 |
 | 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 / DSLR Cameras / November 2006Two weeks!
Til the K10D is released. :-( The wait is killing me. I have just about worn out the manual I had printed. http://www.flickr.com/photos/pete_with_pentax/285444496/ > Til the K10D is released. :-( > > The wait is killing me. I have just about worn out the manual I had printed. > > http://www.flickr.com/photos/pete_with_pentax/285444496/ What is so compelling about having one, please? >> Til the K10D is released. :-( >> [quoted text clipped - 4 lines] > > What is so compelling about having one, please? Have a look at the specs. Makes the offerings from Nikon and Canon look very second rate. http://www.pentax.co.uk/_uk/photo/products/index.php?cameras&gruppe=digital%20sl r&produkt=19095&id=uebersicht Have a look at the sample photos. Expecially the DR of the last one. http://www.pentax.co.uk/_uk/photo/products/index.php?cameras&gruppe=digital%20sl r&produkt=19095&id=bilder For the average hack this is a very serious camera, . >>> Til the K10D is released. :-( >>> [quoted text clipped - 15 lines] > > For the average hack this is a very serious camera, . DNG!!! and I like the battery grip better than it's competitors. And there are several other features that I like better than my Rebel XT. I'm awfully tempted to make the switch. Greg  Signature"All my time I spent in heaven Revelries of dance and wine Waking to the sound of laughter Up I'd rise and kiss the sky" - The Mekons >>>> Til the K10D is released. :-( >>>> [quoted text clipped - 21 lines] > > Greg Luckily I do not have to switch, my 16-45mm goes straight on, my Sigma DG 500 Super flash goes straight on and uses extra features that were not available on my DS, actually about 10 lenses that I have will go straight on and work just the same. > Luckily I do not have to switch, 8< > actually about 10 lenses that I have will go straight on > and work just the same. ah yes, I see it now. ;-) >> Til the K10D is released. :-( >> [quoted text clipped - 4 lines] > > What is so compelling about having one, please? http://www.dpreview.com/reviews/compare_post.asp?method=sidebyside&cameras=canon _eos400d%2Cnikon_d80%2Cpentax_k10d&show=all >>> Til the K10D is released. :-( >>> [quoted text clipped - 6 lines] >look very second rate. >http://www.dpreview.com/reviews/compare_post.asp?method=sidebyside&cameras=canon _eos400d%2Cnikon_d80%2Cpentax_k10d&show=allCompelling specs? Let's see, it has environmental seals,and...umm...umm...well, that's about it really.There is nothing else there that the others don't offer in similarfashion. It has built-in image stabilization, but the others have itin their lenses - no big deal. It has DNG support but I don't considerthat a plus, just another step in workflow. It may have a goodviewfinder like the Nikon, which I think is the most compellingfeature. Everything else is essentially the same between all of the10mp entry models, and it's the most expensive model of those three.I guess it's compelling to a Pentax user who wants a competitivecamera that can use the Pentax lenses they already own. I understandPentax has been behind the curve. This model seems fairly decent andit brings Pentax in line with the others. >>>> Til the K10D is released. :-( >>>> [quoted text clipped - 18 lines] > >behind the curve. This model seems fairly decent andit brings Pentax in > >line with the others. I don't know where you are buying but in Australia it is priced somewhat under the other 10MP cameras. In camera IS offering 3-4 stops sounds pretty compelling to me. Superior focusing system (9 cross points) sounds pretty compelling to me. Continuous shooting till the memory card is full is superior to both Canon and Nikon, compelling, maybe. Bracketing choices. Killer ergonomics compared to the 400D, the D80 is okay good too. In what area has Pentax been behind the curve? >>>>> Til the K10D is released. :-( >>>>> [quoted text clipped - 26 lines] > choices. Killer ergonomics compared to the 400D, the D80 is okay good too. > In what area has Pentax been behind the curve? Size , weight, build quality. [] >> I don't know where you are buying but in Australia it is priced >> somewhat under the other 10MP cameras. In camera IS offering 3-4 [quoted text clipped - 6 lines] > > Size , weight, build quality. Yes, those do appear to be the weak points of the Pentax K10D. To me, size and weight are important, and I had this impression of Pentax producing some of the most compact and lightweight DSLRs, so I was rather shocked on seeing the specs for the K10D. Yes, it looks a nice camera, and having the IS would be important to me (in reducing lens purchase cost), but I don't think the K10D is for me. I might try holding it if I saw it in the shop, though. David > [] >>> I don't know where you are buying but in Australia it is priced [quoted text clipped - 17 lines] > > David They were all plus's. But anyway, I do know someone that used one during some testing here in Aust and he could not say enough superlatives about the handing and feel of the camera. >> [] >>>> I don't know where you are buying but in Australia it is priced [quoted text clipped - 21 lines] > some testing here in Aust and he could not say enough superlatives about > the handing and feel of the camera. For me knowing what has been improved over previous Pentax cameras and having used previous Pentaxes extensively and knowing the quality of the 12 x 18 prints I have from my 6MP DS this camera has got to be the one for me. > > [ ] > >>> I don't know where you are buying but in Australia it is priced > >>> somewhat under the other 10MP cameras. In camera IS offering 3-4 > >>> stops sounds pretty compelling to me. Perhaps, but it depends on if: (a) it actually delivers on this claim, (b) the implimentation doesn't cause too huge of a power drain, and (c) it is mechanically robust. > >>> Superior focusing system (9 cross points) sounds > >>> pretty compelling to me. Unfortunately, the "More = Better" is a common paradigm assumption, even though it is frequently found to be very wrong. I generally find that the more focus spots that I have available in a particular camera, the greater my tendency is to need to manually override to use a single focus point. And when there's more points, the single point is by necessity smaller, which makes the camera less reliable in achieving focus. > >>> Continuous shooting > >>> till the memory card is full is superior to both Canon and Nikon, Actually, this is a good example of a failure to read the fine print and understand its implications: the Canon actually offers a longer burst using RAW (10 vs 9): why is that? The pipe's not going to change, so I suspect that the way that the Pentax was able to get to the "infinite" holy grail was by making their JPEG's smaller - - in other words, they increased their JPEG compression to get the file size down. And as we all know, JPEG compression rates affect image quality. > >>> ... In what area has Pentax been behind the curve? > >> [quoted text clipped - 5 lines] > > They were all plus's. Sorry, but I'm going to have to agree with David as well: a camera body that weighs 40% more than one of its competitors is hardly a plus by any reasonably objective means...it may "feel good" in your hands during ~30 seconds of fat-fingering it at a trade show booth, but carrying it out into the field and then holding the camera up in position for 2 minutes while quietly waiting on a wildlife subject ... hand-held, since that's the selling point of IS ... that extra half pound (8oz) will become quite noticable. And as David says, I too find it a disappointment characteristic when it is coming from the company whose reputation has been to build ligher, more compact cameras than their compeitors. -hh >> >>> Superior focusing system (9 cross points) sounds >> >>> pretty compelling to me. [quoted text clipped - 9 lines] > less > reliable in achieving focus. I agree...I always disable the focus points and use the center one only. My first DSLR had seven focus points, and it would often pick the wrong one, which meant wasted time adjusting it. I find it's faster and easier to focus then recompose than it is to hope that one of the focus points gets it right. And one of the things I discovered I like about the Nikon D80, it has the option to set the center focus point to a wider area which makes hitting the subject focus a tad easier. I never even considered it until I played with it. >> >>> Continuous shooting >> >>> till the memory card is full is superior to both Canon and [quoted text clipped - 3 lines] > and understand its implications: the Canon actually offers a longer > burst using RAW (10 vs 9): why is that? You know, I really don't see the sense in being able to fill a card, unless you're leaving the camera to capture something and HOPE it gets a shot or two of what you wanted, aka paparazzi. Having said that, all cameras can do it, but it's the frames per second rating that varies with some models. As it happens, the Nikon D80 with a fast card can run continuous up to a rated 100 shots in large/normal mode at its rated 3 frames per second. Comparing fine and normal JPG images, normal is essentially as good and is the setting I use when taking snaps. If you want the highest detail and post-processing control, you'll shoot RAW anyway, so that feature is kind of a gimmick to me. As for the number of RAW burst shots you can get in the buffer, I've never used more than about 3-4. With a brief pause of a second or two, I can then fire off another 3-4 without slowing down. That's plenty for me, although I can see others wanting more. > The pipe's not going to change, so I suspect that the way that the > Pentax was able to get to the "infinite" holy grail was by making [quoted text clipped - 3 lines] > > And as we all know, JPEG compression rates affect image quality. That's a possibility. But I think Pentax is smart enough to maintain quality and use a faster processing engine. Apparently they use a fast proprietary processor and DDR2 memory which should give it a fast buffer. That might be how they managed the feature. > carrying it out into the field and then holding the camera up in > position for 2 minutes while quietly waiting on a wildlife subject > ... > hand-held, since that's the selling point of IS ... that extra half > pound (8oz) will become quite noticable. You wouldn't want to use a pro camera then. :-) > And as David says, I too find it a disappointment characteristic > when > it is coming from the company whose reputation has been to build > ligher, more compact cameras than their compeitors. I'm not happy about the weight, but I'm glad Pentax has a more normal sized camera body. But I'm still concerned about the grip - side-by-side images show that the grip might be a little bit shorter yet thicker than some other models, so it might yet be too small for many people such as myself. >>> >>> Superior focusing system (9 cross points) sounds >>> >>> pretty compelling to me. [quoted text clipped - 13 lines] > to focus then recompose than it is to hope that one of the focus points > gets it right. That is sad, I have never had that problem with Pentax cameras, > And one of the things I discovered I like about the Nikon D80, it has the > option to set the center focus point to a wider area which makes hitting [quoted text clipped - 16 lines] > frames per second. Comparing fine and normal JPG images, normal is > essentially as good and is the setting I use when taking snaps. For me this simply means that I will never have to worry about hitting that barrier. > If you want the highest detail and post-processing control, you'll shoot > RAW anyway, so that feature is kind of a gimmick to me. The jpegs are that good that I will not need to use RAW so aften. > As for the number of RAW burst shots you can get in the buffer, I've never > used more than about 3-4. With a brief pause of a second or two, I can [quoted text clipped - 23 lines] > > :-) I guess not. >> And as David says, I too find it a disappointment characteristic when >> it is coming from the company whose reputation has been to build [quoted text clipped - 5 lines] > some other models, so it might yet be too small for many people such as > myself. Your kidding, right?? >> I agree...I always disable the focus points and use the center one >> only. My first DSLR had seven focus points, and it would often pick [quoted text clipped - 3 lines] > > That is sad, I have never had that problem with Pentax cameras, I've used Canon, Nikon, and Olympus cameras, and they have ALL had the same problem with focus point selection with varied contrast subjects. If you've never had the problem, then you're either too inexperienced or you shoot very static subjects that offer high contrast in only one location to lock focus. >> some models. As it happens, the Nikon D80 with a fast card can run >> continuous up to a rated 100 shots in large/normal mode at its [quoted text clipped - 4 lines] > For me this simply means that I will never have to worry about > hitting that barrier. Whatever...it's just a gimmick feature to me. If I need to capture lots of frames of a subject over a period of time, I'll use a different tool that's properly designed for this purpose and heck of a lot easier to use, like a camcorder. :-) >> You wouldn't want to use a pro camera then. > > I guess not. Or a pro lense in the 200mm or longer range, since they weigh more than the camera. That really limits your options for wildlife. Ever heard of a tripod? :-) >> I'm not happy about the weight, but I'm glad Pentax has a more >> normal sized camera body. But I'm still concerned about the grip - [quoted text clipped - 3 lines] > > Your kidding, right?? I use smileys when I'm kidding. I have yet to handle a K10D since it's not available here yet, but I'll be visiting the shop when they arrive to see how they work. For now, I have only photos to compare dimensions, and the grip looks a bit short (didn't I say this already?). I recently sold a Canon XT partly because my little finger would dangle off the bottom of the camera body. I much prefer a larger body that lets me hold the camera with all four fingers. >>> I agree...I always disable the focus points and use the center one only. >>> My first DSLR had seven focus points, and it would often pick the wrong [quoted text clipped - 54 lines] > finger would dangle off the bottom of the camera body. I much prefer a > larger body that lets me hold the camera with all four fingers. You will be pleased with the K10D then. >> > [ ] >> >>> I don't know where you are buying but in Australia it is priced [quoted text clipped - 10 lines] > Unfortunately, the "More = Better" is a common paradigm assumption, > even though it is frequently found to be very wrong. Actually the number is the same, more of them are cross point though so work in both planes. > I generally find that the more focus spots that I have available in a > particular camera, the greater my tendency is to need to manually > override to use a single focus point. And when there's more points, > the single point is by necessity smaller, which makes the camera less > reliable in achieving focus. See above. >> >>> Continuous shooting >> >>> till the memory card is full is superior to both Canon and Nikon, [quoted text clipped - 7 lines] > JPEG's smaller - - in other words, they increased their JPEG > compression to get the file size down. Actually they have used a larger faster buffer using lots of DDR2 memory. The jpegs are of course larger from the larger sensor. > And as we all know, JPEG compression rates affect image quality. > [quoted text clipped - 17 lines] > hand-held, since that's the selling point of IS ... that extra half > pound (8oz) will become quite noticable. Having used S2Pros extensively and 20D's a little, the weight is not a problem. > And as David says, I too find it a disappointment characteristic when > it is coming from the company whose reputation has been to build > ligher, more compact cameras than their compeitors. Lets face it, with the list of features this camera is competing with the D200 and 30D not the 400D or D80. > -hh > > Unfortunately, the "More = Better" is a common paradigm assumption, > > even though it is frequently found to be very wrong. > > Actually the number is the same, more of them are cross point > though so work in both planes. You misunderstand: I'm referring to current cameras which figuratively have a "bazillion" focus points versus cameras that had significantly fewer. For example, a decade ago some SLR's had but 3 focus points, or maybe 5. The problem is that having "many" focus points can be good for some applications, but there are also applications where it becomes "too many", whereupon you then have to start to override the automatic features of the camera to turn things off. This was what lead to the next point: > > I generally find that the more focus spots that I have available in a > > particular camera, the greater my tendency is to need to manually [quoted text clipped - 3 lines] > > See above. Yes, please do that: read my statement again in the context intended. Effectively, how well can you perform when you go to a single focus point, and it is always as narrow as a spot meter? Granted, there probably is a camera out there with 27 focus points that allow you the flexibility to select a single or a group of 3 or 5, but this is adding a lot of complexity that has a very rapidly diminishing return: from a work taskloading perspective, it is substantially better to simply aim, lock, recompose, and shoot. > > Actually, this is a good example of a failure to read the fine print > > and understand its implications: the Canon actually offers a longer [quoted text clipped - 6 lines] > > Actually they have used a larger faster buffer using lots of DDR2 memory. "Larger and Faster" compared to what? The same camera when its shooting in RAW? See, this is the problem: I really doubt that you're claiming that the Pentax somehow physically upgrades its data pipe's level of performance (buffer/pipe), if the shooter is shooting a JPEG versus a RAW, because this would be an absurd design feature: there's no reason to not make it as fast as possible all the time. >From an engineering perspective, the physical hardware's buffers & bandwidth capacity is going to be fixed at some value. The maximum burst rate will be a function of how much data can be passed before the system (buffers & bandwidth) becomes saturated and the shutter has to stop, to stop feeding new data. What this means that the fact that the Canon can shoot a longer RAW burst than the Pentax suggests that the Pentax does not indeed have the "fatter pipe" (with buffer), regardless of what the Pentax's longer JPEG burst performance seems to suggest. > The jpegs are of course larger from the larger sensor. This is why I then said: > > And as we all know, JPEG compression rates affect image quality. ...because the only way that the system can operate in that it does not eventually fill the buffer is if the size of of 3 JPEG's/sec is equal to the bandwidth per second, which we have an idea of its capacity from its RAW limit. Since Pentax is not reducing the number of MegaPixels being passed when the camera's in JPEG mode (versus RAW), the only direction left for them to push to get the bandwidth consumption down to below their limit is to make the file smaller through applying more JPEG compression. And based on the RAW burst limits (10 vs 9), it is clear that Pentax's JPEG compression has to be slightly more than the amount of JPEG compression being used by Canon. If you don't agree with me, just keep saying to yourself: "There's no such thing as a free lunch" as you try to logically reconcile how the effective total system bandwidth for the Pentax can be better at JPEG's while simultaneously being worse for RAW's without hitting a technological self-contradiction. > > ... that extra half pound (8oz) will become quite noticable. > > Having used S2Pros extensively and 20D's a little, the weight is not a > problem. It ultimately will depend on your application. When I hiked the Inca Trail in Peru, I carried an SLR, but pretty much the lightest one that I owned. If I were to do it again, I would again leave behind my heavier bodies/lenses and tripod(s), as the altitude and distance does not make this a trivial walk in the park for most Westerners. > > And as David says, I too find it a disappointment characteristic when > > it is coming from the company whose reputation has been to build > > ligher, more compact cameras than their compeitors. > > Lets face it, with the list of features this camera is competing with the > D200 and 30D not the 400D or D80. Doesn't matter. In its day, products such as the Pentax K-1000 was outclassed by the Nikon FM in many ways, but its small size/weight was effectively a *compensating* factor that made for an interesting trade-off decision, and functionally a viable competitor for a portion of the market. Because Pentax isn't the lightest, they've lost this attribute. -hh >> > Unfortunately, the "More = Better" is a common paradigm assumption, >> > even though it is frequently found to be very wrong. [quoted text clipped - 29 lines] > work taskloading perspective, it is substantially better to simply aim, > lock, recompose, and shoot. Mind you the D2X of one of my friends seem to do rather well with an improved focusing system, I expect the low light performance will also improve with the improvements on the K10D. >> > Actually, this is a good example of a failure to read the fine print >> > and understand its implications: the Canon actually offers a longer [quoted text clipped - 4 lines] >> > JPEG's smaller - - in other words, they increased their JPEG >> > compression to get the file size down. >> Actually they have used a larger faster buffer using lots of DDR2 memory. > > "Larger and Faster" compared to what? The same camera when its > shooting in RAW? Their previous cameras. > See, this is the problem: I really doubt that you're claiming that the > Pentax somehow physically upgrades its data pipe's level of performance > (buffer/pipe), if the shooter is shooting a JPEG versus a RAW, because > this would be an absurd design feature: there's no reason to not make > it as fast as possible all the time. Of course not, perhaps you better ask Pentax for the answer here, just the way it is designed I am guessing. >>From an engineering perspective, the physical hardware's buffers & > bandwidth capacity is going to be fixed at some value. The maximum > burst rate will be a function of how much data can be passed before the > system (buffers & bandwidth) becomes saturated and the shutter has to > stop, to stop feeding new data. You are probably right! > What this means that the fact that the Canon can shoot a longer RAW > burst than the Pentax suggests that the Pentax does not indeed have the > "fatter pipe" (with buffer), regardless of what the Pentax's longer > JPEG burst performance seems to suggest. The difference is trivial. >> The jpegs are of course larger from the larger sensor. > [quoted text clipped - 18 lines] > while simultaneously being worse for RAW's without hitting a > technological self-contradiction. One assumes that while the pipe is long it is not as fat as it could be, I will get over it. >> > ... that extra half pound (8oz) will become quite noticable. >> [quoted text clipped - 6 lines] > heavier bodies/lenses and tripod(s), as the altitude and distance does > not make this a trivial walk in the park for most Westerners. When I hike the Inca trail whereever it may be I will probably shoot film. >> > And as David says, I too find it a disappointment characteristic when >> > it is coming from the company whose reputation has been to build [quoted text clipped - 11 lines] > > -hh Still shooting with my KM and MZ-50 and DS, the K10D will be a great addition to my shooting choices. Nice chatting with you but I am now off for the weekend with my Scout Troop. Be back Sunday night Australian Eastern Standard time. Cheers. Pete D > > ["more=better" paradigm fallacies] > > Mind you the D2X of one of my friends seem to do rather well with an > improved focusing system, I expect the low light performance will also > improve with the improvements on the K10D. The effective proof of low light focusing performance would show up in the specifications with an advertised ability to autofocus at f/8 instead of the typical f/5.6 How does its autofocus performance thus then compare to cameras that have advertised just such a capability, such as the Canon EOS 3? > >> Actually they have used a larger faster buffer using lots of DDR2 memory. > > > > "Larger and Faster" compared to what? The same camera when its > > shooting in RAW? > > Their previous cameras. Yes, and which is utterly and completely irrelevant to my point. > > See, this is the problem: I really doubt that you're claiming that the > > Pentax somehow physically upgrades its data pipe's level of performance [quoted text clipped - 4 lines] > Of course not, perhaps you better ask Pentax for the answer here, just the > way it is designed I am guessing. I have no need to go ask Pentax: I now how things are designed, and can read both a product's published specifications, as well as to read between the lines of the same. All I'm suggesting here is for you to also "read between the lines". > >From an engineering perspective, the physical hardware's buffers & > > bandwidth capacity is going to be fixed at some value. The maximum [quoted text clipped - 3 lines] > > You are probably right! Merely Occam's Razor in a reverse-engineering analysis of what wasn't said in the Spec's. > > What this means that the fact that the Canon can shoot a longer RAW > > burst than the Pentax suggests that the Pentax does not indeed have the > > "fatter pipe" (with buffer), regardless of what the Pentax's longer > > JPEG burst performance seems to suggest. > > The difference is trivial. Yet it means that the Pentax's claim of 'infinite' JPEGS is trivial too. > > If you don't agree with me, just keep saying to yourself: "There's no > > such thing as a free lunch" as you try to logically reconcile how the [quoted text clipped - 4 lines] > One assumes that while the pipe is long it is not as fat as it could be, I > will get over it. Technologically today, we can make the pipe as fat as we effectively want, as well as put in as large of a buffer as we desire: it just sucks money, power, cube, etc. >From a practical design standpoint, a customer is not going to go spend the money for an Über-camera unless they have a very specific need for it. For example, a Vision Research Phantom5 digital camera is only a 1MP, but it shoots 10,000 full frames/sec. With maximum RAM upgrades, its buffer is capable of holding 40,000 MP worth of data before it fills. Just the toy you want...if you happen to have 5-6 digits worth of spare change laying around :-). FWIW, I bought one of these for my lab at work a couple of years ago; it uses Nikon mount for its lenses. > >> > ... that extra half pound (8oz) will become quite noticable. > >> [quoted text clipped - 6 lines] > > When I hike the Inca trail whereever it may be I will probably shoot film. Its in Peru, and is the classical walking approach to Machu Picchu. The better known trail is the highlands one, which is roughly 3-4 days hiking at high altitudes, but there's now a 'Sacred Valley' trail open now too that's slightly shorter, but more importantly, at only a 'mere' 8,000ft average elevation by skipping the high passes (Dead Woman Pass: 4200m/13,779ft). Here's a trail elevation map of the highlands trail: <http://www.ladatco.com/PER-Inca%20Trail%20-%20Map%20Altitudes.htm> For relative comparison, the tallest point in Australia (Mt Kosciuszko in New South Wales) is 2228m, so even the 'easy' Inca trail variant is 3 days of hiking at or above this elevation. All in all, its just some insight hopefully on the relative value of minimizing your photo equipment weight, as this is a case where you do have to carry it yourself the whole way. Wish I didn't have to make such trade-offs, but its IMO better to skip some photo capability in order to minimize the risk of having to abort en route due to altitude sickness. -hh >>>["more=better" paradigm fallacies] >>Mind you the D2X of one of my friends seem to do rather well with an [quoted text clipped - 6 lines] > then compare to cameras that have advertised just such a capability, > such as the Canon EOS 3? Whether or not these sensors work with lenses with a max aperture of f/8 has nothing to do with the sensitivity of the AF system. Think about it (or try it): If the problem was the intensity of incoming light, wouldn't AF work, but only at higher incident light levels? And yet, it simply doesn't, at slow enough apertures. The reason is that the AF sensors are positioned so that if the exit pupil of the lens is smaller than something then they basically can't work (they see nothing). Some cameras have two sets, one which works up to (say) f/2.8 but not slower, and another set for slower apertures. The ones which work up to f/2.8 can detect smaller focus differences for reasons that become obvious if you think about it. > > The effective proof of low light focusing performance would show up in > > the specifications with an advertised ability to autofocus at f/8 [quoted text clipped - 7 lines] > wouldn't AF work, but only at higher incident light levels? And yet, it > simply doesn't, at slow enough apertures. AF systems work on contrast and edge detection - this is why everyone has experienced the fact that they don't work in darkness (without an IR/red helper), no matter how fast the lens. > The reason is that the AF sensors are positioned so that if the exit > pupil of the lens is smaller than something then they basically can't > work (they see nothing). Some cameras have two sets, one which works up > to (say) f/2.8 but not slower, and another set for slower apertures. The > ones which work up to f/2.8 can detect smaller focus differences for > reasons that become obvious if you think about it. Ignore shallow DOF; just try to reconcile with the most common application of the "too slow for AF" f/stop, which comes about when adding a teleconverter to a telephoto lens. We know that simply denying the camera body the information of what the actual f/stop is - typically by the taping of contacts on the TC "fools" the camera into trying to AF, and AF performance gets erratic, but it generally will work under good lighting conditions. This would only be possible if the sensor's position is indeed still within the useful despite the 'too slow' f/stop, and that the manufacturer chose his firmware settings to explicitly block its use for some reason. YMMV on how much of it is to assure reliability of AF performance at various EV's, and how much is to create a tangible difference in his specifications in his product line. -hh >>>The effective proof of low light focusing performance would show up in >>>the specifications with an advertised ability to autofocus at f/8 [quoted text clipped - 10 lines] > has experienced the fact that they don't work in darkness (without an > IR/red helper), no matter how fast the lens. AF in compact digital cameras uses contrast detection: it moves the focusing point of the lens in and out until detected contrast is maximised, as read by the image sensor itself. In SLRs it uses phase detection, which is better in that it tells you how much out of focus (and in which direction) you are (and it does not use the image sensor, but separate sensors that you can see by locking up the mirror of an SLR and looking below it, at the bottom of the mirror chamber). It works by using light from diametrically opposed edges and using a rangefinder principle (well, roughly). The effective rangefinder base scales with the lens diameter, so the faster the lens, the more accurate the focus. However, since the prisms that split the light cannot be moved, you can't just assume you have eg a f/1.4 lens and point your system appropriately, since then if you have eg a f/5.6 lens they will not see anything. So compromises are made. >>The reason is that the AF sensors are positioned so that if the exit >>pupil of the lens is smaller than something then they basically can't [quoted text clipped - 6 lines] > application of the "too slow for AF" f/stop, which comes about when > adding a teleconverter to a telephoto lens. I did not say a single thing about shallow DOF! I like discussion, but at least reply to what I say, not what you expected me to say. Much easier this way :) > We know that simply denying the camera body the information of what the > actual f/stop is - typically by the taping of contacts on the TC [quoted text clipped - 6 lines] > various EV's, and how much is to create a tangible difference in his > specifications in his product line. OK. Since you paid no attention to what I said, I'll repeat: Put a TC on an f/8 lens (eg a mirror lens) and tape the contacts. See if it autofocuses. Then, take an f/5.6 lens and see if it autofocuses in light three stops lower (so that the brightness of the image on the sensor would be the same for the same exposure). Does it? Why not, if it works the way you think? I am not saying that the hard-wired AF limits of eg Canon's cameras are where they should, just that they are not there for the reason you think they are. OK, a quick Google search turns this up: http://www.a1.nl/phomepag/markerink/af_expla.htm It looks like a good explanation (well, I did not read it carefully, just glanced at it, but it has all the right words in it...). Of course, I do not see why you would believe that guy if you don't believe me, but there you go. Cheers. > OK, a quick Google search turns this up: > http://www.a1.nl/phomepag/markerink/af_expla.htm > It looks like a good explanation (well, I did not read it carefully, > just glanced at it, but it has all the right words in it...). OK, it's slightly inaccurate, in that it appears to claim that the accuracy of the AF system will be improved for lenses with larger f-stops. This is only true if the AF system is specifically set up to switch to a wider base for faster lenses. > > OK, a quick Google search turns this up: > > http://www.a1.nl/phomepag/markerink/af_expla.htm [quoted text clipped - 3 lines] > f-stops. This is only true if the AF system is specifically set up to > switch to a wider base for faster lenses. And this was the 'special case' exception which you had already previously mentioned and which was why I asked we "ignore Shallow DOF", where there may (or may not) be additional sensors to increase the system's base length and thus, improve its performance potential. In any event, I was merely looking more to the generalized case of a system and what is the "largest minimum" base length design attributes. If we assume that a lens has been designed such that no part of the image collection frame ("film", or sensor) are ever blocked at any f/stop, the worst case offset that has to be illuminated would be the diagaonal to the corner of the frame, which for a full frame (24mm x 36mm) is roughly 21mm. So if the 'film' corner is never blocked, then because of radial symmetry, nothing inside this 21mm radius will ever be blocked. > OK... Put a TC on an f/8 lens (eg a mirror lens) and tape the contacts. > See if it autofocuses. Then, take an f/5.6 lens and see if it autofocuses > in light three stops lower (so that the brightness of the image on the > sensor would be the same for the same exposure). Does it? Why not, > if it works the way you think? Sorry, I don't have an f/8 mirror lens, but do try this one: a) Take f/5.6 telephoto on a camera that claims AF at up to f/5.6: AF works. b) Put a 1.4x TC on the above (net is f/8) AF now does not work. c) Tape over (select correct) pins on (b): AF now *sometimes* works. Also see: <http://www.fredmiranda.com/TipsPage/#0> The simple fact that (c) works at all is the proof that the sensor locations aren't being blocked geometrically, for at least one full stop (f/8) higher than what the system was officially rated for. -hh >> > OK, a quick Google search turns this up: >> > http://www.a1.nl/phomepag/markerink/af_expla.htm [quoted text clipped - 40 lines] > > -hh My Sigma 70-300mm DG APO lens at 300mm focuses on the moon perfectly with a Tamron 1.4x TC added. http://www.flickr.com/photos/pete_with_pentax/263774304/ > In any event, I was merely looking more to the generalized case of a > system and what is the "largest minimum" base length design attributes. [quoted text clipped - 5 lines] > because of radial symmetry, nothing inside this 21mm radius will ever > be blocked. I think it's obvious that no part is ever blocked at any f/stop. And if there is vignetting, it's at faster (more open) f/stops, not slower. Vignetting, however, is completely irrelevant. Anyway, I explained that the problem is not what you seem to think it is. Read the link I gave a bit more carefully, I think it'll clear things up. > > OK... Put a TC on an f/8 lens (eg a mirror lens) and tape the contacts. > > See if it autofocuses. Then, take an f/5.6 lens and see if it autofocuses [quoted text clipped - 15 lines] > locations aren't being blocked geometrically, for at least one full > stop (f/8) higher than what the system was officially rated for. I already explained what is going on. It has to do with the size of the exit pupil of the lens. If you feel that you have proved that it is an incorrect explanation, ok. You seem to have the wrong picture in your mind. Or maybe I misunderstood what you are imagining. Either way, it's no big deal. Cheers! > > In any event, I was merely looking more to the generalized case of a > > system and what is the "largest minimum" base length design attributes. [quoted text clipped - 3 lines] > > I think it's obvious that no part is ever blocked at any f/stop. Except for the 'special case' for addtional sensor pairs that were set wider apart for better focusing precision at fast stops, unless your intent for saying that they 'see nothing' is for some reason other than their optical pathway simply being geometrically blocked: "The reason is that the AF sensors are positioned so that if the exit pupil of the lens is smaller than something then they basically can't work (they see nothing)." Overall, its merely YA system design factor: putting the sensors closer to the centerline eliminates shadowing, but this simultaneously reduces its focusing resolution performance too. No such thing as a free lunch. > I already explained what is going on. It has to do with the size of the > exit pupil of the lens. If you feel that you have proved that it is an > incorrect explanation, ok. You seem to have the wrong picture in your > mind. Or maybe I misunderstood what you are imagining. Either way, it's > no big deal. Its probably just minor miscommunications, but if you want to pursue it, do feel free to make it up in ZEMAX and email it to me to review. -hh > > > In any event, I was merely looking more to the generalized case of a > > > system and what is the "largest minimum" base length design attributes. [quoted text clipped - 8 lines] > intent for saying that they 'see nothing' is for some reason other than > their optical pathway simply being geometrically blocked: OK, this guy has written something up here: http://doug.kerr.home.att.net/pumpkin/Split_Prism.pdf Again, I have not read it, but the diagrams seem ok. In particular, see fig. 12 (and later). > "The reason is that the AF sensors are positioned so that if the > exit pupil of the lens is smaller than something then they [quoted text clipped - 13 lines] > Its probably just minor miscommunications, but if you want to pursue > it, do feel free to make it up in ZEMAX and email it to me to review. Well, I do not know how to use ZEMAX or anything similar as I am not an optical designer. But since you are an expert on these things, please explain the following (that you wrote): >AF systems work on contrast and edge detection - this is why everyone >has experienced the fact that they don't work in darkness (without an >IR/red helper), no matter how fast the lens Do they work on contrast detection in SLRs? Also, if you are going to tell me that the mentioned effect is indeed the reason you had in mind for why AF does not work for high f-stops (as opposed to too little light due to the slow lens), please reconcile it with this, that you wrote earlier: >The effective proof of low light focusing performance would show up in >the specifications with an advertised ability to autofocus at f/8 >instead of the typical f/5.6 How does its autofocus performance thus >then compare to cameras that have advertised just such a capability, >such as the Canon EOS 3? To me, this is saying that the reason for AF problems with slow f/stops is lack of light, not baseline width. Sorry: I normally try to avoid these mudslinging matches, but your casual mention of ZEMAX is too much, given what you wrote before (not that I doubt that you know how to use it). Or maybe it is a minor miscommunication, as you say. I suggest we do not continue this (although I realise that with the mud I just threw at you, this will be harder for you than for me, so probably this will go on and on and on...). >>> >There is nothing else there that the others don't offer in >>> >similarfashion. It has built-in image stabilization, but the others >>> >have itin their lenses - no big deal. It is if you do not already own stabilized lenses. By putting the stabilization in the body, *every* lens becomes a stablilized lens. You can quibble over whether or not it works *quite* as well, but still, considering htat the body itself is also considerably cheaper than the competion, the camera with a set of stablized lenses is literally going to cost *thousands* of dollars less than with "the others". This is hardly "similar fashion". >> In what area has Pentax been behind the curve? > > Size , weight, build quality. Not sure what you mean here. These are generally considered strong points of Pentax in this price range. Those who prefer larger cameras may find the opposite, but I've never heard anyone wish their camera were heavy or not built as well. --------------- Marc Sabatella marc@outsideshore.com Music, art, & educational materials Featuring "A Jazz Improvisation Primer" http://www.outsideshore.com/ > Til the K10D is released. :-( > > The wait is killing me. I have just about worn out the manual I had printed. > > http://www.flickr.com/photos/pete_with_pentax/285444496/ It may be the best entry level DSLR yet. It's build, features certainly hint at it. >> Til the K10D is released. :-( >> [quoted text clipped - 5 lines] > It may be the best entry level DSLR yet. It's build, features > certainly hint at it. Idiot. Get out of my thread. Oly shill. > >> Til the K10D is released. :-( > >> [quoted text clipped - 7 lines] > > Idiot. Get out of my thread. Oly shill. Did I say something wrong? Oly shill??? >> >> Til the K10D is released. :-( >> >> [quoted text clipped - 9 lines] > > Did I say something wrong? Oly shill??? Certainly nothing constructive. There you go again. > >> >> Til the K10D is released. :-( > >> >> [quoted text clipped - 11 lines] > > Certainly nothing constructive. There you go again. If I choose it out of the current crop of the 10 megs, you'll know it's the best choice. Why? Because I'm not tied to any system whereas people swimming in Canon, Nikon, etc, glass seem reluctant to sell it off to change to a better body. That's one reason why people will suffer with the Rebel XTi's miserable ergonomics. > That's one reason why people will suffer with the Rebel XTi's miserable > ergonomics. What's wrong with the Rebel XTi's ergonomics? It feels about the same as my Rebel XT except slightly heavier. It might be not quite as good, but very close. I would consider upgrading the the XTi if I could get a good price for my XT. > > That's one reason why people will suffer with the Rebel XTi's miserable > > ergonomics. > > What's wrong with the Rebel XTi's ergonomics? It feels about the same as my > Rebel XT except slightly heavier. You just made my point. >> > That's one reason why people will suffer with the Rebel XTi's miserable >> > ergonomics. [quoted text clipped - 4 lines] > > You just made my point. Actually I am with Rich here about the cheap Canons they are crappy to use even though they take good photos. > Til the K10D is released. :-( > > The wait is killing me. I have just about worn out the manual I had printed. Will make the experience sweeter... can't wait to hear your impressions. Cheers, Alan Signature-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm -- [SI] gallery & rulz: http://www.pbase.com/shootin -- e-meil: Remove FreeLunch. >> Til the K10D is released. :-( >> [quoted text clipped - 5 lines] > Cheers, > Alan Hi Alan, Can't wait to experience it first hand. Keep an eye on my Flickr and Shuttertalk posts. Cheers. Pete > "Alan Browne" <alan.browne@FreeLunchVideotron.ca> wrote in message >>Will make the experience sweeter... can't wait to hear your impressions. > Hi Alan, > > Can't wait to experience it first hand. Keep an eye on my Flickr and > Shuttertalk posts. I'll look for your links here... I don't visit those other places. Signature-- r.p.e.35mm user resource: http://www.aliasimages.com/rpe35mmur.htm -- r.p.d.slr-systems: http://www.aliasimages.com/rpdslrsysur.htm -- [SI] gallery & rulz: http://www.pbase.com/shootin -- e-meil: Remove FreeLunch. |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2009 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.