What are your experiences with quality control issues and the major e-mount manufacturers? I had so many troubles over the last year.
My History:
Loxia 21: good 1. Copy
Loxia 25: good 2. Copy (sold it after a while), 1.Copy was decentered
Loxia 25 repurchase: tried 2. Copies, both were decentered, ordered a 3. one and hope that one is better
Loxia 50: good 1. Copy
Loxia 85: good 1. Copy but loose focus ring, 2. Copy was decentered (very bad, even visibly at 1:2), good 3. Copy
CV Nokton 58mm: good 3. Copy, 1. Copy it was not possible to focus to infinity
CV 65 Apo: good 1. Copy
Sony Zeiss 50 ZA: 1. Copy was affected by the fooging problem, 2. Copy was decentered, 3. Copy was good
Sony Zeiss 35 ZA: gave up after 4 bad Copies
Sony 16-35 GM: good at 16-20, at 35mm it was completly off
Just bad luck or is this standard with a short flange distance, 42MP and (maybe) cheaper manufacturing processes and quality control?
Most lenses I test are decentered to some degree, and so I try and only buy lenses I've been able to test beforehand. Took me three copies of the Loxia 21 to get a gem, five copies of the CV15, never did test a good 12-24; my 16-35GM was good 16-20 like yours but terrible at 35, etc.
But I think the vast majority of people don't perform centering tests, or even know what decentering is, and so the issue flies under the radar outside of quite passionate communities like this one.
Aztatlan wrote:
But I think the vast majority of people don't perform centering tests, or even know what decentering is, and so the issue flies under the radar outside of quite passionate communities like this one.
Exactly.
I wish we didn't have to deal with all these decentered lenses, because it really takes away the fun of buying a new lens. It's rare to get a good one at first try.
To name a few of my decentered lenses since I started using Sony:
- 21 Loxia: 2 copies decentered, third one was fine.
- 85 Loxia: 3 decentered copies, I gave up.
- FE 55: 2 copies decentered, third one was fine.
- FE 24-105: 2 badly decentered copies, gave up.
- FE 24-70 GM: 2 decentered copies, gave up.
- FE 35/2.8: 1 decentered copy
- FE 85/1.8: first copy decentered, second was fine
- RX1: first copy decentered, second was fine.
- RX1RII: first copy decentered, second was fine.
... and so on.
The strange thing is that I sometimes got fairly lucky with lenses that are considered to have very bad sample variation.
The Voigtlander 15mm, FE 35/1.4 and FE 16-35 GM were acceptable at first try.
Sony FE 16-35 f4 OSS: 1 copy, fine.
Sony FE 50mm Macro: 1 copy, significantly decentered.
Voigtlander 65mm Macro: 1 copy, fine.
Sigma FE 50mm ART: 1 copy, fine (I'll be able to update this soon, I have a second one waiting in the studio for testing).
Sigma FE 70mm ART Macro: 1 copy, fine.
It's really annoying having to worry about these issues without having Amazon, etc for immediate no quibble picked up returns and quick shipping of new copies. I have a 2nd Sigma 70mm Art Macro on order, I've been quoted weeks for it to arrive in country by the importer. If it's problematic I'm screwed. I wanted to buy a Samyang 24mm f2.8, store told me to forget easy returns and replacement with the importer.
I wish we didn't have to deal with all these decentered lenses, because it really takes away the fun of buying a new lens. It's rare to get a good one at first try.
To name a few of my decentered lenses since I started using Sony:
- 21 Loxia: 2 copies decentered, third one was fine.
- 85 Loxia: 3 decentered copies, I gave up.
- FE 55: 2 copies decentered, third one was fine.
- FE 24-105: 2 badly decentered copies, gave up.
- FE 24-70 GM: 2 decentered copies, gave up.
- FE 35/2.8: 1 decentered copy
- FE 85/1.8: first copy decentered, second was fine
- RX1: first copy decentered, second was fine.
- RX1RII: first copy decentered, second was fine.
... and so on.
The strange thing is that I sometimes got fairly lucky with lenses that are considered to have very bad sample variation.
The Voigtlander 15mm, FE 35/1.4 and FE 16-35 GM were acceptable at first try. ...Show more →
I never thought of having to worry about the RX series though of course it makes sense. Yikes.
Can people from other mounts chime in that this is not an issue with other mounts? I'm curious how common this is. I'm trying my best to start moving away from pixel peeping. There are some rare situations a decentered lens affects my shots, it's not often enough that I want to go through the hassle of purchasing the same lens over and over though.
I’ve had numerous issues with my Nikon lenses. None with my Canon and only one issue with my Sony lenses. My 16-35 was not great at 35. I returned it. All my other Sony lenses seem pretty good. I’m not shooting a lot of landscapes now so decentering is less a worry for me now. Sony lenses owned. 70-200 f2.8, 85 f1.4, sigma 35, Sony 50 f1.8, sony 24 f1.4, and 15 f2
I dont test,haven't noticed. I've bought many Sony lenses. If the area I focus on is sharp,I'm good to go...and I shoot weddings,not pretty skies or birds eating other birds
I don't pixel peep any more and I barely tested my lenses; if then just for decentering and only on wide angle lenses of wider than 28mm. The only FE lens out of some 25 Sony, Zeiss and CV or so that encountered being subpar was the FE 16-35mm 4 years ago when it first came out. Sony replaced that lens and the replacement was excellent. Between you and me and the fence post, I am not worried about that aspect and I enjoy photography more so ever since. However, I realize that folks take on this varies widely and to each his/her own.
I guess I have been lucky with the Loxia's as my 21, 35, 50, and 85 were all sharp and very well centered. I tested them all at infinity and near focus.
All of my lenses are fine, only one of my two 35/1.4ZA was strongly decentered. The 2nd copy was fine. Overall the 21 Loxia, 24GM, 25Batis, 35/1.4 (2x), 50/1.4 (2x), 85Batis, 85GM, 135 Batis. At work we use the 28-75 Tamron and we had a 55/1.8 in the past. Both are fine.
Generally I think some here are obsessed with getting a perfect copy. In a print, I bet most won't notice "issues" at all. That is not to say that decentering couldn't be a problem. Sometimes it is. However I can't believe that the good copies are in the minority, esp. when we deal with Zeiss lenses.
R Cicala: "Another point that I’ll expand on in a future post is for those who spend a lot of time and effort in ‘getting a great copy.’ Some lenses have such variation that you probably will have to go through at least dozens of copies if your definition of ‘great copy’ is ‘near perfect copy.’"
What are your experiences with quality control issues and the major e-mount manufacturers? I had so many troubles over the last year.
My History:
Loxia 21: good 1. Copy
Loxia 25: good 2. Copy (sold it after a while), 1.Copy was decentered
Loxia 25 repurchase: tried 2. Copies, both were decentered, ordered a 3. one and hope that one is better
Loxia 50: good 1. Copy
Loxia 85: good 1. Copy but loose focus ring, 2. Copy was decentered (very bad, even visibly at 1:2), good 3. Copy
CV Nokton 58mm: good 3. Copy, 1. Copy it was not possible to focus to infinity
CV 65 Apo: good 1. Copy
Sony Zeiss 50 ZA: 1. Copy was affected by the fooging problem, 2. Copy was decentered, 3. Copy was good
Sony Zeiss 35 ZA: gave up after 4 bad Copies
Sony 16-35 GM: good at 16-20, at 35mm it was completly off
Just bad luck or is this standard with a short flange distance, 42MP and (maybe) cheaper manufacturing processes and quality control?
With cv Norton 58 it was likely a cumulative tolerance between adapter and lens issue.
As for the rest, it depends on what your standards are.
If what you want is four identical corners then I’d say this is par for the course.
Whether it’s worth your while shooting for that is up to you.
But I suspect most manufacturers don’t even expect the majority of their lenses to be like that wide open. It’s within tolerance if there is a little variation.
My personal tolerance is no variation between left edge and right edge, and no corner much worse than others (subtle difference wide open on a fast lens is acceptable to me)
It’s also worth bearing in mind that sometimes lenses are perfectly centred, but are poor copies centrally. So if you want perfection you need to know how the particular lens ought be in the centre!
I am not too picky, but when I pay 1000$€+ I expect 4 nearly identical corners (at 1:1 magnification). My worst lens was the 2. copy of the Lox 85, one corner was crazy off (even when viewed in lightroom at no magnification it was clearly visible) and the center perfectly sharp. When a lens this bad is leaving the factory something went really wrong at the Zeiss quality control. The Lox 25 was not that bad, but when the right side has noticeably corner smearing something is also wrong.
Our testing method might not be perfect either, the selection of the focusing plane and filed curvature may play a role in throwing the results off. While I am not claiming that are no bad apples as clearly they are but it may be not as many as one thinks. Like I mentioned above, my testing no, evaluation of lenses mostly pertain to the way those lenses are intended to be used, namely to create images, and real-life images at that and not test images. That's my honest opinion and I am sticking to that. Only you have to decide how to handle the situation.
I am not addressing this to anybody, of course....
I think we all appreciate the effort Fred, Guy as well as David, Phillip, Bastian and Co. put into those detailed and unbiased lens reviews. Same goes for Roger and his bench tests.
Honestly, I'm not sure whether I had the patience for that type of rigorous testing with all the different aperture series and in case of zooms, different FL's. I'm with Joshua on this one. For me, too much testing seriously takes the fun out of photography. Also, I don't have the time to spend hours checking whether a lens is perfect, good, bad or ugly. I do a quick infinity test after the lens arrives to make sure there is nothing seriously wrong, like one side/corner looking horrible.
I have had only one or two issues with the two dozen or so Sony lenses I've owned across a-mount and e-mount. I had to sent back two copies of the 24-70 GM until I got one that was good on both the left and right side above 60mm. Out of the two copies of the 35 1.4 that I'd owned, the first one had a slightly weaker lower-right corner, the second one was well centered and did not come with any observable tilt.
Between Loxia/Batis and zm, I've owned half a dozen Zeiss lenses, not a single issue.
My first copy of the Samyang 14 2.8 was only OK-ish, but my first copy of the 14 2.4 is fine.
I guess we all can agree that playing the lottery is a PITA, but those who believe that every lens should be tested, might want to read Roger Cicala's excellent take on the matter of quality assurance.
Chris_88 wrote:
I think we all appreciate the effort Fred, Guy as well as David, Phillip, Bastian and Co. put into those detailed and unbiased lens reviews. Same goes for Roger and his bench tests.
Honestly, I'm not sure whether I had the patience for that type of rigorous testing with all the different aperture series and in case of zooms, different FL's. I'm with Joshua on this one. For me, too much testing seriously takes the fun out of photography. Also, I don't have the time to spend hours checking whether a lens is perfect, good, bad or ugly. I do a quick infinity test after the lens arrives to make sure there is nothing seriously wrong, like one side/corner looking horrible.
I have had only one or two issues with the two dozen or so Sony lenses I've owned across a-mount and e-mount. I had to sent back two copies of the 24-70 GM until I got one that was good on both the left and right side above 60mm. Out of the two copies of the 35 1.4 that I'd owned, the first one had a slightly weaker lower-right corner, the second one was well centered and did not come with any observable tilt.
Between Loxia/Batis and zm, I've owned half a dozen Zeiss lenses, not a single issue.
My first copy of the Samyang 14 2.8 was only OK-ish, but my first copy of the 14 2.4 is fine.
I guess we all can agree that playing the lottery is a PITA, but those who believe that every lens should be tested, might want to read Roger Cicala's excellent take on the matter of quality assurance. ...Show more →
I absolutely agree. Shooting for absolute perfection is a pointless PITA unless you kind of enjoy all the work.
It's important though to work out what matters to you. No-one wants a lens that's overall soft (and such exist). And a corner that is *massively* worse wide open is likely going to be a fair bit worse stopped down for landscape, so that might matter.
Someone whose main concern is Astro might care more about wide open corner performance than someone who uses the lens for people wide open, and landscape stopped down. (Though I have a strong hunch that there are more people who care that their lens is good for Astro than actually do much Astro)
So for the average user, an infinity test looking at edges and midfield as well as corners makes sense. reject a lens with different midfield left and right - that may well mess you up. if the extreme edges are marginally different, it's a judgement call. You don't want the corners to be HUGELY different, but rejecting lenses with subtle differences is a recipe for the misery of endless lens replacement. Those are pretty much my personal criteria too.
The only defective lenses I've had – out of more than one hundred – were two copies of the RX1 Sonnar with tilted image plane (not necessarily the lens of course). I got frustrated and never tried a third.
stripedrex wrote:
Can people from other mounts chime in that this is not an issue with other mounts? I'm curious how common this is. I'm trying my best to start moving away from pixel peeping. There are some rare situations a decentered lens affects my shots, it's not often enough that I want to go through the hassle of purchasing the same lens over and over though.
This whole talk appeared more prominently lately with the introduction of full frame mirrorless and high megapixel sensor. Testing can be done much easier with magnification on a mirrorless so I don't think Canikon people have really been doing that. I do believe all brands have this issue, even Leica and their Cron 50 APO lens. As IQ demand increases, precision in element placement is even more critical. QC issues like decentering can be easily seen with sharp modern lenses than the aberration-ridden vintage lenses
Depends on the degree of decentering, it might not matter that much at all (hidden with shallow DOF at WO and close distance or masked by deep DOF stopping down to f/8). I have had Samyang lenses that are so badly decentered that stopping down still produces noticeably softer side. But you should test your lenses regardless for future sales.
It does of course depend on usage. How many people shoot landscape at f1.4? Or shoot portraiture that close to the corners?
However if you are or if the problems are apparent even stopped down, this might be a real problem. I run a large reproduction studio photographing ancient manuscripts, documents, etc in conjunction with the national library. We're shooting close to the edges to maximise resolution and we cannot have the handwriting at the end of a line softer than in the middle.
The Sony 50mm Macro we tried was decentered even at f8, our shooting aperture and to the extent that there was significant longitudinal CA on the right side as it slipped just out of the focal plane, a disaster with text on a page as you can imagine. It was a sea of green outlines.
On the other hand when I tested my personal Sony 16-35mm f4, I took some shots outdoors and had a look, seemed fine side to side throughout the aperture range, didn't see any problems and moved on to real world shooting. It wasn't the rigorous testing we do in the studio with new lenses on copy stands with test charts, etc. There isn't really the need, I'm not shooting for the corners wide open. If I need the corners I'll be at f5.6 or f8 because of DOF anyway.
I think it boils down to the problem being relevant is based on specific personal needs however that doesn't mean that there isn't a problem that might bother someone else who has different criteria or simply has the time, patience and ability to keep testing until they get a perfect lens. The more information we have the better to decide our own criteria.
