 |
 | 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 / Digital Photography / Digital Photo / August 2005Digital SLR's - expensive hobby
I have never owned an SLR, always seemed to much of a camera for my skills (and too pricey). But I am wondering about the thoughts on digital technology from those who do have them. Someone could own a film SLR for 10, 20 years and just upgrade the lenses as and when necessary. They would be good workhorses and never need the body to be upgraded. Now I see a financial problem with digital in that because the megapixel rate is forever increasing, SLR owners are more like to change the camera body every couple of years to keep up with the current technology. We have seen cameras go from measly 1megapixel up to 16 megapixel in a rate of what...5-7 years? is that all? Probably be only a year before we see 32 megapixel cameras and people will want to upgrade again. What do you all think about the quickly changing technology in digital photography? Are you concerned that financially for you it is moving too quickly? >I have never owned an SLR, always seemed to much of a camera for my skills >(and too pricey). But I am wondering about the thoughts on digital [quoted text clipped - 14 lines] > photography? Are you concerned that financially for you it is moving too > quickly? We're already pushing the limits of existing 35mm lens technology (the glass can only resolve so much detail until they become the barrier to resolution, rather than pixel count), so going much higher than 16MP is going to being increasingly problematic. I think we're actually nearing the *beginning of the end* of MP wars for the small-ish 35mm DSLR format (at 16.7MP). To go much higher and have it pay off, they'll soon have to come up with better lens solutions. Mark >> I have never owned an SLR, always seemed to much of a camera for my >> skills (and too pricey). But I am wondering about the thoughts on [quoted text clipped - 23 lines] > To go much higher and have it pay off, they'll soon have to come up > with better lens solutions. Oh right I did not know that. I thought it would be like memory where it would become never ending. I suppose that is good news it means people can finally settle down to a level and not be bothered by what is round the corner. I am sure not many people need that many pixels anyway. The average home user is only going to want to print off maximum 8x6 prints anyway and not poster size or more. >>> I have never owned an SLR, always seemed to much of a camera for my >>> skills (and too pricey). But I am wondering about the thoughts on [quoted text clipped - 31 lines] > user is only going to want to print off maximum 8x6 prints anyway and not > poster size or more. I should add... ...More MPs DO mean you can crop images and still have enough resolution to do larger prints. Also... In spite of the lens resolution limitations, don't think the technology will stay stagnant. :) Canon, Nikon, and the rest have other ways of making things attractive...like noise levels, sensitivity (ISO), frame-rate, and on and on. Your basic question, though, I think DOES have a fairly happy answer, in that DSLRs are reaching a point where they remain worthy for long periods--so long as the user can resist the upgrade itch. A 4x6 print originally shot on the 16.7MP 1Ds Mark II isn't going to look any better than the same shot taken with a Canon Digital Rebel. Once you get into HUGE prints, though, the high-dollar units start to show their stuff... > I should add... > ...More MPs DO mean you can crop images and still have enough [quoted text clipped - 12 lines] > you get into HUGE prints, though, the high-dollar units start to > show their stuff... I think it is good to know they will eventually level out though. We are only just seeing the release of 16megapixel cameras. In fact I think there is only one model at the moment? If 16 megapixel is the peak, then people can start to settle down and not continue to upgrade their bodies. >> What do you all think about the quickly changing technology in >> digital photography? Are you concerned that financially for you it [quoted text clipped - 8 lines] > To go much higher and have it pay off, they'll soon have to come up > with better lens solutions. I suspect we will see the megapixel race go quite a bit further before it settles. It's pretty much gone too far already on the P&S models. Very few people will ever blow up their P&S above 8x12 and a lot won't even go past 6x4. The problem manufacturers face is what *single* thing differentiates the better camera. Consumers know to look for that 1 magical number which reveals all. ;-) We are seeing it happen with processor speeds now. Manufacturers are having to come up with a new "feature" that allows lazy consumers to judge which is the best. -Mike >>> What do you all think about the quickly changing technology in >>> digital photography? Are you concerned that financially for you it [quoted text clipped - 16 lines] > camera. Consumers know to look for that 1 magical number > which reveals all. ;-) But its not the megapixel count which is most important is it? I would say the lens is most important. There is no point in having say an 8mp camera that has a shitty lens as you might as well have a 2mp one. >> I suspect we will see the megapixel race go quite a bit further >> before it settles. It's pretty much gone too far already on the [quoted text clipped - 8 lines] > say an 8mp camera that has a shitty lens as you might as well have a > 2mp one. Sorry. Should have enclosed that in <sarcasm> tags :-) My point is that most consumers don't want to spend time researching their potential purchase and would rather have one feature that tells them which is best. -Mike > Sorry. Should have enclosed that in <sarcasm> tags :-) > > My point is that most consumers don't want to spend time > researching their potential purchase and would rather have > one feature that tells them which is best. And the average person probably goes for the pixel count and not the other features ;-) There's yet another way to consider this. If the camera and accessories that you own suit your needs and produce photographs that please you, then that should be the criterion for staying with what you have. Excellent photographs are made every day using equipment that has long-since been outmoded. The urge to replace what you already have with something more recent can quickly become competitive as you see another photographer's equipment and get that sudden flush to keep yourself current. Remember too that camera manufacturers are in business to sell you cameras, lenses and accessories. Their advertising campaigns constantly highlight the "superior" qualities of the latest equipment hoping to instill in you the idea that what you currently own is now inferior. It's up to you to decide how pleased you are with what you have. >> Sorry. Should have enclosed that in <sarcasm> tags :-) >> [quoted text clipped - 4 lines] > And the average person probably goes for the pixel count and not the other > features ;-) Excellent > photographs are made every day using equipment that has long-since been > outmoded. The urge to replace what you already have with something more > recent can quickly become competitive as you see another photographer's > equipment and get that sudden flush to keep yourself current. The first photography course my wife took, at a junior college, the instructor required all students to be a real cheap, plastic 120 format box camera, and shoot, process, and print 10 roles of film per week. Some really nice stuff came from that class. > I suspect we will see the megapixel race go quite a bit further > before it settles. It's pretty much gone too far already on the [quoted text clipped - 3 lines] > camera. Consumers know to look for that 1 magical number > which reveals all. ;-) I wouldn't say that P&S models have gone "too far". You could also say that programs such as Photoshop (and Word and Excel) all have features that are esoteric enough that they're rarely used. But the extra features are worth adding, both for the few that really want them, as well as providing ad campaign fodder for the rest of us. I don't care much about features such as video, but maybe that's because today's digital cameras are still too inadequate. But as far as MPs go, I find large heaps of them occasionally quite useful. Not so much for carefully composed shots, but for the spontaneous ones where you don't have time or the ability to control what's taking place. I've often taken pictures where something later seen off to one side was more interesting than what I was focused on (pun intended). With enough MPs I've been able to make decent prints of interesting portions of images, even after discarding more than 75% of the pixels. Mind you, Mark, that existing CMOS might change also... Marcel > >I have never owned an SLR, always seemed to much of a camera for my skills > >(and too pricey). But I am wondering about the thoughts on digital [quoted text clipped - 25 lines] > > Mark > To go much higher and have it pay off, they'll soon have to come up with > better lens solutions. > > Mark There are some really good lenses out there now, if you are willing to foregoe infinite zoom ratio and f/0.5 speed :-) >> To go much higher and have it pay off, they'll soon have to come up with >> better lens solutions. >> >> Mark > There are some really good lenses out there now, if you are willing to > foregoe infinite zoom ratio and f/0.5 speed :-) I'm already using nearly all L lenses...and those that aren't are as sharp as Ls (50 1.4, 100 2.8 macro). I'm not saying that all DSLRs are approaching the limit...but the 16.7MP 1Ds Mark II clearly is getting close to a point where it becomes an optical limitation that is lens based. > We're already pushing the limits of existing 35mm lens technology (the glass > can only resolve so much detail until they become the barrier to resolution, > rather than pixel count Is there a physical law which limits lens resolution or is it a manufacturing issue, such as costs exploding when you increase the lens resolving power beyond a certain limit ?  SignatureAlfred Molon ------------------------------ Olympus 4040, 5050, 5060, 7070, 8080, E300 forum at http://groups.yahoo.com/group/MyOlympus/ Olympus E300 resource - http://myolympus.org/E300/ >"Alfred Molon" <alfredREMOVE_molon@yahoo.com> wrote in message > >news:MPG.1d702f3adfa6699898ace1@news.supernews.com... [quoted text clipped - 8 lines] >manufacturing issue, such as costs exploding when you increase the lens >resolving power beyond a certain limit ? Others here cas address that better than I can, but yes...there are "laws" which limit lens resolution. >> We're already pushing the limits of existing 35mm lens technology (the >> glass can only resolve so much detail until they become the barrier to [quoted text clipped - 3 lines] > manufacturing issue, such as costs exploding when you increase the lens > resolving power beyond a certain limit ? It's more that the film most people use has a limited resolution so there was no need to make a lens that resolved more than the film could record. Most dSLR's are using film lens technology and these systems are starting to show this limitation. To go past this limit, new lenses will be required. SignatureStacey >Is there a physical law which limits lens resolution or is it a >manufacturing issue, such as costs exploding when you increase the lens >resolving power beyond a certain limit ? Diffraction sets an absolute upper limit on the resolution of any lens. If you take just about any SLR lens, set it to its smallest aperture (probably f/16, f/22, or something similar) and focus accurately, the resulting image will be diffraction-limited. However, every time you double the aperture diamater (by opening the diaphragm 2 stops), the diffraction-limited resolution limit also doubles. So with a "perfect" lens, resolution at f/8 will be double that at f/16. F/4 would give another doubling of resolution, and f/2 double again. Real lenses aren't perfect though - they have aberrations. As you open the aperture the aberrations get worse, sometimes dramatically so. Thus at large apertures, lens performance is determined by aberrations, and there can be dramatic differences between lenses. Trying to get lenses to approach the diffraction limit at large apertures may well cause exploding costs, since with more money you can build more complex designs with more elements and tighter manufacturing tolerances, use exotic glass, etc. Dave >>Is there a physical law which limits lens resolution or is it a >>manufacturing issue, such as costs exploding when you increase the lens [quoted text clipped - 19 lines] > designs with more elements and tighter manufacturing tolerances, use > exotic glass, etc. Dave, I think a clarification is needed. While everything you said is correct referring to the angular resolution. But the diffraction spot size in the focal plane is constant for a given f/stop. For example, at f/8 the diffraction spot diameter (the diameter of the first dark ring) is 10.3 microns for 6300 angstrom light. This is true for a 24mm focal length lens at f/8 and a giant 1000mm diameter, 8000 mm focal length telescope. The diffraction spot size scales linearly with f/stop. e.g. f/2 is a 2.6 micron diameter. So 2 to 3 micron pixel sizes are real practical lower limits. And since people want to control f/ratio to have some depth of field, f/2 simply will not cut it, so much larger pixels are better. The high end DSLRs with 8 microns pixels are wonderful in this regard. More on diffraction: http://www.clarkvision.com/imagedetail/scandetail.html#diffraction I predict DSLRs will plateau but with pixel sizes in the 6 to 9 micron range. The Nikon D2X is only 5.5 micron/pixel spacing. The canon DSLRs are in the 6 to 8 range, The Nikon D2Hs is 9.4 microns. That 6-9 micron range is the performance sweet spot, and that will not change. More pixels/micron means less photons per pixel, thus lower signal to noise ratios, and more softness due to diffraction. Roger >Dave, >I think a clarification is needed. While everything you said >is correct referring to the angular resolution. But the diffraction >spot size in the focal plane is constant for a given >f/stop. I was assuming a constant focal length. The original question was about whether there is a physical limit on increasing the resolution of lenses. Diffraction is that limit. But supposed focal length can change: Angular and linear resolution are related by focal length in exactly the same way as aperture diameter is related to f/number. If you increase focal length while keeping aperture diameter constant, the f/number decreases in inverse proportion. If you keep f/number constant instead, the aperture diameter increases proportional to focal length. Thus, if you increase focal length while keeping aperture diameter constant, the angular and subject-space linear resolution limit remains the same, while the linear resolution in image space decreases along with f/number. On the other hand, if you increase focal length while keeping f/number constant, the angular and subject-space resolution limit increases along with increasing aperture diameter, while the linear resolution limit in image space stays the same. >For example, at f/8 the diffraction spot diameter >(the diameter of the first dark ring) is 10.3 microns >for 6300 angstrom light. This is true for a 24mm focal length >lens at f/8 and a giant 1000mm diameter, 8000 mm focal >length telescope. So if you want increased lens resolution to match smaller pixel sizes, you need a lens that is diffraction limited at larger f/numbers, which is a problem because it's harder to reduce aberrations at large f/numbers. >The diffraction spot size scales linearly with f/stop. >e.g. f/2 is a 2.6 micron diameter. So 2 to 3 micron >pixel sizes are real practical lower limits. And since >people want to control f/ratio to have some depth of >field, f/2 simply will not cut it, so much larger pixels >are better. Are you talking about obtaining large or small DOF? If you want *large* DOF, there's no problem. Suppose you decrease pixel pitch by a factor of 2. To keep the same field of view, the lens FL is also divided by 2. To double the resolution in the image (in order to get the same resolution in a same-size print), open the lens up to twice the f/number (which is the same actual aperture diameter). You also need to reduce the circle of confusion size to half of what it was. The effect of all of these changes (lens FL, f/number, CoC diameter) is to keep the depth of field the same as it was before. On the other hand, if you want *small* DOF, you just can't get it with a small sensor and short focal length lens. For shallow DOF, the larger the image plane the better. The other problem with shrinking the sensor is you narrow the range of useful f/stops. If a full-frame sensor gets acceptable sharpness at f/22, then you have the range from f/22 up to the lens maximum aperture (f/1.4 or f/2 or whatever) to use to increase shutter speed or reduce DOF. But a sensor that's 1/5 this size gets the same maximum resolution at f/4, so you end up with only the range f/2 to f/4 as usable aperture range. Full-frame film SLR users are used to lenses having a range of "good" apertures. The largest one or two stops are usually soft because of aberrations, while the smallest one or two apertures show a reduction of sharpness due to diffraction, but the rest are quite sharp on-axis at least. As pixel pitch is reduced, the "not sharp due to aberration" and the "not sharp due to diffraction" f/numbers pull closer to each other, leaving less and less ground in between. >That 6-9 micron range is the performance sweet spot, >and that will not change. More pixels/micron means less >photons per pixel, thus lower signal to noise ratios, and >more softness due to diffraction. And these are physical limitations, not something that's going to change with better semiconductor processing. Dave > The other problem with shrinking the sensor is you narrow the range of > useful f/stops. If a full-frame sensor gets acceptable sharpness at [quoted text clipped - 3 lines] > resolution at f/4, so you end up with only the range f/2 to f/4 as > usable aperture range. The diffraction limit is a bit higher. There is a rule of thumb which states that the diffraction limit is given (approximately) by 1600/F lp/mm: http://bobatkins.photo.net/info/resolution.html At F8 a lens will be diffraction limited at 200 lp/mm (1600/8). A 5MP 1/1.8" CCD with 2.8 micrometer pixels has 178 lines/mm and will be below the diffraction limit at F8. SignatureAlfred Molon ------------------------------ Olympus 4040, 5050, 5060, 7070, 8080, E300 forum at http://groups.yahoo.com/group/MyOlympus/ Olympus E300 resource - http://myolympus.org/E300/ >> The other problem with shrinking the sensor is you narrow the range of >> useful f/stops. If a full-frame sensor gets acceptable sharpness at [quoted text clipped - 3 lines] >> resolution at f/4, so you end up with only the range f/2 to f/4 as >> usable aperture range. >The diffraction limit is a bit higher. >There is a rule of thumb which states that the diffraction limit is >given (approximately) by 1600/F lp/mm: >http://bobatkins.photo.net/info/resolution.html The diffraction limit is the upper limit of resolution under some set of measurement conditions. From the graph shown on that page, this particular diffraction limit is defined as the resolution where the MTF has dropped to 1 or 2% - the limit of visibility. So high frequency information at the limiting frequency has been attenuated 50 or 100 times. That's an interesting place to measure, but for photographic purposes it's more interesting to note the frequency where the MTF drops below 50%. As you can see from the graph, that's at less than half the limiting frequency even for a perfect diffraction-limited lens, much less when the lens isn't perfect. On the other hand my statement above is about the subjective effects of diffraction on sharpness: "acceptable sharpness at f/22". What's acceptable to me might not be acceptable to you. But whatever f/number you decide is your own personal lower limit, when you shrink the sensor size by a factor of 5, you need to multiply the f/number by 5 as well to keep *the same* amount of degradation due to diffraction. >At F8 a lens will be diffraction limited at 200 lp/mm (1600/8). A 5MP >1/1.8" CCD with 2.8 micrometer pixels has 178 lines/mm and will be below >the diffraction limit at F8. Actually, it has 357 pixels/mm which can resolve a maximum of 178 line *pairs* per mm. Although the diffraction limit for the lens at f/8 is above this, the 50% contrast resolution is somewhere around 100 lp/mm. Thus f/8 is probably visibly soft compared to f/5.6 with even the best lens at this sensor size. At f/4 the 50% MTF should be above the sensor resolution limit. Dave [A complimentary Cc of this posting was sent to Dave Martindale <davem@cs.ubc.ca>], who wrote in article <deac3h$e5n$1@mughi.cs.ubc.ca>: > The diffraction limit is the upper limit of resolution under some set of > measurement conditions. From the graph shown on that page, this [quoted text clipped - 4 lines] > purposes it's more interesting to note the frequency where the MTF drops > below 50%. Your argument would be OK in the age when the main technology was film-based. In the age of DSP, it does not make a lot of sense. (In its absurdity it is similar to using the Rayleigh criterion for digital sensors - both the used metrics are not preserved by post-processing.) On the spacial frequencies where MTF of the lens (+ the sensor) is 50%, all that is changed (from the digital processing point of view) is the narrow-band S/N ratio; e.g., if you work with 100ISO setting of contemporary dSLRs, instead of the S/N ratio which is 5 times better than one of the best film, you get one which is 2.5 times better than the best film. Where MTF is 10%, you get the quality (w.r.t. noise) of the speedy film. Let me remind you that in most high-quality photographic systems, the priciest factor is the lens (maybe this is not exactly true during the last couple of years, but I believe this to be a temporary glitch). To fully use the investement into lenses, the sensor should better resolve up to the cut-off frequency of the lens (in the best-resolution setting). (Again, *today* this is not price-effective; one need to wait a couple of years why processing power of camera firmware is up to demands of 50MP sensors.) > Actually, it has 357 pixels/mm which can resolve a maximum of 178 line > *pairs* per mm. Although the diffraction limit for the lens at f/8 is > above this, the 50% contrast resolution is somewhere around 100 lp/mm. > Thus f/8 is probably visibly soft compared to f/5.6 with even the best > lens at this sensor size. At f/4 the 50% MTF should be above the > sensor resolution limit. Another data point: the current 2/3'' cameras with about 7x zooms have lenses which max the resolution at about f/4. They easily achieve about 60% throughput MTF at 150 lp/mm (with proper DSP - which still produces the level of noise comparable to film). Such lenses in 24x36mm format are out of question; but one still needs sensor step about 2.5microns to do proper favor to good 24x36mm lenses. Hope this helps. Ilya > Dave, > I think a clarification is needed. While everything you said [quoted text clipped - 18 lines] > I predict DSLRs will plateau but with pixel sizes > in the 6 to 9 micron range. interesting. if pixels are the unit of measurement for digital sensors, what is its film equivalent? grain? if so, what is the average micron size of a film with an iso of 100? The Nikon D2X is only > 5.5 micron/pixel spacing. The canon DSLRs are in the > 6 to 8 range, The Nikon D2Hs is 9.4 microns. > That 6-9 micron range is the performance sweet spot, > and that will not change. More pixels/micron means less > photons per pixel, thus lower signal to noise ratios, and > more softness due to diffraction. what if they improved pixel performance and lowered noise as well as decreasing its size? wouldn't that move the performance sweet spot in the years to come? > Roger >if pixels are the unit of measurement for digital sensors, what is its >film equivalent? grain? Yes. >if so, what is the average micron size of a >film with an iso of 100? Here's an interesting comparison of film grain to digital pixels, which attempts to explain why film's apparent higher resolution on test patterns doesn't seem to agree with actual results ... http://www.luminous-landscape.com/essays/clumps.shtml > >if pixels are the unit of measurement for digital sensors, what is its > >film equivalent? grain? [quoted text clipped - 9 lines] > > http://www.luminous-landscape.com/essays/clumps.shtml hmm. grain records light on a binary basis!? so you need a clump of 30-40 grains to record what one digital pixel can record on an analog basis?? calling dr. clark! dr. roger "nasa scientist who knows more than this entire ng combined" clark. would you be so kind to respond on this matter? >>>if pixels are the unit of measurement for digital sensors, what is its >>>film equivalent? grain? [quoted text clipped - 16 lines] > calling dr. clark! dr. roger "nasa scientist who knows more than this > entire ng combined" clark. No, I strongly disagree. I've learned a lot from many in this group. We all have our expertise, and I'm still trying to catch up to others who post in this NG, like Bill Hilton, and many others (too numerous to name). > would you be so kind to respond on this matter? Hi, Lucky I noticed this (I skip a lot in this newsgroup). My response, will be similar to the scientific theory that proves that bumble bees can't fly. The problem with that theory is that it is incomplete. Similarly the claim on the clumps.html article is that film grain in "A very fine-grain film has grain particles that are about 2 microns in size." That is incorrect. The article then states that film grains are binary, which is true. The problem is that film grains are sub-micron, not about 2-microns. A 2-micron grain-looking thing in an enlargement is really a "clump" of many grains. One needs an electron microscope to see the real grain in fine grained film. So the article is flawed in its argument. But the conclusion is correct. While the grain clumps may be around 2 microns, and each clump (and clump really is not a correct description), does have intensity range, the problem with film is that the clumps themselves are randomly distributed, with varying sized clumps, with imperfections in the film base, all adding to reduce the image quality on the few micron scale. You need multiple clumps to make a smooth looking image, so that is why 6-micron high signal-to-noise DSLR pixels make for a good image compared to fine grained film with 2-micron "clumps." People generally refer to the "noise" in film images as grain, which leads to confusion if one wants to discuss technical aspects of the medium. Color film gets even deeper into lore versus technical accuracy as you have dye clouds in layers and not film grain like in black and white film. And I have probably been a little loose with definitions here too and a scientist who actually makes film could certainly give a lot more technical detail and set us straight as to correct definitions. Roger thanks for the reply. > >>>if pixels are the unit of measurement for digital sensors, what is its > >>>film equivalent? grain? [quoted text clipped - 21 lines] > to catch up to others who post in this NG, like Bill Hilton, > and many others (too numerous to name). i stand corrected. > > would you be so kind to respond on this matter? > [quoted text clipped - 11 lines] > is really a "clump" of many grains. One needs an electron > microscope to see the real grain in fine grained film. that sounds about right. i found the following on google: Silver Halide Materials By Dinesh Padiyar http://holographyforum.org/files/grains.htm Silver Halide is the most common holographic material currently used by everyone from hobbyists to professional scientists. This is because of it's high sensitivity and the fact that it's easily available commercially. Silver Halide is a salt of Silver and a Halide, which are a group of elements that include Fluorine, Chlorine, Bromine and Iodine, of which the last three are light sensitive. The first, accidental, discovery of the light sensitive properties of Silver salts came in 1727 when J Schulze mixed chalk, Nitric acid and Silver in a flask. He noticed that the side of the flask turned toward the sun turned dark. The first deliberate application of this effect, a photograph of a scene, was carried out by Niepce in 1827. This was a long exposure of about 8 hours. In 1841 Henry Fox Talbot patented a "calotype" process which made a permanent negative image on paper soaked in Silver Chloride. The present day emulsion was created by Richard Leach Maddox, who proposed using gelatin and Silver Bromide in what he called the "dry plate process". Shortly after that Eastman Kodak coated a flexible film with this as a thin emulsion, mass produced it, and launched large scale photography. Silver Bromide is only sensitive in the UV and blue regions and so dyes need to be added to it in order to make it sensitive to other colors. In holography, specific dyes are added to give the emulsion a high sensitivity at particular, commonly available laser lines. Commercial SilverHalide materials consist of crystals, or grains, of Silver Halide dispersed in a layer of gelatin, each grain consisting of many thousands of molecules of Silver Halide. When exposed to light individual molecules of Silver Halide on those grains that were illuminated break down to Silver which, due to it's small size, appears black. Since the number of molecules of Silver Halide that convert to Silver on any one grain depends on the intensity of light hitting the grain, the darkness, or density, of any one grain depends on the light hitting it. Thus the emulsion converts the variation of light that hits it, the exposure, to a density variation. After exposure the emulsion consists of exposed and unexposed Silver Halide grains. The action of converting the exposed emulsion to the same variation of density as the variation of light intensity at exposure by amplifying the exposed grains of Silver Halide is known as processing. The removal of the unexposed grains is known as fixing. Thus the entire process consists of exposure, development and fixing. There is also a bleaching step that is no longer widely used in photography but is common in holography. The fixing step is sometimes omitted. more later. > So the article is flawed in its argument. But the conclusion > is correct. While the grain clumps may be around 2 microns, [quoted text clipped - 17 lines] > > Roger > i found the following on google: > [quoted text clipped - 35 lines] > on the light hitting it. Thus the emulsion converts the variation of > light that hits it, the exposure, to a density variation. not that i don't want a d2x (i think i'll eventually get one), but when you're talking about film granularity from about 2 micron clumps down to silver halide crystals at the sub-micron, molecular level i think there's a lot of info that film is able to pack in, that perhaps current scanning technology isn't fully exploiting, whereas with even a high-end dslr like the d2x, with its remarkable 6 micron pixels and 12 meg image, has a more definitive ceiling in terms of what is extractable. > not that i don't want a d2x (i think i'll eventually get one), but when > you're talking about film granularity from about 2 micron clumps down [quoted text clipped - 4 lines] > meg image, has a more definitive ceiling in terms of what is > extractable. But the sub few micron range is at a very low MTF, which only records very high contrast subjects. And remember that diffraction spot diameter at f/4 is already 5.2 microns (green light). So few lenses can deliver much information below 2 microns anyway. Drum scanners use microscope objectives and, and record up to 11,000 dpi, (2.3 microns/pixel), so scanners can pretty much "get all the detail." Roger > > not that i don't want a d2x (i think i'll eventually get one), but when > > you're talking about film granularity from about 2 micron clumps down [quoted text clipped - 15 lines] > > Roger well, i was thinking more in terms of "prosumer" scanning technology that i could afford :), but point taken. btw, i was just looking at your site again, and i realized i have a copy of your phenomenal "great egret" picture (via nature's best photography magazine, fall 2004)! an absolutely stunning image, and well-deserving of a full-page layout (though there seems to be a slight warm cast to it as compared to the jpg). what stood out immediately when i first saw it last year was the near-perfect composition of the beautiful bird -- almost as if it were hand-drawn -- and how sharp it was. personally, i think your pic and others (e.g. the emperor penguins or the gray wolves & grizzly) was more impressive than the winning picture of the polar bears. couple of questions w/regards to the pic: what kind of tripod set-up did you use? wimberly? also, why the negative 1/2 stop exposure compensation? > btw, i was just looking at your site again, and i realized i have a > copy of your phenomenal "great egret" picture (via nature's best [quoted text clipped - 12 lines] > > also, why the negative 1/2 stop exposure compensation? Thanks. That image was with the lowly D60. I bought the D60 and a 500 mm f/4 for astrophotography, and then people in this group, notably Bill Hilton, helped my along with wildlife photography, which I dabbled in for many years without any really great images. But the 500mm was a life changing lens, opening up so many new possibilities, I'm still having fun with it. The tripod was a gitzo carbon fiber 1325, but at that time I was using a pan head. Everyone was telling me I needed a Wimberly, and they were right. The Wimberly had a little more vibration than my pan head, but it made tracking moving subjects so much easier. Between the gitzo and Wimberly, camera plates, the tripod setup comes to about $1500, but it sure makes a difference. But the Wimberly was critical, regardless of head in supporting a big telephoto. The bird is white on a dark background. Standard metering would have overexposed the bird. I metered on the bird before take-off so I new what to use (checking histograms). The composition is a crop from the right half of the frame. The vertical extent is the full image from the horizontal frame. So it is 3 megapixels out of 6. With careful upsampling in multiple steps, I made a 61.6 megabyte TIF file that prints 16 x 18.3 inches at 305 ppi. For those wondering, here is the image: http://www.clarkvision.com/features/natures.best.win.2004 The quality is good enough that it is hanging in 2 art galleries in Colorado. The shot was pretty lucky. I've tried thousands of other takeoff shots and never gotten anything as beautiful, in my opinion, even when using top end equipment, like the 1D Mark II. Beginner's luck? It was my first visit to the Venice Rookery. Roger > > couple of questions w/regards to the pic: > > [quoted text clipped - 19 lines] > But the Wimberly was critical, regardless of head > in supporting a big telephoto. you have a wimberly now but didn't back then... that's worth about a 1/2 point for those keeping score (<-- don't mind that, it's a little game we play.) > The bird is white on a dark background. Standard metering > would have overexposed the bird. I metered on the bird > before take-off so I new what to use (checking histograms). > > The composition is a crop from the right half of the frame. ok, the negative compensation makes more sense to me now. i didn't realize the picture was heavily cropped... > The vertical extent is the full image from the horizontal > frame. So it is 3 megapixels out of 6. With careful upsampling [quoted text clipped - 4 lines] > The quality is good enough that it is hanging in 2 art > galleries in Colorado. as it should. still... i frequently find myself caught in a dilemma: i'm more likely to take a great shot if i'm shooting digital because of its instant feedback, large capacity memory cards, variable iso, etc., but when i do make one of those killer pictures -- the kind i want to keep forever (which only happens about 2% of the time) and so want to enlarge a thousand fold -- i'm always regretting it wasn't done on velvia 50. > The shot was pretty lucky. I've tried thousands of other > takeoff shots and never gotten anything as beautiful, > in my opinion, even when using top end equipment, like > the 1D Mark II. > > Beginner's luck? It was my first visit to the Venice Rookery. you can't win if you don't play, and it certainly seems like you know how to play. :) > Roger >I have never owned an SLR, always seemed to much of a camera for my skills >(and too pricey). But I am wondering about the thoughts on digital [quoted text clipped - 14 lines] > photography? Are you concerned that financially for you it is moving too > quickly? On the one hand, if today's 6MP DSLR's are taking a good enough photo for you today, why does it suddenly become obsolete because an 8MP or higher is released? On the other hand, I doubt todays DSLR's are built with anything like the longevity of some of the older film cameras. There are plenty of 40yo or older film cameras still in everyday service - I doubt many 300D's will still be going in 5 years. Another big difference is that those 40yo film cameras are taking far better photos today than they did 40 years ago, due to advances in film technology. When recording technologies advanced, all that was required was to load the camera with the new film. In the digital world, to take advantage of improvements in recording technology requires replacing the entire camera. > On the one hand, if today's 6MP DSLR's are taking a good enough photo > for you today, why does it suddenly become obsolete because an 8MP or > higher is released? It doesn't for me. MP numbers to me are not that important. On the other hand, I doubt todays DSLR's are > built with anything like the longevity of some of the older film > cameras. There are plenty of 40yo or older film cameras still in [quoted text clipped - 5 lines] > world, to take advantage of improvements in recording technology > requires replacing the entire camera. Is this build quality of todays cameras really that bad? >> On the one hand, if today's 6MP DSLR's are taking a good enough photo >> for you today, why does it suddenly become obsolete because an 8MP or [quoted text clipped - 14 lines] > > Is this build quality of todays cameras really that bad? They are actually built better. However, they are unlikely to last anywhere near as long. There are two reasons for this: Firstly, manufacturers are able to better predict the life of individual components compared to 40 years ago. This means there are less product failures in the short term but they wear out quicker. The other thing with dSLRs compared to film SLRs is that because the "film" is so cheap, people take *many* more pictures. This causes mechanical parts to wear out in a shorter time. -Mike >>On the one hand, if today's 6MP DSLR's are taking a good enough photo >>for you today, why does it suddenly become obsolete because an 8MP or [quoted text clipped - 15 lines] > > Is this build quality of todays cameras really that bad? No, but dslr owners might shoot 50 frames a day instead of 50 per month. If the shutter or mirror mechanism of a dlsr lasts 50,000 shots, then you got good value - equivalent to about 50 cents per 36 shot roll of film - a fraction of the cost even excluding processing. Chuck it out, get a new one, and remember the old days. The way things are going, you can be fairly sure that the next 50,000 shots of digital film you buy will be better and probably cheaper than the 50,000 you just used. >> Is this build quality of todays cameras really that bad? > No, but dslr owners might shoot 50 frames a day instead of 50 per month. > If the shutter or mirror mechanism of a dlsr lasts 50,000 shots, The real issue is that we don't actually know if the componentry of today's digital bodies is the equal or better than that of the current generation of film bodies. Essentially we are the Beta group for the these cameras. No one will really know anything about the effective duration of these bodies until a large group of users has had them in hand and been using them for 15 to 20 years. In that sense even cameras like the F5 are still being "field tested" for reliable longevity. My suspicion is that we will not see digital bodies with active 20 year lifespans. I think the processing and computing component requirements will fail long before dslr's reach the level of reliability previous film cameras have proven to provide. They still can't, or won't, build a laptop that will last 15 years over a wide group of users and within that realm they have more room and more ability to manage heat and surge and other things that slowly damage processors and boards. >> Is this build quality of todays cameras really that bad? > No, but dslr owners might shoot 50 frames a day instead of 50 per month. [quoted text clipped - 4 lines] > fairly sure that the next 50,000 shots of digital film you buy will be > better and probably cheaper than the 50,000 you just used. The trouble with this analogy, it's not the 50 shots per day you take that is important, it is the number of keepers. From what I've seen of digital photographers, they snap, snap, snap away, whereas film photographers tend to pause and think. I've been on a few day trips with friends who are keen photographers who are using digital. The digital guys have shot 100's, maybe even 1000 shots, I've shot 2 or 3 rolls of chromes. At the end of the day I've come out with more keepers than they have. Some of these digital guys have in the past been good photographers with film who would have come out with near 100% keepers on film, but it seems that the shoot-and-delete of digital causes a mind-set that stops them from thinking before clicking. And they will brag that it doesn't matter that they took 1000 photos and only had 20 good ones, because it cost them nothing - "it would have cost me $500 to do that with film". How soon they forget that a few months prior they would have shot 2 rolls and come away from the day with 60-70 good photos. So if my $300 film body costs 40c for every photo (for chromes it works out about 40c, it's about 20c for colour negative - this is what it costs me for film+process+mount for chromes, and film+process+scan for colour neg), and their $1500 digital body takes 50,000 images (3c/photo) but most are throw-aways, then the equation is a little more evenly balanced toward film. But will the camera take 50,000 images? And how many times and at what cost will they have to clean their sensor in that time? If they do have a camera breakdown prior to 50,000 images but out of warranty, what will the repair cost be compared to a film camera? When all this is taken into account, I'm convinced that there really isn't a significant cost difference between film and digital. If the digital user can press the shutter button 36 times and get 30 keepers, as I normally do with chromes, then digital is far cheaper. My experience with how others use them, and even my own use of them when i've borrowed a DSLR indicates that that isn't the case. > Justin Thyme writes ... > >Some of these digital guys have in the past been good photographers >with film who would have come out with near 100% keepers on film, >but it seems that the shoot-and-delete of digital causes a mind-set >that stops them from thinking before clicking. My experiences and observations of other photographers has been exactly the opposite of this. The guys who were good film photographers are better photographers using digital. I think there are several reasons for this ... one is that you can switch ISO on the fly and so get shots you would otherwise miss. Another is the value of the quick feedback, whether from the LCD or, better, from looking at the images on a computer shortly after shooting them. Several times I've noticed ways I could improve on a scene and gone back after reviewing digital shots on the laptop, something I couldn't do on the road with film. Also, since there is essentially no additional cost for shooting most of us are more apt to experiment and try something new than we were with film. Finally with the digital editing tools like Photoshop you basically have your own advanced darkroom that's much easier to learn than a wet darkroom, so guys are more involved in the process up to the final print. This is especially true if you're shooting RAW files and learn how to use one of the best converters. YMMV but I've never known a photographer who didn't bring back better images with digital than with film, assuming similar gear like a dSLR compared to a film SLR. Bill >> Justin Thyme writes ... >> [quoted text clipped - 6 lines] > the opposite of this. The guys who were good film photographers are > better photographers using digital. Bing! >>> Is this build quality of todays cameras really that bad? >> No, but dslr owners might shoot 50 frames a day instead of 50 per month. [quoted text clipped - 32 lines] >My experience with how others use them, and even my own use of them when >i've borrowed a DSLR indicates that that isn't the case. I don't think you can make this kind of comparison, because your definition of "keeper" will obviously differ from mine. Since I keep many more photos out of a shoot than you do, then, by your reasoning, my DSLR is less expensive than yours, even if we pay the same for them. Your reasoning also assumes that film shooters are just more particular (and thus shoot fewer photos) than digital shooters because, well, just "because". Film cost (and, therefore, cost of the camera, including processing cost of the medium) doesn't enter into it. I strongly disagree with this. When I shot film, cost was "always" a factor, and this kept the number of shots down. I won't presume to say this is a factor for everyone, but I will dare to say it's a factor for most photographers. There's a reason pros shoot more film than amateurs; it produces more keepers. It does the same for digital, too. SignatureBill Funk Replace "g" with "a" funktionality.blogspot.com > Is this build quality of todays cameras really that bad? It's certainly nowhere near as solid as the old stuff. Electronics should in theory be more reliable than the old mechanical cameras, but experience of other electronic gear tells me other-wise. The mechanical parts of modern cameras are almost exclusively plastic, which doesn't last anywhere near as well as metal would in the same bits. The last-ability of a camera comes not only from the build quality but also from the repair-ability of them. Older film cameras are still repairable. Digital cameras are much more expensive to repair, and parts availability is far less. Through my work I often have to deal with extended warranties on cameras - if a camera fails at more than 2 years, it is invariably scrapped and replaced. Even things like D100's or D60's have been scrapped at under 2 years for failures that on a film camera would be considered a minor repair (eg shutter jam). Parts are not available, and even if they are it is not viable to do the repair. To be fair to DSLR's - the same also applies to many of the modern film SLR's also. Anything older than early 90's seems to be viable to repair, anything newer you just throw it away. This scrap and replace is good for those who bought extended warranties - they are getting their cameras replaced with newer and better models, but it becomes expensive for those who didn't buy the extended warranty and have the same failures. > It's certainly nowhere near as solid as the old stuff. Electronics should in > theory be more reliable than the old mechanical cameras, but experience of [quoted text clipped - 4 lines] > film cameras are still repairable. Digital cameras are much more expensive > to repair, and parts availability is far less. Since technology at the moment progresses so rapidly, changing cameras every couple of years is not so terrible in my opinion. Besides the costs have come down substantially (here in Germany for instance you can get the body of an 8MP DSLR for 630 Euro). SignatureAlfred Molon ------------------------------ Olympus 4040, 5050, 5060, 7070, 8080, E300 forum at http://groups.yahoo.com/group/MyOlympus/ Olympus E300 resource - http://myolympus.org/E300/ > It's certainly nowhere near as solid as the old stuff. Electronics should in > theory be more reliable than the old mechanical cameras, but experience of > other electronic gear tells me other-wise. Even as someone who spent his professional career developing high rel aerospace avionics, I agree. One big factor- shutter always closed on non-electric cameras, regardless of what else went wrong. We have electronic film cameras now that need power to both OPEN and CLOSE shutter. Usually low battery has enough power to open shutter, then it stays open. Can't rewind film when that happens. Battery needs to be replaced NOW! We do carry a changing bag occasionally, since film can be rewound in changing bag if near end of roll. > Is this build quality of todays cameras really that bad? On some of the low end models, yes it is. SignatureStacey > Another big difference is that those 40yo film cameras are taking far better > photos today than they did 40 years ago, due to advances in film technology. > When recording technologies advanced, all that was required was to load the > camera with the new film. In the digital world, to take advantage of > improvements in recording technology requires replacing the entire camera. And we are still waiting on a solution for sensor dust problems. This is a non-issue with film ... even the oldest of old film cameras. SignatureThomas T. Veldhouse Key Fingerprint: 2DB9 813F F510 82C2 E1AE 34D0 D69D 1EDC D5EC AED1 Spammers please contact me at renegade@veldy.net. > And we are still waiting on a solution for sensor dust problems. This > is a non-issue with film ... even the oldest of old film cameras. Dust has always been a problem. In the camera it could be the cause of tramlines and spots on the negatives/transparencies. On prints it could cause myriads of white spots especially on large prints - could cause a great deal of time to be spent 'spotting'. SignatureNeil Delete delete to reply by email > Dust has always been a problem. > In the camera it could be the cause of tramlines and spots on the > negatives/transparencies. > On prints it could cause myriads of white spots especially on large prints > - could cause a great deal of time to be spent 'spotting'. I was referring to "sensor dust", not just "any old dust". Sensor dust is what stops many digital SLR photographers from changing lenses, in fear of allowing dust into the camera and onto the sensor. Once on the sensor, it requires an action by the owner to get the sensor cleaned, and very few options are actually cheap [unless a little air is enough]. With film, if you happen to get dust on the "sensor", or rather, on the film, you take the next picture and the dust problem is gone for all subsequent pictures. No action was required to clear the dust. In short, it is MUCH easier, cheaper and safer to clean dust out of a film camera and certainly there is no such hassle as cleaning the sensor as there is with digital cameras. For the record, I use both a Nikon N80 and a Nikon D70. SignatureThomas T. Veldhouse Key Fingerprint: 2DB9 813F F510 82C2 E1AE 34D0 D69D 1EDC D5EC AED1 Spammers please contact me at renegade@veldy.net. >And we are still waiting on a solution for sensor dust problems. This >is a non-issue with film ... even the oldest of old film cameras. Not used sheet film, have you? > Not used sheet film, have you? No, but even with sheet film, the problem is transient; it doesn't require a special sensor cleaning to fix the problem. SignatureThomas T. Veldhouse Key Fingerprint: 2DB9 813F F510 82C2 E1AE 34D0 D69D 1EDC D5EC AED1 Spammers please contact me at renegade@veldy.net. >> Not used sheet film, have you? > > No, but even with sheet film, the problem is transient; This REALLY shows you've never shot with sheet film! Dust is a MAJOR problem shooting with 4X5 and larger formats. SignatureStacey >> Not used sheet film, have you? > >No, but even with sheet film, the problem is transient; Transient, eh? Let's see the first exposure is covered in dust. The second exposure is covered in... ...different dust. Well there's a *massive* improvement. >it doesn't require a special sensor cleaning to fix the problem. No, it requires that you go and do it *manually*, for each and every exposure, in Photoshop, and despotting images which are typically scanned at resolutions in excess of 100 megapixels is somewhat time consuming. If you're not using something like Quickload film (expensive, not widely available, limited choice of emulsions), then even DigitalICE in your scanner won't help that much, becuase it won't deal with the dust that got there before you took the shot (and thus left shadows on the slide/negative). If you ever do get round to shooting sheet film, somewhere through working with your second or third shot, you'll come to the realisation that people who cry about dust on DSLRs really have absolutely nothing to complain about in comparison. > If you ever do get round to shooting sheet film, somewhere through working > with your second or third shot, you'll come to the realisation that people > who cry about dust on DSLRs really have absolutely nothing to complain about > in comparison. I agree. My years of working with film scans prior to switching to a DSLR gave me the view that DSLR dust REALLY isn't that big of a deal. I remember the hours I used to kill fixing a REALLY dusty print from someone else's scanner. Signaturehttp://www.pbase.com/bcbaird > And we are still waiting on a solution for sensor dust problems. We are? My camera already has this solved. SignatureStacey [snip] > When recording technologies advanced, all that was required was to load > the camera with the new film. In the digital world, to take advantage of > improvements in recording technology requires replacing the entire camera. I wonder would it be that difficult to design bodies with replaceable sensors? Perhaps local camera shops could change the sensor in a dust free environment and upgrade the firmware as a while you wait service. This would save the big manufacturers the trouble of having to sell us complete new cameras with such frequency. SignatureT Ritchie (Sr) Save a rainforest: Eat Brazil nuts - www.mongabay.com/1003.htm > I wonder would it be that difficult to design bodies with replaceable > sensors? Perhaps local camera shops could change the sensor in a dust free > environment and upgrade the firmware as a while you wait service. Agreed. I'd love an 8MP, full-frame, _monochrome_ sensor with no IR cut filter, and with pixels that big, one could almost make it from two chips (if it had a low-pass filter, you wouldn't see the join at all). I wouldn't want to pay even US$1500 for a dedicated body with that chip in it, though. But at US$999.95 for the body, US$1500 for a 16.7MP chip, and US$750 for the 8MP chip, and I might almost buy the body + B&W first and play with them for a while... > This would save the big manufacturers the trouble of having to sell us > complete new cameras with such frequency. I doubt that they see it as any trouble at all<g>. David J. Littleboy Tokyo, Japan > [snip] > > When recording technologies advanced, all that was required was to load [quoted text clipped - 4 lines] > sensors? Perhaps local camera shops could change the sensor in a dust free > environment and upgrade the firmware as a while you wait service. That does sound like a good idea if you like to "keep up with the Joneses" as far as resloution goes, but it seems a bit silly from a practical photographic viewpoint. There is more to it (as far as hardware goes) than resolution to making great digital photos, take look at the stunning images that can be made with (older) DSLR cameras like the Fuji S1, Sigma SD 9, Nikon D1 or Canon EOS 30D. > This would save the big manufacturers the trouble of having to sell us > complete new cameras with such frequency. I get the distinct impression that it is the cost of developing the technology and not the cost of manufacture that is more of an influence on the price of a digital camera. Each sensor probably cost only a few cents to manufacture, _BUT_ cost millions of dollars to develop and the companies want to make back their money (plus profit) by offering a complete product and not just a (mere) replacement part. > I get the distinct impression that it is the cost of developing the > technology and not the cost of manufacture that is more of an influence > on the price of a digital camera. > Each sensor probably cost only a few cents to manufacture, Maybe more than a few cents... SignatureAlfred Molon ------------------------------ Olympus 4040, 5050, 5060, 7070, 8080, E300 forum at http://groups.yahoo.com/group/MyOlympus/ Olympus E300 resource - http://myolympus.org/E300/ > [snip] >> When recording technologies advanced, all that was required was to [quoted text clipped - 13 lines] > us > complete new cameras with such frequency. There's a concept to dream on: just like Dell or HP, configure-your-own camera. Don't waste money on features you don't need. SignatureFrank ess >I have never owned an SLR, always seemed to much of a camera for my skills >(and too pricey). But I am wondering about the thoughts on digital [quoted text clipped - 14 lines] > photography? Are you concerned that financially for you it is moving too > quickly? Don't catch Upgrade Disease! That's the really expensive bit. In the years to come whatever you buy now will continue to produce the results you're (presumably) perfectly happy with - whatever else develops in the meantime. So relax and enjoy... RM > Don't catch Upgrade Disease! That's the really expensive bit. I leaned with computers, don't "upgrade" until the newest stuff is old enough to be discounted as they are selling it off to put the latest stuff on the shelf. SignatureStacey >> Don't catch Upgrade Disease! That's the really expensive bit. > >I leaned with computers, don't "upgrade" until the newest stuff is old >enough to be discounted as they are selling it off to put the latest stuff >on the shelf. I know you're right, but.... EOS 5D... shiny... >>> Don't catch Upgrade Disease! That's the really expensive bit. >> [quoted text clipped - 5 lines] > > EOS 5D... shiny... Put that credit card away. It's not even here yet. :-) -Mike >I have never owned an SLR, always seemed to much of a camera for my > skills (and too pricey). But I am wondering about the thoughts on [quoted text clipped - 14 lines] > photography? Are you concerned that financially for you it is moving > too quickly? The equipment today has gone far enough that it will fulfill the needs of most people for a long time. I would not have said that long ago. You do not need to go beyond your needs and future cameras will be only minor improvements for most user's needs. Of course we all need to remember that most of the readers here are not most users; they are a bit more advanced. SignatureJoseph Meehan Dia duit > I have never owned an SLR, always seemed to much of a camera for my skills > (and too pricey). But I am wondering about the thoughts on digital [quoted text clipped - 14 lines] > photography? Are you concerned that financially for you it is moving too > quickly? The problem is not just that the technology is moving too quickly. The problem is that people are thinking too quickly. Keeping up with the Jones's. etc. Many are just looking for digital perfection and ignoring artistic perfection. It is just like the horsepower race that is going on with automobiles right now. Regarding SLR's, there is an artistic difference to the medium which will not make them obsolete. > The problem is not just that the technology is moving too quickly. The > problem is that people are thinking too quickly. Keeping up with the [quoted text clipped - 4 lines] > Regarding SLR's, there is an artistic difference to the medium which > will not make them obsolete. Indeed that is a problem. You could give me the best camera in the world but it would not make me a good photographer. Its not the camera that makes the pictures but the person using it. When film SLRs first came out they were pretty pricey too. I can remember watching prices drop until we finally could afford one. We used what was the equivalent then of a point and shoot and my cheapo double lens reflex (spartaflex) until we could afford our first SLR. > I have never owned an SLR, always seemed to much of a camera for my skills > (and too pricey). But I am wondering about the thoughts on digital [quoted text clipped - 14 lines] > photography? Are you concerned that financially for you it is moving too > quickly? > Beck writes ... > [quoted text clipped - 3 lines] >What do you all think about the quickly changing technology in digital >photography? Twice I've bought dSLR's only to see them replaced by better models a few months later, once the new model was actually cheaper (D30 replaced by D60) with double the pixel count. I think I've learned three lessons ... 1) buy early in the cycle, not late. That is, most of these bodies (at least from Canon and Nikon) are on an 18 month upgrade cycle (some 12 months, some 24 months, but 18 months is typical) so buy as soon as it comes out, not a year or so later, to get full value. The new model announcements are typically made prior to the big shows in September and February so don't buy anything in say August or January unless you're prepared to see your new purchase replaced soon after. 2) One of my pro friends, faced with upgrading again to keep up with her competition, said wisely "I used to buy film for $12/roll but now I buy digital bodies instead" ... so the more images you shoot the more this equation works in your favor. So divide the cost of the digital body by your cost per roll of film (film + processing) and see how many shots you need to take to break even. Anything after that means you're saving money (useful argument if you have to justify the purchase to a spouse :) In other words, digital makes sense financially if you shoot a lot of film and makes less sense financially if you don't shoot it often. If you don't shoot many photos in a year you'll save money keeping the film camera. 3) Try to make do with keeping your gear thru two or even three upgrade cycles instead of just one, this should give you 3-5 years of use for your purchase, assuming it will do what you need. As others are pointing out, with 8 Mpixels (for some, even 6 Mpix) you are able to print beautifully at most popular print sizes up to 11x14" or maybe a bit larger, so do you really need 11 or 13 or 16 Mpixels right away? Bill I believe that sensor technology will evolve in other ways than just pixel count. How about dumping Bayer interpolation in favor of a Foveon approach? It is clear that is fundamentaly superior. I bet that will happen at some time and everyone will want to upgrade from their 'crappy Bayer' sensors. How about dynamic range? Probably clever solutions for that will appear. Both the Foveon approach and dynamic range could be improved with multilayer sensors....three layers for RGB coincident capture (a la Foveon) and perhaps more layers for clever dynamic range solutions. While today's sensor technology is fine for vacation pictures. I predict the bar will keep being raised in different ways for semi-pro and pro applications. > I believe that sensor technology will evolve in other ways than just > pixel count. How about dumping Bayer interpolation in favor of a Foveon > approach? It is clear that is fundamentaly superior. Except for one flaw: color accuracy. The eye's color response functions of red green and blue overlap. The more accurate the color response to red green and blue on a man-made sensor that closely matches the eyes response will mean better color accuracy. Bayer sensors are very good (especially in comparison to film). The Foveon technology has not done well in this regard and it will be extremely difficult to do it right (I won't say impossible). But likely improvements will probably reduce throughput, so reduce sensitivity and increase noise. It is not clear if the idea is superior. Roger >I have never owned an SLR, always seemed to much of a camera for my skills >(and too pricey). But I am wondering about the thoughts on digital [quoted text clipped - 14 lines] > photography? Are you concerned that financially for you it is moving too > quickly? What is going on now closely resembles what happened to film cameras in the 50s and 60s. I suppose you don't remember when Canon had yet another version of a Leica clone almost every year. You don't remember that Canon introduced the Canonflex line about 1960 and replaced all of them with the Canon FL line by 1967. You don't remeber that Nikon introduced yet another finder for the Nikon F with more elaborate metering every couple of years or so. They finally got it right with the AI cameras in 1977. My point is that this is the very early stages of the so called digital revolution. Cameras will be replaced with new and improved models every year or so. But, you don't have to get on this bandwagon. Just buy the current camera that meets your needs, and only replace it when it dies. Until that time, the camera will continue taking as good an image as it ever did. Jim > But, you don't have to get on this bandwagon. Just buy the current > camera that meets your needs, and only replace it when it dies. Until that > time, the camera will continue taking as good an image as > it ever did. I certainly won't be getting on that bandwagon, I can't afford it for starters :-) I had an Olympus 720 and recently made an upgrade to a Fuji S5500. The main reason I upgraded was the shutter lag on the Olympus was pretty bad and was useless for action shots. This Fuji is pretty quick on the shutter so I am happy. At a certain point, continuing technology advance is not a 'necessity', but a 'frill'. At that point there is no reason to buy newer. In the case of computers, if your needs are only office automation, the computers of 7 years ago are plenty! If you're a gamer, faster is always better. Same for cameras. I think that we have arrived at the point where what is on the market for DSLR is all you really need (not true prior to a year or two ago) and newer technology is 'frills'...8megapixels for photo enthusiasts. I think that the P&S market hit that point a couple of years ago, in terms of size vs. cost vs. performance...how many P&S pictures get enlarged enough to need more than 5megapixels??? For pros, "gimme more" megapixels. I have shot professionally using 35mm SLR, medium format SLR, and large format monorail..each has all that I need, and I had no reason to want 'more features'. I didn't buy DSLR until just a couple months ago, because prior to then it cost too much money for the features and performance I wanted...so I waited. Now I have all I need, and something really radically better would need to come along (smaller size, less weight) would need to come along to tempt me enough...I think the DSLR of today is at the same point all my film stuff reached...plenty good enough. The early adopters of digital paid princely amounts for performance that is not even on par with entry level P&S today. We're past that point now. Megapixel rating...at a point, more megapixel only matters for supersized photos or dinky sections of photos enlarged to final size you want. Did film photographers beat on manufacturers to make Kodachrome or Velvia more fine grained? I think megapixel rating is like film granularity...at a certain point no one needs more (other than a very small minority for special purposes). I disag |
Enable EMail Alerts
Start New Thread
Reply to this Message