 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
| PhotoKB HomeDiscussion Groups Digital Photography Digital PhotoDSLR CamerasZLR CamerasPoint & Shoot CamerasFilm Photography 35 mmLarge FormatMedium formatDarkroomFilm and LabsOther EquipmentPhoto Technique Nature PhotographyPeople PhotographyTechnique GeneralGeneral Photo Topics General TopicsAustralian PhotographyUK PhotographyDirectoryPhoto Clubs |
Photo Forum / Film Photography / 35 mm / October 2006Question About Film Scanning Resolution
I recently purchased a film scanner (Pacific Image PF3650Pro3) which has a maximum "Optical Resolution" of 3600 ppi. When I scan at that resolution I end up with image files whose property sheets say that they are 5148 x 3420. I would like to know how it is that I appear to be getting higher resolution than the scanner is capable of delivering. Is my film scanner creating scans equal to 17.6 MP (5148 x 3420)???? I thought that my images would be 3600 x 2400, based on a 3:2 aspect ratio. So how am I getting 50% more ppi than that? If someone could help me understand these numbers I'd appreciate it. > I recently purchased a film scanner (Pacific Image PF3650Pro3) which > has a maximum "Optical Resolution" of 3600 ppi. [quoted text clipped - 10 lines] > > If someone could help me understand these numbers I'd appreciate it. You're gonna hate yourself when I say this, Norman... ;-) Resolution is at pixels per *inch*. You're scanning something that is 36 x 24mm, or (roughly) 1.4 x 1 inches So, using the file measurements you gave above, your actual scanning area measured 1.43 x 0.95 inches, or 36.32 x 24.13mm. Which is just fine. - Al.  SignatureTo reply, insert dash in address to match domain below Online photo gallery at www.wading-in.net "Al Denelsbeck" <news@wadingin.net> wrote in message > Resolution is at pixels per *inch*. You're scanning something that is > 36 x 24mm, or (roughly) 1.4 x 1 inches > > So, using the file measurements you gave above, your actual scanning > area measured 1.43 x 0.95 inches, or 36.32 x 24.13mm. Which is just fine. Great! Thanks for clarifying that for me. Now, about the 17 MP, am I getting images that are approximately equivalent to what a 17MP digital camera would produce? Frankly, I don't see any margin of superiority over digital, but this could be because I'm scanning negatives from 20 years ago. I've seen articles claiming that a film scan is equivalent to 20, 25 even 35MP digital, but I can't see anything so great when I view my digital scans at, say, 300% zoom on screen. > "Al Denelsbeck" <news@wadingin.net> wrote in message > [quoted text clipped - 12 lines] > is equivalent to 20, 25 even 35MP digital, but I can't see anything so great > when I view my digital scans at, say, 300% zoom on screen. Hi. First of all, Your scan is a second generation image but a digital image is a first generation image. Every time You reproduce the image the quality suffers, this is inevitable. There are also other things, You need a good lens and good technique to use all the quality of a good slide film. Scanning will not improve on image quality, on the contrary. Also not all scanners are equal. Scanner resolution may be high but image quality may be something else. Scanner lens and light are important, also how You scan. Scanning programs are not equal, You may get much better results with program B. After all that, I have a Nikon Coolscan V ED I am very, very happy with. See, for example http://shutterbug.com/equipmentreviews/scanners_printers/0504sb_nikons/ 35 mm scan versus digital first generation image.. hmm. The amount of information 35 mm slide includes might be about 20 MP or 15 MP or 10 MP. But final image quality is something else. You might wish to visit http://www.clarkvision.com/imagedetail/scandetail.html or http://www.clarkvision.com/imagedetail/film.vs.6mpxl.digital.html . Clark says, among other things: "Digital 9.8 Mpixels, monochrome: Intensity detail similar to or slightly better than 35mm film detail. Color detail still lacking film." And "Digital 17 Mpixels, monochrome: The intensity detail is now clearly better than the 35mm film, but the color detail is just beginning to be comparable." Väinö Louekari > "Al Denelsbeck" <news@wadingin.net> wrote in message >> [quoted text clipped - 16 lines] > but I can't see anything so great when I view my digital scans at, > say, 300% zoom on screen. Digital camera images are pretty much an entirely different thing, and usually appear sharper. This is because camera pixels are unaffected by any neighboring pixel (usually), and thus can have sharper contrast differences between them. This is what we see as sharpness and resolution. Futher, the camera software often does some interpolation to produce the finished product. Film, on the other hand, produces a chemical response through three or more overlapping layers of emulsion. And they respond in a linear manner - rather than "on" or "off", there is progression of exposure. It's a bit like the light that a flashlight throws: The beam may be brightest right in the focused center, but there's still some residual light that drops off outside of that. So film emulsion has a progression of change to all details, and the goal is to keep this progression as rapid or narrow as possible, producing a higher level of contrast and what we perceive as "sharpness." At high magnification, digital looks better, true enough. This is great if you really need to see the individual pixels. Now, I don't know about you, but I can't say that's ever done anything for my photos - people generally want to see the subject and how it's portrayed, not what it looks like through a microscope ;-) And if you want to do some post-scan software tweaks to alter the contrast closer to a digital file, go for it. But otherwise I would suggest not worrying about it too much and concentrate instead on the photo itself - composition, lighting, timing, and so on. Don't get suckered in by the pursuit of statistics and specifications - those don't make the image. I'm speaking a bit of heresy in this day and age, but then, I like to live dangerously ;-) By the way, viewing the image at 300% won't really tell you much - anything above 100% doesn't reveal the detail any better, and all you're doing is telling the computer to make one pixel go across multiple dots of your monitor. It does have its uses when doing touchups, since you can translate your mouse movements into smaller and finer increments of the image, but that's about it. Hope this helps. Good luck! - Al. SignatureTo reply, insert dash in address to match domain below Online photo gallery at www.wading-in.net > "Al Denelsbeck" <news@wadingin.net> wrote in message > [quoted text clipped - 12 lines] > is equivalent to 20, 25 even 35MP digital, but I can't see anything so great > when I view my digital scans at, say, 300% zoom on screen. Digital camera noise is in one axis: color Film noise is in three axis': color, x and y. To get high res from film, the original image has to be from high quality, sharp glass, perfectly focused, at the sharpest aperture, high shutter speed, perfectly isolated camera, mirror lockup, in high contrast lighting (preferably flash) with a sharp, highly detailed subject. Better yet, hold the shutter open with a cable release in a perfectly dark room and expose with flash only, then let the shutter close. Only deliberate efforts to bring all of these elements together at one time will get you that resolution at scan time. Alan. Signature-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm -- [SI] gallery & rulz: http://www.pbase.com/shootin -- e-meil: Remove FreeLunch. > Digital camera noise is in one axis: color > Film noise is in three axis': color, x and y. [quoted text clipped - 8 lines] > Only deliberate efforts to bring all of these elements together at one > time will get you that resolution at scan time. So what is your theory as to why you have to go through all of that with film when with digital you can just snap the photo? My 350D has a pixel pitch of 3954 ppi, almost exactly what a 4000 ppi film scanner has. So pretty much we can compare directly a shot from my camera with a scan of film, I will get a cropped view due to the smaller image but the center part of the image we are comparing apples to apples. So this was shot in poor light at 1/250 second with a $70 lens hand held. http://www.pbase.com/konascott/image/68928171/original And yet that has more detail then any 4000 ppi scan from film I have ever seen. Scott >So this was shot in poor light at 1/250 second with a $70 lens hand >held. >http://www.pbase.com/konascott/image/68928171/original > >And yet that has more detail then any 4000 ppi scan from film I have >ever seen. Ah, but only 38% of the area of a 35 mm frame. We agree of course -- per unit area, silicon makes a much better sensor than film. rafe b www.terrapinphoto.com > >So this was shot in poor light at 1/250 second with a $70 lens hand > >held. [quoted text clipped - 7 lines] > We agree of course -- per unit area, silicon makes > a much better sensor than film. And the point I was making was that with a DSLR I don't have to do all of the rather extreme measures that Alan outlined as been needed to get a good sharp image on film. I can just hold my camera by hand and use a cheap lens and still get an image that is sharper then film can for the same area. Sure I have less of the frame covered with my sensor but that should not matter when talking about hand holding the camera or how good the lens is. Scott >>>So this was shot in poor light at 1/250 second with a $70 lens hand >>>held. [quoted text clipped - 17 lines] > > Scott It is not quite that simple. With a bad lens you get a low-quality image to begin with. Capturing it digitally does not improve it. Good film or good digital can handle more than a bad lens can deliver. Väinö Louekari >> It is not quite that simple. With a bad lens you get a low-quality image > to begin with. Capturing it digitally does not improve it. Good film or > good digital can handle more than a bad lens can deliver. But then I did not say I used a bad lens, I said it used a cheap lens. Canons 50mm 1.8 may only cost $70 but it is pretty darn sharp none the less. Scott >>>It is not quite that simple. With a bad lens you get a low-quality image >> [quoted text clipped - 6 lines] > > Scott You are right in that, of course. Someone might think that cheap always means bad. And so someone might use a cheap and bad lens thinking that digital capture will somehow make it all right. See my point? Currently I mostly shoot with Pentax 50 mm f1.4 that costs (used) next to nothing and is very, very good indeed. Väinö >>>So this was shot in poor light at 1/250 second with a $70 lens hand >>>held. [quoted text clipped - 15 lines] > sensor but that should not matter when talking about hand holding the > camera or how good the lens is. Of course it does. Look up edgfe sharpness. Signature-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm -- [SI] gallery & rulz: http://www.pbase.com/shootin -- e-meil: Remove FreeLunch. > > Digital camera noise is in one axis: color > [quoted text clipped - 25 lines] > And yet that has more detail then any 4000 ppi scan from film I have > ever seen. What you have seen is a tiny proportion what's been shot and an even tinier proportion of what's been shot at the limits of resolution. Signature-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm -- [SI] gallery & rulz: http://www.pbase.com/shootin -- e-meil: Remove FreeLunch. > > > Digital camera noise is in one axis: color > > [quoted text clipped - 28 lines] > What you have seen is a tiny proportion what's been shot and an even > tinier proportion of what's been shot at the limits of resolution. Well in area it is about 39%. But so what, are you saying that you only need to do all the things you outlined to get the full frame sharp? that if you hand held the shot and used a slower shutter speed and did not use a strobe that the center of the film shot would be sharp but not the edges? Scott >>> > Digital camera noise is in one axis: color >>> [quoted text clipped - 34 lines] > not use a strobe that the center of the film shot would be sharp but > not the edges? No. See my other post. And what I meant is the there are a lot of exceptionally sharp film images that have been made. You haven't seen many of them. Signature-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm -- [SI] gallery & rulz: http://www.pbase.com/shootin -- e-meil: Remove FreeLunch. > No. See my other post. And what I meant is the there are a lot of > exceptionally sharp film images that have been made. You haven't seen > many of them. Well since this thread is about the sharpness of a scanned film image it should be easy for you to show me a link to one of these images. Again I am looking for an image that is sharp when viewing at the pixel level at 4000 ppi. Scott > > Digital camera noise is in one axis: color > [quoted text clipped - 12 lines] > So what is your theory as to why you have to go through all of that > with film when with digital you can just snap the photo? And, I neglected to add, that all of those steps are cumulative in terms of results. Each step resulting in yet finer detail. > So this was shot in poor light at 1/250 second with a $70 lens hand > held. Poor light? What are you talking about? ISO 200 f/9 for 1/250 is just 1.3 stops from sunny-16. That would make it, probably, a mid day shot with high, not very thick cloud cover or lots of low cumulus and open sky with the sun hidden. At that the highlights look washed out, so that 1/3 stop was probably not needed ... more like EV 14. > http://www.pbase.com/konascott/image/68928171/original 1/250s with a 50mm lens is hardly setting up a tough handheld shot. And for that matter, even reduced as shown does not look particularly sharp. Signature-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm -- [SI] gallery & rulz: http://www.pbase.com/shootin -- e-meil: Remove FreeLunch. > Poor light? What are you talking about? ISO 200 f/9 for 1/250 is just > 1.3 stops from sunny-16. That would make it, probably, a mid day shot > with high, not very thick cloud cover or lots of low cumulus and open > sky with the sun hidden. At that the highlights look washed out, so > that 1/3 stop was probably not needed ... more like EV 14. It was diffuse light, which tends to make things look dull. > > http://www.pbase.com/konascott/image/68928171/original > > 1/250s with a 50mm lens is hardly setting up a tough handheld shot. And > for that matter, even reduced as shown does not look particularly sharp. This is not a reduced image, that is a 100% crop would should be compared to a 4000 ppi scan. Sure 1/250 with a 50mm lens is not hard, that is the whole point. You gave a list of requirments for getting a sharp film image that was pretty exterme, like locking up the mirror and using a strobe if posible. So let's see your 100% crop when you have done everything right and see how it compares. Scott >>Poor light? What are you talking about? ISO 200 f/9 for 1/250 is just >>1.3 stops from sunny-16. That would make it, probably, a mid day shot [quoted text clipped - 3 lines] > > It was diffuse light, which tends to make things look dull. But certainly not "poor" light. Difuse light is quality light and in this case sufficiently bright. >> > http://www.pbase.com/konascott/image/68928171/original >> [quoted text clipped - 8 lines] > pretty exterme, like locking up the mirror and using a strobe if > posible. No, I listed the chain to maximize detail. Each step leads to greater and greater detail. Of course that's how you get full detail on film at a level that an 8 Mpix sensor cannot. > So let's see your 100% crop when you have done everything right and see > how it compares. Sure. I probably have several. But I'm not exercizing this mutt for that purpose alone. Next time I have the right film in the Maxxum 9 and I have the studio setup, I'll shoot frames just for this purpose. Signature-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm -- [SI] gallery & rulz: http://www.pbase.com/shootin -- e-meil: Remove FreeLunch. > No, I listed the chain to maximize detail. Each step leads to greater > and greater detail. > > Of course that's how you get full detail on film at a level that an 8 > Mpix sensor cannot. Just to make this a bit easier there is an image from a 0.25 inch square of my sensor. http://www.pbase.com/konascott/image/68975516 BTW this was taken when it was sunny. This is a square that is 989 pixels on a size, about 2% smaller then what a 4000 ppi film scan would be for the same area. This is a useful size because it can be compared to a set of scan samples that Rafe B. has on his site. http://www.terrapinphoto.com/jmdavis/ The lens I am using is the Canon 50mm f/1.8, a lens that fits my film camera just fine. The question is if the only way to get a sharp image with film is to take the step you gave then how is it that this image is sharp and yet it was hand held with no mirror lock up and a used a very cheap lens? Why is it that hand holding a camera will ruin a film shot but not a digital? Scott > > No, I listed the chain to maximize detail. Each step leads to greater > > and greater detail. [quoted text clipped - 22 lines] > Why is it that hand holding a camera will ruin a film shot but not a > digital? Umm, that isn't what Alan said, but it did sound like it was what you were saying (even if, as we surely hope, you didn't mean to). This thread seems to have got very badly lost in the woods somewhere. Peter > Umm, that isn't what Alan said, but it did sound like it was what you were > saying (even if, as we surely hope, you didn't mean to). This thread seems > to have got very badly lost in the woods somewhere. Well just to be clear this is what Alan said "To get high res from film, the original image has to be from high quality, sharp glass, perfectly focused, at the sharpest aperture, high shutter speed, perfectly isolated camera, mirror lockup, in high contrast lighting (preferably flash) with a sharp, highly detailed subject. " Now clearly all of these items that he is saying you have to have to get high res from film don't seem to be needed with a digital sensor. So the questin is why are they needed for film by not digital? Scott > http://www.pbase.com/konascott/image/68928171/original > > Scott Ok, so I'm going to ask a silly off-topic question ... what are those white boxes at the driveway of every house? I've also seen them in other photos of recently-built Hawaiian housing. > > http://www.pbase.com/konascott/image/68928171/original > > [quoted text clipped - 3 lines] > white boxes at the driveway of every house? I've also seen them in other > photos of recently-built Hawaiian housing. Those are the house numbers. He reason to put them up on boxes like that is to give people something to back over with their cars. Scott >>>http://www.pbase.com/konascott/image/68928171/original >>> [quoted text clipped - 6 lines] > Those are the house numbers. He reason to put them up on boxes like > that is to give people something to back over with their cars. I thought they would at least contain junction boxes for underground power, phone, cable access... Signature-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm -- [SI] gallery & rulz: http://www.pbase.com/shootin -- e-meil: Remove FreeLunch. > >>>http://www.pbase.com/konascott/image/68928171/original > >>> [quoted text clipped - 9 lines] > I thought they would at least contain junction boxes for underground > power, phone, cable access... Nope, and this is what they look like when they are backed into, in this case by a cement truck. http://www.pbase.com/konascott/image/68978430 This is the power box http://www.pbase.com/konascott/image/68978610 Note it is shifted a few inches after the earthquake. Scott > This is the power box > http://www.pbase.com/konascott/image/68978610 Almost identical to ours. Signature-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm -- [SI] gallery & rulz: http://www.pbase.com/shootin -- e-meil: Remove FreeLunch. >> >>>http://www.pbase.com/konascott/image/68928171/original >> >>> [quoted text clipped - 14 lines] > this case by a cement truck. > http://www.pbase.com/konascott/image/68978430 It's significant that the truck driver was careful to avoid the fire hydrant........:^) > > > http://www.pbase.com/konascott/image/68928171/original > > > [quoted text clipped - 5 lines] > Those are the house numbers. He reason to put them up on boxes like > that is to give people something to back over with their cars. Wish I had one of those. Then the truck that knocked down my garden wall could just have driven over the little box instead... Peter > > > http://www.pbase.com/konascott/image/68928171/original > > > [quoted text clipped - 7 lines] > > Scott LOL. And I thought that it was something like cable or electrical connections. Thanks for answering my silly question. >I thought that my images would be 3600 x 2400, based on a 3:2 aspect ratio. >So how am I getting 50% more ppi than that? Because a 35mm film frame is not 1" wide. A 35mm frame is 36mm x 24mm, or 1.42" x 0.94", and there's typically a bit of overscan to account for frames that are slightly misaligned. At 3600 dpi, 1.42" x 0.94" yields 5112x3384 pixels. Whether or not there's actually 16Mpx of visual information there to capture depends on the film, lens, and processing. SignatureCentral Maryland Photographer's Guild - http://www.cmpg.org Strange, Geometrical Hinges - http://sgh.rnovak.net |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2009 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.