 |
 | 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 / General Photo Topics / Australian Photography / November 2005Canon 350D
So I haven't had an SLR before and I am thinking of getting a Canon 350D with the standard 18- 55mm lens and just wonder if it is worth spending the extra to get a 55-200mm lens? I found a 75-300mm lens to go instead of the 200mm would this be a better option? Comments apprieciated thanks. > So I haven't had an SLR before and I am thinking of getting a Canon 350D > with the standard 18- 55mm lens and just wonder if it is worth spending the > extra to get a 55-200mm lens? I found a 75-300mm lens to go instead of the > 200mm would this be a better option? > > Comments apprieciated thanks. It really depends on what you think you might be photographing. Do you need a longer lens or do you just want people to approach you and say " Gee, that's a long lens" or something equally as inane. Given the choice between a long or wide lens I would always go for the wide as I find them to be far more useful for many picltures than a long lens. Again, it depends on what you are photographing. nb > It really depends on what you think you might be photographing. Do you need > a longer lens or do you just want people to approach you and say " Gee, [quoted text clipped - 5 lines] > > nb So don't worry about the extra lens and save to buy a wide one when I get used to the camera? Thanks > So I haven't had an SLR before and I am thinking of getting a Canon 350D > with the standard 18- 55mm lens and just wonder if it is worth spending [quoted text clipped - 3 lines] > > Comments apprieciated thanks. Tamron & Sigma have 18-200 lenses to suit Canon digital SLRs. Not sure whether that's an option for you. The reviews I've read suggest the Sigma is the better lens, but at the price don't expect anything like "L" quality. http://www.dpreview.com/news/0502/05021402sigma18-200dc.asp >>So I haven't had an SLR before and I am thinking of getting a Canon 350D >>with the standard 18- 55mm lens and just wonder if it is worth spending [quoted text clipped - 9 lines] > anything like "L" quality. > http://www.dpreview.com/news/0502/05021402sigma18-200dc.asp Actually... The Sigma 24 -70 f2.8 'DG' lens is a stellar performer on 5D or an EOS film camera. You only use the centre portion of the lens with a 350D so it wouldn't be much different. This lens most definitely *IS* better than Canon's 'L' series of the same proportions. This picture http://www.auspics.com/big/IMG_0017_01 and the next one in the line show just how good this particular Sigma lens is. It costs just $1000 compared to a Canon lens at $2500. My own opinion is that it is so much better than the Canon because it is designed specifically for digital cameras, not film. I shot a 25 member family session last week with this lens and it really will take something out of bag from Canon to better it's image quality. This unfortunately does not apply to all Sigma lenses anymore than it applies to all Canon lenses. It would seem from what I can find on the Internet that a lens needs to be designed to direct light rays at 90 degrees to the sensor to avoid purple fringing. A lens not digital specific directs light at variable angles and this can overload the photo receptors which spill over onto neighboring pixels and cause the dreaded fringe fringe. Chromatic aberrations (red and purple fringing) are caused by lenses with no or poorly engineered apochromatic capabilities. As a general rule of thumb, it is best to avoid lenses with plastic (acrylic actually) lens elements on Canon DSLR cameras. The God awful "kit" lens from Canon has such elements and right up until I had the opportunity to shoot a bride with the kit lens and a 'L' lens, I would have believed sharpness equated to image quality... It doesn't!  SignatureDouglas... Specifications are good to read but When it comes to judging Digital Cameras... I'm in the "how do the pictures look" category. > a lens needs to be designed to direct light rays at 90 > degrees to the sensor to avoid purple fringing. Just another urban myth ... Apart from the lenses made specifically for the so-called APS-C format digital SLRs, the only real difference between a 'film' lens and a 'digital' one is the coating on the rear element. 'Digital' versions have better coatings to cope with the fact that digital sensors reflect more light back onto the lens that film does. With the so-called APS-C format lenses, which have the same coatings, the rear element sits so much closer to the lens that the angle of attack around the edges of the frame is much the same as it is for conventional lenses in a 135 format system. Cheers >>a lens needs to be designed to direct light rays at 90 >>degrees to the sensor to avoid purple fringing. > > Just another urban myth ... "Lenses designed to be digital" As the pixel resolution grew and pixel density became more of an issue many cameras faced issues with their lens choices. In 2000 with the introduction of the E-10 Olympus pioneered the first super high-quality zoom lens for digital photography. This 4x zoom lens features ED glass, Aspherical, and Low dispersion glass elements in a design which focuses all the light from edge-to-edge on the CCD to give extremely sharp, smooth and clear images." I've seen the results of light fall off on all DSLR sensors from the 2x crop factor Kodak sensors used in Olympus cameras to the full frame sensors of 5D and 1Ds cameras and all in between. You may choose to believe it is all bullshit but I have the evidence in images that say it is not. I'm not an Optician. I wouldn't know how to tell the difference between fluorite lens material and a coke bottle base but I sure as hell know if it swims, has gills and scales, it is probably a fish. Likewise I know if the edges of a picture are less well defined and have purple fringing in the corners but not the centre, it's the lens design that's at fault. SignatureDouglas... Specifications are good to read but When it comes to judging Digital Cameras... I'm in the "how do the pictures look" category. > "Lenses designed to be digital" > As the pixel resolution grew and pixel density became more of an issue [quoted text clipped - 4 lines] > all the light from edge-to-edge on the CCD to give extremely sharp, > smooth and clear images." Nothing in that Olympus marketing hype (http://www.olympusamerica.com/cpg_section/cpg_innovations.asp) about changing the angle at which light strikes the sensor or to "direct light rays at 90 degrees to the sensor to avoid purple fringing", which is what you claimed. Having changed the lens mounts to go with their adoption of the 4/3rds system, of course Olympus needed new lenses! The they needed something good to say about them ... a 4mp camera with a mere 4x zoom 9 to 36mm lens (35 to 140mm equivalent on a 35mm camera) is hardly pushing the optical limit. > I'm not an Optician. You don't need to be an Optician to be able to work out that the only way to reduce the angle at which the light from a rear element of a given diameter strikes the sensor is to move the element further away from the sensor. The so-called APS-C format lenses do the exact opposite! > I wouldn't know how to tell the difference between > fluorite lens material and a coke bottle base but I sure as hell know if > it swims, has gills and scales, it is probably a fish. Shark, maybe? > Likewise I know > if the edges of a picture are less well defined and have purple fringing > in the corners but not the centre, it's the lens design that's at fault. It might. It might also be poor micro-lens design on the front of the sensor itself. If the lens doesn't produce these results with film, the resolution of which still far surpases even the best digital SLRs, it's unlikely to do so with digital. Think about it. >>"Lenses designed to be digital" >>As the pixel resolution grew and pixel density became more of an issue [quoted text clipped - 14 lines] > 36mm lens (35 to 140mm equivalent on a 35mm camera) is hardly pushing the > optical limit. Give it up mate! It took me all of 20 seconds to find this: http://www.olympus-esystem.com/dea/technology/dl/index.html It should suit you, it's all in pictures! :-) Next time do some research of your own. SignatureDouglas... Geez, it pains me to look *almost* like I am supporting Douglas.. But, macropod, I have to ask - are you involved in CCD/CMOS design? I have seen a number of references scattered across the web in regard to microlens problems that *DO* mean that some devices have issues, especially with any lens (usually wide-angle of course) that has highly slanted light paths. Here's a quote from the first link I found on this topic, from a producer of cinematic CCD and CMOS lens/sensor systems.. ==========Quote============= Microlenses can boost effective fill factor to approximately 70%, improving sensitivity (but not charge capacity) considerably. The disadvantage of microlenses (besides some additional complexity and cost in fabrication) is that they make pixel response increasingly dependent on lens aperture and the angle of incident photons. At low f-numbers, microlensed pixels can suffer from vignetting, pixel crosstalk, light scattering, diffraction (Janesick, 2), and reduced MTF-all of which can hurt their resolving power. Some of these effects can be minimized by image processing after capture (which is what happens in most digital still cameras using microlensed sensors). ================== This quote comes from: http://www.dalsa.com/dc/documents/Image_Sensor_Architecture_Whitepaper_Digital_C inema_00218-00_03-70.pdf (a long and quite interesting article) In case you are wondering about the 'Janesick' reference - he (James J..) is a CCD imaging pioneer. And from dpreview (not that i would call them a scientific and indisputable source.. (O;) "We'll call it 'purple fringing' but it's clearly a combination of chromatic aberrations caused by the interaction of the lens and the microlenses on the sensor and enhanced by a certain amount of blooming which carries the color out further than the original artifact." And yes, there's even Douglas' Olympus link, which quite clearly says: "Unlike silver-halide film, an image sensor's photodiode for converting light into electrical signals is below the sensor surface. Consequently, light will not reach the photodiode unless it hits the sensor nearly head-on. This can impact the image in a variety of ways, including darkness and blurring. No consideration is given to the angle of incoming light with lenses for silver-halide cameras, because the angle at which light strikes the film has no impact on its photo- reception....The ZUIKO DIGITAL lens.. (is)..specially designed for digital applications, it allows light to strike the image sensor nearly head-on, minimizing degradation of resolution and light intensity, even in the image periphery or when using a wide-angle lens." .. And the link has nice little diagrams. And Olympus has quite a reputation for it's series of 'digital' lenses. Is Olympus bullshitting? I'm genuinely interested if you really do know better, but would appreciate some *real* references and links. It seems to me that there is some truth in this 'myth', and of course your average camera manufacturer (except perhaps Oly?) is not exactly going to be leaping out and telling everyone about the problem and how they deal with it. ..Or fail to.., eg in the case of the poor old Sony DSCF828 (which IMO was still a very fine camera) and many others. Are you guys for real? We're getting way OT, but what the heck! Both of you have been sucked in by good old-fashioned marketing hype and 'chrlz' even cites someone who doesn't even know what exposure latitude is as an authority on photography. He might know his stuff about imaging sensors, but I'd leave it at that. Without placing too fine a point on it, the Olympus web page with its lovely diagrams is disregarding the laws of physics. The only other way to reduce the angle at which light from a given lens reaches the film/sensor is to move the rear element further away. 'nearly head-on' is not the 90-degree angle Douglas claimed - for all we know, the Olympus definition might mean 60-90 degrees, or more. In order for all the light from a lens, in this case the rear element, to strike the film/sensor at a 90 degree angle, two conditions would have to be satisfied: 1. The diameter of the lens would have to be at least as large as the diagonal of the film/sensor; and 2. The lens would have to be constructed so that all the light passing through it left it at a 90 degree angle to the film/sensor. I know of no lens where either of these is true, let alone both of them. Every lens I know of has a smaller diameter than the diagonal of the film/sensor; any light reaching the film/sensor outside the diameter of the lens can't be reaching it at a 90 degree angle. Furthermore, every lens ever made has an attribute called 'focus', which is achieved when the light from the whole surface of the lens *converges* on the focal plane. This convergence involves light from many angles, including 90 degrees. Were it possible to achieve focus without convergence, which is what the 2nd criterion demands, we'd be able to have macro lenses with infinite DoF at any aperture. But hey, don't take my word for it. You can prove it for yourself with very little effort. All you need is a low-powered laser pointer and the (unmounted) lens of your choice. Set the lens up so that its rear element faces a wall at a distance of, say, 1m. Adjust the lens for the close focusing. It doesn't matter whether the focusing distance matches the distance from the wall; you just need to avoid the light being cut off by the mount at the rear. Now shine the light from the laser pointer on the front of the lens so that the light passes through it and onto the wall at the rear. If the lens is constructed so that all the light passing through it leaves at a 90 degree angle to the film/sensor, you should be able to establish that no matter what you do, moving the pointer around in front of the lens is incapable of projecting an image that defines an area larger than the film/sensor. What you'll soon find is that you're if fact able to project a beam through the lens that defines a much larger area than the film/sensor size. That's because the light doesn't only leave it at a 90 degree angle to the film/sensor. QED Cheers PS: The 'wide-angle' issue cited by 'chrlz' is another red-herring - the angle in front of the lens makes no difference at all to how light strikes the film/sensor - that's the rear element's job. What you typically find is that the rear element of a wide-angle lens is smaller than for a telephoto lens, and that means less of the film/sensor has light striking it from directly overhead. > Are you guys for real? We're getting way OT, but what the heck! > [quoted text clipped - 29 lines] > criterion demands, we'd be able to have macro lenses with infinite DoF at > any aperture. snipping like mad here > Cheers > PS: The 'wide-angle' issue cited by 'chrlz' is another red-herring - the [quoted text clipped - 3 lines] > lens, and that means less of the film/sensor has light striking it from > directly overhead. Chrlz is my Pet troll. Someone said a while ago: "You've never really made it on the Internet until you get your own personal troll". I got him, by Golly! and he's a f.cking dork :-) It would have been good if he could get things right now and then but hey, you can't have everything in a troll! Anyone who believes his demented rantings really has to take a long hard look at themselves and decide if maybe it's time to give up newsgroups. This guy's main claim to fame is his unrepentant determination to "expose" me as some sort of crook using a workstation at a South Australia youth and unemployment centre. I'm surprised he still has access to it. When I said I can enlarge a digital image to proportions he doesn't believe I can and some of the people who work for me, posted something to substantiate my claims from PCs on a network inside my router which displays a single (static) IP identity for 8 different PCs and a server, he got the idea I was pumping up the price of a print centre I was selling by fraudulently impersonating my factory foreman. Yeah, right... Like everyone following newsgroups is gonna fork out $40k because of a post here about our algorithm deal with Samsung Tecwin... *NOT*! According to this loser of the primary kind who can't bother with a phone book to find someone ...he has contacts high up in the investigative arm of just about every Government authority in Australia whom he has at one time or another threatened to get onto my case. Curious creature this one. No one seems to know this mysterious Mr Stevens from the Ports of South Australia... Up in smoke like his joints, perhaps? What case? Buggered if I know. His imagination is more fertile than seagull sh.t. Come to think of it... Hmmm. Anyway, I wasted enough spare cash on a PI to track him down and discover he's just a little sardine in a muddy puddle making noise just to be heard. He can thank his lucky stars he hasn't got zilch worth suing for or I'd have it by now for his defamation and slander. As for Olympus verses Macropod? So far it's Olympus 1. Macropod 0. That could change at any moment. All that's needed to reverse the score is for Macropod to 'fess up with some optical engineering qualifications to back up his claims the Optical Engineers at Olympus have got it wrong and like a thunderbolt, I'll switch sides. I gotta tell you tho... It's not looking good for the little Aussie battler right now. I wait with painted breath for the next exciting episode of "do digitals cameras need special lenses to avoid overloading adjacent photo receptors and stop producing purple fringes on back lit objects"... Sort of like Nikon's explanation of why they have 2 same colour photo receptors at each pixel and 4 channel RGB when there can only ever be 3 channels in their D2X... The old feed 'em peanuts and the monkey's will thrive thing, no doubt. SignatureDouglas... Specifications are good to read but When it comes to judging Digital Cameras... I'm in the "how do the pictures look" category. >> Are you guys for real? We're getting way OT, but what the heck! >> [quoted text clipped - 104 lines] > channels in their D2X... The old feed 'em peanuts and the monkey's will > thrive thing, no doubt. at least ther was a white paper on fuji's sensor! > Without placing too fine a point on it, the Olympus web page with its lovely > diagrams is disregarding the laws of physics. The only other way to reduce > the angle at which light from a given lens reaches the film/sensor is to > move the rear element further away. 'nearly head-on' is not the 90-degree > angle Douglas claimed - for all we know, the Olympus definition might mean > 60-90 degrees, or more. my thoughts exactly when looking at the brochure... it's crap! actually if it was possible, I beleive the light would not be focusing in a point and hence you would get nothing but blur! kosh >> Without placing too fine a point on it, the Olympus web page with its >> lovely [quoted text clipped - 12 lines] > > kosh Oh... Kosh. Here we go again, eh? You really are starting to sound like Chrlz and Macropod here. Do you really think a company - a leader in the field of Optical engineering - is going to come out publicly with a new type of lens and it all be "crap"? Get on the program man, hallucinations are controllable. If you are going to have any credibility in your assessment, you need to have some evidence to contradict the evidence I produced when Macropod prodded me with his similar claim... The only difference between him and you is he choose to have a go at me directly, instead of looking first to see if what I said 'could' have any truth to it. In your case you read the proof and dispute that. Fine by me mate. Which University did you get your degree in physics from? It might help if you also had some (any) qualifications which could put your opinion out of the "another aussie with attitude" area and into some proper perspective in the field of optics. As far as Nikon's D2X sensor goes... Read the brochure. http://www.nikonusa.com/fileuploads/pdfs/D2X.pdf It says in very contradictory terms and (thankfully for some) pictures that the sensor has 2 separate photo receptors for some colour pixels and it also says Nikon have done the impossible... Produced 4 RGB channels from 3! It goes on to say you can also switch off the outer half of the sensor and shoot with the centre half. It's a *Dual sensor*. My pet; Chrlz and some joker with a microchip for a brain in the USA refuses to accept this as being a dual sensor... Their problem, not mine. I take photographs for a living, not design the cameras I use. If a manufacturer of gear I use tells me the lens I need is "Designed specifically for digital cameras", unless someone with as much credibility and as many qualifications can produce evidence (any evidence) to contradict the claim, I'll believe the manufacturer. Olympus said the quality of their photos is because of the design of their lenses - it sure as hell couldn't be because of the Kodak sensors. Nikon says their sensor in the D2X has dual photo receptors for colour and the sensor can switch between outer and inner pixels. It therefore has to be a "dual Sensor" by definition. No one contradicting this has offered any evidence they are anything but newsgroup junkies as qualifications for dumping on me. None of the f.cking w.nkers who set out to belittle me because I choose to believe people or companies with published qualifications or leaders in their field instead of a newsgroup troll, have got a hope in hell in this argument. Don't get into their class Kosh. Have an opinion for sure but when the chips are down, normal intelligent people will look to see who is qualified to dispute other qualified information. Right now it's not looking good for the little Aussie battlers, Kosh and Macropod. It's Olympus 2, Aussies 0. Nikon 1, a bunch of yankie w.nkers (and Chrlz)0. Me? I'm just a spectator taking pictures. :-) SignatureDouglas... Specifications are good to read but When it comes to judging Digital Cameras... I'm in the "how do the pictures look" category. Douglas will do anything to try to get out of his error about the Nikon sensor. >As far as Nikon's D2X sensor goes... Read the brochure. I did. Carefully. >http://www.nikonusa.com/fileuploads/pdfs/D2X.pdf It says in very >contradictory terms No, they are not contradictory at all. > and (thankfully for some) pictures that the sensor has 2 separate >photo receptors for some colour pixels Sighs wearily. Douglas, did you not know that the typical digital sensor is actually NOT an rgb sensor, it is an rgbG sensor. There are two (2) (TWO) greens in every array. So *all* cameras have 4 'channels' of data. But how they process those channels may differ. F'rinstance, the Sony sensor in the F828 uses a different colour for the second green ('Emerald'..!) to allegedly get better colour, but it is the only camera I've heard of that does this. All the rest are RGBG. FOUR channels are read by the sensor, but that fact is irrelevant to normal imaging because the two G's are combined before you can get at the data. Hence everyone refers to RGB, not RGBG. (Feel free to do a little of that research you are so proud of, and find out why a second green is used. But most of *us* know.) >Nikon have done the impossible... Produced 4 RGB channels from 3! Bwahaha! Like every other camera before them. Have a close look at the upper part of the diagram, and you will note that the CONVENTIONAL system also has 4 streams - RGBGRGBGRGBG. Just LOOK. BOTH types of sensor have the same array of 4 sensors for each site. One red, one blue and TWO greens. Nikon are just processing it differently. And that has nothing to do with your silly 'dual sensor' claim. It is NOT a dual mode sensor. You *can* get the camera to not process the outer periphery of the read data - which increases its shooting rate at the cost of resolution - but this is not a sensor design issue, it is a data-processing issue. >It goes on to say you can also switch off the outer half of the >sensor and shoot with the centre half. NO IT DOES NOT. Stop bullshitting. It says, and I QUOTE: >High-speed continuous shooting >Optimized camera systems realize unprecedented continuous >shooting at full 12.4 megapixel resolution of 5 fps for >15 consecutive NEF images. When the situation demands yet >faster frame rates, the 6.8 megapixel High Speed Cropped >mode allows 8 fps shooting for 26 NEF images. It does NOT say the sensor is partially turned off, because that is NOT how cropped mode works. The sensor still reads the full data, but then only the centre portion of the data is *processed*. It is that reduction in PROCESSING that gives increased frame rate. You have had this explained to you many times, yet you persist, misquoting, ranting and generally showing your complete lack of knowledge. You simply cannot ever admit error. It's a sure sign of psychosis. Sigh. Staying offtopic.. (O; Just do your experiment again, and then close the aperture right down, then consider the light paths *again*. You have NOT given a single contrary reference to the three provided so far. That's 3-0. Shall I dig up more? Can you show me *any* optics expert disputing the Olympus site? - according to you it should be dead easy, and think of the kudos they would gain from doing so. And please feel free to deal with the basic issues, instead of talking about lenses that don't exist, and supposing an experiment that no-one is going to do, yourself included. Looking at a single light ray doesn't help much, although it will support what I have said above, especially when you compare light paths from *small rear elements* (which you concurred with below!), and small apertures. So let's just go back to the absolute basics. I think you will have to concede that a ccd or cmos sensor DOES require the light to strike it at an angle that is not too shallow. Whereas film, given its relatively clear substrate and small but still three-dimensional 'grains' has little trouble with quite oblique rays. And microlenses on ccds and cmos's make this issue worse. Agreed? Yes or No? If NO, give reasons. If you concede that.. I can pretty easily prove the remaining issue - namely that wide angle lenses and small apertures result in increased numbers of oblique light rays at the outer peripherals of the film/sensor. Obviously the olympus diagram was not enough, but I think I can do better if I need to.... By the way, *you* said this: "What you typically find is that the rear element of a wide-angle lens is smaller than for a telephoto lens, and that means less of the film/sensor has light striking it from directly overhead." If you concede that a digital sensor requires relatively non-oblique light, and the rear element is smaller on a wide angle, then *of course* the light rays at the outer edges of the film/sensor will have MORE oblique rays than right-angled ones. That doesn't require a picture. So which bit do you dispute? Be concise, please. PS - My apologies for Douglas - he's lost the plot. He gets so uptight about me that he doesn't even notice when I'm supporting him! (O: ======================= Ok, that was the short version. Read on only if you are a glutton for punishment. I'd hate to be accused of selectively quoting, or throwing red-herrings... So here is every point dealt with... >Are you guys for real? We're getting way OT, but what the heck! But it's interesting, and the myth is not a myth IMO. >Both of you have been sucked in by good old-fashioned marketing hype >and 'chrlz' even cites someone who doesn't even know what exposure >latitude is as an authority on photography. Hullo? Can you *quote* whatever the heck you are talking about? You criticise me for red-herrings, yet you blithely claim this, with not a single reference. Instead of attacking someone's reputation (and without evidence), just quote what he got wrong, and give the reasons WHY. >He might know his stuff about imaging sensors, but I'd leave it at that. Who? The entire Dalsa Corporation? Olympus? J.J. Janesick? Like I said - LINKS and contrary evidence please. >Without placing too fine a point on it, the Olympus web page with its >lovely diagrams is disregarding the laws of physics. You keep saying this, but where is the actual wrong part? Be specific. > The only other way to reduce the angle at which light from a given lens.. 'Other way'? Which was the first way? >reaches the film/sensor is to move the rear element further away. Whaaat???? The angle of the light striking any given part of the imaging area depends on the lens design *and* the aperture. This is in Optics 101.. It's why you can fry an ant (aren't kids horrible!) with a magnifying glass. BUT if it is a low-powered one (ie different focal length, just like tele versus wide), or a flat piece of glass, *at the same distance*, it won't work. Why? Because the angle of those light rays changed. Yes, distance away does affect the angles also, but it is by no means the 'big one'. And it is largely irrelevant to the argument - if the angle is made greater by *anything*, in a specific area of the image sensor/film, then the problem exists. >.. 'nearly head-on' is not the 90-degree angle Douglas claimed - for all we know, >the Olympus definition might mean 60-90 degrees, or more. Your point being? >In order for all the light from a lens, in this case the rear element, to strike >the film/sensor at a 90 degree angle, two conditions would have to be >satisfied: Whaaat??? I didn't say for a moment that all the light had to be at 90 degrees. About the only way to get that ludicrous situation is to have no lens and an infinitely distant source. Or perhaps a projected image, retransmitted.. We are talking about *normal* lenses. For a normal lens, the area of the image plane that gets most of those 90 degree rays is the dead centre. Here's *my* experiment. Use a standard SLR lens. Stop it down to f16 or smaller. Hold it dead still, then move your eye behind it across an imaginary rectangular area.. When your eye is at the corner or outer edge of that area, what angle are those rays coming from, through that little hole? (hint - a quite OBLIQUE angle) When your eye is right behind it, what angle? (hint - 90 degrees) >1. The diameter of the lens would have to be at least as large >as the diagonal of the film/sensor; [quoted text clipped - 3 lines] > >I know of no lens where either of these is true, let alone both of them. No-one said it was. This is irrelevant. But you CAN design a lens that provides some degree of correction for the outermost light ray paths (badly worded, but what the heck), and improves upon the angle that the light strikes *all* of the sensor. >Every lens I know of has a smaller diameter than the diagonal of the >film/sensor; any light reaching the film/sensor outside the diameter of the >lens can't be reaching it at a 90 degree angle. EXACTLY. Which *proves* what I have been saying. Think about it. At the centre of the image area, what angle is the majority of the light hitting the sensor - 90 degrees!! You said it yourself. At the outer edges, what angle is most of the light hitting the sensor? You just said the lens is smaller than the sensor, so it cannot possibly be getting as many 90 degree rays. And if you stop it down, the ratio, and therefore problem, increases. >Furthermore, every lens ever made has an attribute called 'focus', which > is achieved when the light from the whole surface of the lens *converges* >on the focal plane. Yes, of course, but this is not the problem. > This convergence involves light from many angles, including 90 degrees. Again, true. But the RATIO of 90 degree versus oblique light VARIES across the image sensor. In general terms, and all things being equal, more oblique light hits the outer edges, for EVERY lens type. And (again all things being equal) wide angle lenses have a higher ratio of oblique versus 90 degree light as you move outwards on the sensor. In oparticular, visulaise that wide angle lens stopped right f\down, so that all of the light from a very wide angle is going through a tiny opening in the lens, and *then* has to be thrown out through the rear element spread out over the sensor/film. > Were it possible to achieve focus without convergence, which is what the 2nd >criterion demands The '2nd criterion' (ie a lens that passes all light through at 90 degrees????) was on eof those red-herrings you referred to earlier. It is not relevant and doesn't exist. >we'd be able to have macro lenses with infinite DoF at any aperture. And we don't. Again, it's irrelevant. >But hey, don't take my word for it. or mine.. Do your own research folks. >You can prove it for yourself with very little effort. All you need is > a low-powered laser pointer and the (unmounted) lens of your choice. You don't need a laser pointer for my experiment above, which is much simpler. Also, low-power does not necessarily mean safe. >Set the lens up so that its rear element faces a wall at a distance of, say, 1m. >Adjust the lens for the close focusing. Why? >It doesn't matter whether the focusing distance matches the >distance from the wall; you just need to avoid the light being >cut off by the mount at the rear. I don't follow this, but go on.. It sounds good.... >Now shine the light from the laser pointer on the front of the lens >so that the light passes through it and onto the wall at >the rear. >If the lens is constructed so that all the light passing through it leaves >at a 90 degree angle to the film/sensor Oh, I get it, you are back to that second criterion, namely that lenses don't emit light at 90 degrees. We KNOW they don't. Why do you keep wanting to prove that something that doesn't exist, doesn't exist? What we are trying to prove, and you seem to be wandering dramatically, is that a TYPICAL lens provides more 90-degree-ish light at the centre than the edges. It does. And your experiment doesn't help with that. >snip< >That's because the light doesn't only leave it at a 90 degree angle to >the film/sensor. QED Exactly. And if it doesn't leave the lens at 90 degrees, and the rear element is smaller or similar in size to the image sensor, then by absolutely trivial extrapolation, then the outermost edges of the sensor will receive more oblique rays than the centre. Which is what this was all about. Once upon a time. >PS: The 'wide-angle' issue cited by 'chrlz' is another red-herring - the >angle in front of the lens makes no difference at all Not a red- herring at all. The 'angle in front' has to be matched by a similar 'angle in back'. Think about it. That means the problem is worse for wide angle than telephoto. > to how light strikes the film/sensor - that's the rear element's job. No, it is the entire lenses job actually. >What you typically find is that the rear element of a wide-angle lens is smaller >than for a telephoto lens YES. which leads to the inevitable... > and that means less of the film/sensor has light striking it from directly overhead. YES, which is what I was arguing!!! The centre of the sensor still gets the vast majority of the 90 degree rays. The outermost edges MUST get more oblique rays - just put your eye behind that lens again! Then compare it with a telephoto.. If it still doesn't get through, I give up. In summary: 1. CCD and CMOS sensors, especially those with microlens technology, encounter image quality problems quite unlike film if light hits them at oblique angles. 2. Wide angle lenses present the outer edges of the image sensor with more oblique light rays than telephotos, all things being equal. 3. Small apertures also result in more oblique rays striking the outer edges of the sensor. 4. Lens design *can* help. (grin) Douglas said "a lens needs to be designed to direct light rays at 90 degrees to the sensor to avoid purple fringing" and cites an Olympus web page referring to having "light to strike the image sensor nearly head-on", saying that proves him correct. Well, it doesn't. All it proves is that Douglas can't read. The rear element on the Olympus E-10 has a diameter of 16mm, compared to a 22.5mm diagonal for the sensor. By way of comparison, the rear element on the Olympus 75-150 f4 lens for the OM series has a diameter of 29mm. So, unless Olympus digital technology includes magic, for which you or Douglas can presumably provide the link to the relevant incantations, there's no way known the light from a 16mm optic can strike a 22.5mm sensor at 90 degrees as Douglas (not Olympus) claimed. Sure, some of it does, but very little overall. What this camera's schematics show is that the angle at which the light strikes the sensor being anything up to 34 degrees from perpendicular to the sensor. That's what Olympus means by its ability to have "light to strike the image sensor nearly head-on". It's about time Douglas studied optics, instead of relying on "sales brochure" web pages for the gullible. But then given his Nikon 'dual sensor' drivel, that's probably too much to hope for. But, just in case Douglas reads this and wants to actually learn something (ha), perhaps he'd like to try this web page for starters: http://www.physics.utoledo.edu/~lsa/_color/15_lenses.htm Even includes a cut-away drawing of the Olympus E-10, but no mention of any supposed ability to have the light striking the sensor at only 90 degrees. I think chrlz, that deals with your 'plea' to "go back to the absolute basics". And it clarifies, for the Olympus E-10 at least, how far from head-on the light rays can be. The reason I said that a smaller rear element can reduce the angle of incidence is that light rays leaving its extremeties and striking the opposite side of the sensor will then be less acutely offset from it. The trade-off is that you also have less light approacing the sensor from a more oblique angle also. The effect is no different than moving a larger sensor further away from the sensor. Getting back to the OM series for a moment, but I can demonstrate the same thing with other makes, the rear element on the 75-150 f4 lens has a diameter of 29mm. This makes for a fairly acute angle when the light from, say, the right side of the lens travels to the left side of the film. With the 35mm f2.8 lens, the rear element is only 15mm wide. Consequently, the wide-angle lens imposes a less acute angle when the light from one side of the lens travels to the opposite side of the film. > Douglas said "a lens needs to be designed to direct light rays at 90 degrees > to the sensor to avoid purple fringing" and cites an Olympus web page [quoted text clipped - 41 lines] > the light from one side of the lens travels to the opposite side of the > film. You're doing well to have hung in the way you have Macro. Your determination to find a reason - any reason, to support your contradiction of Olympus's claims is like clutching at straws to keep afloat. I still don't have a clue what you are on about. Olympus put up a diagram to demonstrate their "made for Digital" lenses and it shows the light emitting from the rear element of a "designed for digital" lens as exiting perpendicular but the light from a "designed for film" - an OM lens, as the way you describe it *has* to be. I'm still with Olympus on this, Sorry Macro. Your are starting to flounder here. Getting closer to that fuckwit Chrlz in your determination to find something about what I repeated to attack me personally. Why not just give it up? I'm with Nikon's factory representatives too on the case of a dual sensor in the D2X. You are so eloquent at explaining why something described by it's designers as doing what they say it does, tell us then, how do you switch off the outer half of your sensor whilst still retaining full use of the inner half if it's not a "dual Sensor"? Almost beyond bladder control waiting this one. :-) SignatureDouglas... Specifications are good to read but When it comes to judging Digital Cameras... I'm in the "how do the pictures look" category. > Are you guys for real? We're getting way OT, but what the heck! Give it up, mate. You are asking a person with a single digit brain cell count and a degree in common bullshit to think. That is way too far fetched. As far as the size of the last element, it doesn't really matter much, as all wide angle lenses have to be made as inverted telephotos (because the back focal distance governed by mirror cavity is far greater than wide angle lens' focal length). Even a 'normal' APS lens (~30mm f.l.) will be made like this. A typical APS wide (say 10-12mm) cannot be made to have angle 90 degrees to sensor with a typical BFL of 40+ mm and be free of coma. Ray bundle must pass the second cardinal point (where lens stop is normally situated), and that HAS to be close to the film - that is you cannot make a functioning wide angle with abnormal BFL (a necessary prerequisite to have 90 deg angle of incidence). In other terms, Olympus page is highly misleading at best. I'd ask them to show a ray trace of some of their APS wide angles (do they make 10mm?) and THEN ask them about "90 deg rule". Plus purple fringing is not caused by angle of incidence anyway, this is what normal people call lateral chromatic aberration, and it is a function of the lens, not detector type. I liked your laser pointer thought experiment. Too bad it flew way overhead of those targeted. Thanks for the insults, Bratislav. In case YOU hadn't noticed (and I won't debate *your* intelligence quotient for that error), macropod clearly implied that wide angle lenses do not cause problems with digital sensors. He also implied that the oblique ray issue was a red-herring. Here is what he said. >You don't need to be an Optician to be able to work out that the >ONLY way to reduce the angle at which the light from a rear element >of a given diameter strikes the sensor is to move the element further >away from the sensor. Do YOU agree with this, Bratislav? Be honest - do you *also* think that is the ONLY WAY to control the angle of incidence across the image area, and that nothing else is relevant? You might want to pause and reflect (or is it 'refract'?) a little. And then he said: >If the lens doesn't produce these results (problems at outer edges of sensor) > with film, the resolution of which still far surpases (sic) even the best digital >SLRs, it's unlikely to do so with digital. So here macropod says that the angle of light incidence is irrelevant, and that digital sensors react the same way as film. Do YOU agree, Bratislav? He then went on about trying to build a lens that has all its light coming out at 90 degrees, yes, fascinating stuff - but the real issue is designing a lens that gives a better balance of right-angle and oblique light, in particular towards the edges. The Olympus site may be exaggerated, but you suggest they are LYING, is that correct? Here's what they say: >Specially designed for digital applications, it allows light to strike >the image sensor nearly head-on Note the words 'nearly head on' - that is not 'everything at exactly 90 degrees'. It is an esaggeration and the diagrams are obviously simplistic and designed to make you think about the problem, rather than show an actual ray-trace. Sheesh. What do you guys expect from a one page advertising grab? They go on.. > minimizing degradation of resolution and light intensity, even in the image >periphery or when using a wide-angle lens Gee, they seem to think there is a problem with wide angles, too. So that's Dalsa, Olympus, Mr Janesick, and I can supply plenty more (see below). Where are YOUR countering references? Neither macropod nor you has come up with any. And yes, us lower forms of intelligence *do* know that it is pretty much impossible to produce a lens that gives 90 degree output. As I said in the first post. Perhaps you should actually *read* it. Then you could QUOTE the wrong parts and give your corrections. Until then, hold the insults. They will be ignored, and i will keep posting *information*. Do feel free to correct my mistakes..... Yes, there are other issues here where macropod is quite correct, but they are not relevant to those two VERY misleading statements, which needed clarification. In my humble and non-insulting opinion. I'm still waiting for a counter reference or statement of specific error.... Why is that? Oh, and you may wish to do a bit more research on purple fringing.... Before dismissing the possibility, wouldn't it be better to keep an open mind? Here's some light reading for you before ya go: http://www.pictureline.com/newsletter/2004/july/purplefringe.html http://www.dpreview.com/learn/?/Glossary/Optical/chromatic_aberration_01.htm http://www.sd3.info/pf828/PFree/PFree0-1.html http://www.photoreview.com.au/Articlexasp/ff89a68a-2a67-4543-9bd0-6626e610deaf/D efault.htm (read the bit about blooming) http://www.bobatkins.com/photography/reviews/18-55d.html (yep, even Bob thinks there might be something to it, but he avoids going deeper!) http://www.dl-c.com/discus/messages/36/2357.html?1085412913 (and Jonathan Sachs too.. Do you know who he is?) And visit this one and tear it apart too.. http://www.shutterbug.com/features/0305digitally/ I can post loads more, but I think most folk will get the picture. An awful lot of manufacturers seem to be doing this optimising-for-digital stuff. But I'm sure you and macropod are right. I guess they think we are all suckers, and they can just play on our insecurities and utter stupidity.... Oh, and I can't resist adding this one, from Imatest (not that they would know anything....) http://www.imatest.com/docs/sfr_chromatic.html =quote= Purple fringing is not chromatic aberration, though it is often mistaken for it. It is a saturation phenomenon in the sensor, caused by the overflow of electrons from highly saturated pixel sites to nearby unsaturated sites. It tends to be worst in cameras with tiny pixels (e.g., 8 megapixel compact digital cameras). IT HAS EVERYTHING TO WITH THE SENSOR and NOTHING TO DO WITH THE LENS. =end= Sorry about the capitalisation... So, Imatest backs me up, but you still reckon I'm wrong. Wow, what a leap of logic that must be! > Oh, and I can't resist adding this one, from Imatest (not that they > would know anything....) [quoted text clipped - 10 lines] > > Sorry about the capitalisation... My final post here, because this is just a waste of time.. macropd said: >So, Imatest backs me up, but you still reckon I'm wrong. That was in regard to PF not being CA, which was BRATISLAV. If *you* disagreed with Bratislav on that point, then yes, I agree 100% on that. But how is it that you didn't comment on where I argued with YOU, eg YOU said: >You don't need to be an Optician.. the ONLY way to reduce > the angle at which the light..strikes the sensor is to move >the element further away from the sensor. WRONG. You even said yourself that this was only for a rear sensor of a given diameter - so you were precluding the lens designer from doing *anything*!! Noone said this was about using an existing lens. The angle of incidence is dependent on distance from centre of sensor, lens design, focal length and aperture. >If the lens doesn't produce.. (problems at outer edges).. with film >.. it's unlikely to do so with digital. WRONG AGAIN. As my links show, and there are hundreds more, digital sensors *ARE* much more sensitive to angle of incidence, and the problems caused are significant to the final image. (end of transmission) > Oh, and I can't resist adding this one, from Imatest (not that they > would know anything....) [quoted text clipped - 10 lines] > > Sorry about the capitalisation... So how come those saturated pixels are always PURPLE ? (capitalisation intended) Normal Bayer mask has twice as many green pixels as it has red/blue. If anything it would be GREEN fringing. Purple color comes from a combination of red+blue wavelengths which in a lens with normal color correction are focused further that green focus (secondary chromatism). In an APO lens, three or more colors are brought to focus, and typically only a violet halo remains. Most noticeable in corners of the image where lateral color is compounded with other aberrations. Anything else I can do for you ? Comment were not targeted at you, but to Douglas. I posted before I saw your answer. > http://www.bobatkins.com/photography/reviews/18-55d.html geez, thanks. Perfect example of a LENS caused purple fringing. Look at the three images supplied. 28-105 - moderate purple fringing. 50mm 1.8 next to nonexistent. 18-55mm SEVERE purple fringing. Different sensors used, eh ? BTW none of the lenses above have apochromatic correction (no ED elements of any sort in them). So chromatic aberration seen is just about expected. And before I go, I am heavily involved with astronomical use of CCDs and CMOS. Even with telescopes like 10m diameter f/1.8 (fast enough for you?) there's not in SLIGHTEST trace of purple fringing. If anything there is RED fringing (caused by a scatter of long wave photons from a glass backing substrate. Not applicable to normal lenses, only instruments with apertures measured in METERS, where photon flux is strong enough). And yes, I'm using DSLR with a fully color corrected telescope (Wright) and there's no purple fringing of any sort. Stars are ideal for this, you know - bight point next to a complete pitch black patch. It will show everything you want (and DONT want) about your lens/telescope sensor, guiding, ambient temperature, light pollution, you name it. > Comment were not targeted at you, but to Douglas. I posted before I > saw your answer. [quoted text clipped - 7 lines] > 18-55mm SEVERE purple fringing. > Different sensors used, eh ? Somewhere the plot has been derailed... Can you keep to the script here by any chance Brat? Chrlz - my pet troll with an attitude bigger than his brain cells... Is on my black list. I only know he's in this thread when someone replies to him. Whatever he has to input is all rubbish, cut and pasted from some often controversial sources he seems to think are not. Has anyone checked out Atkins' qualifications in optical engineering or physics? There are none yet Chrlz would have you believe he is a source of indisputable information in direct contrast to that from highly qualified Professors of Physics and the world's best Optical engineers. Anyone replying to my troll and claiming to be aiming at me ought to stop sniffing all those Subaru fumes and get a grasp on what's going on here. Schoolies week is over tomorrow so maybe we'll see some highly informed and pertinent comments to out class even Chrlz's. Purple fringing and Chromatic Aberration are not the same thing. Purple fringing from the sensor is caused by pixel overload and the shedding of energy to adjacent pixels. Chromatic Aberrations are lens faults. Chrlz seems to be able to dig up false information and paste it into his replies to discredit anything. I guess where total w.nkers like him are let loose, there will always be this sort of thing going on. If he ever got a grasp on what he pastes about, it would be a first. SignatureDouglas... Specifications are good to read but When it comes to judging Digital Cameras... I'm in the "how do the pictures look" category. >> Comment were not targeted at you, but to Douglas. I posted before I >> saw your answer. [quoted text clipped - 34 lines] > let loose, there will always be this sort of thing going on. If he ever > got a grasp on what he pastes about, it would be a first. But Douglas that what .NET stands for - Not Everything's True. (grinning cheshire cat-like) Noite, this post has little to do with the 350D, and is more about sensor problems, and purple fringing.. >Has anyone checked out Atkins' qualifications in optical engineering or >physics? There are none yet Chrlz would have you believe he is a source >of indisputable information in direct contrast to that from highly >qualified Professors of Physics and the world's best Optical engineers. No, that's not quite what I said, but if I have to pick from Bob or Douglas or Brat or macro, all of whom who still selectively quote (if at all) and who continue to post errors (see below)... Gee, difficult choice. >Purple fringing and Chromatic Aberration are not the same thing. Purple >fringing from the sensor is caused by pixel overload and the shedding of >energy to adjacent pixels. Chromatic Aberrations are lens faults. YES, Douglas. That is what *I* said. So Douglas and I agree (scary), as do all those links above, on that one. >Chrlz seems to be able to dig up false information But noone has yet been able to post a quote and then a contradiction. Not once, and not a single countering link/reference. So the information seems to be true.. Perhaps Douglas *should* read my posts, then he might notice he is actually agreeing with me.. (O; >From Bratislav: >So how come those saturated pixels are always PURPLE ? >(capitalisation intended) Normal Bayer mask has twice as > many green pixels WOAH!! - You blew it there! Brat, you even said it yourself.... >Purple color comes from a combination of red+blue wavelengths >which in a lens with normal color correction are FOCUSED FURTHER >that (sic) green focus (capitalisation added) So *even* if that was the only effect, the fact there are twice as many greens does not affect the fact that the red and blue GETS BENT FURTHER. But WAIT - it's worse than that, Bratislav - one of those Greens is used for extra INTENSITY/CONTRAST information, NOT color!!!!! So straight away your argument falls over - of course the MAIN image and the green channel will overlap exactly! Sheesh. For someone who claims high ground in sensor knowledge, the fact you didn't know that the second green does NOT contribute color information is astonishing. And revealing. BUT it's irrelevant anyway - the purple fringing issue is actually more about what LEAKS (electron wise) from one sensor to another. Read those links again. And yes, the problem DOES get tangled up with CA, but that doesn't mean CA is the only problem. You can stick your fingers in your ears and yell 'lalalalala', but the problem remains real. >Anything else I can do for you ? Nope. I can't make the horse drink. My links and comments are self explanatory, and I'll wait to hear from those who have more reputation in this area. This is getting offtopic and silly. You may know your optics, but you don't know your sensors. In summary: 1. You CAN design a lens to have better performance in regard to the angle of incidence. Olympus, and many other manufacturers HAVE done so. 2. The angle of incidence is far less important for film than it is for a digital sensor, especially if that sensor has 'deep' wells and/or microlenses. Oblique light rays can cause numerous problems including low output, diffraction, fringing, softness. Oblique light rays are more of a problem at the outer edges of the image area. 3. Purple fringing is not CA. Although by its nature it loooks similar, and is often found in conjunction with CA (and other Aberrations). > Here's what they say: > >> Specially designed for digital applications, it allows light to strike >> the image sensor nearly head-on And THAT is bullshit. Because you cannot design a wide angle lens (and there's ONLY problem with wide angles, telephotos by definition cover narrow FOV so light WILL be nearly head on anyway - there's nothing to improve on in long f.l. lenses), again CAN NOT design a functioning wide angle lens with BFL far greater than effective focal length to have light in corners hit sensor "nearly head on". Yes, they may improve it (by making a last element larger and more positive, which will add its own dose of headaches to the designer), but not to the point where it is going to make huge difference, and certainly nowhere near what that awful diagram shows. Olympus is playing catching game with Canon and Nikon (who have cream of the crop of Japanese optical designers) and have resort to car salesman tactics. Which obviously works for some. BTW, I've been designing optics for quite some time (went through Beam 2, 3 and 4, OSLO and now using Zemax EE). Quoting few misleading web sites (even YOU agreed Olympus graphs are crap) doesn't work for me, sorry. If a purple fringin was sensor problem, lots of CCD/CMOS manufacturers would be already hard at work trying to find a cause and then fix. Show me ONE site that talks about purple fringing in scientific terms, not non-peer reviewed hearsay crap. Like this one for example, which talks about RED halos (IR photon substrate scatter): http://acs.pha.jhu.edu/instrument/detectors/halo/ (in a nutshel a reprint from SPIE article SPIE 6655 p.608) or http://www.edpsciences.org/articles/aa/abs/2002/12/aa1856/aa1856.html (Astronomy & Astrophysics 384, 763-771 DOI: 10.1051/0004-6361:20011813) or http://www.stsci.edu/largefiles/hst/HST_overview/documents/calworkshop/workshop2 002/CW2002_Papers/walsh.pdf Any references to purple fringing found there ? > If a purple fringin was sensor problem, lots of CCD/CMOS manufacturers > would be already hard at work trying to find a cause and then fix. [quoted text clipped - 19 lines] > > Any references to purple fringing found there ? Hey Brat... Aren't you the one who pointed me to some site boldly proclaiming you could take a hex wrench to your 300D (maybe 350D) and adjust the back focus yourself? Did not that same site describe in detail how to dismantle your DSLR and remove the Infra Red filter over the sensor for some astro reason? Isn't the French Government now investigating the owners of that site with a view to a law suit against them or shutting them down? That is/was hardly a world best standard web site you pointed to for a reference mate, so don't come on now with the demand for scientific evidence from me. I wouldn't know a red Proton from a green one except they are made in Malaysia by company founded by the son of it's former dictator. What I do know is that Olympus simply wouldn't risk ridicule by falsely claiming they can redirect light in a world where they are leaders in optical goods for the medical profession which do bend light around corners and take pictures up people's arses. Now as for the sites you point to for a (cough) scientific reference about purple fringing... Astro Photography is so different from the photography we all practice with cameras here on earth as to be totally irrelevant to the discussion. Go do some surfing on Kodak's site and then head on over to Samsung Tecwin, Korea's largest camera maker and see what they are doing in the development of a sensor pickup point that won't shed it's excess charge and cause it's near neighbors to record a false purple. If you have qualifications as an Optical engineer or Physicist and intend to dispute Olympus for claims about their range of "made for digital" lenses - and keep in mind they make the widest of wide (7mm) lenses in this range, let's have some evidence from you that it is bullshit. Otherwise I'll keep believing the people who actually develop the products, not some alien tourist searching the heavens for ride home. SignatureDouglas... Specifications are good to read but When it comes to judging Digital Cameras... I'm in the "how do the pictures look" category. Douglas, Why didn't you simply abbreviate your last post to: "Don't confuse me with the facts, I've a prejudice to uphold"? All that drivel amounts to the same thing. > Douglas, > > Why didn't you simply abbreviate your last post to: "Don't confuse me with > the facts, I've a prejudice to uphold"? > > All that drivel amounts to the same thing. Since when, Macro... Did you let the truth spoil a good story? SignatureDouglas... Specifications are good to read but When it comes to judging Digital Cameras... I'm in the "how do the pictures look" category. >Are you guys for real? We're getting way OT, but what the heck! Do you guys actually ever take any pictures, or are you too busy arguing and disagreeing back and forth about the minutiae of how digital cameras work.....? > >Are you guys for real? We're getting way OT, but what the heck! > > Do you guys actually ever take any pictures, or are you too busy arguing and > disagreeing back and forth about the minutiae of how digital cameras work.....? I don't know about anyone else but it helps to while away the time taken by your PC to develop 600 or so RAW files by having a few lines on the news groups which doesn't consume much processor time. Just say the magic words David and if it offends you, I'll read a book instead... In fact.... SignatureDouglas... Specifications are good to read but When it comes to judging Digital Cameras... I'm in the "how do the pictures look" category. >I don't know about anyone else but it helps to while away the time taken >by your PC to develop 600 or so RAW files by having a few lines on the >news groups which doesn't consume much processor time. > >Just say the magic words David and if it offends you, I'll read a book >instead... In fact.... I WAS of course just stirring..... >>So I haven't had an SLR before and I am thinking of getting a Canon 350D >>with the standard 18- 55mm lens and just wonder if it is worth spending [quoted text clipped - 9 lines] > anything like "L" quality. > http://www.dpreview.com/news/0502/05021402sigma18-200dc.asp yeah but there is a quality compromise with these all in one lenses... and often close focus. granted you would expect similar results from the 55-200 anyway most mega zoom lenses I have tested need to be stopped down several stops to get any real contrast in the photo. > So I haven't had an SLR before and I am thinking of getting a Canon 350D > with the standard 18- 55mm lens and just wonder if it is worth spending the > extra to get a 55-200mm lens? I found a 75-300mm lens to go instead of the > 200mm would this be a better option? > > Comments apprieciated thanks. I can't comment on the 55-200 but the 75-300 is hardly a world-beater. (I have the optically similar 100-300). There is a new 70-300IS but it is pretty expensive for one UD element and IS. I have read that the 55-200 is OK at f8 and looking at Canon's MTF curves the 55-200 is better at 200mm than my unloved 100-300 is at 300mm, but *nothing* like as good as the 70-200 f4L at 200mm. I'm waiting for Canon to put IS on the 70-200 f4L and then I will buy one. I also have a 350D and the kit lens and I'm happy with those. >>So I haven't had an SLR before and I am thinking of getting a Canon 350D >>with the standard 18- 55mm lens and just wonder if it is worth spending the [quoted text clipped - 6 lines] > have the optically similar 100-300). There is a new 70-300IS but it is > pretty expensive for one UD element and IS. but great if you hav enot had an SLR before.... one does not drive a ferrari when they are only learning.... didn't Ansel Adams use a box brownie to learn????? 75-300.... better than the 55-200 and not a $1000 around $300 (please don't start with the I can find it chaeaper stuff... as a guide only!) beyond this it's a bit like music.... after listening to $5000 speakers for awhile you can never go back.... but if you are used to a $300 outfit, all that extra money will seem like a waste! But after prolonged exposure..... don't do it man! I once said I would just try a Leica lens to see what all the fuss was about....... a bucket of money later..... kosh > So I haven't had an SLR before and I am thinking of getting a Canon 350D > with the standard 18- 55mm lens and just wonder if it is worth spending the > extra to get a 55-200mm lens? I found a 75-300mm lens to go instead of the > 200mm would this be a better option? > > Comments apprieciated thanks. the 75-300 is a better lens..... also the 55-200 is discontinued in anything but 350D kits. as the ccd is smaller your focal length is x1.6 meaning a 300mmm lens would be quite nice indeed. My tip get a tamron 70-300 rather than the canon 75-300 kit lens tamron lens blows the canon images are a lot sharper. > So I haven't had an SLR before and I am thinking of getting a Canon 350D > with the standard 18- 55mm lens and just wonder if it is worth spending the > extra to get a 55-200mm lens? I found a 75-300mm lens to go instead of the > 200mm would this be a better option? > > Comments apprieciated thanks. Just another question I just looked at the price of the Pentax ist dl and it is tempting($1000)... is the canon the wise choice in your mind for the beginner I am? > Just another question I just looked at the price of the Pentax ist dl and it > is tempting($1000)... is the canon the wise choice in your mind for the > beginner I am? I woul ddo it for the cmos sensor alone... plus the extra 2 megapixels. If your on a tight budget hoewver... the pentax is excellent value fo rmoney if you woun't require exoctic accesories. kosh if i can ad my 2 cents pentax ist dl is a worthy camera, I have used a few of them and last sunday took one down to a small aifield and took some shots. after using it, it's almost an old canon EOS film clone in operation, similar mode dial, command dial on the right, the same focus beep etc. The beauty about this camera is the fact that you can use a 30 year old lens on it. That I have done, I put a cosina 28-50 K mount (mid-late 70's vintage from a K-1000) on the IST and the overall picture quailty blew the sigma kit lens out of the water. what chance have you got of getting a 30 year old FD mount on a 350D ... none (unless you want to get a FD-EF adaptor) Anyway back to last sunday I was shooting away on the IST with a tamron 70-300 with UV filter and cokin P circular polariser and shots were better than I was expecting. when i popped the memory card in the PC and looked at what I took I couldn't believe it. one drawback is the ISO, its slowest is 200. If budget is an issue (it is with me :-) ..... ) definitly consider it. >> Just another question I just looked at the price of the Pentax ist dl >> and it [quoted text clipped - 6 lines] > > kosh |
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
|
©2008 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.