 |
 | 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 / Film Photography / 35 mm / July 2005Canon ef 28-105 and dark corners
Hi all, After a lot of reading in this newsgroup and in the sites that are mostly linked here (photo.net, photodo etc), I decided that the lens mentioned in the subject line would be good for me (3.5-4.5). I got lucky on ebay and bought one very cheap, less than one year old, and the person who sold it to me told me not to use the rubbery lens hood he gave me with it at 28 mm as it would show in the picture. I just shot one film yesterday to check it out, no lens hood or any filters, and I noticed that in most of my sky shots (I live close to an airport) I see dark corners. I don't see it in all of them, but since I was using my rebel ti on full auto I didn't bother writing down apertures etc. I did google the subject a bit, but most problems refer to this problem when using filter(s) or hoods, which I didn't. Also, most problems mention the wide end, when I see the dark corners (vignetting?...) throughout the range, even on the long end. I cannot say that it is severe, but I find it very strange that a newish design lens that is generally accepted as 'good', 'above average' etc has that problem. In fact I was surprised to see that some people where talking about vignetting even on L lenses. Can somebody confirm (or tell me otherwise) that this can indeed appear not only in the wide end of this lens? I don't know if it could be that have a bad piece, but I cannot see how a bad lens can have this particular problem. I will scan a print or too to show the problem tomorrow; don't have my (cheap) scanner at home today. Otherwise, how can I rid of this problem ? Don't 28mm primes have vignetting ? If no, why not ? :-) Thanks in advance. Which 28-105 is it? There are at least 3 lensess of that length and the f3.5-4.5 is the best. The 4.5-5.6 versions are cheapies  Signaturehttp://www.chapelhillnoir.com home of The Camera-ist's Manifesto The Improved Links Pages are at http://www.chapelhillnoir.com/links/mlinks00.html A sample chapter from "Haight-Ashbury" is at http://www.chapelhillnoir.com/writ/hait/hatitl.html > Hi all, > [quoted text clipped - 27 lines] > > Thanks in advance. > Which 28-105 is it? There are at least 3 lensess of that length and the > f3.5-4.5 is the best. The 4.5-5.6 versions are cheapies It is indeed the 3.5 - 4.5 I'm afraid... Is it normal, dark corners on the tele end ? > > Which 28-105 is it? There are at least 3 lensess of that length and the > > f3.5-4.5 is the best. The 4.5-5.6 versions are cheapies > > It is indeed the 3.5 - 4.5 I'm afraid... Is it normal, dark corners on the > tele end ? I shoot with this lens all the time on my Elan IIe, and do not experience any vignetting at all. With a lens hood, you might catch some of it in the frame on the wide end -- but the one I used was a generic brand, I think, and not one issued by Canon, so I don't know if the Canon hood would have caused the same problem. Patrick Not in my experience. I owned one for about a year. It could well be that I never shot a situation where it would be noticable though. Signaturehttp://www.chapelhillnoir.com home of The Camera-ist's Manifesto The Improved Links Pages are at http://www.chapelhillnoir.com/links/mlinks00.html A sample chapter from "Haight-Ashbury" is at http://www.chapelhillnoir.com/writ/hait/hatitl.html > > Which 28-105 is it? There are at least 3 lensess of that length and the > > f3.5-4.5 is the best. The 4.5-5.6 versions are cheapies > > It is indeed the 3.5 - 4.5 I'm afraid... Is it normal, dark corners on the > tele end ? > Hi all, > [quoted text clipped - 27 lines] > > Thanks in advance. First off, all lenses, of whatever quality, vignette. There are good reasons for this, the first being that compound lenses are constructed in a sort of 'tube'. When you look through a lens with your eye on the lens axis, the lens appears round, but if your eye is off-axis, you will see an ellipse, the width of the ellipse varying with the viewing angle. This is caused by two factors; the length of the lens barrel if the lens is wide open, and simply by the oblique view of the diaphragm if the lens is stopped down. Clearly, the shape (area) of the ellipse will pass less light than the full circular aperture seen on axis, so there will be light fall-off towards the edges of the image cast by the lens. The second reason is called the Cos^4 law (which also takes into consideration the off-axis effects mentioned above), which says that theoretically the light intensity at any point on an image is reduced by the fourth power of the cosine of the angle subtended with the lens axis by the point under consideration. The practical effect depends on lens design, but becomes more apparent with wide-angle lenses. The Ilford Manual of Photography quotes a 44% fall-off at the corners of a normal lens, and for a 90-degree wide-angle lens, the corner fall-off is 75%. Vignetting, I believe, is more of a problem with digital printing. When using an optical enlarger, the enlarging lens displays vignetting just as did the camera lens, but this time the light fall-off tends to counteract the vignetting in the negative - the thinner corners of the negative are printed lighter because of the enlarger vignetting. No such counteraction occurs when printing digitally, from either a scanned or direct digital image, so measures may have to be taken during the photoshop phase if vignetting is apparent. Vignetting is not all bad, a small amount does tend to subtly frame the image, preventing your eye from straying off the picture through a light-toned area on the print edge. Printmakers often 'burn-in' the edges of an image for just this reason. After all that, the short answer to your question is, you can't get rid of the problem. The Canon 28-105 f/3.5 - f/4.5 lens is a pretty good lens, much better than the f/4 - f/5.6 version, and better than the 28-80mm predecessor. Colin > Can somebody confirm (or tell me otherwise) that this can indeed appear not > only in the wide end of this lens? I don't know if it could be that have a > bad piece, but I cannot see how a bad lens can have this particular problem. > I will scan a print or too to show the problem tomorrow; don't have my > (cheap) scanner at home today. You should only see this at the wide end and at wide open settings. If you're seeing it at 105mm or at a few stops closed down (even wide), there is something strange going on. (Or it's a crappy design which isn't out of the question). Redo your tests a little more carefully to be sure. Use manual settings. 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. There are some discussions about this problem: http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=003avC http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=003vP2 Regards > Hi all, > [quoted text clipped - 27 lines] > > Thanks in advance. > There are some discussions about this problem: > http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=003avC > http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=003vP2 I read the comments on the links, thanks. Since I am not the only one who has the problem, and with the results of the test that Alan suggested still pending, I find it incredible that a lens which is famous for being above average on tests, without serious distortion and contrasty, can on the other hand have such a serious fault. And it's really weird to say the least that it is so highly recommended by so many people here in this group and elsewhere. I don't care too much about detail, but in my sky pictures the problem is too much there to ignore. I realise that some lenses may not have this problem but this thing about 'within tolerances' from Canon's side sounds ridiculous. It's very weird that my kit 28-90 that was sold shortly after bought did not have dark corners on any aperture/focal length but a lens that costs 2 to 3 times as much has it. I agree with some of the posters (in the links) that the lens should not be advertised as 3.5-4.5 but as 8-11 (assuming that my test confirms that stopping down reduses/eliminates the problem). In the end I am lucky, as I bought this one used and very cheap; I'ld be totally gutted if I had payed it as new and discovered that. By the way, the person who sold it to me used it only on a 300d; is it possible that he had not seen the problem because his sensor is smaller than the 35mm film? Do all manufacturer's lenses have this problem ? What about wide-angled primes ? Thanks > Do all manufacturer's lenses have this problem ? What about wide-angled > primes ? Many zooms also suffer from vignetting at the wide angle, wide open end. To help, remove the filter, but there's usually something. Rounding up my lenses (all Minolta): 20mm f/2.8 (Minolta), even with a 72mm throat, vignettes if I leave a filter on when wider than about f/4.5 With a circ-pol it is even more suscpetible (esp. my monster sized Minolta circ-pol). This lens is really in its element for landscapes, so removing the filter is always a benefit. It's a bad candidate for circ-pol use in most circumstances in any case. 80-200 f/2.8 vignettes slightly at the wide end wide upon. Again, removing the filters helps a little. For this lens it's more apparent on slide film than negative film (or seems so... OTOH I rarely use that lens with negative film so maybe they're the same and I haven't noticed). My 28-70 f2.8 is prety well behaved, but I believe I've noticed soem vignetting on circ-pol shots. 100 f/2.8. Perfect. 50mm f/1.7 ... hmm, not sure really. I don't use that little gem enough. 300 f/2.8, never noticed any vignetting. OTOH I tend to shoot it at about f/5.6 and slower. 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. > Hi all, > [quoted text clipped - 27 lines] > > Thanks in advance. What film are you using that is giving you this result? High-contrast film like slow slide films will make the vignetting appear worse than if you are using a negative film with lower contrast. I have used a 28-105 on my EOS 10 for years, and haven't had the sort of problem you describe - but I always shoot color negative material. Colin > What film are you using that is giving you this result? High-contrast > film like slow slide films will make the vignetting appear worse than if > you are using a negative film with lower contrast. I have used a 28-105 > on my EOS 10 for years, and haven't had the sort of problem you describe > - but I always shoot color negative material. Fuji Reala 100... No slide film, no filters. Have you ever shot a light colored surface or sky ? But then again, you may belong to the lucky ones that own a better sample of this lens. Thanks, Kostas > Hi all, > I just shot one > film yesterday to check it out, no lens hood or any filters, and I noticed > that in most of >>>>>>>>>>>>>>>>my sky shots <<<<<<<<<<<<<<<<<<<<<< (I live close to an airport) I see dark > corners. This is very common and has more to do with light polarisation across the sky than your lens. I see it all the time when processing. It is especially pronounced when exposure is incorrect. Most cameras do not correctly expose for a 'point in the air' sky shot so you get light fall off around the edges as the processor attempts to correct the density of your print when processing. If you don;t get the same effect in normal photos then there is nothing wrong Hi , After a long delay I managed to find time to scan & post the results of my simple demonstration of vignetting of the (favourite over here) Canon EF 28-105/3.5-4.5. If interested, have a look here: http://www.pbase.com/melissakos/281053545 I don't know if there is an alternative for this lens (since I don't have either the money or the patience to buy and use primes) but if I had seen this before I am not sure I would have bought it. Might be useful to people considering to buy it. Kostas Look, friend, cheap zooms have fall-off. They are intended to be used by amateurs who shoot color negative film. With such films, ordinarily, fall-off is not so noticeable. If you are that critical, put your money where your mouth is (into single-focal length primes), or shut the hell up. > Hi , > [quoted text clipped - 13 lines] > > Kostas > Look, friend, cheap zooms have fall-off. They are intended to be used > by amateurs who shoot color negative film. With such films, ordinarily, > fall-off is not so noticeable. If you are that critical, put your money > where your mouth is (into single-focal length primes), or shut the hell > up. I am not your friend, and you are missing the point; what I am talking about is a problem of the lens that wasn't mentioned each time somebody recommended it. Also, what you say doesn't answer why the kit lens does not vignette at all; and this is the one that is even more "amateurcolornegative" oriented. Looking at the pictures I posted, I don't consider my self critical. Is that what you call not so noticable? In the end it seems not only cheap zooms have fall-off. Kostas > In the end it seems not only cheap zooms have fall-off. The GR1v has fall off in the corners, and it does not have a zoom. Signaturehttp://www.petezilla.co.uk 28-105 is a rather large ratio. It would be remarkable if there were not significant fall-off. Sometimes lens designers deliberately choose falloff in illumination at the corners to solve other problems, including manufacturing tolerances and cost contraints. In other words: YOU GET WHAT YOU PAY FOR. If you don't like the fall-off, get SINGLE focal length lenses. > > Look, friend, cheap zooms have fall-off. They are intended to be used > > by amateurs who shoot color negative film. With such films, ordinarily, [quoted text clipped - 12 lines] > > Kostas > Hi , > [quoted text clipped - 13 lines] > > Kostas You don't say what film you used here. The inherent contrast of the film is a factor in how visible vignetting is. High-contrast transparencies will show it much more than negative films, not to forget that if a negative is printed optically, the vignetting of the printing lens works to counteract that in the negative. Some advanced digital printers like Frontiers will compensate for vignetting, but home printing from a digital file - either a scanned negative or transparency, or direct digital - will not have any compensation for vignetting unless you use an advanced plug-in for Photoshop, PSP, or other image-handling software which corrects for the problem. *All* lenses vignette to some degree. Any light bundle through a lens barrel will have less area as the incident angle goes off-axis. A simple single-element lens with no lens barrel at all does this, as the cross-section of the light bundle goes from circular when on-axis, to elliptical - with less area - as the angle with the axis increases. With practical lenses mounted in barrels of significant length, you then have cutoff due to the barrel length as well, and this effect is worse with zooms, as the barrel length is longer than would be required with a prime lens. These effects (excepting barrel length) are summed up in the cos^4 law, that is, the light intensity striking the sensitive surface varies with the 4th power of the cosine of the angle the particular point on the image makes with the lens axis. To illustrate with simple numbers, consider a perfect 28mm lens projecting an image onto a 24 x 36 mm surface. The corners of that image are 21.6 mm off-axis. With the focal length of 28mm, that makes an angle off-axis of 37.65 degrees. Cosine of 37.65 is 0.792, and 0.792^4 is 0.3929. That says the light intensity from a perfect 28mm lens in the corner of a 35mm frame is 39% of the intensity in the centre of the frame. In photographer speak, the corners will be underexposed by a stop and a half, more or less. Note this is for a perfect lens, which no practical lens is. Barrel length will cause even less intensity in the corners. I have the same lens, and used it for years on an EOS 10, entirely with negative film, and the performance was no worse than my 28-80 or the 35-135 that the camera came with. Tell us some more about your film and printing set-up. Colin > You don't say what film you used here. The inherent contrast of the I noticed the problem on Fuji Superia 400. The film I used to shoot these pictures is super-market brand... Both color negatives. These are developed in the lab I always take my films too, but in the end, the dark corners are visible even when looking through the viewfinder, so I don't think the film is the problem. > *All* lenses vignette to some degree. Any light bundle through a lens I haven't owned many so I'll take your word for it; but then, how do people use this partuicular one, WITH filters ? Do you ever shoot wide open ? I find it surprising. > Note this is for a perfect lens, which no practical lens is. Barrel > length will cause even less intensity in the corners. > > I have the same lens, and used it for years on an EOS 10, entirely with > negative film, and the performance was no worse than my 28-80 or the > 35-135 that the camera came with. But then ,and this is just out of curiosity as I learned to live with it, have you had vignetting problems? And would I have to expect vignetting even from the EF 28mm/2.8 ? Or the 24mm ? Thank you for the detailed reply, Cheers Kostas >These effects (excepting barrel length) are summed up in the cos^4 law, >that is, the light intensity striking the sensitive surface varies with [quoted text clipped - 12 lines] >Note this is for a perfect lens, which no practical lens is. Barrel >length will cause even less intensity in the corners. You are correct about the cos^4 law for a simple perfect lens, but incorrect if you are implying that any practical lens will perform worse than this. In fact, techniques available to designers enable them to produce significantly less vignetting than that suggested by cos^4. To give one example, the over-large negative elements used at front and rear of most wide angle lenses are there for precisely this reason. Agreed it is very difficult, with a rectilinear lens, to eliminate vignetting entirely when the lens is at maximum aperture; it is though possible with fisheyes. David SignatureDavid Littlewood I have the same lens which I use with my Elan IIe, which I use all the time, without a hood, and do not experience any vignetting whatsoever. Patrick |
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.