Guy, I see your point, but experienced photographers who bought this lens are posting problems they are having.
Which follows numerous prior posts from ecstatic owners when the lens was initially released.
Savas K wrote:
Guy, I see your point, but experienced photographers who bought this lens are posting problems they are having.
Which follows numerous prior posts from ecstatic owners when the lens was initially released.
My understanding is that this is a known issue with this lens as well as the non-IS version. It bothers me though as I like to use my zooms for close-ups. Although, I just started using my 70-200 non-IS with ext. tubes and so far the results have been good.
Today, I tested three 70-200 f/4.0 L IS lenses with the date codes:
UV03, UV12, UW04
and all of them exhibited the problem (UV03 was the sharpest at 135 mm but the least sharp at 200mm - all at MFD). I also tested 70-200 f/4.0 L (UW01) and it was less sharp at 135 mm but also the sharpness was reduced slightly less at 200 mm (but the issue was the same).
I would have probably chosen the lens with date code of UW04 but it already had dust (three specs) inside!
I thought that lenses were expected to be a little sharper away from their MFD? The only way I could think of it would be a way that I saw the "lens body calibration" explained.
Imagine a "Unit" doesn't matter what, but we will say "canon units". Your camera ships and to be in spec is -1/+1 unit (or 0).
Lenses are calibrated to be at -1/+1 units (or 0). You may try one that is at 0, and your camera is a +/-1, making it seem "softer" at the MFD, but moving it a bit further back and the performance increases. You could try one, and it's at +1/-1 and your camera is the same, and it seems better than another you tested. Then finally, you check another and it's at -1 and your camera is at 0, and it seems better a little before MFD, but sucks at the long/short end of the focal length.
I totally stole the "unit" from another thread, but I imagine it would be applicable to this scenario.
nkguy wrote:
Dangit. When people are speculating, could you guys put a question mark in the subject line? The subject line that this thread has is totally misleading.
For example: "70-200 F4 IS bad batch?" says to me, "I am wondering if there is a bad batch."
Whereas "70-200 F4 IS bad batch" says to me, "I am claiming that there is a bad batch." But there is no such bad batch - it's all just Internet speculation.
Bad batch was an understatement, this is a flaw as far as I'm concern. I guess like you said it's all internet speculation and we all are having Illusion about this MFD issue....
RobertLynn wrote:
I thought that lenses were expected to be a little sharper away from their MFD? The only way I could think of it would be a way that I saw the "lens body calibration" explained.
Imagine a "Unit" doesn't matter what, but we will say "canon units". Your camera ships and to be in spec is -1/+1 unit (or 0).
Lenses are calibrated to be at -1/+1 units (or 0). You may try one that is at 0, and your camera is a +/-1, making it seem "softer" at the MFD, but moving it a bit further back and the performance increases. You could try one, and it's at +1/-1 and your camera is the same, and it seems better than another you tested. Then finally, you check another and it's at -1 and your camera is at 0, and it seems better a little before MFD, but sucks at the long/short end of the focal length.
I totally stole the "unit" from another thread, but I imagine it would be applicable to this scenario....Show more →
I don't think so, in MFD some of these lenses not only were back/front focus, the subject usually had some sort of haze/halo around it, I think this is a lens diffraction that someone was referring to. I don't think calibration is going to help in this regard at all.
mfurman wrote:
Today, I tested three 70-200 f/4.0 L IS lenses with the date codes:
UV03, UV12, UW04
and all of them exhibited the problem (UV03 was the sharpest at 135 mm but the least sharp at 200mm - all at MFD). I also tested 70-200 f/4.0 L (UW01) and it was less sharp at 135 mm but also the sharpness was reduced slightly less at 200 mm (but the issue was the same).
I would have probably chosen the lens with date code of UW04 but it already had dust (three specs) inside!
man that's a bummer.....I guess getting a new replacement probably not going to make much difference, either I'll get another one just like this, or the one that is not as soft at 200mm but inferior at 135 or 70mm. Given those options I would rather have a very sharp 70 to 135mm....
Are the folks who are seeing issues, using crop bodies, full bodies or both? Anyone testing the same lens against both at the same time, under the same conditions? I only have a crop body, so I can't make that comparison.
Someone in a different thread linked a Nikon story that mentioned that Nikon 70-200 zoom lenses were "optimized" for crop bodies and that Canon might be optimizing for full bodies, but this seemed speculative.
If true though, they should make a 70-200 1.6 and a 70-200 full and quit with all this guessing!
I understand this is not a close up or macro lens, it has a very nice zoom range and generally a very sharp lens with 4 stops IS, light weight and all the good stuff. But in my mind IF none of the 4.0 IS can do well at >135mm in close range then I think Canon probably should at least warn us about it in the user manual, otherwise this lens isn't up to the spec, and in a way it is false advertisement.
I wonder how the MTF tests were carried out at photozone and www.the-digital-picture.com, anybody know how far the test subject was?
yauyi wrote:
I understand this is not a close up or macro lens, it has a very nice zoom range and generally a very sharp lens with 4 stops IS, light weight and all the good stuff. But in my mind IF none of the 4.0 IS can do well at >135mm in close range then I think Canon probably should at least warn us about it in the user manual, otherwise this lens isn't up to the spec, and in a way it is false advertisement.
I wonder how the MTF tests were carried out at photozone and www.the-digital-picture.com, anybody know how far the test subject was? ...Show more →
I remember somebody had contacted Photozone about this and that the tests were all conducted at a large distance (certainly not at MFD).
Daan B: remember somebody had contacted Photozone about this
I asked this question and they (he) said that it was either 5 times, or even more, MFD. If one puts the focus limiter to 3 m (>9 feet) everything is fine anyway.
As you can see, I'm guessing in MFD @f/4.0 is so blurry could this due to the DOF being practically nonexistent? @f/5.6 DOF open up to only about 1/5 inch, @f/8.0 DOF is about 1/3 inch, @f/11 DOF increase to 1/2 inche, and @f/16 the DOF is almost 1 inche.
Focus point was on number 4, it appear that 200mm MFD is useful in the field only when I stopped down to atleast f/8.0, but man....it needs f/16 to cover 1 inche of DOF, plus the back focus issue pretty much render it useless on subject in the same plane as the sensor, maybe that's why if the test subject is something like the bank note or newspaper they all came out blurry regardless how small the f stop was. I'm starting to wonder if this has anything to do with the diffraction of my 40D sensor instead of the lens, too bad I don't have a backup body to test with, I'm guessing this is going to be worst in 1.3x crop or fullframe camera, maybe smaller sensor like the 20/30D or xti might get better result?
Daan B and mfurman, your lens exhibit similar behavior as mine right? can you guys or anyone else try similar test on a ruler or something to see if you can get the same result? I used 580EX to aid AF, in manual mode with shutter speed set at the max sync speed 1/250s, spot metering.
I prefer lenses which are sharp at infinity, a bit less sharpness at MFD is something I can live with.
You need to enlarge an image really a lot before a it looks unpleasant only because its not tack sharp, for Portraits anyway.
When I bought that lens, I had hoped that the MFD was closer though. I still prefer the 135L because of its ability to shoot close up
yauyi:
I decided to test MFD @200mm of my lens and made some interesting discovery....
Interesting but how do you explain that the lens is dramatically better at MFD and 135 mm (or even 160 mm)? I took tens of shots at 200 mm and MFD (6 lenses as of today) and did not get a single sharp one. In my mind, the lens is not sharp at 200 mm (and MFD) and your test show it too (not a single line is sharp at f/4.0 or f/5.6)
That's what got me curious to test a wide range of f-stops in MFD @200mm because 135mm did not have this problem, I don't know why and can't explain it, but I wonder how the same lens is going to behave on different camera, like the xti, 20D or 30D. I can only imagine it's going to get even worst on full frame or 1.3x crop camera due to the nature of those sensors(narrower DOF). I do not get this lens to do macro but I do need to use 200mm in very close range for portraiture sometime, I guess I could close it down to f/11~f/16 and focus on the subject's nose so the eyes can be in focus, but man....f/16!
I just got my replacement today, this copy is WAY better than the one I returned. Wide open in MFD @200mm is very sharp and I do not see any haze/halo like the previous copy. I also tested it briefly around the house from 70~135mm in MFD, it's safe to say that I've got a good copy this time and am very satisfied.
This was my first 'L' lens - date code UW0305 - I had high expectations from all the glowing reviews. I have been very disappointed at results past 180mm - with 40D, 30D and XTi and under 10 feet.
Andi Dietrich wrote:
I prefer lenses which are sharp at infinity, a bit less sharpness at MFD is something I can live with.
You need to enlarge an image really a lot before a it looks unpleasant only because its not tack sharp, for Portraits anyway.
When I bought that lens, I had hoped that the MFD was closer though. I still prefer the 135L because of its ability to shoot close up
135 f/2 is not sharp at infinity at f/2 but then it's ok from f/2.8
for closeups it's great tho'
but then it's ok from f/2.8
