, a modern user interface and a big 3" LCD!
    Ah, but now I'm day-dreaming...and WAY off-topic!
     
  25. I think for 6x6-6x9 format film a practical scanning resolution is 2000-3000ppi (I use an LS-9000). Going higher than that although you can get slightly increased sharpness the grain will be more obvious. So, at 3000 ppi you get approx. 65 megapixel resulting scan. I'm not suggesting this number is some kind of absolute truth but it's what I find practical (sometimes a smaller file, a few times I have used the full resolution of my scanner leading to 600MB files...). If you want comparable results in digital you need a larger sensor than FX / 24x36mm but exactly how large (i.e. what type of MF digital system; is 40MP sufficient, or do you need full frame 645 digital) I cannot say.
    I recently sold my 6x7 camera (Mamiya 7) and got a 24MP FX camera in its place, and while sharpness is satisfactory (for my purposes), the richness of the tonality is not quite there (when comparing to 6x7 black and white film).
     
  26. I would want about 25-30mp of non AA filtered digital. Medium format digital would do the job no problem. It would likely do better too.
    The M9 at 18mp with no AA filter looks very good at 1mX80cm, but perhaps not quite as good as 6x7 or 6x9 film.​
    Stuart, your figures agree very well with mine above: I reckoned 6x9 film at around 25 MP of non AA filtered digital; you say 25-30mp. I put 6x6 film at around my 16.7 non AA filtered digital back; you say that the 18MP non AA filtered M9 is not quite as good as 6x7 [from which I infer that your 18 MP is about as good as 6x6]. Perhaps we're onto something.
     
  27. It is commonly stated that a good 12-24MP DSLR image is, for all practical purposes for most people, as good as, or better than, a conventional (silver/"analog") medium format print of the same size up to 16x20, and this is my observation too. There are sometimes differences noted between the two media. The upside of digital is its ease of producing excellent output. I feel obliged to say that I like film too!
     
  28. The lens on my Fujifilm 6x9, although quite good, is admittedly inferior to a comparable lens (approx. 28mm) on a Leica M9, but Ilkka makes a point that I tend to agree with, although without having at hand any images to prove it. The tonality and gradation possible with B&W film and the Fuji are likely better than that which the M9 (or another high megapixel FF 35mm digital camera) can produce, even if resolution appears similar at 16 x 20 or 20 x 24 (cropped) prints. One should look at prints to gauge it. Color may be another thing, but I am hesitating to trade up from an M8 to an M9 (might have to sell something to do it), as much of my work is in B&W and I am less critical when it comes to color digital work. Perhaps if the FF 35mm (a possible higher MP M10?) starts to compare with the better MF digital cameras in terms of tonality and resolution of details I may make the leap.
     
  29. This reminds me of when audiophiles abandoned vinyl in favor of CD's because they thought they heard more detail, etc... Before that they abandoned tube amplifiers in favor of solid state. No one tusted their ears, they believed the "golden ears" and succumbed to marketing. Now there are people willing to spend thousands of dollars on a turntable and tube amplification, etc...
    I love shooting digital, it's convenient and it's good enough and it keeps getting better.
    But if I look at enough digital photos on Flickr, full frame stuff and then look at pictures shot from a Mamiya7 or Pentax67, etc.. there is the "Wow" factor that I never see with digital. Same as listening to music from a CD and then listening to the same on vinyl.
    But a poorly scanned negative is going to result in a poorly scanned photo just as vinyl played on poor equipment is going to sound awful.
    And I would guess that many of the comparisons that people are making are not from the same lens, etc... I'm finding that lenses make a huge difference with digital. Look at images from the Epson RD1 with low megapixel count but usually used with fantastic lenses.
    I have on a calibrated monitor, rangefinderforum.com/forums/printthread.php?t=76371&pp=40. Scroll through pages from there.
    Making judgements based on resolving a crack in a wall is hardly a defining moment. Quality of scan. Different lenses. Someone might use the same test to compare lenses on the same digital camera and see the similar differences. Lot's of detail because there are things missing.
    CD's offer a lot of detail with music because things are missing.
    We look at specs and decide what we want to see or hear. Instead of trusting our ears and eyes.
     
  30. Lots of comparisons of scanned res, MTF and other techie issues...That I do not pretend to know a lot about.

    Strictly from the dimension driven math of this comparison, I figure a 96Mpxl sensor would be the right physical dimensions for a Full Frame 6x9 image area.

    How is that you say?...MF digital backs today have a pixel pitch of about 7 microns (.007mm). The 6X9 image area is 84mm X 56mm, that comes out to 12,000 pixels X 8,000 pixels = 96 Mpxls.
    A nice round number...and not on the market yet at any price!
    Wait... I believe the Iranian's capture of the RQ-170 stealth drone has some mighty fine huge Mpxl optical digital sensors in it for sale to highest bidder right now! :)
     
  31. So show us your "vinyl versus CD" samples Faustin. It seems to me that those people that are posting samples and trusting their eyes are getting unwarranted criticism for doing so, while everyone else is just theorising and making meaningless analogies.
    It's strange that so many professional photographers who spent years working with film were so happy to abandon it so quickly. And strangely, most professional electronic engineers, who had the equipment to closely examine the quality of output of tubes against solid state, were so ready to abandon tubes to the dustbin of history where they belong.
     
  32. Speaking to vinyl, I prefer my VPI table and the LP versions to my Krell CD player and the CD versions. That's just me.
    As to film vs digital, if you're scanning the 6x9 on a flatbed, then most 15mp DSLRs and up will match it in detail. On a good scanner, I agree with about a 30mp figure on print. Ya, you'll see the film having more detail at 100% on screen...but on print, where it counts, the digital will look sharper even though is has less absolute resolution.
    It's pretty amazing what even the Sony A900 and Nikon D3X can do when compared with MF film.
    As to dynamic range, every DSLR on the market has more dynamic range than the best scans of any chrome film available. For color neg, the best DSLR with 12-14 stops can pretty much equal it. For B&W film, processed in dilute concentrations and metered for shadows, one can achieve about 15-18 stops....and no DSLR or digital back can touch it. That said, a 3 exposure bracket on the DSLR and combined to HDR will get you that 15-18 stops.
     
  33. And strangely, most professional electronic engineers, who had the equipment to closely examine the quality of output of tubes against solid state, were so ready to abandon tubes to the dustbin of history where they belong​
    However, many recording musicians and most guitarists keep using valves (tubes) because they prefer the sound.
    Technically, solid state is much better than valve but musically, the valve sound wins out. To me, the same is true with film. It might not be as perfect as digital but I prefer it.
     
  34. Indeed.<br>But don't be to quick to give in on the "technical bit" of the argument when saying you prefer film.<br>Both, because it's by no means so that digital is technically better (whatever "better" may be), and because there's a technical reason too why the (technical) quality of both tubes and film is prefered by many.<br>(And yes, may do not. But that may be because many do not even know tubes, or film.)
     
  35. 155.7 mega-pixels on a 4625.6 square mm sensor
     
  36. First off, tubes are far from having been left in the dustbin. To the contrary, someone needs to get up to speed. Most of your high end amplifiers are tube, not solid state. Vinyl has come back strongly and with the younger generation. In the last few years, tube guitar amplifers have become more and more important. Guitar entusiasts have learned to improve and modify tube circuits. After tubes were sent to the audio dustbin, it was an engineer who reconsidered and decided that people were measuring for the wrong things and weren't listening with their ears. That was probably 25 - 30 years ago and tube popularity has grown over that time. Now the guitar people have started to pile on and some never left. There were some audiophiles and gutiar players that didn't buy into the hype and trusted their ears. When CD's first came out they players were horrible by any standard and still people were tossing their vinyl and turntables.
    When I see postings of the same scene posted for both digital and film and comparisons made I think that we could see the same differences between two film cameras or two digital camers, or two different lenses. While the comparisons might be valid they need to be qualified.
    A few years ago I turned in a Cibachrome that I had printed maybe 25 years ago to a competition and it won. For those that don't remember or are two young a Cibachrome is done in a darkroom using a slide instead of a negative. Less expensive and far more efficient than digital. Of course, everything else turned in were inkjets. It was amazing how many people came up to me wanting to know about the printer and paper that I used. I really hesitated to turn in the Cibachrome because I knew it would so change the playing field and no one knew until after the competition. It used to be uncomplicated and inexpensive. A camera that you would keep for years, some slide film and a relaxed hour or two in the darkroom. Of course, we are now all posting our pictures on the internet and we mostly all have a computer anyway.
    Also when comaring film to digital images when they both come out of a computer is not always a level playing field. So many people are not clued into workflow.
    I know someone locally that has a lab and she does photo work for photographers all over the country and she demonstrated to me the differences on a monitor between a quality film scan and a digital image. It was easy to see how much more film could be enlarged without falling apart. This where I learned the importance of good workflow. The scanner is one extra element in the mix. The quality of my work improved dramatically when I got this all together. All the good equipment in the world doesn't assure anyone of quality work. There will always be someone out there that will produce better images with an Olympus XA than someone with the most expensive full frame digital camera and two thousand dollar lens.
    That said, I'm into the digital thing, considering the Sony A850 to use with some of the old Pentax M42 lenses that I collected years ago. Though I wish someone would come out with something in digital like the Hexar AF. Instead of the GW690III I would consider the Mamiya 6 or 7. I think for medium format film there are some affordable choices. The Fuji GA645 would be something simple and inexpensive to consider.
    I didn't mean to be critical toward anyone that wants to see things through the technical side, whatever works.
     
  37. Just one last thing :)
    In my example comparison on p2, I think I made my detail crops a little bit too big because they are the size of a 60x40in print. I think that might be bigger than John Dowle might like to print (but of course, I don't know)
    I tried reducing them in size (to a 30x20in print equivalent) and it's really hard to tell them apart from a viewing distance of 18in. This magnification also reduces the appearance of grain in the film image to a more pleasing amount too.
    So, if you want to experiment, download the little detail images, reduce the size by 50% and see for yourself what a 30x20in print would look like.
     
  38. I just bought a 24mp digital camera and the image quality is clearly not twice as good as its predecessor, which was 12mp.
    That's only because you did the math wrong. In order to obtain twice the resolution, you have to double both X and Y axes, not total resolution. So you'd need something like 48 mp to double resolution from 12 mp.
     
  39. Still not seeing any side-by-side examples from you filmophiles! We can all make spurious claims with nothing to back them up, and Faustin, there will always be believers of those spurious claims that will spend a fortune on an Emperor's tailoring.
     
  40. Most people who use at least a Coolscan 9000 or a Flexitight X5 scanner and actually print large routinely can confirm that 35mm film (I use Velvia or Tmax) gives you similar or higher detail (visible) than the highest end DSLR today. 6x7 or 6x9 film will give you at least five tome the detail as a top end DSLR.
    I own and use:
    - Scanner: Coolscan 9000
    - Printers: 44"-Epson 9890, 24"Epson-7880, 17"-Epson 3800
    - Cameras: Mamiya 7II and Mamiya RZ67II
    - Most used papers: Epson USFA, Epson VFA, Epson EW Canvas Satin, Ilford Fibre Gold
    I have posted plenty of test and anyone interested can email me to request them.
    Those who argue that MF film and glass do not provide several orders of magnitude of the detailed captured by an AA 35mm DSLR either have a problem with their camera, their lenses, their scanner or their printer.
    Grain can be removed from scans if desired the same as a camera removes digital noise. Sharpening can be applied at will.
    DSLRs have higher acutance (contrast at pixel level) and abrupt detail extinction.
     
  41. The vast majority of landscapes will contain high frequency of medium/high contrast detail. (eg, branches, hay, grass, etc).
    A 35mm DSLR (today) is IMO inadequate for prints larger than 16x20 and I print 24x30 and 30x40 routinely.
    A direct answer to the OP is difficult to provide but instead I can give a range from 40MP to 200MP depending on film and scanner.
     
  42. Here I tested 4 lenses and 2 camera systems. This will give you an idea of the resolution you can get from 6x7 lenses, film, and the limitations of my Coolscan 9000.
    00Zid3-423231584.jpg
     
  43. Most people who use at least a Coolscan 9000 or a Flexitight X5 scanner and actually print large routinely can confirm that 35mm film (I use Velvia or Tmax) gives you similar or higher detail (visible) than the highest end DSLR today.
    Those who argue that MF film and glass do not provide several orders of magnitude of the detailed captured by an AA 35mm DSLR either have a problem with their camera, their lenses, their scanner or their printer.​
    Ok, show us your comparison. I did. In fact, I was surprised a little at the results. I'll admit, I don't shoot Velvia, but 35mm Velvia vs. 24mp DSLR? I would be very, very surprised to see a more pleasing large print from 35mm Velvia scan. But I don't really know, show us since you say you have the tools to do so. Thanks!
     
  44. Maruo, a test chart is not a real continuous tone photograph. Show us a real photographic comparison please :) With Digital Capture vs. Film / Scan capture. Yes, I know that if I put film under a microscope, I'll see more detail. But what about the quality of a large print of the same subject by digital and film? That is what we need to see.
     
  45. These are the progressive bottleneck's in my workflow (fromwhich I can directly speak):
    In order of resolving power:
    - On a 6x7 frame, my Mamiya lenses project over 300 megapixels of true detail. (measured on fresnel and 100x microscope)

    - On a 6x7 frame of TMAX there are more 150 megapixels of true detail. (light table and 100x microscope)

    - The Coolscan effectively captures 90 MP (3850 effective resolution). (above)

    - The V500 effectively captures 23 MP (1950 effective resolution). (tested separately)

    - For comparison a 20MP DSLR effectively captures approximately 15MP. (see everywhere ~ dpreview tests)
     
  46. Mauro, you've never used a DSLR have you? I was really surprised to see how much detail a 13mp DSLR captured compared to 6x9 color negative film. But detail isn't everything, I think we need to look more closely at an "overall" "image to noise" ratio.
    I think Maruo, if you're making such large prints routinely, you should seriously look at a MF digital camera. I think you'll like it more than scanned Velvia.
     
  47. Mauro, I use a Coolscan LS 8000 with the same resolution as the LS 9000, and have scanned all types of fine-grained 35mm negative and transparency films at the scanner's full resolution, 4000 d.p.i. Without getting into an argument about the resolution math, if you told me I'd need to make an 11x14 inch, or larger, print, and gave me a choice of shooting a Nikon F6 with the whatever you think is the best film for detail OR shooting a Nikon D3x, I'd choose the D3x any day of the week.
    I've been clear that I believe MF film beats 35mm-style DSLRs for biggish enlargments. However, IMHO, if you're comparing any 35mm piece of film with a FF 24MP chip, the film is going to lose that battle if the end game is detailed big prints.
     
  48. BG, "But what about the quality of a large print of the same subject by digital and film?"
    Here is how I print film large (It is hard to show you the direct prints unless you live near Atlanta).
    00Zidi-423249584.jpg
     
  49. Eric,
    The following test was done with Velvia and a 10MP DSLR. The film was scanned with an Imacon at 8,000dpi and a Coolscan 9000 at 4,000dpi.
    Both the Coolscan scan and the 10MP DSLR shot were upsized to match the size of the 8,000 dpi scan.
    Note that a 24MP DSLR will give you aprox 50% more linear resolution than a 10MP DSLR.
    00Zidr-423251584.jpg
     
  50. Also using the DSLR and 35 mm Velvia I shot an assortment similar of objects. Velvia was scanned with the Coolscan.
    00Zidy-423255584.jpg
     
  51. And 100% crops.
    00Zidz-423255684.jpg
     
  52. Now, if possible, would you post your side by side examples that made you arrive to different conclusions?
     
  53. Thank you Maurio. There is far too often a workflow issue. People are sloppy with workflow. I thought just by hooking a good scanner and printer was going to automatically get me good results, it's not going to happen. I went to one of the best people out there and paid for a lesson on good work flow and it made all the difference. She was able to show me on her monitor the differences between a good film scan and digital. When she started enlarging the images the film clearly was better.
    We can take an assortment of digital cameras, from a full frame down to a point and shoot. Someone is going to like the image from the point and shoot better. There are photographers that would always favor a Holga image over any digital image and that's easy for me to understand.
    I have a friend who likes digital prints because to her they look like lithographs where digital colors are screened. She isn't into the subtle transitions between colors that she sees from film.
    Look at images from the Epson RD1, 6.1 megapixels. It leaves me amazed. Almost always used with a fantastic lens.
    And if you want to get the digital crowd excited, including the scanned film crowd, have them gather around some large prints from film done in a darkroom. Find some well done Cibachromes from 35mm Velvia or Kodachrome, it's another world.
    It's not the image, it's what we do with the image afterwards. For some of us the first and last viewing will be on the LCD screen on the back of our camera.
     
  54. Mauro, sorry, I didn't spell your name correctly in my previous response.
    BG, John never asked to compare DSLR to a scanned negative, Velvia or otherwise. If you can find a good Cibachrome produced from Velvia and compare it to a really good image printed from a DSLR.
    Someone correct me if I am wrong but does the Fuji = megapixels or is John asking about the 6 x 9 film negative or positive compared to a full frame digital image or otherwise? Or is the question from John not qualified correctly? I think there has been enough written about the equivalent megapixels from various films, discounting a scan. Perhaps a question to be answered are the potential megapixels for a scanned negative or positive compared to the best that a DSLR can produce? It's not all about megapixels but everything else being equal, the lens etc. John has narrowed it down to megapixels and is equating megapixels to quality. Would then 6 megapixels be enough or 10 or 25 or more, given the full potential of film? There are a lot of other factors that go into a quality image, lens, etc... Given that those other things are satisfied, at what point do more megapixels not matter? Can we compare a 12 megapixel point and shoot to a 6.1 megapixel Epson RD1?
    Eric, I understand, it depends on the end game.
     
  55. Pretty much the opposite of my results. This is an 8000dpi scan of a Kodachrome 64 slide taken on a Leica M7 with a 75mm APO summicron at f/5.6. From the same tripod, I took the same photo with the same lens at the same aperture on a Leica M9. The M9 has significantly more detail. I did not want to believe it, but I have done the test many times with many different cameras and situations by now. I shoot both film and digital side by side and have no bias for one or the other. They are both excellent and wonderful ways to take pictures. But, the real world resolving power per unit area is much higher now on digital, at least on the equipment I have been using. Please click the links to see the actual comparison sizes...photo.net compresses them down to smaller sizes.
    http://stuartrichardson.com/kodachrome-vs-m9.jpg
    [​IMG]
    Here is another test, this time comparing a 10 megapixel camera to E100G scanned on an Imacon 646 at 6300 dpi:
    [​IMG]
    [​IMG]
    The inline images don't seem to be working too well, so I will just post the links:
    Digital 100%
    http://stuartrichardson.com/hvalfjordur-dmr-crop.jpg
    Film scan at same visual size (pixel dimensions are larger, but there is no more detail)
    [​IMG]
    Or finally, there was the time I compared a 22mp digital medium format back to 6x7 Fuji Acros:
    http://stuartrichardson.com/fuji-acros-whole-image.jpg
    http://stuartrichardson.com/fuji-acros-100crop.jpg
    [​IMG]
    http://stuartrichardson.com/22mp-whole-image.jpg
    http://stuartrichardson.com/22mp-100crop.jpg
    [​IMG]
    Again, I am not saying film is worse, and indeed, it looks better in several of these images. It also enlarges better than digital in certain cases due to the way that it is sharpness falls off (gradually, whereas digital falls off like a cliff). But, in terms of perceived detail and sharpness, assuming equal imaging area, digital won that battle awhile ago. So I will stand by my 25-30mp of non-AA filtered digital with a good lens will be roughly equivalent to a 6x9 image scanned on the best scanners.
     
  56. I decided to do a quick statistical analysis of the estimates that were offered. Some people gave estimates of the pixel count a DSLR would need to have to equal the resolution of a 6x 9 film camera, which was what John Dowle wanted to know. There were also some estimates of the scanner resolution needed to do justice to a 6 x 9 film. I debated whether to included those (I can see how they could be relevant). But I decided to stick to the ones that more directly addressed the original question. A couple of the responses were stated as a range of MP values, or as a possible minimum. John Crowe gave a range of 80 to 120MP. I used the 120MP figure since he seemed to feel that it could take that much. And Ilkka suggested a minimum of 40MP or more. I used 40. With that in mind, the data points were:
    Roy Butler: 40MP
    Rob F: 75MP
    Rodeo Joe: 3000 dpi (13.5MP, I believe)
    BG 13MP
    John Crowe: 120MP
    John Thurston: 96MP
    Dave Luttman: 15MP
    John Narsuitus: 155.7MP
    Ilkka: 40MP
    Some of these were arrived at mathematically, and others perhaps by subjective impressions of quality. As far as I am concerned, both approaches count. Although one person's estimate may or may not prove anything, I think the impressions of many have a way of zeroing in on the truth. So I felt that a statistical analysis of all these impressions and calculations might point to a result that is in the correct ballpark. So with that in mand, here goes:
    n=9
    mean (average) =63.1MP
    median (midpoint) =84.4MP
    standard deviation=48.7MP
    Two of the data points were "outliers," meaning that they were pretty far outside the range of the other data offered. These were John Crowe at 120MP, and John N. at 155.7MP. I'll do a second run, using John C.'s 80MP estimate, and leaving out the 155.7 data point. That may tighten things up a bit:
    n=8
    mean=46.6MP
    median=54.1MP
    standard deviation=31MP
    Leaving out the outliers doesn't make it more accurate: how do I know they weren't the right answer? It's just to see how the results might look that way.
    These results don't "prove" anything (before someone asks). The standard deviation is large compared to the mean, in other words, the answers are very spread out (from 13MP to 155MP). So they don't strongly cluster in a central region. But if we could know the "right" answer, perhaps it would lie somewhere in a range from around 46MP to 63MP. I would really need more than the 8 or 9 data points to get a stable, reproducible result. 60 or 80 would be nice. Anyone want to get some more people in on this?
     
  57. Stuart, a couple questions on your samples.
    Kodachrome: what 8,000dpi scanner did you use? could you post a link to the scan (or the crop) at 8,000dpi without reducing its size?
    E100G: could you post a link to the crop as it came from the scan without reducing its size?
    Fuji Acros: what scanner did you use?
     
  58. Rob, thank you for the stats on opinions. If you reduce the sample to those who have posted examples or have actual direct experience you should arrive to the only objective and valid answer:
    1- 6x9 film outresolves every single MF scanner named across this entire thread.
    2- Because of 1, people are not discussing 6x7 or 6x9 film. They are just discussing their scanners.
    Specific to the numbers:
    - My Coolscan gives me aprox 90 megapixels of true measured detail (lower than its 100 pm full RGB nominal capture - because of the scanner not the film).
    - A DSLR gives you aprox 70% of its nominal resolution in true detail (although the color is interpolated).
    -----> 90 / 70 % = 128 MP required from a DSLR to match 6x7 film (and this is a 128MP medium format SLR due to lenses and no-AA requirements).
     
  59. So John, after all these comments, what do you think?
     
  60. You can get 50 megapixel with this Fujifilm gsw690III.take a look to this picture.
    00Zimn-423399584.jpg
     
  61. Fujifilm gsw690III,50 megapixel camera.
    00Zimq-423401584.jpg
     
  62. Rob, I never said 15mp is the best that film could do with a good scan.....I said that was about as good as you get from scanning with a V700 Epson. A good scan of 6x9 would take in the area of 35 to 40mp to equal.
     
  63. depends on the film, sensor, and lenses.
    if we ignore the lens-or imagine that the imaging item (film or sensor) is looking through essentially a perfect lens, then we have just the two left to compare.
    http://www.cambridgeincolour.com/tutorials/camera-sensors.htm
    instead of copy pasting and stealing their work, I suggest reading that.
    nearly every camera sensor works that way- foveon (i think that's who it is) sensors work more like film, in that they layer three colors in each sensor site.
    however, foveon (if it is them) does exaggerate their resolution a bit, but their sensor is closer to how film works.
    film? well, i'll choose my favorite film to compare-velvia 50 4x5.
    transparency film generally has a higher resolution than regular film, and lower speeds have more fine detail.
    well, if you look through all the data and resolution of that particular film, you do need a metric ton of megapixels to get equal quality (defined here as just how fine of a detail it can register, nothing more, nothing less)
    standard digital isn't at the level of 4x5 yet-probably is approaching medium format though-I haven't really stayed on top of the latest digital information.
    4x5 scan backs can likely be very close to film, if not equal to film in pure resolution, but cost such an extraordinary amount, and are very slow to use.
    to answer your question (which i guess I wasn't really) it really depends on what film you are usng and what lens.
    but i would guesstimate that for a good transparency film and a good lens, perhaps an 80mp or greater back would be roughly equivalent? I think sinar makes one like that.
     
  64. I did some pretty extensive testing a couple years back. I found my 7D (18 MP) pretty much matched 35mm Velvia 50, scanned on an Imacon, on high contrast detail, and was a bit better on low contrast detail. To match or exceed 6x7 Velvia 50 I found it takes a 3 frame stitch from the same 7D.
    Of course Velvia 50 is considerably higher resolution than most films. Still, this is the same ball park as reported by others here.
    In print I don't think it matters up through about 24".
     
  65. Mauro -- the Kodachrome scan was made on a Hasselblad X5, the second photo was scanned on an Imacon 646, and the Acros one was also scanned on an X5. I have been using the Imacon and Hasselblad scanners for approximately 10 years without interruption. I also earn a living printing exhibitions for people, so I see all different film formats and digital cameras. I am not saying that makes me right by default, but it does mean that I have a lot of experience looking at all sorts of digital and film originals and seeing big prints from them.
    For example, I just printed 15 2 meter long panoramas from 6x17 for an exhibition at a Norwegian gallery. There is nothing I have seen out there from digital that can equal this in an instant capture. There is of course stitching, but that is another matter.
    In any case, I will try to find the files you mention, but currently my lab is under construction, so I have limited access to the computer with all the actual files in it, and it does not have the internet at the moment. I would ask that you take my word for it that there is no more detail to be had, just more pixels of the same softness, but I understand how you need to see for yourself.
     
  66. Wow, thanks to everyone for your replies to my question.
    I'm new to MF photography and have a lot to learn and obviously a lot to learn about scanning MF negs which to me after reading all the replies several times over seems to be a very important part of MF photography when wanting to print big. I think that when I eventually get photographs that I want printing large I will be sending them to a Pro to be scanned properly after reading all the information in the replies.
    Again many thanks.
    John.
     
  67. John,<br><br>Learning about scanning is only an important part of MF photography when wanting to print big if you want to insert a redundant digital stage into the process, instead of printing your MF negatives directly.
     
  68. just how many mega-pixels in a modern DSLR do you think would give the equivalent quality​
    Quality is not determined by megapixels. Many other parameters affect the quality like: colors, WB, noise,...One important factor is the lens and how to get correct focus but the 35mm DSLRs use different lenses from the lens of your Fuji
     
  69. Some of these film vs. Digital comparisons are the same ones from a few years ago. The DSLRs are old models and are not fair comparisons as the technology is constantly changing. So here is a link for a comparison of a Canon 5Dii against against film. Now in a year there should be a new version of the 5Dii which will mean this comparison will have to be done again. LOL
    http://www.twinlenslife.com/2011/01/digital-vs-film-canon-5d-mark-ii-vs.html
     
  70. For reference here's a drum scan from a 645 frame at about 4500 dpi rez (9975 x 7504 pixels = approx 75 Mpix). This is from a Pentax 645, 55mm f2.8, Provia 100F. You should be able to double this for a 6x9 frame I would think. ~1/10 crop to follow...
     
  71. For reference here's a drum scan from a 645 frame at about 4500 dpi rez (9975 x 7504 pixels = approx 75 Mpix). This is from a Pentax 645, 55mm f2.8, Provia 100F. You should be able to double this for a 6x9 frame I would think. ~1/10 crop to follow...
    00Zixv-423585584.jpg
     
  72. Crop from 645 Frame.
    00Ziy0-423587584.jpg
     
  73. Thank you Stuart. I look forward to seeing the files if you can locate them.
     
  74. For reference here's a drum scan from a 645 frame at about 4500 dpi rez (9975 x 7504 pixels = approx 75 Mpix). This is from a Pentax 645, 55mm f2.8, Provia 100F. You should be able to double this for a 6x9 frame I would think. ~1/10 crop to follow...
    Crop from 645 Frame.​
    Here is Donald's crop, resized to 50% to halve the amount of (potential) information per dimension, or quarter it in total; saved and closed; then reopened and resized to 200% to restore it to the original size (948 x 710 pixels - 1 pixel was lost in each dimension because the original had odd numbers).
    There is no discernable difference in the detail. All the little stem hairs and leaf structures look the same - the only detail to come out poorer is the dust/hair artefact, which was not part of the image captured on film by the camera. This illustrates the difference between having lots of scanned pixels, and having lots of actual information on the film. It's merely unnecessary oversampling.
    So his 75 Mpix image can be represented in 1/4 the scan size (approx 19 Mpix). Double that 645 area for 6x9, and you get 38 MPix. We are back at the 35-40 Mpix that we had estimated above!
    (The full frame of Donald's photo is a lovely image, BTW, and a reminder that resolution is certainly far from being everything!)
    00Zj9I-423781584.jpg
     
  75. I have a sunflower picture too just like that one.
    00ZjBG-423813584.jpg
     
  76. crop from 5000 pixel high file.
    00ZjBH-423815584.jpg
     
  77. I'm seeing a lot of content drift away from the original question of the resolution a modern DSLR would need to compete with a 6 x 9 film camera for image quality. Many posts are, instead, about file sizes generated by scanning a 6 X 9 film.
    That's fine.
    But I think we should be aware that there are really two discussions going on. And that the MP figures reported for scanning film are a lot higher than reasonable estimates for how much a DSLR really needs. That's because it's easy to get such high numbers when scanning. As this thread moves along, we ought to be aware of the direction in which the thread is going, and how much our answers have to do with the original question.
     
  78. And also be aware that "the MP figures reported for scanning film are a lot higher than reasonable estimates for how much a DSLR really needs" is not a neutral observation, but one of usual the positions in the DSLRs-are-all-you-need 'debate' that has been raging ever since the first 2 MP DSLRs appeared.
    (And with that, we should also be aware that, since that claim has indeed been made since those first 2 MP cameras appeared, it's awfully hard to know when such a debate will change from the usual reciting of the DigiDo Mantra into something that has to be given serious attention.)

    The thing is, Rob, that while scanning can indeed produce high pixelcounts, what matters is whether those pixelcounts represent something more than just the number of sensor elements on the scan-bar. Showing and comparing results trying to answer that, and is not topic drift. If we would find the number of scanned pixels per area unit where the transition from true info to empty pixels would occur, we will also get close to knowing (there are other issues to deal with too) how many MP a DSLR would need to equal the results produced by 6x9 film cameras.
     
  79. Then in order to learn something from scanning 6 x 9 films that can help answer the question of how large a pixel count a DSLR needs to equal a 6 x 9, we might say that we need the largest pixel count that still reveals additional detail in a very fine-grain film, such as Velvia. Ray Butler, above, suggests that 35 to 40MP ought to do it. And he bases this on an experiment, not just on armchair analysis.
    I can only think of one thing wrong with that approach. Scanning a film combines two media that are both resolution-limited. I believe that the combined resolution of two media is given by:
    (R1 x R2)/(R1 + R2)
    where R1 is the resolution of medium 1 (in this case, the digital process);
    and R2 is the resolution of medium 2 (the film, in this case)
    So if for example, the film can resolve (say) 100 lines per millimeter, and the digital process can resolve 100 lines per millimeter, then:
    (100 x 100) / (100 + 100) = 10,000/200 = 100/2 = 50 lines per millimeter.
    This would suggest that no matter what resolution you need to get the most out of a 6 x 9 film (35 to 40 MP in Ray's experiment), you have not extracted the full detail in that film. This is because the combined resolution of film and the digital copy of that film must always be lower than that of the film alone. You can come as close as possible to extracting the maximum detail if the digital process has much higher resolution than the film. For example, if the film resolves 100 lines per millimeter, and the digital process resolves 1000 lines per millimeter, then we have:
    (100 x 1000)/(100 + 1000) = 100,000/1100 = 90.9 lines per millimeter.
    By selecting a digital resolution ten times that of the film's we get a result that is reasonably close to the film's 100 l/mm. But that is misleading, because you don't need 1000 lines per millimeter to equal the film; only to copy the detail in a picture already taken with that given film. So does this method (digitizing a piece of film) really address the original question of the resolution of the DSLR needed to equal the film? I think not. I don't think we can answer the question in this way, owing to what I presented above. The formula I offered for combined resolution needs to be reviewed for relevance by others who are more knowledgeable than I in this area. I think it's right, but I will be interested to see what others say.
     
  80. And someone else gave different, non-armchair figures.<br><br>The calculations for system resolution do not apply here at all. You are using a device with a (limited) ability to discern thing's sizes to look at the size of things. The result is not a "system" result.<br>Or can you explain why these calculations would apply?<br>Simplified (not taking things like shape, adhesion and what not into account): if you use a sieve with a 1 mm mesh, you can be sure that things smaller than that will fall through, things bigger will not. Reduce the mesh size until nothing comes through, and you will know the size of the smallest thingies you are sifting.<br>If you want to see at what level the resolved detail in an image peters out, all you have to do is look at it with something able to see detail smaller than the smallest detail present in the thing you are looking at. If nothing more reveals itself when looking at even smaller detail you will know what you want to know.<br>Scanning with a 1000 lp/mm resolution will be fine enough, of course, to see that detail in a given film image is not finer than 100 lp/mm. It does not reduce that 100 lp/mm to 90 lp/mm. It does not reduce your inspection instrument's ability to see detail at the 100 lp/mm level.
     
  81. The calculations for system resolution do not apply here at all. You are using a device with a (limited) ability to discern thing's sizes to look at the size of things. The result is not a "system" result.
    Or can you explain why these calculations would apply?​
    I'd like to see a rigorous explanation to support the position that the combination of film and digital is not a system. Remember, the object is to determine the digital resolution needed to capture the same scene at the same level of detail as a 6 x 9 film can do. If we take the picture with a digital camera, that's not a system (for our present purposes). If we take the picture with a film camera, that's not a system. (Or, if you prefer, it is not a cascading of two systems.) But when we take a digital "picture" (scan) of an analog picture, now we have a system (or system-on-system) whose performance is the composite of both elements of the system; and will be inferior to either one alone.
    If we want to know how many megapixels are beneficial to get the most out of a piece of film, or to measure the size of the smallest elements on that film, then certainly the method of successive sieves you suggested above is not merely appropriate; it is an elegant solution! But if we want to know what it takes to capture the original scene with a digital camera, without an analog intermediary (because this was not the original question), we'll have to take the picture twice: once with a digital camera, and once with that (now familiar) 6 x 9. And then take a look. We are certainly on good ground to attempt a theoretical estimate by calculation; but no theory is in good standing until there has been some experimental verification.
    If I had a 6 x 9 camera, and a digital camera that had a whole bunch of pixels,I would do it myself. But my D700 has only 12MP, and my only MF is a Hasselblad, which is 6 x 6. I'd welcome some more examples from anyone who may have such examples on hand, or can do the experiment.
    A final thought: it is possible that we are asking a bad question that has no correct answer. One might ask how high the sampling rate has to be before a CD can sound like a vinyl record. But it just might be that a CD is not going to sound like a vinyl record, no matter what we do to it; and a digital picture might never look just like an analog one, no matter what. (Just a moment of doubt.)
     
  82. The explanation, Rob, is that you are not using the scanner to image the detail as present in an original scene through the medium of a lens and film, or anything like that. We are not trying to figure out the ratio between that input's (the scene's) detail and how much there is left on it on film, needing to sum the limiting effects of lens and film through such calculations.
    Though "the object is to determine the digital resolution needed to capture the same scene at the same level of detail as a 6 x 9 film can do" all we are doing is to zoom in on a bit of film to see what scan resolution would still make sense, and what (higher) scan resolution would only produce more pixels but not reveal more detail/differences present in the subject (the bit of film). The only mediator, the only limiting factor is the scanner. Not the bit to be examined, i.e. the film. No system.

    Should we establish that at a scan resolution of, say, 6000 ppi the extra pixels are (apart from differences caused by noise, etc.) the pixes are mere duplicates of what we alraedy had at lower scan resolutions, i.e. not reveal more detail, we could say that that would be already beyond the MP limit of what a bit of film can deliver. Going backwards we could establish the exact MP number that corresponds with the 'true' image content of the film image.
    And that would be part of the object as stated. (All else involved, such as lenses, taking technique, etc. assumed to be equal, of course.)
    The other part is to make people understand that - thanks to Bayer patterns, anti-alias filters (which in itself is an alias: they used to be called soft focus filters) and the like - a, say, 22 MP camera does not capture 22 MP worth of true image content. So if a bit of film would show to hold only 22 MP, a 22 MP camera would not be enough to equal it.

    I don't think it is a bad question, without an answer. Just that the history of it, which it drags along with it, is not a good one.
    Like i said before (and it's not a joke, not an exageration, or anything like that), ever since the first 2 MP cameras were thrown at us, people have been claiming they were as good as film.
    And ever since then, people have vehemently defended the 'truth' of that claim, using all sorts of calculations and concepts borrowed from elsewhere, whether they apply or not, and redefined or invented terms (like that anti-alias filter thing, or telecentric, or - a personal favourite - optimized for digital) to shroud inconvenient truths or make them sound less bad.
    That has all become a bit of a tradition, with people repeating things religiously instead of thinking about what it's all about.
    And that makes it a difficult question. But not one that cannot be answered.

    But of course there's so much more than pixel count. Whether we like film or digital better is something that really can't be determined by measurements and/or calculations. There are still answers though. Correct ones too. But what is correct, and why to each his own
     
  83. Well, I think we have both said what there is to say about our respective approaches to this. I'd like to get both our approaches "peer reviewed," so to speak, by sitting back now and waiting to see what others who have given thought to this matter have to say. It seems to me that I want to answer the original question by comparing pictures taken independently with film, and then again with digital; while Q.G. wants to use the digital process as a an instrument to measure what film can do, when expressed in digital terms.
    Maybe both ideas have merit. Let's see what others say.
     
  84. Yes, let's.

    But not before we put one thing straight: "comparing pictures taken independently with film, and then again with digital" as happens here, and other places on the internet, is using "the digital process as a an instrument to measure what film can do, when expressed in digital terms".

    The tricky bit in the original question ("how many mega-pixels in a modern DSLR do you think would give the equivalent quality that you get from [film]") is in the "equivalent quality" part of it. What would that mean exactly?
    The original question suggests that matching the quality is a matter of the right number of megapixel. And takens as such, using "the digital process as a an instrument to measure what film can do, when expressed in digital terms" is the only way to settle the question.
     
  85. And takens as such, using "the digital process as a an instrument to measure what film can do, when expressed in digital terms" is the only way to settle the question.​
    I don't agree with this (and you are welcome to disagree with my disagreement!). For me, photography is about prints. The only valid way to make a comparison is to compare a digitally capture image printed digitally and a film derived image printed optically. Otherwise it's just a digital vs. scanner comparison with the scanner being the weakest link.
     
  86. You know, i don't disagree, Steve.
    It's a perfectly valid way to fill in what "equivalent quality" could mean: looks as good as in a print.
    And for that we need indeed not worry about the number of MP you could eke out of film (which - i believe - is a rather difficult question anyway, if you do not just look at the "resolved and recorded information" content of an image).
    So i was wrong.
     
  87. A couple of things possibly worth reading:
    http://toyotadesigner.wordpress.com/tag/resolution-mf-slides/
    2007-04-vitale-filmgrain_resolution.pdf
     
  88. I've done an extensive test of the Mamiya 7 vs various digital - Canon 5Dmk2, Nikon D3X, Phase IQ180 - and the results showed that in terms of clear resolution, the Mamiya 7 resolved more lines than the IQ180 but when you looked at overall rendering it was somewhere between the DSLR's and the IQ180. From this I would estimate that you would need about 40-50 megapixels to emulate T-Max or Velvia 100F and about 30-40 to emulate Portra 160/400.
    http://static.timparkin.co.uk/static/tmp/cameratest-2/800px.html
    The one thing that came out of this is that digital files can take more sharpening and can look sharper than film but at a certain level of enlargement they start to look wrong (about 240 dpi on close inspection) whereas film has fine detail beyond this. So for large enlargement factors film can look better as long as you don't mind grain. Adding grain to digital helps to a point but it still breaks down at a certain level. We saw this with the IQ180 which looked sharper than 5x4 until you got to 40x50 inch prints at which point 5x4 looked better.
    Here's a direct comparison of the IQ180, Portra 160, T-Max and a Nikon D3X
    http://bit.ly/Nwzvzp
     

Share This Page