 |
 | 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 / May 2007Sigma SD14 review/comparison
Mike Chaney published a not-quite review of the Sigma SD14 on 03/16/2007 at http://www.ddisoftware.com/sd14-5d/. This is the Digital Domain web site, where the Qimage printing software resides. The article appears quite unbiased and gives a favorable review of the SD14, mostly comparing it with the Canon EOS 5D. I don't own or have an interest in buying a SD14, since I'm quite happy with my Pentax K10D. I just found the article interesting, and quite at odds with the random flaming that seems to follow Sigma cameras. A quote from the article follows my sig. Steve . . . "The bottom line in the debate about where to place the SD14 among other (Bayer based) cameras is that I believe the SD14 to be about equivalent to a 10 MP Bayer dSLR as far as pure (maximum) resolution. When taking into account how the camera achieves that resolution, however, I would have to say for image quality, the SD14 compares well to standard dSLR's a little closer to 12 MP, that is, more comparable to something like the Canon 5D. When taking equivalent shots of "real" subjects and examining SD14 and 5D photos side by side, SD14 photos compare nicely to photos from the 5D. I've done a number of these tests and in scrolling around with my "pixel peeping" hat on, I can always find some areas that I like better on the SD14 and other areas that I like better from the 5D photos. For image quality alone, it's a toss-up for me when comparing the SD14 and 5D. The SD14 seems to have a little less consistent/controllable color than the 5D but the SD14 produces that 3D presence that no other standard dSLR can match. In the race to get the best image quality, I suspect some will like 5D photos better than SD14 photos and vice versa. The mere fact that the SD14 compares so well to cameras like the 5D is a testament to how good the SD14 really is!" > Mike Chaney published a not-quite review of the Sigma SD14 on 03/16/2007 > athttp://www.ddisoftware.com/sd14-5d/. This is the Digital Domain web [quoted text clipped - 28 lines] > the SD14 compares so well to cameras like the 5D is a testament to how > good the SD14 really is!" Get he get paid for the number of times he wrote SD14 in this article? >Mike Chaney published a not-quite review of the Sigma SD14 on 03/16/2007 >at http://www.ddisoftware.com/sd14-5d/. This is the Digital Domain web >site, where the Qimage printing software resides. The article appears >quite unbiased and gives a favorable review of the SD14, mostly comparing >it with the Canon EOS 5D. 1) The "review" only compares image quality and the objective numbers put it lower than the 5D in terms of resolution. 2) A camera is much more than just a sensor, and there is nothing about the overall performance. 3) A telling sentence: "The new camera with 1.01 firmware only locks up occasionally such as when shooting buffered shots quickly so it is certainly not as bad as the initial 1.00 firmware."  SignatureRay Fischer rfischer@sonic.net > Mike Chaney published a not-quite review of the Sigma SD14 on 03/16/2007 > at http://www.ddisoftware.com/sd14-5d/. This is the Digital Domain web [quoted text clipped - 28 lines] > the SD14 compares so well to cameras like the 5D is a testament to how > good the SD14 really is!" I had both SD9 and SD10 cameras. If this SD14 is as good as he claims in all areas of use (which I doubt) then it certainly will be refreshing. The real problem for Sigma is that there must be a hell of a lot of earlier SD buyers who (like me) wouldn't touch another Sigma camera with a barge pole. Mike's camera was a dud out of the box and he got run around on warranty just like I did. So far he's on camera number two. I had three SD10s (the first at a huge discount to replace the SD9) and they were still rubbish. Personally I think Sigma ought to send a new SD14 to every buyer of earlier SD cameras. Right along with an apology for the rubbish they passed off on them. Douglas >Mike Chaney published a not-quite review of the Sigma SD14 on 03/16/2007 >at http://www.ddisoftware.com/sd14-5d/. This is the Digital Domain web [quoted text clipped - 5 lines] >with my Pentax K10D. I just found the article interesting, and quite at >odds with the random flaming that seems to follow Sigma cameras. The Sigma SD14 was reviewed in a recent issue of "Amateur Photographer", a weekly magazine published in the UK that has an excellent reputation for unbiased reviews. The SD14 gained a 72% rating, which is the lowest I can recall seeing in that magazine for a Digital SLR. Most DSLRs achieve over 80% and the best of them are in the >90% range. >> Mike Chaney published a not-quite review of the Sigma SD14 on 03/16/2007 >> at http://www.ddisoftware.com/sd14-5d/. This is the Digital Domain web [quoted text clipped - 13 lines] > in that magazine for a Digital SLR. Most DSLRs achieve over 80% and > the best of them are in the >90% range. I wonder if 72% = Below Average in DPReviewspeak? >>> Mike Chaney published a not-quite review of the Sigma SD14 on 03/16/2007 >>> at http://www.ddisoftware.com/sd14-5d/. This is the Digital Domain web [quoted text clipped - 15 lines] >> >I wonder if 72% = Below Average in DPReviewspeak? That would probably depend on whether Sigma were a generous sponsor. Only joking. ;-) > Steve King <wheelgat@c-zone.net> wrote: >> Mike Chaney published a not-quite review of the Sigma SD14 on >> 03/16/2007 at http://www.ddisoftware.com/sd14-5d/. This is the >> Digital Domain web site, where the Qimage printing software >> resides. The article appears quite unbiased and gives a >> favorable review of the SD14, mostly comparing it with the Canon >> EOS 5D. > The Sigma SD14 was reviewed in a recent issue of "Amateur > Photographer", a weekly magazine published in the UK that has an [quoted text clipped - 3 lines] > seeing in that magazine for a Digital SLR. Most DSLRs achieve > over 80% and the best of them are in the >90% range. Reviews of the SD14 in the UK press is mixed, to say the least. Below is some quotes from six recent reviews. British Journal of Photography, 2007-05-02: "In Summary, the SD14 produces outstanding photographs at low ISO and appears to be optimised for performance at low to moderate sensitivity. The camera is unlikely to hold great appeal for sports and action photographers because of the disappointing high ISO performance, but others will appreciate its undoubted qualities." "This suggests that for real-life shooting situations at ISO100 the SD14 is possibly equivalent to a 12-14 million pixel Bayer camera, just as the advertisement says." Digital Photo Pro, May 2007: "My final thought however was that no-one currently using, say a Canon or Nikon system would in their right mind want to ditch those options." Digital Camera Magazine June 2007 - 82%: "You could fill a page with the SD14 flaws." "But in the right conditions this camera produces image quality you can't get from a conventional sensor". What Digital Camera, June 2007: "Rudimentary spec, dodgy performance and iffy images." "Weak auto focus, poor LCD, poor Hi-res Jpegs, inconsistent white balance." "I hate to say it, but the SD14 is the most disappointing camera I have come across for a few good years. Digital SLR User, June 2007 - 3/5: "Excellent Colour reproduction, good build, affordable lens range, effective anti dust measure, superb bundled RAW editing software." "Out-resolved and out performed by cheaper 10 MP models, tardy AF system, small viewfinder, restrictive low light performance." "Ultimately without the Foveon sensor, little to make it stand out." "Enjoyable to use, but won't appeal to everyone." Practical Photography, June 2007 - 3/5: "Superb quality at lower ISO settings, but overall performance lets it down." "Superb results from what some are calling a 5MP camera. Supplied Software better than most". "Write speed and review times far too slow for many users, leaving it suitable only for those who need great quality and aren't bothered by speed." "£1000 is a little too much". Signature- gisle hannemyr [ gisle{at}hannemyr.no - http://hannemyr.com/photo/ ] ------------------------------------------------------------------------ Sigma SD10, Kodak DCS460, Canon Powershot G5, Olympus 2020Z ------------------------------------------------------------------------ >> Steve King <wheelgat@c-zone.net> wrote: > [quoted text clipped - 60 lines] > by speed." >"£1000 is a little too much". Thank you very much for posting those quote, Gisle. I had read the reviews in "Amateur Photographer" and "British Journal of Photography" but not the others. I have never found the BJP reviews in any way worthwhile, as they often seem to bear no resemblance to reality. Long experience has taught me to trust "Amateur Photographer" reviews beyond all others, but the variable nature of the reviews in the other magazines is amusing, to say the least > > Steve King <wheel...@c-zone.net> wrote: > >> Mike Chaney published a not-quite review of the Sigma SD14 on [quoted text clipped - 63 lines] > Sigma SD10, Kodak DCS460, Canon Powershot G5, Olympus 2020Z > ------------------------------------------------------------------------ When oh WHEN are they going to standardize testing on a SCIENTIFIC basis so we don't get this near-useless scattergun reporting from people who simply and CLEARLY do not understand the underlaying technology but only understand snapping pictures? Photographers as a group are the most divorced from understanding their technology as anyone in any discipline I've ever heard of. > When oh WHEN are you going to not post the same thing five times? nospam <nospam@nospam.invalid> wrote in news:230520071439346532% nospam@nospam.invalid: >> When oh WHEN are > > you going to not post the same thing five times? Not only that, he's using a broken newsreader that does not insert the "References:" line, so there is no threading and his replies appear to be first generation. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > > Steve King <wheel...@c-zone.net> wrote: > >> Mike Chaney published a not-quite review of the Sigma SD14 on [quoted text clipped - 63 lines] > Sigma SD10, Kodak DCS460, Canon Powershot G5, Olympus 2020Z > ------------------------------------------------------------------------ When oh WHEN are they going to standardize testing on a SCIENTIFIC basis so we don't get this near-useless scattergun reporting from people who simply and CLEARLY do not understand the underlaying technology but only understand snapping pictures? Photographers as a group are the most divorced from understanding their technology as anyone in any discipline I've ever heard of. > > Steve King <wheel...@c-zone.net> wrote: > >> Mike Chaney published a not-quite review of the Sigma SD14 on [quoted text clipped - 63 lines] > Sigma SD10, Kodak DCS460, Canon Powershot G5, Olympus 2020Z > ------------------------------------------------------------------------ When oh WHEN are they going to standardize testing on a SCIENTIFIC basis so we don't get this near-useless scattergun reporting from people who simply and CLEARLY do not understand the underlaying technology but only understand snapping pictures? Photographers as a group are the most divorced from understanding their technology as anyone in any discipline I've ever heard of. > > Steve King <wheel...@c-zone.net> wrote: > >> Mike Chaney published a not-quite review of the Sigma SD14 on [quoted text clipped - 63 lines] > Sigma SD10, Kodak DCS460, Canon Powershot G5, Olympus 2020Z > ------------------------------------------------------------------------ When oh WHEN are they going to standardize testing on a SCIENTIFIC basis so we don't get this near-useless scattergun reporting from people who simply and CLEARLY do not understand the underlaying technology but only understand snapping pictures? Photographers as a group are the most divorced from understanding their technology as anyone in any discipline I've ever heard of. > > Steve King <wheel...@c-zone.net> wrote: > >> Mike Chaney published a not-quite review of the Sigma SD14 on [quoted text clipped - 63 lines] > Sigma SD10, Kodak DCS460, Canon Powershot G5, Olympus 2020Z > ------------------------------------------------------------------------ When oh WHEN are they going to standardize testing on a SCIENTIFIC basis so we don't get this near-useless scattergun reporting from people who simply and CLEARLY do not understand the underlaying technology but only understand snapping pictures? Photographers as a group are the most divorced from understanding their technology as anyone in any discipline I've ever heard of. snip.............................................................................. On When oh WHEN are they going to standardize testing on a SCIENTIFIC basis so we don't get this near-useless scattergun reporting from people who simply and CLEARLY do not understand the underlaying technology but only understand snapping pictures? Photographers as a group are the most divorced from understanding their technology as anyone in any discipline I've ever heard of. Apparently you do not know many audiophiles-high-endists. Those people are **really** divorced from logic and comprehension. > snip.............................................................................. > [quoted text clipped - 7 lines] > Apparently you do not know many audiophiles-high-endists. > Those people are **really** divorced from logic and comprehension. I would have thought a good Photographer to be like a good race car driver. Understanding the way to get the best from their equipment without having any knowledge of the underlying technology which allows it. Let's face it. Photography is an art form, not a science. Why then would a an artist need to also be a scientist to make good pictures? Douglas >> Apparently you do not know many audiophiles-high-endists. >> Those people are **really** divorced from logic and comprehension. [quoted text clipped - 5 lines] > Let's face it. Photography is an art form, not a science. Why then > would a an artist need to also be a scientist to make good pictures? It definitely isn't a requirement, but all things being equal, it can sometimes contribute to better pictures. Ansel Adams is an example of the "rocket scientist" that was able to routinely squeeze out more of his camera/film's potential image quality than most other photographers. But he also had lots of time to envision and compose his shots, and then to carefully coax the best from his prints. Not knowing his other abilities, he may well have been a run of the mill photographer if he went into sports photography. Many of the greatest images taken haven't pushed their camera's potential nearly as much, nor need they. I suspect that the best photojournalists rely much more on innate talent, lots of experience and practice rather than knowing about photons, full well capacity, shot noise and statistics. This is true in many other fields as well. There are singers considered to be wonderful or great and others that are merely very good, and it's not unusual for the formers to possess 'instruments' that aren't nearly as good as the latters. Examples of these are Maria Callas (stretching it here a bit), Billie Holiday, Dylan and Louis Armstrong (based only on his singing, obviously). Exceptions that prove the rule by having it all are Ella, Bing and Nat King. :) >Ansel Adams is an >example of the "rocket scientist" that was able to routinely squeeze [quoted text clipped - 3 lines] >prints. Not knowing his other abilities, he may well have been a >run of the mill photographer if he went into sports photography. If only you took a little time to learn about him, you would know that Ansel Adams worked across many genres of photography as a full service commercial photographer. He brought the same standard of excellence to every genre. Adams' skills in social and documentary photography were every bit as outstanding as in landscape work. His work with Japanese Americans who were interned during WW2 was particularly remarkable. Clearly, his landscape work in the National Parks, especially Yosemite, was what brought him fame, even in his own lifetime. He is credited with having popularised Yosemite, a remote wilderness, to the point where it is overcrowded with visitors who rarely stray far off the roads. But he was much, much more than "just" the world's greatest landscape photographer. If you genuinely wish to learn more about him, do a Google search and ignore the landscape-related links. To find out more about the project with the Japanese internees, search on "ansel adams japanese internment" and you will find many links to his superb body of work. >>Ansel Adams is an >> example of the "rocket scientist" that was able to routinely squeeze [quoted text clipped - 8 lines] > commercial photographer. He brought the same standard of excellence > to every genre. I don't doubt that one bit. But you seem to have missed my point. None of the examples you gave (social and documentary photography) require the skills of a good sports photojournalist, where speed, anticipation and reaction time are critical. And note that I did *not* say that he wouldn't have had those skills had he tried his hand at sports photography. Just that if he had, there's no reason to assume that he would also have risen to the top in that category, despite having a formidable standard of excellence. >>>Ansel Adams is an >>> example of the "rocket scientist" that was able to routinely squeeze [quoted text clipped - 17 lines] >to assume that he would also have risen to the top in that category, >despite having a formidable standard of excellence. I think it is you who missed the point. The genre of sports photography was almost created by the advent of 35mm "system" SLRs. Prior to that, sports photography was a tortuous process involving cumbersome equipment that was quite unable to represent sport as we see it now. By the time 35mm "system" SLRs were introduced, starting with the 1959 Nikon F, Adams' fame as landscape photographer had been well established, and Adams is never recorded as having used the 35mm format to any great extent. He considered the Hasselblad as a small format camera, and does not seem to have been tempted by the Leica. So your assertion that he may not have made a good sports photographer is somewhat irrelevant. It is like saying he may not have made a good Baptist minister, or a good airline pilot. We will never know. But what is clear about Adams is that he was the master of a very wide range of genres - advertising, product, architectural, portrait and wedding photography along with his exquisite landscapes. There is no reason to suggest that he would not have mastered sports photography, given the right equipment at the right time. But he did not have that opportunity, so unless time travel becomes a possibility, we will never know. > I think it is you who missed the point. Nonsense. Or as is more often said on your side of the pond, rubbish! :) > The genre of sports photography was almost created by the advent of > 35mm "system" SLRs. Prior to that, sports photography was a tortuous [quoted text clipped - 18 lines] > opportunity, so unless time travel becomes a possibility, we will > never know. Yes, we'll never know. But despite being a master of a wide range of genres, do you think that if he was interesting in pursuing a career as a baseball or football player he would have been a perennial MVP? Or if he wanted to be an Air Force pilot he would have had the right stuff to become a combat ace? The type of photographic skill that sports shooting requires, as I spelled out for you, is partly mental, but those lacking the best physical attributes won't rise to the top of the field. Despite having bulkier, slower equipment to work with, it didn't negate the attributes needed by a sports photographer unless you go back far enough in time to where equipment was so slow that shots had to be posed. Desire and mental strength can be beneficial in the genres that you spoke of. The skills that I already mentioned (anticipation, quick reflexes, etc.) might help somewhat with wedding photography, but are not essential for it. As much as I appreciate and admire Adams' work, I'm not yet ready to deify him. > The genre of sports photography was almost created by the advent of > 35mm "system" SLRs. Prior to that, sports photography was a tortuous > process involving cumbersome equipment that was quite unable to > represent sport as we see it now. My father appeared leaping hurdles (he was an Oxford Blue in this back-destroying sport and paid for it later in life - NEVER become a hurdler!) on the cover of Amateur Photographer. The shot is similar to any modern sports action pix, taken from close quarters and fairly dramatic. I don't know who it was by, the cover is framed (somewhere) and the date must be around 1931-33 when he was at Oxford. It is in colour, but it's hard to tell whether it is tinted or real as the colours are so muted. It looks as if it could be rollfilm, maybe even the new 35mm. J H Dallmeyer's book 'Telephotography' which predates even this leaves the reader in no doubt that sports photography was being undertaken, and even with zoom lenses - the New Large Adon was technically a varifocal telephoto, the Bigma of its era. Compared with a current Canon DSLR outfit as toted by most sports pros, a VN Press camera with a longish lens (210mm) would be a lightweight bit of equipment. It was the cut film holders or glass plate carbine magazines that really added the weight! Camera, one kilo. Film, requires a handcart... David > >> Apparently you do not know many audiophiles-high-endists. > >> Those people are **really** divorced from logic and comprehension. [quoted text clipped - 12 lines] > other photographers. But he also had lots of time to envision and > compose his shots, Not always. His "moonrise" was a grab shot. > Not always. His "moonrise" was a grab shot. Not surprising. Moon shots are usually spur-of-the-moment. :) "Sorry officer. I was drinking, not thinking." > snip.............................................................................. > [quoted text clipped - 7 lines] > Apparently you do not know many audiophiles-high-endists. > Those people are **really** divorced from logic and comprehension. Perhaps the SD14 is the dslr equivalent of the "Shakti Stone". http://www.shakti-innovations.com/audiovideo.htm > Perhaps the SD14 is the dslr equivalent of the "Shakti Stone". > http://www.shakti-innovations.com/audiovideo.htm I've seen worse, about £300 for a wooden volume knob. Signaturehttp://www.petezilla.co.uk > > snip.............................................................................. > [quoted text clipped - 9 lines] > > Perhaps the SD14 is the dslr equivalent of the "Shakti Stone".http://www.shakti-innovations.com/audiovideo.htm Only because it's wired with 6 nines copper. > > The SD14 gained a 72% rating, which is the lowest I can recall seeing > > in that magazine for a Digital SLR. Most DSLRs achieve over 80% and > > the best of them are in the >90% range. > > I wonder if 72% = Below Average in DPReviewspeak? at the moment, i'm thinking dpreview will give it an 'average' rating. the camera has a lot of issues that would rank it low, however, phil liked the results from the previous two versions. on the other hand, numerous people are reporting corrupt files (which is unacceptable for a camera at any price, let alone one that sells for $1600), camera lockups, inconsistent image quality and the sigma photo pro software is slow and crashes a lot. if phil encounters any of that more than once or twice, the camera stands a good chance of getting a 'below average' or maybe even be the first to get a 'not recommended.' >>Mike Chaney published a not-quite review of the Sigma SD14 on 03/16/2007 >>at http://www.ddisoftware.com/sd14-5d/. This is the Digital Domain web [quoted text clipped - 13 lines] > in that magazine for a Digital SLR. Most DSLRs achieve over 80% and > the best of them are in the >90% range. I have just got our copies of May/June Master Photo Digital with Richard (son) reviewing the SD14 and the DPS repro of one shot is exceptional. It certainly has no trouble with A3+ in print at 175-200 screen. The only issue is that for some reason we had a 24-70mm non-digital lens on the camera (I think to match some tests I was shooting on the Fuji S5) and you can really tell where this lens is 'off' in the top corners - as the bottom corners (foreground) are sharp, curvature of focus field seems to be the culprit. But where it's sharp, it is amazingly detailed and realistic. I have enough copies of Master to give a few away - an A4 envelope with 60p stamp affixed to our address below will cover UK. Anyone out of the UK who wants to see a copy, £3 by Paypal to david@maxwellplace.demon.co.uk will cover stationery/postage. David SignatureIcon Publications Ltd, Maxwell Place, Maxwell Lane, Kelso TD5 7BB Company Registered in England No 2122711. Registered Office 12 Exchange St, Retford, Notts DN22 6BL VAT Reg No GB458101463 Trading as Icon Publications Ltd, Photoworld Club and Troubadour.uk.com www.iconpublications.com - www.troubadour.uk.com - www.f2photo.co.uk - www.photoclubalpha.com - www.minoltaclub.co.uk Tel +44 1573 226032 > But where it's sharp, it is amazingly detailed and realistic. That's exactly what I don't find about aliased imaging; it looks sharp, but *FALSE*. The real world of analog vision does not have implied cartesian biases, with "snap to grid" effects. The array of pixels should not be hinted at in a proper capture, IMO. The idea that you should be able to have a white pixel next to a black one, or a saturated red one next to a saturated blue one is pure anti-scientific fantasy. You can not possibly image accurately and most of all, consistently, with that possibility in the palette. Proper imaging would have so many pixels that the maximum MTF of the lens is oversampled, with high contrast edges taking several pixels to occur. Aliased imaging is a dirty shortcut that tries to force resolution that can't really happen. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>But where it's sharp, it is amazingly detailed and realistic. > [quoted text clipped - 12 lines] > Aliased imaging is a dirty shortcut that tries to force resolution that > can't really happen. Sounds like a legitimate rationalization for antialiasing filters: so the individual pixels are always hidden. It's possible that by the time you print, the jaggies aren't visible. Also, I guess that's why they resample up from 5MP to 14MP (similar effect as capturing at 10 or 12MP with an antialiasing filter). The RAW files ought to be smaller & the jpegs larger, that's the main difference I can see. It's not a fair comparison to the 5D because the Sigma gets noisy in low light where the 5D excels. P&S even look pretty good in good lighting. SignaturePaul Furman Photography http://www.edgehill.net/1 Bay Natives Nursery http://www.baynatives.com Paul Furman <paul-@-edgehill.net> wrote in news:zV65i.9414$RX.6678 @newssvr11.news.prodigy.net: > Also, I guess that's why they > resample up from 5MP to 14MP (similar effect as capturing at 10 or 12MP > with an antialiasing filter). It's not a similar effect. It may take more than one pixel to go full modulation, but an AA-filtered capture can have weighted edges at sub-pixel locations, while an aliased image has them snapped to a coarser grid if there is a lower MP count. That is exactly what I find wrong with these low-MP, aliased images; I can see that no matter how much Jaggies are smoothed, the sampling points are clearly few indeed. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< [] > Proper imaging would have so many pixels that the maximum MTF of the > lens is oversampled, with high contrast edges taking several pixels > to occur. Aliased imaging is a dirty shortcut that tries to force > resolution that can't really happen. John, agreed with you on that one. So should the spatial cut-off frequency of an anti-alias filter for a Bayer camera be half that for an SD14? In practice, is it, or do the Bayer manufacturers also compromise the cut-off frequency to produce a "sharper" image? David > [] >> Proper imaging would have so many pixels that the maximum MTF of the [quoted text clipped - 8 lines] > Bayer manufacturers also compromise the cut-off frequency to produce a > "sharper" image? The Leica M8 has no AA filter. The RAW files have tremendous color moire; the converters ignore high-frequency color, at least when moire is detected. The Canon 1DmkIII, that Imaging Resource has, has an AA filter, but a very weak one, and has more color moire in the RAW data than most Canons (the mkII might have a weak one, too, IIRC). For B&W, and for RGB photosites, the AA filter need not be as strong as it need be for avoiding moire with a CFA. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< [] > The Leica M8 has no AA filter. The RAW files have tremendous color > moire; the converters ignore high-frequency color, at least when [quoted text clipped - 6 lines] > For B&W, and for RGB photosites, the AA filter need not be as strong > as it need be for avoiding moire with a CFA. Point agreed about CFA versus "point" sensors. I draw from other comments that the impression that the AA filter strength in various cameras is a compromise between what would be "correct" in signal processing terms [softer], and what is "acceptable" in marketing or end-user values [sharper]. All other things being equal, "you pays your money and takes your choice.....". Cheers, David On May 24, 6:24 pm, "David J Taylor" <david-tay...@blueyonder.not-this- bit.nor-this-part.co.uk> wrote: > [] > [quoted text clipped - 16 lines] > end-user values [sharper]. All other things being equal, "you pays your > money and takes your choice.....". But it's not that simple to settle. What does correct mean? The smoother the rolloff, the less artifacts (ringing etc), but the less the resolution. If "correct" means absolutely no artifacts (no ringing, no aliasing etc) then I don't see how it can be done. eg take the fourier transform of a delta function (which is a constant), truncate at some frequency and transform back: it rings. Then instead of truncating, multiply eg by exp(-(q/r)^2) and transform back (here, r is the range): it doesn't ring, but passes arbitrarily high frequencies, so it'll show aliasing (although you can reduce them as much as you want, at the cost of also removing lower frequencies). And so on. There just isn't a perfectly correct way, as far as I can tell. You need to balance the two somehow, and various cameras make different compromises. Actually I may be wrong here, but I just don't see how you can completely cut off everything above some frequency (to avoid aliasing) without having ringing. [] > But it's not that simple to settle. What does correct mean? The > smoother the rolloff, the less artifacts (ringing etc), but the less [quoted text clipped - 13 lines] > completely cut off everything above some frequency (to avoid aliasing) > without having ringing. I must confess to being more familiar with electronic signal processing than optical, but the AA filters used in audio can be quite complex. Additionally, they sometimes oversample (partially to ease the filter design), a luxury not yet possible with digital cameras. I'd welcome a more informed viewpoint. David On May 25, 12:27 pm, "David J Taylor" <david-tay...@blueyonder.not- this-bit.nor-this-part.co.uk> wrote: > achilleaslazari...@yahoo.co.uk wrote: > [quoted text clipped - 22 lines] > Additionally, they sometimes oversample (partially to ease the filter > design), a luxury not yet possible with digital cameras. Well I was trying to indicate that it is mathematically impossible to completely avoid artifacts. Of course this does not mean that in practice it is not possible to get results practically free of artifacts, but I think it shows rather clearly that there will always be a tradeoff, unless, as you say, you can oversample (which we do not do with digital photography at the moment). As for the low-pass filter's passband, I do not know how it looks, but I'd be surprised if it's that complicated (but it would be interesting if someone who knows can tell us). On May 25, 12:27 pm, "David J Taylor" <david-tay...@blueyonder.not- this-bit.nor-this-part.co.uk> wrote: > I must confess to being more familiar with electronic signal processing > than optical, but the AA filters used in audio can be quite complex. > Additionally, they sometimes oversample (partially to ease the filter > design), a luxury not yet possible with digital cameras. Here http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=1450934 is a quantitative and thorough discussion of optimising the tradeoffs between resolution and artifacts, although they don't call it that: they study ways in which to minimise the difference between a function and its bandlimited representation (obviously, both loss of resolution and artifacts increase this difference). I hope it is available to people not in subscribing institutions (I have no way to check unfortunately). > I hope it is available to people not in subscribing institutions (I > have no way to check unfortunately). Nope. It seems to be available two ways. First way: > You are not logged in. > You must log in to access: [quoted text clipped - 3 lines] > • Full Text PDF > • Full Text HTML Second way: > You have elected to buy the following document: > The optimization of bandlimited systems [quoted text clipped - 10 lines] > Once you access the document, you have 72 hours to download the > Full Text PDF. I'll pass for now. :) > > I hope it is available to people not in subscribing institutions (I > > have no way to check unfortunately). > > Nope. It seems to be available two ways. Hmm, I was afraid so. I'll see if it's freely available somewhere else (unlikely, as it's rather old so not on electronic preprint archives), or if the thing is reported elsewhere (more likely). Thanks for the feedback. achilleaslazarides@yahoo.co.uk wrote in news:1180037352.487698.317260 @g4g2000hsf.googlegroups.com: > eg take > the fourier transform of a delta function (which is a constant), > truncate at some frequency and transform back: it rings. An AA filter in a camera doesn't do filtering based on FFTs. It is an optical device that makes some of the light that would hit a pixel hit its neighbors instead. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > achilleaslazari...@yahoo.co.uk wrote in news:1180037352.487698.317260 > @g4g2000hsf.googlegroups.com: [quoted text clipped - 6 lines] > optical device that makes some of the light that would hit a pixel hit its > neighbors instead. Well of course it doesn't do FFTs or fourier transform in any other literal way. However, any signal can be resolved into a sum of various frequencies, and what happens to each frequency individually can be studied, and the result pieced back together. Unless the processing (digital or analog, ie with a physical filter) is nonlinear, in which case one cannot consider each frequency individually. But I don't think AA filters operate nonlinearly (in this sense); I may be wrong, though. For example, fourier transforming a signal, multiplying by a function that decreases with frequency and inverse fourier transforming has the effect of blurring the signal (because you have removed higher frequencies). eg doing that to a delta function (a point) spreads it out. On May 26, 3:41 am, achilleaslazari...@yahoo.co.uk wrote: > > achilleaslazari...@yahoo.co.uk wrote in news:1180037352.487698.317260 > > @g4g2000hsf.googlegroups.com: [quoted text clipped - 15 lines] > think AA filters operate nonlinearly (in this sense); I may be wrong, > though. Maybe I was not clear: I am talking about the process in fourier (frequency) space because there it is easier to think about what it means to do something to frequencies: in real space, blurring is a nonlocal process: as you said, light that would hit a pixels is made to hit its neighbours; but in frequency space, it is merely the attenuation of particular frequencies, with a factor depending on that frequency, so can be understood locally (in frequency space: you do not need to think about any frequency other than the one you are currently dealing with). If the process is nonlinear, it is more complicated, as it becomes nonlocal in frequency space too.. > Mike Chaney published a not-quite review of the Sigma SD14 on 03/16/2007 > at http://www.ddisoftware.com/sd14-5d/. This is the Digital Domain web [quoted text clipped - 5 lines] > with my Pentax K10D. I just found the article interesting, and quite at > odds with the random flaming that seems to follow Sigma cameras. Comparing to a 5D implies it's full frame & I don't think that's true. If it was full frame & mounted Nikkors, I'd be curious but otherwise no interest. SignaturePaul Furman Photography http://www.edgehill.net/1 Bay Natives Nursery http://www.baynatives.com > > Mike Chaney published a not-quite review of the Sigma SD14 on 03/16/2007 > > at http://www.ddisoftware.com/sd14-5d/. This is the Digital Domain web [quoted text clipped - 9 lines] > If it was full frame & mounted Nikkors, I'd be curious but otherwise no > interest. it is a 1.7x crop factor. |
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.