 |
 | 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 / February 2006Canon 5D Sharpness Question
I bought a Canon 5D and a Canon 20mm lens at Christmas. I've recently bought a Canon 28mm F2.8 and a Canon 50mm F1.8 Mk 2. The quality of the picture with the 20mm is always great. The quality of the picture with both the 28mm and the 50mm is great sometimes and not great the other times which is very annoying. I don't think depth of field is an issue and it also happens when the camera is on a tripod. I think the pictures taken with the 28mm and the 50mm are not focused correctly. I have set the camera to focus on the central point only. I really rely on the camera to focus due to the lack of split screen or Fresnel ring. Has anyone else experienced problems with autofocus? It could also be a case of me not using the camera correctly - it's a lot more complex than SLRs made 20 years ago which I'm used to. I'm considering buying one of the replacement 3rd party focusing screens that do offer split screen or Fresnel ring to help me out. :I bought a Canon 5D and a Canon 20mm lens at Christmas. : I've recently bought a Canon 28mm F2.8 and a Canon 50mm F1.8 Mk 2. [quoted text clipped - 17 lines] : I'm considering buying one of the replacement 3rd party focusing screens : that do offer split screen or Fresnel ring to help me out. 5Ds are like all EOS cameras. They have an unusually small distance between the sensor and the real lens element. This is called the back focus area. Because of this short distance, back focus on Canon SLRs is more prone to be incorrect from the factory than with other brands. The other Canon unique focus issue is the anti alias filter they put over the sensor to stop moiré effect. this produces a permanently soft image which is only correctable with the use of artificial sharpening after the shot. If you set the sharpness in the camera, all it does is just perform the artificial sharpening itself. You need to see if the out of focus thing happens when manual focusing as well as on auto focus. If you can manually focus the camera better than the auto focus, the back focus is off and needs to be re-calibrated by Canon. Otherwise, it is not uncommon for Canon brand lenses to also need calibration WITH the body to obtain correct focus all the time. These errors usually are not noticeable past about a metre (3 feet) distance. The 1.8 Mk II lens is a long way from being a great lens as far as focus reliability goes. Warranty claim time and take your lenses back too. The adjustment is a simple one but unfortunately Canon refuse to release the software to independent repairers do this. My 5D was away for over 3 weeks. The 20D before it, 10 days and the EOS 3 before that, a week. Wouldn't you think things would improve with time? The only Canon SLR I've had in the past 4 years which didn't need back focus adjustments was a 10D. Oddly this is the camera everyone else reported focus errors with. Other Canon models I've sent back for back focus repair include the A1 and a Video camera. Before you send the 5D back, do some exposure checks. They also have a habit of not white balancing properly on auto and another often overlooked problem is their inability to meter for a 5 stop darker scene after taking a picture in a bright scene. All these things are fixable with a warranty claim. > The other Canon unique > focus issue is the anti alias filter they put over the sensor to stop moiré > effect. all dSLRs have AA filter that does exactly this. The problem OP mentions could be as well result of poor telecentricity (which is fine for film cameras, but produces soft picture on edges on 24x36mm sensors). But I agree with you that BF/FF is first thing to be checked. B. : > The other Canon unique : > focus issue is the anti alias filter they put over the sensor to stop moiré [quoted text clipped - 6 lines] : : B. As I under stand the specifications of DSLRs, Nikon do not use an anti alias filter. Also my experience at printing large images from all brands of DSLRs is that the Nikon images are the much sharper than Canon out of the camera. Most really crisp image from Canon DSLRs have both Unsharp Mask and a high pass sharpening layer from Photoshop before they rival an out of camera image from a Nikon camera. I'm not knocking Canon. I own a 5D (as well as several other brands of DSLRs) and I like it fine. I just think people will become better at using their cameras if they understand some critical issues instead of worship the bullshit from the God of EOS. > As I under stand the specifications of DSLRs, Nikon do not use an > anti alias filter. Also my experience at printing large images from > all brands of DSLRs is that the Nikon images are the much sharper > than Canon out of the camera. I assure you, they do - otherwise picture quality would badly suffer from moire. Almost all brands of cameras use antialias filter. It's just that AA filter in D70(s) was rather weak, the one in D200 is stronger (ie. produces more visible blur). Only Sigmas do not have AA filter, effectively they suffer from luminance moire (but not chrominance moire, which is result of unusual sensor construction and lack of Bayer mask). > Most really crisp image from Canon > DSLRs have both Unsharp Mask I can not discuss with it, not knowing what models you compared and if you used RAW or JPEG (with some in-camera sharpening applied). > I'm not knocking Canon. and I'm not defending Canon, in fact I do not like them nor own one ;) I have Pentax *ist DS. B. >I assure you, they do - otherwise picture quality would badly suffer >from moire. Almost all brands of cameras use antialias filter. It's just >that AA filter in D70(s) was rather weak, the one in D200 is stronger >(ie. produces more visible blur). Only Sigmas do not have AA filter The Leica (Imacon) Digital Modul-R (DMR) has no anti-alias filter. Images from the DMR are superbly sharp and render just as much detail as a 16 MP Canon DSLR. >I'm not knocking Canon. I own a 5D (as well as several other brands of >DSLRs) and I like it fine. I just think people will become better at using >their cameras if they understand some critical issues instead of worship the >bullshit from the God of EOS. While I question Canon's judgement in many areas, I can not fault them for their AA filters. These cameras aren't made to make posters from 1MP crops of an 8 to 16MP image. I can't see any softness I could pin on the AA filter with 13*19 prints from my 10D or 20D. If you like the look of poorly- or non-filtered images at great magnification, perhaps you can't tell the difference between artifacts and subject detail.  Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >Most really crisp image from Canon DSLRs have both Unsharp Mask and a high >pass sharpening layer from Photoshop before they rival an out of camera >image from a Nikon camera. That is because the sequence of events in correct information processing is to sufficiently low pass filter the analogue bandwidth to prevent aliasing *prior* to sampling and then high pass filter the digital data to flatten the frequency response *after* sampling. In imaging terms that translates directly into blurring the image before it gets to the sensor using a combination of the AA filter and the spatial response of the sensor pixels themselves and then re-sharpen the image in the digital image. Inadequate AA filtering creates a sharper *effect* without any post processing, at the expense of creating false detail that is not present on the original. Once this false detail has been created it cannot be distinguished from real detail and consequently cannot be removed from the image without also removing real detail - which is why you need to filter before it is created. >I'm not knocking Canon. I own a 5D (as well as several other brands of >DSLRs) and I like it fine. The 5D actually has less AA filtering than is necessary to avoid aliasing completely. For example, you can see it in this image David Littleboy has on his gallery: http://www.pbase.com/davidjl/image/52118414/original SignatureKennedy Yes, Socrates himself is particularly missed; A lovely little thinker, but a bugger when he's pissed. Python Philosophers (replace 'nospam' with 'kennedym' when replying) > As I under stand the specifications of DSLRs, Nikon do not use an anti alias > filter. As we already knew, you do not understand the specifications (or use) of DSLRs. Of course Nikon uses an anti-alias filter. You were also the one insisting crazy things about the D2x sensor, which had no basis in anything other than your overactive imagination. SignatureJeremy | jeremy@exit109.com : > As I under stand the specifications of DSLRs, Nikon do not use an anti alias : > filter. [quoted text clipped - 4 lines] : You were also the one insisting crazy things about the D2x sensor, which : had no basis in anything other than your overactive imagination. Ah yes... The troll from the frozen waste. A waste of space too. > Ah yes... The troll from the frozen waste. A waste of space too. You're the one saying Nikon doesn't use anti-alias filters. Who's the one wasting space? SignatureJeremy | jeremy@exit109.com : > Ah yes... The troll from the frozen waste. A waste of space too. : : You're the one saying Nikon doesn't use anti-alias filters. Who's the one : wasting space? Idiot Jeremy. If I had specified some models.... If only Jeremy had a life but in the frozen wasteland he lives in, this is his life. Every year the same useless comments. Step outside and freeze the icicles off you brain. It might work for awhile then. > Idiot Jeremy. If I had specified some models.... Okay -- tell us which Nikon DSLR doesn't use an anti-alias filter. Go on, we're waiting. > If only Jeremy had a life but in the frozen wasteland he lives in, this > is his life. Every year the same useless comments. Step outside and freeze > the icicles off you brain. It might work for awhile then. You know, something is dawning on me, and this is getting pretty funny. A while back, you emailed me directly in reply to a message posted here. When I replied asking why you were mailing me directly, you said that you wanted me to reply so that you could look at my headers and find out who and where I am. I pointed out that a few moments at Google would get you that and more, and you said that information at Google was unreliable. Then you said that, had you known where I was from, you wouldn't have bothered replying to me at all. I couldn't figure out what you meant by that; I thought perhaps you'd seen "The Sopranos" and were wary. But, dude -- did you decide I was from *Canada*? That's priceless! So much for Google being unreliable, eh? SignatureJeremy | jeremy@exit109.com > :I bought a Canon 5D and a Canon 20mm lens at Christmas. > : I've recently bought a Canon 28mm F2.8 and a Canon 50mm F1.8 Mk 2. [quoted text clipped - 62 lines] > with > a warranty claim. I've done some tests using centre auto focus point and all three lenses at maximum aperture. Camera shake was also ruled out by fast shutter speeds. Manual focus at close distance at a focus test page (i.e. a flat object at approx 50cm) on the 50mm lens is less accurate than auto focus. I placed three cans of soup in a line, but staggered in distance to the camera by the width of one can. I auto focused on the middle can. I enlarge images from a 50mm f1.8, 28mm f2.8 and a 20mm f2.8 so that the cans were all the same size. The image quality from the 50mm and 20mm was equal even though the 20mm had to be enlarged by a greater degree. The best quality was easily the 28mm. What appears to be the case is that the camera has actually focused on the can to the right of the one I had covered by the middle focus indicator. Could it be that the focus indicator in the view finder and where the camera actually focuses is slightly out? |
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.