This topic is in the 2009 archive. You can read it and like its posts here. It is read only.
Archive 2009Canon 5D MKII and Live View
p.2 #1Canon 5D MKII and Live View
burningheart wrote:
Went out today and was able to reproduce including shooting upside dwon. Used the Angeneiuex 180, Leica 180, Canon 180 Macro with same results. I also used the coastal Optics 60mm and Nikon 28mm lens Intersetingly enough the Nikon did not produce this. It occurs only on the opposite side of the mirror, not as noticable when camera in landscape positioning but there none the less. Anyways I see this as an inconvience glitch, I'll just turn off Live View after focusing for non Canon lens. Todays shots were near infinity focusing.
Upside down Leica 180 F2.8 1/5000
http://www.robert-chisholm.com/fred_miranda/upside_down.jpg
Correct me if I'm wrong, but I believe I read that the Canon 5Dii's LV implementation skips the first curtain when a shot is taken while it's active. Since we don't see any evidence of the vignetting being non uniform horizontally, even when shooting at 1/5000, I think we can assume that the mirror is getting in the way throughout exposure rather than dropping prematurely.
If so, that's interesting. Maybe the mirror assembly is floating a bit while in the up position, intruding into the mirror box? You might be able to see this by visual inspection, keeping live view active while rotating or (gently) shaking the camera with no lens on it. Or perhaps it just pops down a bit when the second curtain fires? Does the phenomenon ever show up while shooting in movie mode?
I have no explanation for why it'd only affect certain lenses, but it does sound like something Canon would be interested in as well.
Went out today and was able to reproduce including shooting upside dwon. Used the Angeneiuex 180, Leica 180, Canon 180 Macro with same results. I also used the coastal Optics 60mm and Nikon 28mm lens Intersetingly enough the Nikon did not produce this. It occurs only on the opposite side of the mirror, not as noticable when camera in landscape positioning but there none the less. Anyways I see this as an inconvience glitch, I'll just turn off Live View after focusing for non Canon lens. Todays shots were near infinity focusing.
Upside down Leica 180 F2.8 1/5000
http://www.robert-chisholm.com/fred_miranda/upside_down.jpg
Correct me if I'm wrong, but I believe I read that the Canon 5Dii's LV implementation skips the first curtain when a shot is taken while it's active. Since we don't see any evidence of the vignetting being non uniform horizontally, even when shooting at 1/5000, I think we can assume that the mirror is getting in the way throughout exposure rather than dropping prematurely.
If so, that's interesting. Maybe the mirror assembly is floating a bit while in the up position, intruding into the mirror box? You might be able to see this by visual inspection, keeping live view active while rotating or (gently) shaking the camera with no lens on it. Or perhaps it just pops down a bit when the second curtain fires? Does the phenomenon ever show up while shooting in movie mode?
I have no explanation for why it'd only affect certain lenses, but it does sound like something Canon would be interested in as well.
Feb 08, 2009 at 11:18 PM
p.2 #2Canon 5D MKII and Live View
i believe that when you are using Live You should turn Off the miror lock up option !
Using a Mirex blad and a Oly 35 PC i don't have this problem... but i've find some strange flare once... maybe lihgt coming from the eye piece ?
Using a Mirex blad and a Oly 35 PC i don't have this problem... but i've find some strange flare once... maybe lihgt coming from the eye piece ?
Feb 09, 2009 at 04:25 PM
p.2 #3Canon 5D MKII and Live View
archivue wrote:
i believe that when you are using Live You should turn Off the miror lock up option !
Using a Mirex blad and a Oly 35 PC i don't have this problem... but i've find some strange flare once... maybe lihgt coming from the eye piece ?
You could try shining a penlight into the viewfinder while taking a long exposure to confirm this possibility, but it shouldn't happen. The mirror's position during exposure should effectively block off this light path. (Though extraneous light entering the viewfinder before the exposure can certainly affect your metering!)
i believe that when you are using Live You should turn Off the miror lock up option !
Using a Mirex blad and a Oly 35 PC i don't have this problem... but i've find some strange flare once... maybe lihgt coming from the eye piece ?
You could try shining a penlight into the viewfinder while taking a long exposure to confirm this possibility, but it shouldn't happen. The mirror's position during exposure should effectively block off this light path. (Though extraneous light entering the viewfinder before the exposure can certainly affect your metering!)
Feb 09, 2009 at 10:29 PM
p.2 #4Canon 5D MKII and Live View
Surprisingly quiet discussion since if this is common 5DmkII problem this basically means that 5DmkII is not usable with alternative lenses (depends on shooting style but I shoot 99% with live view). Is burningheat the only person suffering from this?
I'm considering 5DmkII but secondhand 1DsmkIII start to seem interesting choice if there is even slight risk of this bug in 5DmkII. 2nd hand 1DsmkIII only about 2k more and has better autofocus / custom functions / viewfinder for the situations when I'm not using alternative lenses. Only bad thing about 1D series cameras is the size and weight. Luckily they still don't have 5DmkII in Finnish stores...
--
Samuli Vahonen
http://www.vahonen.com
I'm considering 5DmkII but secondhand 1DsmkIII start to seem interesting choice if there is even slight risk of this bug in 5DmkII. 2nd hand 1DsmkIII only about 2k more and has better autofocus / custom functions / viewfinder for the situations when I'm not using alternative lenses. Only bad thing about 1D series cameras is the size and weight. Luckily they still don't have 5DmkII in Finnish stores...
--
Samuli Vahonen
http://www.vahonen.com
Feb 13, 2009 at 12:33 AM
p.2 #5Canon 5D MKII and Live View
olyacme wrote:
Always in portrait orientation, or in landscape as well? How about when the camera is inverted in landscape orientation - is the problem worse?
I think it is only in portrait mode or similar "off horisontal" shooting. Havent seen it in a while though. The problem just appears every now and then for me.
ovredal73 wrote
I have this same problem, and have gotten a few shots ruined due to it. Sometimes it is quite severe. I thought it was just me...I think it is only in portrait mode or similar "off horisontal" shooting. Havent seen it in a while though. The problem just appears every now and then for me.
Feb 13, 2009 at 12:44 AM
p.2 #6Canon 5D MKII and Live View
Samuli Vahonen wrote:
Surprisingly quiet discussion since if this is common 5DmkII problem this basically means that 5DmkII is not usable with alternative lenses (depends on shooting style but I shoot 99% with live view). Is burningheat the only person suffering from this?
I'm surprised discussion abated, too, considering the furor generated by an issue that, unlike this one, cannot be seen at web resolution. It would be aggravating in the extreme to find one shot spoilt in the middle of a mosaic sequence or focus stack come processing time. I'm also contemplating buying a 5Dii, and I'd really like to find a reason to blame live view exclusively as then at least there'd be a definitive workaround. Even at that, given my reliance on LV, I know I'd still get burned by this from time to time. It'd be particularly awful if it affected movie mode, though with the widespread use of manual lenses for this I expect it would have gained wider attention by now if it did.
I remain curious as to what happens when the camera's inverted (or gently shaken!) when taking a movie through a manual lens.
Surprisingly quiet discussion since if this is common 5DmkII problem this basically means that 5DmkII is not usable with alternative lenses (depends on shooting style but I shoot 99% with live view). Is burningheat the only person suffering from this?
I'm surprised discussion abated, too, considering the furor generated by an issue that, unlike this one, cannot be seen at web resolution. It would be aggravating in the extreme to find one shot spoilt in the middle of a mosaic sequence or focus stack come processing time. I'm also contemplating buying a 5Dii, and I'd really like to find a reason to blame live view exclusively as then at least there'd be a definitive workaround. Even at that, given my reliance on LV, I know I'd still get burned by this from time to time. It'd be particularly awful if it affected movie mode, though with the widespread use of manual lenses for this I expect it would have gained wider attention by now if it did.
I remain curious as to what happens when the camera's inverted (or gently shaken!) when taking a movie through a manual lens.
Feb 13, 2009 at 01:13 AM
p.2 #7Canon 5D MKII and Live View
ovredal73 wrote:
I think it is only in portrait mode or similar "off horisontal" shooting.
This is the worst possible - panorama heads are build so that camera is on vertical orientation. I have also found out that due to my shooting style I shoot almost 45% of the photos vertical so for me this would be show stopper even without panoramas.
I think it is only in portrait mode or similar "off horisontal" shooting.
This is the worst possible - panorama heads are build so that camera is on vertical orientation. I have also found out that due to my shooting style I shoot almost 45% of the photos vertical so for me this would be show stopper even without panoramas.
Feb 13, 2009 at 01:37 AM
p.2 #8Canon 5D MKII and Live View
I am able to reproduce the problem with a Voigtlander 40/2, an Olympus 50/1.2, but a Zeiss ZE 50/1.4 Planar and EF 50/1.8 show no issue.
Nor do I observe an issue on the EOS 1Ds Mark III, using the problem lenses.
Based on my inspection of the 5D Mark II, I do not believe this is a mirror issue; I think it's a software bug.
Nor do I observe an issue on the EOS 1Ds Mark III, using the problem lenses.
Based on my inspection of the 5D Mark II, I do not believe this is a mirror issue; I think it's a software bug.
Feb 15, 2009 at 03:25 PM
p.2 #9Canon 5D MKII and Live View
André - did you use chipped or non-chipped adapters?
Thanks, Samuli Vahonen
Samuli - I only use cheap, non-chipped adapters. I will try to test this some more this week. I only use the rokkor 58 now and it does not have any issues. Havent seen this problem for a while, maybe due to lens choice...
André
Thanks, Samuli Vahonen
Samuli - I only use cheap, non-chipped adapters. I will try to test this some more this week. I only use the rokkor 58 now and it does not have any issues. Havent seen this problem for a while, maybe due to lens choice...
André
Feb 15, 2009 at 03:54 PM
p.2 #10Canon 5D MKII and Live View
I was asked for additional information. Summarized my findings
1. Only happens with alternate lens with unchipped adapters for Lieca, Olympus and F mount. Use cameraquest adapters. Has not happened with any Canon lens.
2. Only appears when aperture is set F1.4 through F5, does not happen F5.6 through F45.
3. Usually when lens is tilted vertically but on opposite side of mirror. Happened on mirror side when the shot camera upsidedown. I saw it a coupke of times on a horizontal picture at the bottom of the frame.
4. Seems most pronounced on non wide angle lens. I have seen it on focal lengths of 60mm(Coastal Optics), 135mm(Nikon DC), 180mm(Angenieux, Lieca), and 250mm(Olympus). Didin't happen on the 28mm(Nikon).
5. Does not occur when Live view turned off before snapping picture.
Other findings I just recently noticed
1. Happened so far only in daylight, irrelevant of suns position.
2. Seems to occur at shutter speeds faster than 1/2500 second.
My solution has been to focus with live view and then turn it off to shoot. I am not spending anytime right now investigating any further as my Mom is in the hospital, so photography is not a priority.
1. Only happens with alternate lens with unchipped adapters for Lieca, Olympus and F mount. Use cameraquest adapters. Has not happened with any Canon lens.
2. Only appears when aperture is set F1.4 through F5, does not happen F5.6 through F45.
3. Usually when lens is tilted vertically but on opposite side of mirror. Happened on mirror side when the shot camera upsidedown. I saw it a coupke of times on a horizontal picture at the bottom of the frame.
4. Seems most pronounced on non wide angle lens. I have seen it on focal lengths of 60mm(Coastal Optics), 135mm(Nikon DC), 180mm(Angenieux, Lieca), and 250mm(Olympus). Didin't happen on the 28mm(Nikon).
5. Does not occur when Live view turned off before snapping picture.
Other findings I just recently noticed
1. Happened so far only in daylight, irrelevant of suns position.
2. Seems to occur at shutter speeds faster than 1/2500 second.
My solution has been to focus with live view and then turn it off to shoot. I am not spending anytime right now investigating any further as my Mom is in the hospital, so photography is not a priority.
Feb 15, 2009 at 04:57 PM
p.2 #11Canon 5D MKII and Live View
A discovery: the Auto LIghting Optimizer feature is not responsible, but it exacerbates the problem. Be sure to turn it off, though with RAW you can do this later in processing.
I've now tried the Leica 90/2 APO-Summicron-R also, with a Novoflex (dumb) adapter. Same problems as with the other lenses.
http://diglloyd.com/diglloyd/2009-02-blog.html#_20090215VignettingLiveView
I've now tried the Leica 90/2 APO-Summicron-R also, with a Novoflex (dumb) adapter. Same problems as with the other lenses.
http://diglloyd.com/diglloyd/2009-02-blog.html#_20090215VignettingLiveView
Feb 15, 2009 at 05:14 PM
p.2 #12Canon 5D MKII and Live View
Burningheart,
Thanks for the summary and in documenting this problem.
Hope your mom recovers soon. Best wishes.
Thanks for the summary and in documenting this problem.
Hope your mom recovers soon. Best wishes.
Feb 16, 2009 at 02:01 PM
p.2 #13Canon 5D MKII and Live View
Thanks for the well wishes she's in for tests now so I had some spare time today and received my cameraquest adapter in the mail that was chipped for 100mm F2.0. Bought from another FM'er. Ran a quick set of tests ISO 200, 1/8000 lens Angeneiux 180mm(f2.3) and Nikon 105mm DC (f2.0) first with chipped adapter and the without. I found the chipped adapter shots do not show the issue. Chipped shots on the right.
http://www.robert-chisholm.com/fred_miranda/chipped.jpg
http://www.robert-chisholm.com/fred_miranda/chipped.jpg
Feb 16, 2009 at 03:32 PM
p.2 #14Canon 5D MKII and Live View
Actually I didn't do a very good job of describing the issue with my 40D: sometimes I see the vignetting while focusing with LiveView, but when I take the shot it's not there and the image is always fine.
Weird.
Weird.
Mar 05, 2009 at 02:52 AM
p.2 #15Canon 5D MKII and Live View
Finally had opportunity to test this with my new 5D mkII: it seems chipped adapters work just fine, or actually if camera has electronical connection to lens, this problem can be regenerated by Canon lenses as well by blocking the electronical contact between camera and lens. Luckily I have all my adapters chipped so this issue does not concern me at this point at all.
1. EXIF chip connected, 2. no electonical connection, 3. EXIF chip connected, 4. no electonical connection
--
Samuli Vahonen
http://www.vahonen.com
1. EXIF chip connected, 2. no electonical connection, 3. EXIF chip connected, 4. no electonical connection
--
Samuli Vahonen
http://www.vahonen.com
Mar 14, 2009 at 07:56 AM
p.2 #16Canon 5D MKII and Live View
just did a quick and extremely unscientific test using 4 lenses:
-cv125 apo macro, which has been converted to ef mount (not the native ef mount version) and does not have a chip or other electronic connector
-zeiss 85/1.4 with happypagehk af-confirm chipped adapter
-leica 19mm/2.8 (ver II, non-ROM) with non-chipped cameraquest adapter
-ef 135L.
I took 5 images per lens, all at max aperture and infinity focus out my apartment window at the 59th st bridge and extremely overcast sky in the distance. these were handheld as I don't feel like getting out the tripod. I took images at 5 shutter speeds: 1/8000, 1/5000, 1/3200, 1/2500, 1/2000.
no vignetting issues at all on the ef 135L or the zeiss 85/1.4 with chipped adapter at any shutter speed.
the cv125 and leica 19 both exhibited substantial vignetting in BOTH portrait and landscape orientation at the bottom of the frame (the right side in vertical orientation) at 1/8000 shutter speed. it got better at 1/5000. Very little but noticeable at 1/3200 and basically gone by 1/2500 and 1/2000.
basically, to me, this means that LV with non-chipped lenses will create vignetting at shutter speeds of 1/3200 and faster. the vignetting is on the bottom of the image, which, as i understand it, is the part of the sensor closest to the flash shoe, correct?
i wonder if somebody else wants to try to replicate these results? while i have no understanding of shutter construction and how LV works, this doesn't make a whole lot of sense to me that the chipped adapter makes a difference.
-damon
-cv125 apo macro, which has been converted to ef mount (not the native ef mount version) and does not have a chip or other electronic connector
-zeiss 85/1.4 with happypagehk af-confirm chipped adapter
-leica 19mm/2.8 (ver II, non-ROM) with non-chipped cameraquest adapter
-ef 135L.
I took 5 images per lens, all at max aperture and infinity focus out my apartment window at the 59th st bridge and extremely overcast sky in the distance. these were handheld as I don't feel like getting out the tripod. I took images at 5 shutter speeds: 1/8000, 1/5000, 1/3200, 1/2500, 1/2000.
no vignetting issues at all on the ef 135L or the zeiss 85/1.4 with chipped adapter at any shutter speed.
the cv125 and leica 19 both exhibited substantial vignetting in BOTH portrait and landscape orientation at the bottom of the frame (the right side in vertical orientation) at 1/8000 shutter speed. it got better at 1/5000. Very little but noticeable at 1/3200 and basically gone by 1/2500 and 1/2000.
basically, to me, this means that LV with non-chipped lenses will create vignetting at shutter speeds of 1/3200 and faster. the vignetting is on the bottom of the image, which, as i understand it, is the part of the sensor closest to the flash shoe, correct?
i wonder if somebody else wants to try to replicate these results? while i have no understanding of shutter construction and how LV works, this doesn't make a whole lot of sense to me that the chipped adapter makes a difference.
-damon
Mar 14, 2009 at 08:58 AM
p.2 #17Canon 5D MKII and Live View
Damon, I forgot to mention the shutter speed effect - I find effect of shutter speed similar as you. And yes this is area of sensor which is closest to the hotshoe. So in Finland this rarely is problem 
Mar 14, 2009 at 09:25 AM
p.2 #18Canon 5D MKII and Live View
SOME CLARIFICATION PLEASE:
1: Does the problem happen when in live-view MODE, or only when the live-view LCD is on?
2: Is this being discussed elsewhere, is it a confirmed problem?
3: If confirmed, is it a defect, found on on some cameras? (i.e. a recall problem)
4: Has it been confirmed that chipped adapters solve the problem, 100% ?
-----------------------
5: Is a specific type of chip required?
In other words, is focus-confirm enough, or does it require maximum aperture exif data? I ask because my chips can't always be programmed for the exact max aperture. (I use the programmable type)
-And if aperture data is required, what happens when you stop down a little?
- If focus-confirm is all that's required, what happens if you purposely have the center of the shot OOF?
1: Does the problem happen when in live-view MODE, or only when the live-view LCD is on?
2: Is this being discussed elsewhere, is it a confirmed problem?
3: If confirmed, is it a defect, found on on some cameras? (i.e. a recall problem)
4: Has it been confirmed that chipped adapters solve the problem, 100% ?
-----------------------
5: Is a specific type of chip required?
In other words, is focus-confirm enough, or does it require maximum aperture exif data? I ask because my chips can't always be programmed for the exact max aperture. (I use the programmable type)
-And if aperture data is required, what happens when you stop down a little?
- If focus-confirm is all that's required, what happens if you purposely have the center of the shot OOF?
Mar 20, 2009 at 06:20 AM
p.2 #19Canon 5D MKII and Live View
Cableaddict wrote:
SOME CLARIFICATION PLEASE:
1: Does the problem happen when in live-view MODE, or only when the live-view LCD is on?
What is live view mode? Either you use live view or don't use - when you do use live view, then image is displayed in LCD and then this happens, there is no half way live view mode that live view is on and LCD is not connected unless you use external monitor via cables, that I haven't tested...and will not. It won't happen if you are not using live view.
Cableaddict wrote:
2: Is this being discussed elsewhere, is it a confirmed problem?
I haven't checked the manual focus forum, but that is about the only place in addition to this forum in which alternative lenses are discussed. Another usage without electronical connection between lens and body is astrophotography (T-mount to telescopes), in which I doubt that higher shutter speeds than 1/2000s are used...
Cableaddict wrote:
3: If confirmed, is it a defect, found on on some cameras? (i.e. a recall problem)
The number of testers in this thread is too small to clearly identify this. Based on how the few testers, who have purchased 5DmkII at different times from different continents, in this thread have so similar symptoms that there is no reason to thing this is camera/batch specific.
This definetly ain't causing recall, why Canon would care people not using their own lenses? And as can be seen from images with chip, Canon lens, slow enough shutter speed or simply live view not used there is no strange vignetting. This is a software problem, when no electronical connection then something goes wrong.
Cableaddict wrote:
4: Has it been confirmed that chipped adapters solve the problem, 100% ?
At least happypagehk and non-brand ebay adapters do work. Canon lens works the same way if you turn it in the mount so it doesn't have electronical connection. 100% sure, wait 5 years and I'll tell you...
Cableaddict wrote:
5: Is a specific type of chip required?
I have tested with happypagehk (10+ adapters to C/Y, Nikon, Leica) and nobrand eBay chipped cheap adapters. To me it seems that there is no difference between chips, but for statistical analysis the sample is too small.
Cableaddict wrote:
In other words, is focus-confirm enough, or does it require maximum aperture exif data? I ask because my chips can't always be programmed for the exact max aperture. (I use the programmable type)
- And if aperture data is required, what happens when you stop down a little?
- If focus-confirm is all that's required, what happens if you purposely have the center of the shot OOF?
Only thing needed is that camera things that there is electronically connected lens in the camera. This has nothing to do with the values provided by the chip or autofocus system of the camera.
It's really missleading to call these chips focus confirm chips, what the heck could be more missleading, the chip doesn't tell if the focus is achieved or not. The chips are simply providing electronical communication to camera, part of the information is EXIF relevant. This electronical communication information is used by camera, for example Canon cameras won't enable the AF system unless lens communicates aperture value to camera.
Some of the adapters I also have wrong max value, for example Zeiss Planar 1.7/50 is coded to 1.8 and Zeiss 3.4/35-70 had to be coded to 3.5, since it's not possible to send 1.8 or 3.5 from lens to camera.
--
Samuli Vahonen
http://www.vahonen.com
SOME CLARIFICATION PLEASE:
1: Does the problem happen when in live-view MODE, or only when the live-view LCD is on?
What is live view mode? Either you use live view or don't use - when you do use live view, then image is displayed in LCD and then this happens, there is no half way live view mode that live view is on and LCD is not connected unless you use external monitor via cables, that I haven't tested...and will not. It won't happen if you are not using live view.
Cableaddict wrote:
2: Is this being discussed elsewhere, is it a confirmed problem?
I haven't checked the manual focus forum, but that is about the only place in addition to this forum in which alternative lenses are discussed. Another usage without electronical connection between lens and body is astrophotography (T-mount to telescopes), in which I doubt that higher shutter speeds than 1/2000s are used...
Cableaddict wrote:
3: If confirmed, is it a defect, found on on some cameras? (i.e. a recall problem)
The number of testers in this thread is too small to clearly identify this. Based on how the few testers, who have purchased 5DmkII at different times from different continents, in this thread have so similar symptoms that there is no reason to thing this is camera/batch specific.
This definetly ain't causing recall, why Canon would care people not using their own lenses? And as can be seen from images with chip, Canon lens, slow enough shutter speed or simply live view not used there is no strange vignetting. This is a software problem, when no electronical connection then something goes wrong.
Cableaddict wrote:
4: Has it been confirmed that chipped adapters solve the problem, 100% ?
At least happypagehk and non-brand ebay adapters do work. Canon lens works the same way if you turn it in the mount so it doesn't have electronical connection. 100% sure, wait 5 years and I'll tell you...
Cableaddict wrote:
5: Is a specific type of chip required?
I have tested with happypagehk (10+ adapters to C/Y, Nikon, Leica) and nobrand eBay chipped cheap adapters. To me it seems that there is no difference between chips, but for statistical analysis the sample is too small.
Cableaddict wrote:
In other words, is focus-confirm enough, or does it require maximum aperture exif data? I ask because my chips can't always be programmed for the exact max aperture. (I use the programmable type)
- And if aperture data is required, what happens when you stop down a little?
- If focus-confirm is all that's required, what happens if you purposely have the center of the shot OOF?
Only thing needed is that camera things that there is electronically connected lens in the camera. This has nothing to do with the values provided by the chip or autofocus system of the camera.
It's really missleading to call these chips focus confirm chips, what the heck could be more missleading, the chip doesn't tell if the focus is achieved or not. The chips are simply providing electronical communication to camera, part of the information is EXIF relevant. This electronical communication information is used by camera, for example Canon cameras won't enable the AF system unless lens communicates aperture value to camera.
Some of the adapters I also have wrong max value, for example Zeiss Planar 1.7/50 is coded to 1.8 and Zeiss 3.4/35-70 had to be coded to 3.5, since it's not possible to send 1.8 or 3.5 from lens to camera.
--
Samuli Vahonen
http://www.vahonen.com
Mar 20, 2009 at 08:57 AM
p.2 #20Canon 5D MKII and Live View
The mirror is just grazing something and getting slowed down enough in its travel as a result to be caught by the shutter when taking the frame - hence the shadow at what would be the bottom of the frame.
The reason not everyone sees it is the infamous variation in mirror box tolerances/mirror placement between 5D/IIs. Also why will not seen with Canon lenses or with FF 1 series. Depending on lens, the problem may only appear as you focus towards infinity and lens mech extends rearward.
You'll get the same effect if you mount say a Leica 19 Mod2 that isn't shaved enough on a FF 1 series. Mirror slows at top of arc as it JUST grazes rear element shroud (remember lens projects upside down) and just gets caught in frame as shutter trips. Mirror will return as if no issue - you won't hear the difference in shutter movement. Try the same lens on a 1 series FF and you'll not see the vignetting - tighter manufacturing tolerances. Same goes if you try them on a cropped body.
Remember, in most cases (A900 excluded) the mirror in a DSLR travels in a straight arc. As such the distance between the bottom edge of the mirror and the lens mount changes throughout that arc. It's why when you shave a lens's rear element shroud you only need to shave it one spot.
Many adapter makers also mill their product a little skinny to "guarantee" infinity focus. On a 5D/II this can cause problems such as you're seeing.
Solutions - Shave mirror (a hair from looks of it) or thicker adapters and risk giving up infinity. I've run adapted lens's from any number of players using chipped, unchipped and custom-chipped adapters on the 30D, 1Ds2, 1D2, 1DS2, 5D. The only time I've ever seen this is in the situation described above. It will always appear at what would be the bottom of the frame in normal landscape mode.
Edited on Mar 20, 2009 at 09:20 AM · View previous versions
The reason not everyone sees it is the infamous variation in mirror box tolerances/mirror placement between 5D/IIs. Also why will not seen with Canon lenses or with FF 1 series. Depending on lens, the problem may only appear as you focus towards infinity and lens mech extends rearward.
You'll get the same effect if you mount say a Leica 19 Mod2 that isn't shaved enough on a FF 1 series. Mirror slows at top of arc as it JUST grazes rear element shroud (remember lens projects upside down) and just gets caught in frame as shutter trips. Mirror will return as if no issue - you won't hear the difference in shutter movement. Try the same lens on a 1 series FF and you'll not see the vignetting - tighter manufacturing tolerances. Same goes if you try them on a cropped body.
Remember, in most cases (A900 excluded) the mirror in a DSLR travels in a straight arc. As such the distance between the bottom edge of the mirror and the lens mount changes throughout that arc. It's why when you shave a lens's rear element shroud you only need to shave it one spot.
Many adapter makers also mill their product a little skinny to "guarantee" infinity focus. On a 5D/II this can cause problems such as you're seeing.
Solutions - Shave mirror (a hair from looks of it) or thicker adapters and risk giving up infinity. I've run adapted lens's from any number of players using chipped, unchipped and custom-chipped adapters on the 30D, 1Ds2, 1D2, 1DS2, 5D. The only time I've ever seen this is in the situation described above. It will always appear at what would be the bottom of the frame in normal landscape mode.
Edited on Mar 20, 2009 at 09:20 AM · View previous versions
Mar 20, 2009 at 09:03 AM