If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
Sony tells DSLR shooters they're idiots
["Followup-To:" header set to rec.photo.digital.]
Chris Malcolm wrote: In rec.photo.digital Wolfgang Weisselberg wrote: Chris Malcolm wrote: They [previews] usually show you what you'll get. In certain circumstances they don't. You're right that learning what those circumsntaces are, and if an inquisitive person, why, is a good and useful idea. A brief glimpse of the taken shot flashed up on the LCD or EVF will immediately show you the difference between preview prediction and result. So you need the same time to assess the correctness of the preview mode as you need to check any other parameters. Not necessarily. For example if the EVF (or LCD panel) is set to show the resulting image after the shot there is an instant flick between pre and post views which highlights small differences rapidly & effectively. Astronomers use that same process to draw the eye to very small changes which are otherwise imperceptible. Blink comparators switch many times between 2 images, not just once --- for a very good reason. And even then you need perfect alignment *and* no blanking. Oh, and does your EVF or LCD show 10x magnification of the whole image at once? Unfortunately, aperture and exposure time is well behaved, but preview prediction is much less so. Thus you need to check the results mor often and for a much longer time. I don't find that to be the case. Interesting. [...] It would be nice if our cameras' auto functions were all infallibly perfect, but none of them are, and it's part of any inquisitive and careful photographer's work to find out when they can't be trusted. Incorrect. There can be an inquisitive and careful photographer who stays with full manual for his work. True, but the more restricted is the range of inquisitive investigation the less it deserves the title. Why should someone who investigates what's effective in his photography and doesn't need automatic functions for his work investigate something so unimportant to his work? Do photographers routinely learn electronics and program firmware and study algorithms for deinterlacing and the theoretical groundwork for image manipulation, lest they be called not inquisitive enough? An optical viewfinder shows the image the lens is presenting to the image sensor. An EVF shows you what the sensor makes of it. An EVF in preview mode shows you in addition what your selected jpeg processing options etc. have on the image. An EVF in preview mode *sometimes* show you an image *downsampled to 0.3-0.4 MPix* of what the sensor and JPEG engine made of the light that arrived through the lens *some time ago* on an *uncorrected* monitor that's usually much too bright in low light situations. The preview mode can equally well be used as an aid to intelligent experiment by the curious. True. Psychoactive drugs can equally well be used as an aid to therapy, too. Unfortunately in the case of drugs, this turns out to be unlikely in most cases of drug use. I fear it'll be the same with preview mode: only a very few will use it "as an aid to intelligent experiment". You may be right. But I don't choose my photographic equipment or develop my techniques with a view to their use in educating the unwilling or incurious. Nor do I or my doctor choose my drugs on the basis of how addicts abuse them. OTOH your doctor doesn't try out a dozen different drug combinations on you until he hits one that sort of works, at least usually. He does chimp, though. [With resepct to finding te right shutter speed to blur moving skaters while keeping stationary skaters sharp] It takes a little skill to judge it in preview mode, true. I found it quicker use it to home in to the right kind of shutter speed and then use chimping for the final refinement (if there was time) than use chimping all the way. That's arguably because I'm not a sports shooter and shoot this kind of ice rink shot about once every two years. I'm curious. My first guess would be 1/500 or 1/1000 on a moderate tele to freeze action. What was your first guess after preview mode and your final shutter speed= The point was not to freeze action but to blur it! The shutter speeds needed were generally in the 1/10th to 1/100th sec range. i.e. just handholdable. Preview mode usually let me get it right, and when it didn't it wasn't more than a stop out. Whatever floats your boat ... you still needed to chimp. It's also possible that all kinds of variable eye and brain physiology comes into this and that some people will find the preview facilities far more annoying than useful. I would find it plainly impossible to see the difference between tack sharp and mostly sharp in preview. Which I expect is why the facility to magnify the preview image up to image sensor pixel level was provided. Which again --- with roller skates --- means you need to have your subject pass the part of whatever 0.3 MPix you're currently seeing at pixel level and judge in 1/60s (or whatever refresh rate you EVF uses) if the subject was tack sharp or only mostly sharp --- or reliably follow the subject at that magnification without camera shake or blur from the camera movement. At, say 150mm (35mm equivalent) and 15 MPix that would mean steadying an effective 1060mm (and that's only ~7x). That's the same as if your skaters filled the frame at 21mm for distance and speed and you'd followed someone's belly button at 150mm at that range. You'd need to *record* the preview image and then zoom in and watch at your leisure, taking a second or 3. Which is chimping and works even better with the shutter button. It would only tell me major blur or no blur --- not only because the EVF doesn't have that much resolution, The preview magnfication of the sensor image is done by computer not lenses, so the resolution of the EVF doesn't put any limits on the magnification. You either get the whole frame and low resolution, or a tiny shred of the whole frame, through which even a stationary object will jump around unless you're well braced and have steady hands or a tripod. (I can do that. Sort of. Sitting. Bracing my heavy 70-200mm lens on my legs and fixating the camera with my hands. Which won't work at all with an EVF or moving subjects I have to track.) It's kinda hard to see 'is tack sharp' or not under these circumstances. but because I can't look that fast. A natural limit which applies to any technology which relies on you seeing what's happening. Which is easily circumvented by softly squeezing the shutter button and chimping. Same with DOF, but I can get an idea of DOF by stopping down the lens. A nice feature of EVFs is that stopping down the lens doesn't dim the view unless you want it to. Very handy when doing long tripod exposures in churches or carefully selecting DoF in interior shots lit by strobes. Yes, *please* show me how you evaluate the light, the JPEG settings and the DoF in a dark interior shot lit by strobes .... in preview mode! [snip] I can't really judge sharpness without magnification, much less in a limited-dot/3=pixel-viewfinder. You seem to think the magnification is done by optically magnifying the view of the EVF screen. It's not. It's done the same way as your computer can zoom into pixel level detail on your image editor, even if it's a 50MP image and your computer monitor has only 1MP. I KNOW. Still, I can't judge the sharpness of a 24 or 36 MPix image on a 0.33 MPix (1 MDot) monitor. I can usually see if such an image is *quite* unsharp on a large 27" monitor (when the image occupies 3 MPix (9 Mdot)). I *still* need magnification to judge critical sharpness there.) [snip] Not with any refresh rate that's needed to even sorta track the subject, and not without a 10x (or more) loupe, which severely cuts your perception of the scene. A 10x loupe?? Why on earth would you ever need a loupe on a camera which can easily magnify any part of the image, That isn't called a loupe? No. A loupe is a lens. Do you describe zooming in and panning around when inspecting a large image on your computer monitor as "using a loupe"? I suggest you look up "loupe" in a dictionary. http://www.gregorybraun.com/Loupe.html http://www.artissoftware.com/screentools/loupe.html http://www.markus-bader.de/MB-Ruler/help/loupe.htm Maybe your dictionary is out of date? If you don't want more than sorta-WYSIWYG, preview mode only, then that's perfectly fine with me, use preview mode, stay with preview mode. But you cannot transfer what you learn there to other modes, not without deliberately and consciously working on it. In other modes you can't get away ignoring the numbers and their meanings --- you learn by default. But underneath the image are displayed the shutter speed, aperture, ISO, plus a lot more which you can choose whether or not to display. And on web pages there are ad banners on the top and on each site. They're habitually ignored. On Windows, there are many cases where you need to confirm a detrimental action. But as such pop ups are used for everything, people are conditioned to ignore the warning that they destroy their installation and click OK. Just having the numbers doesn't mean that they're being seen. That facilitates the learning if learning is what you want to do. You might be right that it also helps you to avoid learning if that's what you want to do. In your skater example, did you look at the numbers, adjusted them and then checked the effect, or did you turn the dials till the effect was sorta what you wanted and then looked at the numbers? (Or did you ignore the numbers completely?) It doesn't bother me if the preview mode of my camera might help the lazy and ignorant to take photographs without learning what's going on. I've managed to avoid having to teach people who didn't want to learn all my life and I don't intend to start now. That's not the kind of person to use preview mode, they use full auto everything mode, not even scene modes. It's harder to follow action with that than simply looking at the straight image through the viewfinder, but it's easier to follow the action with it than with a 10x loupe. Yep, try that sometime. Take a flock of birds flying overhead, track one bird with a long lens and see in the viewfinder if his eye is tack sharp. So what? I'm talking about using a specific camera feature and method to help solve a particular photographic problem -- selective speed blurring of moving skaters. You're right, preview mode is just applicable to moving skaters. :-) Phew! Now I'm relieved! I do know about the reciprocal of the focal length for shutter speed, that you have to adapt that to digital sensor size and resolution, add in the image stabilising factor when appropriate, adapt it to the holding method employed (e.g. elbows on wall, monopod, tripod), factor in wind, factor in unusual rotational inertias (e.g. long reflective vs refractive lens), etc etc. Yep. So how many bodies and lenses do you regularly use? Depepnding on how you define "regularly" anywhere between five and ten lenses. I'm not counting lenses I use less than once a year. I usually carry at least three. I only use my backup camera when I want to reduce lens changes. It doesn't do anything better than my best camera so it never goes out alone. That varies between more than once a week to less than once a month so I wouldn't call my use of two bodies regular. Do your know your regular lenses well? Really? I try a new technique, I see what comes out, I adjust as necessary to my goal, I remember what works and what not for next time. Sounds like you're blessed with a much better memory than I've ever had. I sometimes solve a problem on the run in a busy shoot and by the time I get to reviewing the results on the computer I've forgotten how I did it. You don't have EXIF in your files? There's everything in there that you can influence by using the preview mode ... You're not using chemical sensors, are you? :-) Can't follow the action at 2 fps --- probably can't even keep the camera steady. Many action shooters employ the both eyes open method for following action. Action shooters generally do not employ preview mode, though: they don't want the additional *variable* lag between photons hitting the sensor and dots lighting up on the EVF. Of course. In that case you'd either avoid using the laggier kinds of EVF processing or avoid preview mode altogether. The fastest and most difficult kind of action shooting I do is birds in flight with a 500mm lens and for that I don't use the camera's viewfinder at all. I use an adapted gunsight which lets me use both eyes on the whole scene. Yet wouldn't it be perfect if you used preview mode to have the body tack sharp but the tips of the wings blurred to show the dynamic movement? Much easier to freeze one frame and evaluate it at leisure, once you have a rough idea (which requires about as much training as using the preview properly) --- and exactly that is what happens when you take a shot. That is indeed much easier, but sometimes there isn't time to do that. Then there isn't time to play with preview modes either. Not my experience. I find I still have time to use preview when there's no time to shoot and review. It's better at that than I expected before trying it. That would mean lots of experience with preview, but low on experience with exposure times and apertures. SURE, if you have NO idea at all, then a preview helps --- I find it also helps in my case where I have quite a good idea, having learned my photgraphy back in the old days before there was even autoexposure let alone autofocus. You're saying you're still surprised by aperture or exposure time settings? Surprised is too strong a word. But where there are conflicting demands and aesthetic trade offs involved I like to experiment with different compromises. Anything which shortens the time between experiment and result is useful, especially when the shortening steps over the important boundary between experiment and check into interactive process control. The best shortening would be knowing pretty well which combination work. It's sorta like phase AF and contrast AF: Phase AF knows the direction and (quite exact) the amount of travel, contrast AF is an interactive control process ... guess which one is still faster. provided you have enough ideas how to *use* the preview mode. And it's limitations. And where it fails. On your camera. In your specific preview mode setting. Perfectly true. These are all things you have to learn for each new camera, just as in the old days you had to learn about different films and developing techniques. But you didn't need to relearn all of exposure time and aperture. Which is what the preview mode supplies. Preview doesn't do newer generations of sensors or digital darkroom. [...] But if you're referring to learning and experience which is specific to certain makes and models of camera you're right. Exactly. You learn --- and then have to throw away most of it when you change cameras. That lack of generality is also true of autofocus. That doesn't stop autofocus being very useful, nor does it stop it being useful to learn exactly how a specific kind of autofocus technology works and where and why it fails. AF is fully automatic. Full auto mode is fully automatic. Preview is fully manual. And slower than full auto. Which means the knowledge doesn't carry over. Your arguments are much too black and white. That not all the knowledge carries over doesn't mean that none of it does. All the knowledge doesn't carry over. But a useful amount of it does. Yet with aperture and exposure time almost all carries over, even switching sensor sizes. A surprising amount of what I learned in my film shooting days has turned out to have been oversimplified and overgeneralised. That's mainly been due to increasing sensor resolution revealing unsuspected problems in earlier cruder generalisations. Rather like the way improved detail resolution in scientific measuring instruments reveals the simplifications and overgeneralisations in earlier mathematical models. Naah. You've increased the enlargement (you print larger or look at 100% with higher resolutions), you needed to factor that in even back when. What has changed is that you are more variable in your ways. But luckily all you need is a simple correction factor. Just as switching from Deutschmark to Euro. In fact angle of view is more independent and useful than "equivalent focal length" which IMHO is a silly fudge of an incomplete generalisation. So you'd write an angle of view on a lens --- which is then attached to a 35mm-sized sensor, a 1.6x crop sensor, a 2x crop MFT and maybe even to a 2.7x '1"' sensor. For which sensor would you write the angle of view? None, for exactly the same reasons I wouldn't write "equivalent focal length" on a lens either. Thus the focal length is most useful, and from there you get trivially to equivalent focal length, but not to angle of view. lot. This year I spent only a few minutea on it. I was stopped by a security guard who was worried that I might have a perverted interest in photographing child skaters or be planning a terrorist attack. And next year you'll be arrested for carrying a camera. I doubt it. In the UK over the last few years photographers and lawyers have succeeded in getting the law clarified and better guidelines issued from the government to police and security personnel. That's led to improved relations between photographers, police, and security guards. I used to get harassed often enough that I carried a copy of the relevant legislation in my gear bag. I don't now because there's much less harassment. All it takes is just one terrorist that also used a camera once. No idea what that argument means. Basically: If you need to chimp for a rather long time to find the right settings, you don't know your camera well. I don't. It's a recent acquisition. It takes me at least six months to get to know a camera well. But on the other hand the exposure parameters I'm playing with are in this case largely camera independent. And exposure --- unlike preview mode --- carries over well. If you don't need to chimp for a long time, Which I don't. I call your ""sufficiently much faster that in five minutes shooting you can come away with many more good shots of a much greater ^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^ variety than without preview" bull. ^^^^^^^ Experiment trumps speculation. My experiment, your speculation. So you did experiment shooting a second five minutes shooting without preview mode and came off with many less good shots and with a much smaller variety? I doubt that! So, let's stay with logic: you'd need to produce many good shots and of a much greater variety in the time you'd spend with chimping otherwise. I dunno about you, but I can't do many *good* shots in 20 seconds. Much less in a very great variety. It does you credit that you're worried about the educational state of lazy or stupid photographers and would rather the market insisted on supplying them with cameras they couldn't work without a proper scientific understanding of camera technology. Where did I require that one enter Maxwell's equations into the camera before the shutter works? Or alternatively, the theory of charge transport in semi-conductors? Or maybe how to design and make a processor for the camera? When you look up "loupe" in your dictionary look up "eristic argument" as well :-) Yep, it's really a good term for your "cameras they couldn't work without a proper scientific understanding of camera technology." Thank you. Unfortunately the market is based on consumer choice. Really? So where are the cameras many ask for? In the shops. So where's the affordable compact camera with a really large sensor, 8 or less huge MPix, a good *fast* lens, an optical view finder ... So why not leave off the preview mode (instead of wasting months or years on it) and start being in a fix and thus learning immediately? Please use short, simple words to explain that ... Because that learning process is less fun, and I'm easily bored. Plus preview lets me get a lot more fairly good shots while I'm doing the learning. Helps my motivation. I also suspect that learning which is more fun works faster and better. But I'm willing to accept that may be a personal idiosyncracy. I see. You want the easiest way, not the fastest or the best way. Not what I said nor meant. I said I wanted the most fun while learning. Yep, the easiest way. By that I meant playful skill acquisition and problem solving. That's always been the best and fastest way to learn for me. A friend and cognitive psychologist tells me that's not just a personal idiosyncrasy of mine. "best" and "fastest" are hard to judge, since you cannot compare well yourself. Funniest/most entertaining way, that's easy to find out. -Wolfgang |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Sony tells DSLR shooters they're idiots | Alfred Molon[_4_] | Digital Photography | 455 | January 16th 13 09:22 PM |
Sony tells DSLR shooters they're idiots | Wolfgang Weisselberg | Digital SLR Cameras | 9 | December 25th 12 02:28 PM |
Sony tells DSLR shooters they're idiots | J. Clarke[_2_] | Digital SLR Cameras | 0 | December 1st 12 07:42 PM |
Sony tells DSLR shooters they're idiots | nospam | Digital SLR Cameras | 1 | November 30th 12 06:45 PM |
Sony tells DSLR shooters they're idiots | nospam | Digital SLR Cameras | 0 | November 30th 12 06:27 PM |