Decentered 24-70mm GM ii

clarnibass · Jul 30, 2022

coastlight said:
I also assumed with GM lenses QC would be very good and sample variation low. It's my first GM lens.

Lenses vary in QC and allowed tolerances. Even in some imaginary world where there was just one bad GM lens slipping out of their factory, there would still be the unlucky person who ends up with it.

That said, your tests here and the overall impression from reading other peoples' experience suggests there is some decentering and it's pretty common, even more than a little bit. The only thing that would probably cause them to improve it is if people returned them in droves for that reason... which is unlikely I guess.

4Photos · Jul 31, 2022

FastJimmy said:
4Photos said:

This one looks REALLY good. The f2.8 is outstanding! What changed between this and your earlier efforts? The very mild softness of the upper right in this example would not worry me (and yes, I am a "pixel-peeper" and have returned lenses for decentering issues).

If you find a better copy of the 24-70 GM let us know, I would be very interested in your results. Thanks!

Click to expand...

Thanks. I think it does as well. I plan on keeping it and if I can get another for testing I will. And I will report back to the forum in a new thread since it appears I may have hijacked this one a bit.

Jim

Oh funny, I didn't realize you were the OP

- so I was surprised how great the corners look in this test. It shows what this GM is capable of. Better than most primes!

(And it confirms that OP probably got a bad copy.... )

coastlight · Jul 31, 2022

It's not clear from your post, but I assume this is also the 24-70 GM ii at 70mm? If so, then this is absolutely interesting for me to compare with mine. Your top right corner appears much better, I'd say.

coastlight · Jul 31, 2022

Actually a bit surprised by your results, your upper right looks almost worse at f5.6 than at f2.8 which makes little sense to me...

At f5.6 I think my lens is at least on par with yours.

coastlight · Jul 31, 2022

Had time to do some more testing of my 24-70 GM ii so did new tests at 24mm (as requested by JohnNEX).

Tests at f2.8, f5.6 and f11.

100% crops:

ISO100 1/3200s f2.8 24mm

ISO100 1/1250s f5.6 24mm

ISO100 1/250s f11 24mm

To me this looks good. Right side marginally better than left at f2.8 but there's almost nothing in it. Surprised to see what I can only assume is diffraction creeping in at f11, making f5.6 marginally superior actually, to my eyes.

fboss · Jul 31, 2022

coastlight said:
Had time to do some more testing of my 24-70 GM ii so did new tests at 24mm (as requested by JohnNEX).

Tests at f2.8, f5.6 and f11.

100% crops:

To me this looks good. Right side marginally better than left at f2.8 but there's almost nothing in it. Surprised to see what I can only assume is diffraction creeping in at f11, making f5.6 marginally superior actually, to my eyes.

At f2/8, the top left is slightly behind but it's more similar than different. Looks good overall.
Since you have some result variation at 70mm, I wonder how you test this. Are you using manual or AF, EFCS, is the camera on a tripod or hand-held (if hand-held, do you have a fast enough shutter speed ? image stabilization On ? etc ). Your 2nd results at 70mm look good too, the degradation on the top right is really minor.

coastlight · Jul 31, 2022

fboss said:
At f2/8, the top left is slightly behind but it's more similar than different. Looks good overall.
Since you have some result variation at 70mm, I wonder how you test this. Are you using manual or AF, EFCS, is the camera on a tripod or hand-held (if hand-held, do you have a fast enough shutter speed ? image stabilization On ? etc ). Your 2nd results at 70mm look good too, the degradation on the top right is really minor.

I guess you have to read the thread carefully now to get the whole picture. First, another user (FastJimmy) also posted his results (presumably also from a 24-70 GM ii) so do not confuse that with mine (don't mind it by the way as I could use the comparison). As to your question, the answer is also provided earlier in the thread. There's a link early in the thread to instructions at phillipreeve.net on how to perform a decentering test.

By the way, I'm finding that I need to be extremely careful when performing the test. Shutter shock is a thing, and it's unclear whether stabilization could induce blur on a tripod. On some tests I had better results at f2.8 than at f5.6 and after some confusion realized the slower shutter speed at f5.6 must have made the camera prone to shutter shock.

So tripod, electronic shutter and a 2 second self timer is how I'm doing it now. And the test probably needs to be done multiple times at the same settings to be absolutely sure. It's looking more and more like the only problem with the lens is at 70mm upper right corner, and any other "issues" I've seen was produced by other factors during testing.

Tristimulus · Jul 31, 2022

coastlight said:
fboss said:

At f2/8, the top left is slightly behind but it's more similar than different. Looks good overall.
Since you have some result variation at 70mm, I wonder how you test this. Are you using manual or AF, EFCS, is the camera on a tripod or hand-held (if hand-held, do you have a fast enough shutter speed ? image stabilization On ? etc ). Your 2nd results at 70mm look good too, the degradation on the top right is really minor.

Click to expand...

I guess you have to read the thread carefully now to get the whole picture. First, another user (FastJimmy) also posted his results (presumably also from a 24-70 GM ii) so do not confuse that with mine (don't mind it by the way as I could use the comparison). As to your question, the answer is also provided earlier in the thread. There's a link early in the thread to instructions at phillipreeve.net on how to perform a decentering test.

By the way, I'm finding that I need to be extremely careful when performing the test. Shutter shock is a thing, and it's unclear whether stabilization could induce blur on a tripod.

Actually it is not unclear (read the user manual).

Turn off image stabilisation when using a tripod.

Turn on image stabilsation when using a monopod or shooting hand held.

coastlight said:
On some tests I had better results at f2.8 than at f5.6 and after some confusion realized the slower shutter speed at f5.6 must have made the camera prone to shutter shock.

So tripod, electronic shutter and a 2 second self timer is how I'm doing it now.

Some flimsy tripods do not calm down within 2 sec. Guess yours is a decent one.

coastlight said:
And the test probably needs to be done multiple times at the same settings to be absolutely sure. It's looking more and more like the only problem with the lens is at 70mm upper right corner, and any other "issues" I've seen was produced by other factors during testing.

Yup - proper testing is precision work. ;-)

fboss · Jul 31, 2022

coastlight said:
I guess you have to read the thread carefully now to get the whole picture. First, another user (FastJimmy) also posted his results (presumably also from a 24-70 GM ii) so do not confuse that with mine (don't mind it by the way as I could use the comparison). As to your question, the answer is also provided earlier in the thread. There's a link early in the thread to instructions at phillipreeve.net on how to perform a decentering test.

So tripod, electronic shutter and a 2 second self timer is how I'm doing it now. And the test probably needs to be done multiple times at the same settings to be absolutely sure. It's looking more and more like the only problem with the lens is at 70mm upper right corner, and any other "issues" I've seen was produced by other factors during testing.

Ah yes sorry, I read the thread a bit too quickly and didn’t see the 2nd results were from Fastjimmy.
Your post is useful to me because I’m considering the 24-70 GM II. If i buy it I will definitely test the corners at different focal lengths while I can still return it. But will resist doing the tests on my current GM lenses (35 and 70-200 II), as I have been very happy so far with the pictures and I don’t want to ruin it

FastJimmy · Jul 31, 2022

coastlight said:
fboss said:

At f2/8, the top left is slightly behind but it's more similar than different. Looks good overall.
Since you have some result variation at 70mm, I wonder how you test this. Are you using manual or AF, EFCS, is the camera on a tripod or hand-held (if hand-held, do you have a fast enough shutter speed ? image stabilization On ? etc ). Your 2nd results at 70mm look good too, the degradation on the top right is really minor.

Click to expand...

I guess you have to read the thread carefully now to get the whole picture. First, another user (FastJimmy) also posted his results (presumably also from a 24-70 GM ii) so do not confuse that with mine (don't mind it by the way as I could use the comparison). As to your question, the answer is also provided earlier in the thread. There's a link early in the thread to instructions at phillipreeve.net on how to perform a decentering test.

By the way, I'm finding that I need to be extremely careful when performing the test. Shutter shock is a thing, and it's unclear whether stabilization could induce blur on a tripod. On some tests I had better results at f2.8 than at f5.6 and after some confusion realized the slower shutter speed at f5.6 must have made the camera prone to shutter shock.

So tripod, electronic shutter and a 2 second self timer is how I'm doing it now. And the test probably needs to be done multiple times at the same settings to be absolutely sure. It's looking more and more like the only problem with the lens is at 70mm upper right corner, and any other "issues" I've seen was produced by other factors during testing.

Thanks for doing the test. I don't know if I had made it clear..... I used a very stable tripod, manual focus, 2-second timer, stabilization off, and electronic shutter. Especially for high MP cams to get the best possible image for any situation, technique obviously plays a major role.

And yes, the images were from the 24-70GM ii.

Jim

FastJimmy · Jul 31, 2022

4Photos said:
FastJimmy said:

4Photos said:

This one looks REALLY good. The f2.8 is outstanding! What changed between this and your earlier efforts? The very mild softness of the upper right in this example would not worry me (and yes, I am a "pixel-peeper" and have returned lenses for decentering issues).

If you find a better copy of the 24-70 GM let us know, I would be very interested in your results. Thanks!

Click to expand...

Thanks. I think it does as well. I plan on keeping it and if I can get another for testing I will. And I will report back to the forum in a new thread since it appears I may have hijacked this one a bit.

Jim

Click to expand...

Oh funny, I didn't realize you were the OP - so I was surprised how great the corners look in this test. It shows what this GM is capable of. Better than most primes!
(And it confirms that OP probably got a bad copy.... )

Nope. Not the OP, but a curious member.

Jim

FastJimmy · Jul 31, 2022

coastlight said:
Actually a bit surprised by your results, your upper right looks almost worse at f5.6 than at f2.8 which makes little sense to me...

At f5.6 I think my lens is at least on par with yours.

That is correct. I will need to retest a few times to see whether or not it is user error.

Jim

coastlight · Jul 31, 2022

I'm finding repeatedly that at 50mm my 24-70 GM ii has sharp corners at f2.8 but at f5.6 the right side is somewhat soft. At f11 everything's sharp again. I tried three times with the same result, using tripod, silent shutter, self timer and stabilization off. I'm thinking it's still user error but I can't possibly imagine how. Is it actually possible that a lens could be stronger at f2.8 than f5.6? To me it makes absolutely no sense because all the stopping down does is increase depth of field.

TheOwl360 · Jul 31, 2022

coastlight said:
I'm finding repeatedly that at 50mm my 24-70 GM ii has sharp corners at f2.8 but at f5.6 the right side is somewhat soft. At f11 everything's sharp again. I tried three times with the same result, using tripod, silent shutter, self timer and stabilization off. I'm thinking it's still user error but I can't possibly imagine how. Is it actually possible that a lens could be stronger at f2.8 than f5.6? To me it makes absolutely no sense because all the stopping down does is increase depth of field.

At this point, that lens is taking up so much of your bandwidth and opportunity cost that you should sell it as soon as possible. Based on your expectations of lenses, you may consider buying several copies at one time and returning those that fail your tests.

4Photos · Aug 1, 2022

TheOwl360 said:
coastlight said:

I'm finding repeatedly that at 50mm my 24-70 GM ii has sharp corners at f2.8 but at f5.6 the right side is somewhat soft. At f11 everything's sharp again. I tried three times with the same result, using tripod, silent shutter, self timer and stabilization off. I'm thinking it's still user error but I can't possibly imagine how. Is it actually possible that a lens could be stronger at f2.8 than f5.6? To me it makes absolutely no sense because all the stopping down does is increase depth of field.

Click to expand...

At this point, that lens is taking up so much of your bandwidth and opportunity cost that you should sell it as soon as possible. Based on your expectations of lenses, you may consider buying several copies at one time and returning those that fail your tests.

I completely disagree with your assessment. Some people like testing, others abhor it. That's fine. Ansel Adams was one who did countless tests in order to refine his darkroom work. It didn't hurt.

Here we have an interesting optical question. How can a lens be sharper at f2.8 than at f5.6 - assuming the tests have been done carefully (and it appears that way), it is very curious.

I wish someone with an interest in optics would comment, and people who have a philosophical dislike of testing in general would just stay silent

101Colors · Aug 1, 2022

Here's a thread of interest.. Which should add fomentation and further data for the OP and potential buyers - particularly those perfectionists.

'Sample Variation of the Sony 24-70 GM ii"

It's a sad post to make. :-D

coastlight said:
It's a sad post to make,

--
//////// Sometimes it's better to be kind than to be right. \\\\\\\

TheOwl360 · Aug 1, 2022

4Photos said:
TheOwl360 said:

coastlight said:

I'm finding repeatedly that at 50mm my 24-70 GM ii has sharp corners at f2.8 but at f5.6 the right side is somewhat soft. At f11 everything's sharp again. I tried three times with the same result, using tripod, silent shutter, self timer and stabilization off. I'm thinking it's still user error but I can't possibly imagine how. Is it actually possible that a lens could be stronger at f2.8 than f5.6? To me it makes absolutely no sense because all the stopping down does is increase depth of field.

Click to expand...

At this point, that lens is taking up so much of your bandwidth and opportunity cost that you should sell it as soon as possible. Based on your expectations of lenses, you may consider buying several copies at one time and returning those that fail your tests.

Click to expand...

I completely disagree with your assessment. Some people like testing, others abhor it. That's fine. Ansel Adams was one who did countless tests in order to refine his darkroom work. It didn't hurt.

Here we have an interesting optical question. How can a lens be sharper at f2.8 than at f5.6 - assuming the tests have been done carefully (and it appears that way), it is very curious.

I wish someone with an interest in optics would comment, and people who have a philosophical dislike of testing in general would just stay silent

Checking zoom lenses for decentering (they’re all decentered) ad nauseum and learning how to develop prints in a darkroom are two totally different pursuits.

Staying silent while reading this thread or ignoring this thread because it’s substantively trifling are two different things as well. I’ll be doing the latter.

Have fun testing.

FastJimmy · Aug 1, 2022

coastlight said:
I'm finding repeatedly that at 50mm my 24-70 GM ii has sharp corners at f2.8 but at f5.6 the right side is somewhat soft. At f11 everything's sharp again. I tried three times with the same result, using tripod, silent shutter, self timer and stabilization off. I'm thinking it's still user error but I can't possibly imagine how. Is it actually possible that a lens could be stronger at f2.8 than f5.6? To me it makes absolutely no sense because all the stopping down does is increase depth of field.

That's a head-scratcher because, as you saw, my lens exhibited the same behaviour. I would not have ever discovered it had I not experimented based upon this thread. I did my testing with an A1. Which cam are you using? I'm going to do some testing on a couple of other lenses and perhaps on my A7r4 just to see what's happening.

There was a comment on the value of testing and the implication that one may be obsessed with testing. I respectfully disagree with that inference because as photographers we need to know the strengths and potential weaknesses of our gear. Granted the softness is near the corner at f/5.6 in our samples but how far towards the center of the frame does that softness "intrude." Group photos, even of small groups, in an event could suffer. No more needs to be said.....other than this thread has caused me to now take some time to further scrutinize the accuracy of my gear. Sadly, as I inferred, I have a ton to test.

Jim

FastJimmy · Aug 1, 2022

101Colors said:
Here's a thread of interest.. Which should add fomentation and further data for the OP and potential buyers - particularly those perfectionists.

'Sample Variation of the Sony 24-70 GM ii"

It's a sad post to make. :-D

coastlight said:

It's a sad post to make,

Click to expand...

Just saw this post while posting my other comment. I subscribe to Lloyd's blog and find his evaluations interesting. He is a perfectionist and, as such, I respect his testing. Many times his nits are things not easily seen with my eyes. However, I use his results as input on some of my cam and lens selections. He did trash Sony's perceived QC and hypothesized a 1 in 5 chance (I think) of getting a good copy of the newest GM24-70/2.8. YMMV.

Jim

4Photos · Aug 2, 2022

TheOwl360 said:
4Photos said:

TheOwl360 said:

coastlight said:

I'm finding repeatedly that at 50mm my 24-70 GM ii has sharp corners at f2.8 but at f5.6 the right side is somewhat soft. At f11 everything's sharp again. I tried three times with the same result, using tripod, silent shutter, self timer and stabilization off. I'm thinking it's still user error but I can't possibly imagine how. Is it actually possible that a lens could be stronger at f2.8 than f5.6? To me it makes absolutely no sense because all the stopping down does is increase depth of field.

Click to expand...

At this point, that lens is taking up so much of your bandwidth and opportunity cost that you should sell it as soon as possible. Based on your expectations of lenses, you may consider buying several copies at one time and returning those that fail your tests.

Click to expand...

I completely disagree with your assessment. Some people like testing, others abhor it. That's fine. Ansel Adams was one who did countless tests in order to refine his darkroom work. It didn't hurt.

Here we have an interesting optical question. How can a lens be sharper at f2.8 than at f5.6 - assuming the tests have been done carefully (and it appears that way), it is very curious.

I wish someone with an interest in optics would comment, and people who have a philosophical dislike of testing in general would just stay silent

Click to expand...

Checking zoom lenses for decentering (they’re all decentered) ad nauseum and learning how to develop prints in a darkroom are two totally different pursuits.

Different pursuits, similar mindset. Chasing perfection (knowing that it is an elusive goal).

All I can say is that I test all my lenses for decentering issues as soon as I receive them, and I have returned one or two and received better copies. Makes me happy

Decentered 24-70mm GM ii

Veteran Member

Veteran Member

Member

Member

Member

Forum Enthusiast

Member

Veteran Member

Forum Enthusiast

Forum Enthusiast

Forum Enthusiast

Forum Enthusiast

Member

Forum Enthusiast

Veteran Member

Leading Member

Forum Enthusiast

Forum Enthusiast

Forum Enthusiast

Veteran Member

Similar threads

Keyboard shortcuts