 |
 | 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 / December 200520D ?? 5D
Has there been any comparison on Hi ISO settings of the two cameras. There was some reports on a banding problem with the 20D , has there any problems with the 5D. If no banding problems are found with the 5D, then that is the way to go. I just hate it when a Product has problems and its covered up. > Has there been any comparison on Hi ISO settings of the two cameras. > There was some reports on a banding problem with the 20D , has there any > problems with the 5D. If no banding problems are found with the 5D, then > that is the way to go. I just hate it when a Product has problems and its > covered up. no banding that i have been able to discern with a 20D. Bare in mind that the 5D has larger individual pixels than a 20D. So resolving power or lines per milimetre would be lower on the 5D. The 5D has equivalent sized pixels to the 300D. They just have more of them. I therefore find it unlikely that picture quality is better unless you are cropping heavily or printing very large pictures indeed. >> Has there been any comparison on Hi ISO settings of the two cameras. >> There was some reports on a banding problem with the 20D , has there any [quoted text clipped - 8 lines] > unlikely that picture quality is better unless you are cropping heavily or > printing very large pictures indeed. Then you would be wrong. There is considerable improvement in image quality in the 5D over the 20D. Fine detail is not the end all and be all. BTW, I have, and use, both cameras.  SignatureSkip Middleton http://www.shadowcatcherimagery.com >>> Has there been any comparison on Hi ISO settings of the two cameras. >>> There was some reports on a banding problem with the 20D , has there any [quoted text clipped - 13 lines] > all. > BTW, I have, and use, both cameras. Smallest individual dot resolution, noise level, colour accuracy are the factors that are 'quality'. I suspect with a larger photosite and 20D noise reduction tech the noise level is even lower. I'm not sure how a larger sensor would equate to better colour accuracy. As for sharpness well the middle part of the lens is the best part and that is the only part a 20D will use on a full size lens and not an ef-s. comparing 17mm ef-s and a 27.2mm full size lens the outcome is more likely the quality of the lens. Given that the wider the lens the harder to make i suppose it would even out seeing as the full size sensor will use the entire lens and not just the centre portion you will get a warts and all rendition of the lens. There is also less depth of field for a given focal length. As to the mechanics of the sensor itself having simply more of the same pixel (300D) will not improve matters. Putting 20D post processing (digic II) and the appropriate hardware noise reduction would reduce overall noise. As to colour accuracy i suspect at the wide end and in a cropped centre circle the 5D sensor won't pick up as much purple fringing, pincushion etc. That taken into account the extra cost, the extra storage requirements, the higher standard of lenses required, the slower shooting rate and the lack of a built in flash (i use an external but do use the onboard one for snapshooting) I don't think the extra costs involved justify the end product. I will probably invest the 24-105F4L eventually. That is more likely to remain at a stable price than the camera. That has dropped £500 since the launch already. I therefore find >>> it unlikely that picture quality is better unless you are cropping >>> heavily or printing very large pictures indeed. [quoted text clipped - 7 lines] > factors that are 'quality'. I suspect with a larger photosite and 20D > noise reduction tech the noise level is even lower. Lower than the 5D? No, it isn't, but noise is roughly equivalent at high ISOs, the 5D has an advantage at lower ISO. >I'm not sure how a larger sensor would equate to better colour accuracy. As for sharpness well the > middle part of the lens is the best part and that is the only part a 20D > will use on a full size lens and not an ef-s. comparing 17mm ef-s and a > 27.2mm full size lens the outcome is more likely the quality of the lens. Compare a 24-70L to the 17-85. No contest. > Given that the wider the lens the harder to make i suppose it would even > out seeing as the full size sensor will use the entire lens and not just [quoted text clipped - 3 lines] > will not improve matters. Putting 20D post processing (digic II) and the > appropriate hardware noise reduction would reduce overall noise. Don't forget, the 5D uses the same processor. As to colour accuracy > i suspect at the wide end and in a cropped centre circle the 5D sensor > won't pick up as much purple fringing, pincushion etc. That taken into > account the extra cost, the extra storage requirements, the higher > standard of lenses required, the slower shooting rate The difference between 5fps and 3fps is hard to discern. My old film cameras had a similar difference, the A2 (EOS5, to you) was at 5fps, the 1n (without booster) was 2.5fps. It never caused a real problem, so I don't expect the 5D to be a cause for concern, either. >and the lack of a built in flash Again, the same difference with my film cameras, and the only time I've ever used the pop up on the 20D was one night, when I forgot my 420EX. I think I used the one on my D30 once, too. I never used the one on my A2, by the way, it was blocked by my 28-135IS, as is the one on the 20D by my current 24-70 f2.8L. > (i use an external but do use the onboard one for snapshooting) I don't > think the extra costs involved justify the end product. I will probably > invest the 24-105F4L eventually. That is more likely to remain at a > stable price than the camera. That has dropped £500 since the launch > already. Now you've hit the nail on the head. It's not that the 5D doesn't produce better images, it's that the improvement may, or may not, be worth the extra investment. That I cannot argue with. We already had the lenses to suit the 5Ds that we bought, only buying the 70-200 because of the "reach" we lost in the switch. See, Ian, you keep saying things like "more likely," "I suspect" and other speculative terms. I use both cameras, extensively, and the image quality is much improved, on prints, especially large ones. I can see it on images on my monitor, too, but one has to be a bit hypercritical. Part of the difference is not quantifiable, it is in the tonal transitions, much like the difference between 35mm and medium format, or between Ilford XP-2 and Kodak TCN. http://www.pbase.com/skipm/image/44869137 http://www.pbase.com/skipm/image/53255498 SignatureSkip Middleton http://www.shadowcatcherimagery.com >Lower than the 5D? No, it isn't, but noise is roughly equivalent at high >ISOs, the 5D has an advantage at lower ISO. Are you talking about sloppy digitization artifacts (like banding), or random sensor noise? I don't see how having less random noise than the 20D at ISO 100 could be beneficial; 12-bit digitization actually requires a little noise to break up quantization artifacts (posterization). Now, if the digitization were clean and 14 bits or more, then lower noise would be a great advantage. I disagree that the 20D and the 5D are about the same at high-ISO. The 5D, from the images I've seen, is clearly cleaner, and the lower banding makes downsampled high-ISO 5D images better yet. Even downsampled images from relatively noisier sensors like the D2X clean up better with downsampling than the 20D, if the 20D is used in conditions that display banding artifacts. Of course, proper RAW conversion could almost eliminate banding effects without losing image detail ... Still waiting. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>Lower than the 5D? No, it isn't, but noise is roughly equivalent at high >>ISOs, the 5D has an advantage at lower ISO. > > Are you talking about sloppy digitization artifacts (like banding), or > random sensor noise? I'm talking about random noise. > I don't see how having less random noise than the 20D at ISO 100 could > be beneficial; 12-bit digitization actually requires a little noise to > break up quantization artifacts (posterization). Now, if the > digitization were clean and 14 bits or more, then lower noise would be a > great advantage. True, but the files aren't perfectly clean, just cleaner... > I disagree that the 20D and the 5D are about the same at high-ISO. The > 5D, from the images I've seen, is clearly cleaner, and the lower banding [quoted text clipped - 3 lines] > banding artifacts. Of course, proper RAW conversion could almost > eliminate banding effects without losing image detail ... Still waiting. I know, but the quantization charts I've seen don't reflect that, so I didn't want to get into a discussion of it. Visibly, they're better, though. SignatureSkip Middleton http://www.shadowcatcherimagery.com >> Smallest individual dot resolution, noise level, colour accuracy are the >> factors that are 'quality'. I suspect with a larger photosite and 20D >> noise reduction tech the noise level is even lower. > > Lower than the 5D? No, it isn't, but noise is roughly equivalent at high > ISOs, the 5D has an advantage at lower ISO. You misunderstood. The 5D has larger photosites than the 20D. As does the 300D. The 300D uses the digic I processor. The 20D has lower noise despite having smaller photosites. So take the 20D level of noise reduction and stick in a 5D you get even lower noise. > >I'm not sure how a larger sensor would equate to better colour > accuracy. As for sharpness well the [quoted text clipped - 3 lines] > > Compare a 24-70L to the 17-85. No contest. So the better result is due to the lens. Not the extra size or any other characteristic of the sensor save that of using a 24mm lens as a 24 and not a38.4mm >> Given that the wider the lens the harder to make i suppose it would even >> out seeing as the full size sensor will use the entire lens and not just [quoted text clipped - 16 lines] > 1n (without booster) was 2.5fps. It never caused a real problem, so I > don't expect the 5D to be a cause for concern, either. It would be for me. the failings of the 300D was the limit to only 4 continuous shots and the 2.5fps shooting time. I doubt going from 2.5 to 3 would help. Jumping for 2.5 to 5fps would mean not only a before and after shot but also 1 or two during. With 9 or more continous shots i can press and hold earlier to make sure i get the whole sequence. i.e goalkeeper diving for goal. Waiting and shooting i'm pressing the shutter but the miniscule delay is enough for the goal keeper to finish flying through the air and have landed. That elusive mid air catch may now be possible. For weddings and the like its not an issue. > >and the lack of a built in flash > [quoted text clipped - 3 lines] > by the way, it was blocked by my 28-135IS, as is the one on the 20D by my > current 24-70 f2.8L. I use an external 550ex for anything important but carrying an extra £250 of kit just in case is uncomfortable. For the odd party of shooting family get together at the local restaraurant the cheap kit wideangle and the built in flash is sufficient. Not a huge issue but a nice 'backup, always there, just in case etc.' >> (i use an external but do use the onboard one for snapshooting) I don't >> think the extra costs involved justify the end product. I will probably [quoted text clipped - 7 lines] > suit the 5Ds that we bought, only buying the 70-200 because of the "reach" > we lost in the switch. Some people buy stuff cos its the best. Might not be able to tell but that warm fuzzy feeling etc with buying the best kit be it cameras or hifi. Me i listen to hifis and when my philistine like ears could no longer hear an improvement and stopped looking. I'll have that one. As for scanners its was 'well that's the best i can afford so that will do' For the 300D it was '10D are like gold dust and the 300D is freely available' As for the 20D it was 'well the 350D better than the 300D but not better enough to justify as i its not that much better. The 20D is same size, faster, better, and most importantly has the extra dials for ease of use that i know i can make use of.' As for image quality i couldn't discern them just by looking at a print. With that in mind even printing A3 i don't think i can justify the purchase even if it cost a mere £1500. > See, Ian, you keep saying things like "more likely," "I suspect" and other > speculative terms. Yes openly saying that i am speculating the benefits using the newsgroups and reviews and my own knowledge of sensors and things. I am not spouting these suppositions as fact and then arguing about it. Seeing as i don't have £2500 lying around making the place look untidy thats the closest i'm going to get. I use both cameras, extensively, and the image quality > is much improved, on prints, especially large ones. I can see it on > images on my monitor, too, but one has to be a bit hypercritical. Part of > the difference is not quantifiable, it is in the tonal transitions, much > like the difference between 35mm and medium format, or between Ilford XP-2 > and Kodak TCN. Yup, if you crop then zoom in 1600% and examine each colour channel minutely you may see the difference. If however i printed two images A3 and took a decent step back for normal viewing could i see it? Its back to the hifi again, yes i have the best on the market and the frequency response is x and the speaker drivers are titanium and i have 22carat gold wires and contacts. I don't have to worry about it sounding odd, i am satisfied that as i am 66 years old if it sounds a bit iffy then my hearing aid battery needs changing ;) > http://www.pbase.com/skipm/image/44869137 > > http://www.pbase.com/skipm/image/53255498 >>> Smallest individual dot resolution, noise level, colour accuracy are the >>> factors that are 'quality'. I suspect with a larger photosite and 20D [quoted text clipped - 7 lines] > despite having smaller photosites. So take the 20D level of noise > reduction and stick in a 5D you get even lower noise. You said you found it unlikely that the image quality would be better in the 5D. "The 5D has equivalent sized pixels to the 300D. They just have more of them. I therefore find it unlikely that picture quality is better unless you are cropping heavily or printing very large pictures indeed. " So I guessed you were continuing in that theme. >> >I'm not sure how a larger sensor would equate to better colour >> accuracy. As for sharpness well the [quoted text clipped - 8 lines] > characteristic of the sensor save that of using a 24mm lens as a 24 and > not a38.4mm I should have been more clear, compare a 24-70 on a 5D to a 17-85 on a 20D. Compare the best vs. the best available for each camera. Or, just compare a 28-135 on a 5D to a 17-85 on a 20D. That's one I'd like to see, but I don't know anyone with a 17-85. >>> Given that the wider the lens the harder to make i suppose it would even >>> out seeing as the full size sensor will use the entire lens and not just [quoted text clipped - 26 lines] > through the air and have landed. That elusive mid air catch may now be > possible. For weddings and the like its not an issue. Don't forget, I also shoot motorsports. And the 5D has a much larger buffer than the 20D. At 4 frames, your 2.5fps fills the buffer in less than 2 seconds. The 5D will fill its buffer, RAW, in 6 seconds. It has a capacity of 60 frames in JPEG, as opposed to the 9 RAW, 27 JPEG of the 20D. >> >and the lack of a built in flash >> [quoted text clipped - 60 lines] > minutely you may see the difference. If however i printed two images A3 > and took a decent step back for normal viewing could i see it? That was my point, print A3+ and you do see the difference. Really. SignatureSkip Middleton http://www.shadowcatcherimagery.com >>>> Smallest individual dot resolution, noise level, colour accuracy are >>>> the factors that are 'quality'. I suspect with a larger photosite and [quoted text clipped - 14 lines] > you are cropping heavily or printing very large pictures indeed. " So I > guessed you were continuing in that theme. Fair enough. The sheer size of the sensor itself doesn't improve the image quality on its own, or so i believe. However being bigger means quality wide angles are more accessible. Then again you are not cropping so you need better glass as the sensor will detect vignetting, purple fringing, edge to edge sharpness etc. >> So the better result is due to the lens. Not the extra size or any other >> characteristic of the sensor save that of using a 24mm lens as a 24 and >> not a38.4mm > I should have been more clear, compare a 24-70 on a 5D to a 17-85 on a > 20D. Compare the best vs. the best available for each camera. Or, just > compare a 28-135 on a 5D to a 17-85 on a 20D. That's one I'd like to see, > but I don't know anyone with a 17-85. The 17-85 is very good except at the widest setting (17mm). Many lenses are comparable if you don't use the widest aperture. It would seem though that the overall combination of 5D and 24-70 is better than 20D and 17mm. But take the 24-70 and set to 24 mm on 20D then shoot. Now take exact same lens place on 5D and shoot at 38.4mm and compare. Do this at a4 and a3. I would expect the A4 of the 20D to contain more resolved detail. I would expect the A3 print to be similar from a straight print. Genuine fractals or bicubic may narrow the gap. It also depends on the printer. Inkjet used to only manage 150dpi in real terms. An owner of the epson 2100 stated that above 240dpi he didn't notice the difference. My R300 epson claim to do 310. Most minilabs go to 300dpi. Being a current 20D owner i can't quite justify the 5D at least till i get the r1800 epson or the canon 9950 A3 printer. Even then it would be a waste unless i have at least one more L lens to add to my collection. I will probably have to invest in having everything professionally profiled too. So probably 24-105f4L will be next purchase unless the 28-135IS will do. Then the A3 printer. Then the profiling. Then the 5D which by that time should be much more affordable. >>> The difference between 5fps and 3fps is hard to discern. My old film >>> cameras had a similar difference, the A2 (EOS5, to you) was at 5fps, the [quoted text clipped - 16 lines] > capacity of 60 frames in JPEG, as opposed to the 9 RAW, 27 JPEG of the > 20D. hmm slower frames but longer burst rate. 27 would be enough. Current frame rates proved in adequate. So 20D wins by a nose. >> Yup, if you crop then zoom in 1600% and examine each colour channel >> minutely you may see the difference. If however i printed two images A3 >> and took a decent step back for normal viewing could i see it? > > That was my point, print A3+ and you do see the difference. Really. Ah. Are we talking a brightly lit room with a loupe or subdued light standing back 3 feet or so? If i had waited 3 weeks i would have said is the extra £1500 than the 20D worth it? Or worse still 'how do you define £1500 worth of quality? Or would £1500 of lenses, profiling and A3 printer garner a better result? It was easier a few years ago. £2500 for a D30 or £500 for a scan elite II and keep my existing kit. I have the scanner and now have a stonking great DSLR. The safest bet/investment would seem to be another L lens. SignatureSkip Middleton http://www.shadowcatcherimagery.com > > I have the scanner and now have a stonking great DSLR. The safest > bet/investment would seem to be another L lens. There you have it. That's probably the best route, given your needs/wants. I keep saying to those who naysay the 5D, if the crop of whatever camera's sensor you're considering isn't a deal breaker, then the 5D probably isn't the camera for you. The image quality is markedly better, and I like the overall feel of the camera better than the 20D, but these features aren't worth $1500 to most people. The full, 35mm sized sensor is what was worth $1500 to me, the image improvement was icing on the cake... >There you have it. That's probably the best route, given your needs/wants. >I keep saying to those who naysay the 5D, if the crop of whatever camera's [quoted text clipped - 3 lines] >worth $1500 to most people. The full, 35mm sized sensor is what was worth >$1500 to me, the image improvement was icing on the cake... For me, the greatest advantage over the 20D is in dynamic range. The 5D can record a range of contrast that is typically 1.5 - 2.0 stops greater than the 20D. This is invaluable to a wedding photographer who almost always faces the challenge of recording detail in a white bridal dress and in a groom's dark suit - in the same shot. Usually, with a DSLR, the detail in the groom's dark suit has to be sacrificed as the dynamic range is insufficient to record both, and the bride insists on seeing as much detail as possible in 'that dress'. But the 5D can record a surprising amount of detail in the darker tones too. The only "35mm based" DSLR that can do better than the 5D is the FujiFilm FinePix S3, which is probably 1 stop better. Tony Polson wrote: >>There you have it. That's probably the best route, given your needs/wants. >>I keep saying to those who naysay the 5D, if the crop of whatever camera's [quoted text clipped - 18 lines] > The only "35mm based" DSLR that can do better than the 5D is the > FujiFilm FinePix S3, which is probably 1 stop better. $2000 12MP camera: http://www.dpreview.com/reviews/fujifilms3pro/ special dual sensor: http://www.dpreview.com/news/0301/03012202fujisuperccdsr.asp >Tony Polson wrote: >> The only "35mm based" DSLR that can do better than the 5D is the [quoted text clipped - 5 lines] >special dual sensor: >http://www.dpreview.com/news/0301/03012202fujisuperccdsr.asp Claims of 12 MP for the S3 are highly misleading. The camera has 6.45 million photosites, each having two diodes. However, this dual-diode system is not used to increase resolution beyond 6.45 MP. Instead, it improves dynamic range, with one diode for highlights and the other for shadows. The output from the camera is therefore 6 MP, not 12 MP. So while the S3 may have a better dynamic range than the Canon 5D, it only produces half the number of pixels. >$2000 12MP camera: >http://www.dpreview.com/reviews/fujifilms3pro/ That's 6MP, really. It has two wells for each pixel. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >> Has there been any comparison on Hi ISO settings of the two cameras. >> There was some reports on a banding problem with the 20D , has there any [quoted text clipped - 8 lines] > unlikely that picture quality is better unless you are cropping heavily or > printing very large pictures indeed. I shoot lots of 1600 ISO near night-time pics with exposures up to a 1/60sec at f11 ish without any banding and ver very low noise on a 5D. 200 ISO at exposures of 1 minute or so and f8 upwards are just superb, in extremely low light, too.. >> Has there been any comparison on Hi ISO settings of the two cameras. >> There was some reports on a banding problem with the 20D , has there any [quoted text clipped - 8 lines] >that picture quality is better unless you are cropping heavily or printing >very large pictures indeed. Banding has nothing to do with pixel sizes. Banding is a read-out issue. There are basically 2 causes; electrical interference, and inconsistent black-point registration on a line-by-line basis. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >The 5D has equivalent sized pixels >to the 300D. Not really. The 300D would be 16MP if it were full-frame. 6.3MP * 1.6 * 1.6 Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>The 5D has equivalent sized pixels >>to the 300D. > > Not really. The 300D would be 16MP if it were full-frame. > > 6.3MP * 1.6 * 1.6 pop over to dpreview and have a look at the diagrams. aps c is half the area of 35mm. so it is only 6.3mp x2. 16mp is the 1dMK2 which came out before the 20D. The 20D uses those pixels but a smaller and thus cheaper sensor chip. Price increases exponentially with size increase not merely doubles. The failure rate is simply higher (too many duff pixels). Same problem with LCD displays. The success rate is getting better which is why prices are comming down. >>>The 5D has equivalent sized pixels >>>to the 300D. >> Not really. The 300D would be 16MP if it were full-frame. >> 6.3MP * 1.6 * 1.6 >pop over to dpreview and have a look at the diagrams. aps c is half the >area of 35mm. so it is only 6.3mp x2. A 1.6x crop factor means that full frame is 1.6x wider, and 1.6x higher, for 2.56x the area. You really need to slow down. You make an awful lot of incorrect statements here every day. >16mp is the 1dMK2 which came out before the 20D. The 20D uses those pixels >but a smaller and thus cheaper sensor chip. No. The pixel pitch on the 1DsmkII is close to that of the D60/10D/300D, not the 20D. If you are talking about the actual diameter of the wells, rather than the pitch, then I don't know; I'd have to look that up. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>>>The 5D has equivalent sized pixels >>>>to the 300D. > >>> Not really. The 300D would be 16MP if it were full-frame. >>> 6.3MP * 1.6 * 1.6 > [quoted text clipped - 6 lines] > You really need to slow down. You make an awful lot of incorrect > statements here every day. Actually i make alot of suppositions and take onboard the corrections. Finally grasped foveon and bayer thanks to the back and forth of the last week. You're confusing my enquiries with the pronouncements such as made by Rich. 2ndly. Alot of the rows start cos egos get involved and smarmy remarks like the one you just made. So bear in mind that the pre snipped statement i made was that the 20D photosites were smaller than the 5D. I therefore deduced that they had opted for 300D size. As your maths point out that would equate to 16mp. In fact it is 12mp or 13 according to who you ask site. (the 1mp variance is not currently relevant we are back to 'what is an effective pixel territory'). Now in order for fewer mps to fill the same size area they have to be bigger. So the 5D sites are even bigger than the 300D. Ironically this was overlooked when you made the 'slow down' remark. So by all means keep up the corrections just leave out the pointed remarks afterwards. >>16mp is the 1dMK2 which came out before the 20D. The 20D uses those >>pixels [quoted text clipped - 5 lines] > If you are talking about the actual diameter of the wells, rather than > the pitch, then I don't know; I'd have to look that up. Obivously need to slow down a bit ;) > no banding that i have been able to discern with a 20D. Bare in mind that > the 5D has larger individual pixels than a 20D. So resolving power or lines > per milimetre would be lower on the 5D. Ian- You are technically correct with reference to the sensor itself. Translated to the final print, lines per enlarged millimeter would be greater for the 5D unless you were cropping. Therefore your statement is not relevant where the photographer frames photos in the viewfinder. Lets see where we are this time next year. I believe the next round of digital improvements will include a full frame sensor with pixel density equal to the 20D, or around 20 megapixels, and at a price competitive with the 5D today. And next year's Rebel may use the 5D sensor in a cheaper package! Fred > Has there been any comparison on Hi ISO settings of the two cameras. > There was some reports on a banding problem with the 20D , has there any > problems with the 5D. If no banding problems are found with the 5D, then > that is the way to go. I just hate it when a Product has problems and its > covered up. There's a review of 5D on dpreview.com that also includes the high ISO comparison and samples that reveal banding. As far as I remember, banding problem in 20D was fixed later by updated firmware. Banding problem with 5D is still there, but most likely it will be fixed later the same way it was done with 20D. Best regards, Andrey > Has there been any comparison on Hi ISO settings of the two cameras. >There was some reports on a banding problem with the 20D , has there any >problems with the 5D. If no banding problems are found with the 5D, then >that is the way to go. I just hate it when a Product has problems and its >covered up. The 5D can exhibit banding in AI servo mode with certain lenses. Generally speaking, however, it has much weaker banding than the 20D does. Just about all digital cameras have banding. It is most visible when it combines with noise in a certain ratio. Cameras that have less random sensor noise tend to show banding moreso than cameras with stronger random noise. IT is generally most visible as banded noise in dark areas of high-ISO images, and as purer color bands (casts) in extremely-under-exposed low-ISO images. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >> Has there been any comparison on Hi ISO settings of the two cameras. >>There was some reports on a banding problem with the 20D , has there any [quoted text clipped - 12 lines] > areas of high-ISO images, and as purer color bands (casts) in > extremely-under-exposed low-ISO images. In otherwords you are not going to notice it. If you shot with film the picture would look much worse. >>> Has there been any comparison on Hi ISO settings of the two cameras. >>>There was some reports on a banding problem with the 20D , has there any >>>problems with the 5D. If no banding problems are found with the 5D, then >>>that is the way to go. I just hate it when a Product has problems and its >>>covered up. >> The 5D can exhibit banding in AI servo mode with certain lenses. >> Generally speaking, however, it has much weaker banding than the 20D >> does. >> Just about all digital cameras have banding. It is most visible when it >> combines with noise in a certain ratio. Cameras that have less random >> sensor noise tend to show banding moreso than cameras with stronger >> random noise. IT is generally most visible as banded noise in dark >> areas of high-ISO images, and as purer color bands (casts) in >> extremely-under-exposed low-ISO images. >In otherwords you are not going to notice it. If you shot with film the >picture would look much worse. In some ways, yes. In other ways, no. With just the right transfer curve from the original RAW to the displayed image, the banding can be extremely distracting; much more so than grain. And banding does not disappear when you downsize, whether you do it quick-and-dirty with nearest-neighbor, or properly with bicubic or Lanczos. Film grain disappears much better than banding, super-imposed upon noise. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > Has there been any comparison on Hi ISO settings of the two cameras. > There was some reports on a banding problem with the 20D , has there any > problems with the 5D. If no banding problems are found with the 5D, then > that is the way to go. I just hate it when a Product has problems and its > covered up. I have not noticed banding on my 20D although I remember a report or two. I have a feeling there was suposed to be a software fix some time back. I think I have heard of one banding report on he 5D but that is all. No idea where though but on a web forum. >I have not noticed banding on my 20D although I remember a report or two. I >have a feeling there was suposed to be a software fix some time back. There is a banding that is present in most digital cameras; it is usually only visible if you push the extreme shadows in an ISO that has a high potential S/N ratio, as smooth colored bands (offsets of color/luminance, such as the Canon 10D at "ISO 100" (actually it is ISO 64)), or low noise at a certain ratio to banded RAW data offsets (like the 20D). Both are most prevalent when the light source is farther from optimal (magenta for the 20D, pink for the 10D). So, an image under-exposed in the forest with green prevailing light, whitebalanced, will be a prime candidate for banding. >I think I have heard of one banding report on he 5D but that is all. No idea >where though but on a web forum. I've seen banded 5D images posted on the Rob Galbraith forums, but they were AI-servo related with certain lenses. The non-AI-servo related 5D banding tends to be very mild, compared to some older Canons. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< |
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.