Home · Register · Join Upload & Sell

Moderated by: Fred Miranda

New fredmiranda.com Mobile Site

New Feature: SMS Notification alert

New Feature: Buy & Sell Watchlist

FM Forums | Sony Forum | Join Upload & Sell

2 end

Archive 2016 · Reasonable x for "best of x"?

stevei
Offline
• • •
Upload & Sell: Off

p.1 #1 · p.1 #1 · Reasonable x for "best of x"?

I've read some people say over many years that it's reasonable to buy 3 copies of a lens and return 2. It has always felt unreasonable to me to do that, but the last two lenses I've bought I have tried two copies and have quite frankly been shocked at how big the differences have been between the two copies. I'm definitely much happier with the lenses I have, knowing how much worse the other copies were. But is it even reasonable to always expect a copy that is in the top 50% of shipped lenses?

If you do routinely choose from multiple copies, how do you go about it? Buy one, see what it's like, and buy another if you're not happy, or buy multiple copies at once and compare them?

For both my recent lenses, I just bought one, then a second when I wasn't happy with the first. For one, I held onto the first till I got the second, as I thought the second could easily be worse, and it was. For the other lens, I returned the first lens and used the refund to buy the second, as the first was clearly unacceptable to me, even if the second turned out to be even worse. But I'm now thinking it might be better overall for both me and the retailer to simply order multiple copies at once. They only have to pay for postage once, and unless they're all shockingly bad and you end up returning them all, they're at least guaranteed the profit margin for one lens and know that you're not one of those people who buys and returns lenses as a free rental service.

Has anyone ever had any feedback from retailers as to what their preference is? I know that for items such as clothing where sizing can be an issue, they seem quite happy for you to buy a range of sizes and return all except the one that fits best. I want to get a CV 15 as I'm finding the Batis 18 a bit long for some situations, but based on posts here when it first shipped I'm thinking it really might be necessary to try 3 copies to get a decent one. I'm just struggling to get over the feeling that it would be a bit unreasonable to put 3 in my online shopping cart and buy them.

Nov 18, 2016 at 04:30 PM

retrofocus
Offline
• • • • • •
Upload & Sell: Off

p.1 #2 · p.1 #2 · Reasonable x for "best of x"?

So far I only ever bought one single copy of any lens which I own or owned. I never encountered a dissatisfactory outlier both for brand new and for used lenses. I believe the whole "my lens is not sharp enough/decentered etc" is overhyped in the internet and most lenses simply are within specification. I doubt that there are visible differences of photos taken between several copies of the same lens - the slight discrepencies might only be seen with lens-specific analysis/refraction tools.

Nov 18, 2016 at 05:06 PM

notherenow
Offline
• • •
Upload & Sell: Off

p.1 #3 · p.1 #3 · Reasonable x for "best of x"?

I clearly have missed something as I have always just got one copy of a lens and just lived with it.

Those I didn't like so much I would use less and maybe sell or trade for something else but not often. I have disposed of lots of gear at times but not lenses for image quality (or lack there of other than destroying some very cheap "coke bottles" as experiments from time to time).

I guess all the lenses I have purchased have been better than I needed with whatever I was using or in most cases have been the lottery of being second hand. IE, how much has their "life" and how they have "lived" been affected ? (been dropped? thrown about in a bag? used excessively? ETC).

I did have two copies of a lens once but never noticed if one was better than the other (one developed stuck blades and it was cheap enough).

I guess with higher resolution cameras like 36mp and above and paying thousands for a brand new lens, I might be more inclined to sweat on any difference but at my level where I got a few top class items but mostly live in the lower reaches of the undergrowth of photography, getting what I can afford that suits, I take what I can.

Nov 18, 2016 at 05:08 PM

uhoh7
Offline
• • • •
Upload & Sell: Off

p.1 #4 · p.1 #4 · Reasonable x for "best of x"?

retrofocus wrote:
So far I only ever bought one single copy of any lens which I own or owned. I never encountered a dissatisfactory outlier both for brand new and for used lenses. I believe the whole "my lens is not sharp enough/decentered etc" is overhyped in the internet and most lenses simply are within specification. I doubt that there are visible differences of photos taken between several copies of the same lens - the slight discrepencies might only be seen with lens-specific analysis/refraction tools.

It's been an issue forever, and plenty of shooters, like kubrick, always tested many lenses and picked one. I'd wager nearly anyone with a big name in the past or now has done. They pretend the equipment is not primary. But most are extremely meticulous.

Lensrentals has documented variations real well. Sony is the worst for full frame. With the rest it's case by case, with some lenses like that nikon 50/1.8 (or is it the canon) which has nearly no variation.

But it's a pain to test carefully, and the vast majority of "normal people" like us don't do it. That's OK.

It's a real issue, but the priority is in the mind of the consumer.

Nov 18, 2016 at 09:11 PM

ggreene
Offline
• • • • •
Upload & Sell: Off

p.1 #5 · p.1 #5 · Reasonable x for "best of x"?

Lensrentals also says that the differences from a bench test are very small between copies and other factors almost always take over first. That said, I think it's time for a much better effort from manufacturers on the QC side of things. Problem is it's only going to further increase the cost of lenses.

Nov 18, 2016 at 09:29 PM

DavidBM
Offline
• • • • •
Upload & Sell: Off

p.1 #6 · p.1 #6 · Reasonable x for "best of x"?

Nov 18, 2016 at 09:31 PM

retrofocus
Offline
• • • • • •
Upload & Sell: Off

p.1 #7 · p.1 #7 · Reasonable x for "best of x"?

uhoh7 wrote:
It's been an issue forever, and plenty of shooters, like kubrick, always tested many lenses and picked one. I'd wager nearly anyone with a big name in the past or now has done. They pretend the equipment is not primary. But most are extremely meticulous.

Lensrentals has documented variations real well. Sony is the worst for full frame. With the rest it's case by case, with some lenses like that nikon 50/1.8 (or is it the canon) which has nearly no variation.

But it's a pain to test carefully, and the vast majority of "normal people" like us don't do...Show more →

As mentioned earlier in other posts, I have no experience with Sony FE lenses (I only borrowed once the 35/2.8 FE lens from a friend of mine for a test comparison, and it performed just fine). I can only tell from a variety of Canon, Voigtlander, Leica, Sigma, Minolta, and Tamron lenses where I have - fortunately enough maybe - never seen an issue. Lensrentals shows the discrepancies between different lens copies - but those differences all rely on sophisticated analytical tools. I doubt you would see a severe difference by looking at magnified photos taken of the same scenery with different lens copies. There will always be a manufacturing discrepancy between camera bodies and also lenses - I doubt also that lens manufacturers put out too sloppy standard deviations of their products in general. This doesn't mean that there are outliers - surely there are, but it will be a very small percentage if the manufacturing process runs correctly. IMO it doesn't justify at all to buy several lens copies, select one and return the others.

Nov 19, 2016 at 09:34 AM

AGeoJO
Offline
• • • • • • •
Upload & Sell: On

p.1 #8 · p.1 #8 · Reasonable x for "best of x"?

Looking at the number of people that complain about their lens performance during the initial test/evaluation, I must say that I consider myself lucky. So far, I only encountered a single bad FE lens in the last 3 years. Knock, knock, knock! Sony replaced that lens and I was happy with the replacement. During my 14-15 years of using Canon gear, I also had a single bad copy.

I used to scrutinize lenses more so during my Canon days but now, I just check for centering issues by flipping the camera upside down and I use it for several real life test shots after that. And that's it. I am happy as a clam and focus more on photography rather than on testing lenses and fret about minute imperfections or about differences that may or may not even show up in real-life images. I fully intend of keeping it my way in the future and that's only my way. I fully realize that some folks are different and no offense to them that test lenses in a way more intense way. To each his/her own, I guess. Cheers!

Nov 19, 2016 at 09:51 AM

charles.K
Offline
• • • • • •
Upload & Sell: Off

p.1 #9 · p.1 #9 · Reasonable x for "best of x"?

I agree the bar has been raised with the ability to view at 100% and see imperfections in lenses which we would have been oblivious to in the past.

All my FE (35/2.8, 35/1.4, 55/1.8), Loxia (21,35 50), Batis (25, 85) and GM (85) lenses I did not have any issues fortunately. Maybe in Australia they are not game to send us below par lenses

I do believe in testing lenses, but I agree with Joshua, no lens will be perfect.

Even Leica with the legendary 50 Lux Asph, one of the sharpest lenses, the variation at f/1.4 was quite amazing. I know from the bench testing in the lab, the lenses would vary dramatically. My lens was probably about in the 70% level, with a glow wide open, but disappeared very quickly when you stopped the lens down. Personally I loved this lens, as it was not clinical when I was using it wide open.

If you are using lenses wide open, the intended use will usually not be for landscapes, so some aberrations maybe even desired

Nov 19, 2016 at 07:38 PM

Chris_88
Offline
• • • •
Upload & Sell: Off

p.1 #10 · p.1 #10 · Reasonable x for "best of x"?

Until I had to go through three copies of the 24-70 GM to get a good one, I haven't had to return any lens in a decade of shooting with DSLR's, and I have used a lot of Sony a-mount and e-mount lenses as well as some Leica, Zeiss, Minolta, Sigma and Tokina glass.

I agree that the internet has a tendency of blowing some of these things out of proportion. After all, people with good copies will just use them and not whine and bitch about their new lens on the net. On the other hand, new sensors for better or worse show lens weaknesses for more readily and this means bad copies are more easily recognizable as such. What's more, the US and many European countries have quite lenient return policies. Here in Japan, you can't just buy three copies and return two.

And btw, we know you're bitter about Sony design choices, Charlie, but Roger on his blog also mentioned things have improved quite significantly; Fred had to return plenty of de-centered Canon 24-70's, and this is before we even get to the oil spill on the D600's sensors, the VR problems of various Nikon lenses or all the issues many (experienced) users are still experiencing with the 7D2 in al servo. The grass, it would seem, isn't always greener in non-Sony land

Nov 19, 2016 at 08:36 PM

stevei
Offline
• • •
Upload & Sell: Off

p.1 #11 · p.1 #11 · Reasonable x for "best of x"?

ggreene wrote:
Lensrentals also says that the differences from a bench test are very small between copies

Small is a very subjective term. If you look at the difference between best and worst in his MTF graphs, it can be bigger than the difference between different models of lens, e.g. the best Sony 50/1.8 can be better than the worst Sony Zeiss 55/1.8.

My 2nd Batis 18 was so poor that all I had to do was zoom to 12.5x while manually focusing in the corner of the frame and it was clearly visible that it was just never coming sharp compared to the 1st copy. Obviously I also took a range of photos with both lenses, including on a tripod with a remote release, and both manual focus and auto focus, but the 2nd copy was simply a dog. The 1st was still a bit disappointing but trying a 2nd copy made me extremely grateful I wasn't stuck with the 2nd one I tried!

Nov 20, 2016 at 08:10 AM

RCicala
Offline
• • • •
Upload & Sell: Off

p.1 #12 · p.1 #12 · Reasonable x for "best of x"?

Well, I'll throw in my $0.02.

First, I think it's important that one man's variance is another man's whatever. Depending on how the lens is used and on what camera what one of us finds affects his photographs in a big way may never be noticeable to another. For example, stopped a bit much variance disappears. If you shoot landscapes, street, portrait, or sports some field of focus tilt makes little or no difference. Shoot large wedding groups or architecture and that becomes a deal breaker.

Second, when we look at lenses we look at two very different things, but those two things tend to get lumped into one in forum discussions. Variance, as we use the term, is the 'white noise' caused my the inevitable small variations in each element and it's assembly. "Outliers" are truly bad lenses, noticeable quickly in a photograph, and usually caused by a single major misalignment in one element or group. An outlier, at least in theory, can be corrected, but variance is something we have to live with.

It would be more useful, really, if I could give you 'outlier' frequency, but we simply don't see enough lenses. If I tell you that 26 of 1,400 copies of lens A were outliers, but 'only' 13 of 900 copies of lens B, I really haven't told you anything statistically significant.

More to the point, variability used to be something you really couldn't detect. Film cameras and 12 mpix cameras were very forgiving. Now we're shooting 34 or 42 mpix and we can actually see it if we look closely.

Nov 20, 2016 at 11:29 AM

uhoh7
Offline
• • • •
Upload & Sell: Off

p.1 #13 · p.1 #13 · Reasonable x for "best of x"?

Excellent discussion

OK.....I'm a little "bitter", but come on, you want to get what is advertised, no?

Nov 20, 2016 at 12:24 PM

retrofocus
Offline
• • • • • •
Upload & Sell: Off

p.1 #14 · p.1 #14 · Reasonable x for "best of x"?

uhoh7 wrote:
What do we think is the fastest, easiest way to test a lens for a normal person?

One way which was mentioned here in FM several times is to shoot a brick wall structure with the camera sensor in the best possible way arranged in parallel to the wall. You would see some decentering effect easily in a final photo even with aperture stops like f/11. In such case, one corner would look more smeary than the one on the opposite side.

Nov 20, 2016 at 02:26 PM

Chris_88
Offline
• • • •
Upload & Sell: Off

p.1 #15 · p.1 #15 · Reasonable x for "best of x"?

uhoh7 wrote:
Excellent discussion

OK.....I'm a little "bitter", but come on, you want to get what is advertised, no?

The crying about bad lenses we have seen or other issues may be a bit painful if you like to "love" your manufacturer, but that noise will also be the impetus for improvement in the future.

What do we think is the fastest, easiest way to test a lens for a normal person?

It's not "painful", it's just annoying, nothing more, nothing less, in particular if it's misleading and unfair criticism, giving the impression that the grass is greener in Canikon land, which in many cases, it is not. The fact, that you don't even own a single Sony lens doesn't make things better.

I prefer to register my complaints directly with Sony, and if they don't follow up, let my wallet do the talking. The Sony guy on the phone wasn't happy when I told him that my second 24-70 GM copy too was unacceptably soft in the corner(s), and that others had complained about de-centering too. However, both lenses went back to Sony for verification and in both cases, they found an issue. Those lenses should at least appear as two points in their databases. Alas, my third copy seems to be decent. Needless to say, here too Sony is not alone, as there are apparently quite a few decentered Canon 24-70 ii's out there as well, as Fred and others had to find out.

I would think that such direct complaints which lead to them actually getting faulty lenses back, is a more effective way to register my complaints than crying wolf on the net.

As for testing, I don't believe in test charts. I just shoot a standard landscape scene at various apertures and then compare the results to a lens whose performance I'm familiar with. If I can't make out flaws in such a "test" that closely resembles real world usage, then I'm most likely not bothered by it.

Nov 20, 2016 at 03:44 PM

notherenow
Offline
• • •
Upload & Sell: Off

p.1 #16 · p.1 #16 · Reasonable x for "best of x"?

I think part of why it might SEEM that Sony is (was) a bit worse is that FF E mount STARTED with a 36mp camera as "flagship" while Canon and Nikon have had several generations of cameras to get there with many lenses used on many lower pixel count cameras.

As I said earlier, I have not had any noticeable issue but I have not really looked for any issues like that and have used 24mp and 12mp FF E mount cameras and often buying second hand, I have been more concerned with peering endlessly for dust (or more importantly for me, fungus).

I don't think Sony is any worse than others and indeed with all the attention that made it appear to be, they may well now be better because they may just have to be regardless due to the perception.

Ideally, they should test all lenses but that would add to cost. Maybe they could test a few and certify the better ones for the 36/42 and future higher count cameras for a small premium?

EDIT and if Sony didn't want to do it, there is a market for Roger and OLAF. You could just use it exclusively to test copies of lenses for pixel peepers with high res cameras (and maybe need a second soon enough and demand will only increase with higher res cameras

Nov 20, 2016 at 04:08 PM

uhoh7
Offline
• • • •
Upload & Sell: Off

p.1 #17 · p.1 #17 · Reasonable x for "best of x"?

Chris_88 wrote:
giving the impression that the grass is greener in Canikon land, which in many cases, it is not. The fact, that you don't even own a single Sony lens doesn't make things better.

Sony never changed because one person called them. They did seem to respond to that petition about the uncompressed RAWs but could not bother to help the older A7 models with their execrable "RAW" files. It's an outright lie to call them RAW, as it was to tout the easy use of other lenses back when they did not make so many, while the design was actually in the other direction. Would have been so easy to just put the D800 thickness cover glass on the sensor and be true to the marketing. Or don't and be honest.

I own 6 Sony lenses. Since 2010 I have bought a number of Sony E mount lenses, but all are APS-C, except the 70400G. I considered FF lenses for my A7r and A7 many times. Few were available at the camera release. It soon became apparent that there were QC issues with some of those. Or I should say some of all of them. Ming had to try 5 55/1.8s before he got one he really liked, and then he loved it. There were many accounts of decentered 35/2.8s.

Those are two of their more consistent ones.

On top of that, the Sony lenses are expensive. So my choice was easy. I am shooting a Sony lens right now I really like, the 35 on the RX1rii. It seems perfectly fine QC wise and outright great with results. I've not hesitated to bring that up in several forums

I have never once bashed anything because it was "Sony". I've said many nice things about them since 2010. When they deserved it. Same with Leica

What Sony lenses are as consistent as the most well made made Canon and Nikon lenses? Not a sarcastic question, I really would like to know.

Nov 20, 2016 at 05:17 PM

Chris_88
Offline
• • • •
Upload & Sell: Off

p.1 #18 · p.1 #18 · Reasonable x for "best of x"?

uhoh7 wrote:
Sony never changed because one person called them. They did seem to respond to that petition about the uncompressed RAWs but could not bother to help the older A7 models with their execrable "RAW" files. It's an outright lie to call them RAW, as it was to tout the easy use of other lenses back when they did not make so many, while the design was actually in the other direction. Would have been so easy to just put the D800 thickness cover glass on the sensor and be true to the marketing. Or don't and be honest.

I own...Show more →

Well, consistency is relative. A few months ago, Roger posted an article detailing sample variation of recent Sony e-mount lenses, which wasn't much different from Canikon's.

You will also find that plenty of a-mount glass is on the same level as the Canikon equivalents. As I said above, I didn't have a bad copy among my 20+ Sony lenses I've owned until the GM.

I'm afraid, we won't be able to agree on this, though. I'm happy with my long Nikon glass, but I don't believe that QC and/or copy variation is a Sony-exclusive issue. There are plenty of examples (some I mentioned above), which you can read up on on the other boards. As Charles and others have mentioned, even Leica isn't free of those.

Nov 20, 2016 at 06:09 PM

Mystik
Offline
• • •
Upload & Sell: Off

p.1 #19 · p.1 #19 · Reasonable x for "best of x"?

I might get flamed for this, but I disagree with the whole notion that you should buy 3 lenses with the intention to return two them. Why? Because retailers are taking a hit when you return gear because they can longer sell it as new.

I understand if you get a lens and is clearly a a bad lens, but very often, if you look hard enough for a problem you will find it. I've had 9 FE lenses since I got my a7r close to 3 years ago, and I suspect that many of the copies I've owned would not pass the scrutiny that many others put their lenses too....but I've produced thousands of photos for myself and for paid gigs not once have I looked at an image and thought it was ruined by decentering or anything of that nature.

Nov 20, 2016 at 08:31 PM

seanj
Offline
• •
Upload & Sell: Off

p.1 #20 · p.1 #20 · Reasonable x for "best of x"?

I have returned 2 - 3 lenses over the years. I had a Nikon D70, Sony NEX 5N, Nikon D600, Canon 6D, and now a Sony A7RII, so I went through a lot of lenses as I shifted from system to system (I am not a serial returning by any stretch). I recently bought the Rokinon 14/2.8 AF for Emount. I might be going back, but I have only done regular shooting and I need to test it more rigorously before I draw any conclusions. I've been struggling with an ultra wide angle lens choice.

I rented the Voigtlander 15 in Emount for a trip to Bali late this past summer and really liked it. But I've followed the threads about the lens closely and I think it seems like a crapshoot to get one that would satisfy me (yes, I pixel peep . . . it's part of the fun for me). I love the size of the Voigtlander 15, but I refuse to buy copies until I get a good one and that seems to be the game.

Nov 20, 2016 at 08:52 PM

1 2 end
FM Forums \| Sony Forum \| Join Upload & Sell 1 2 end

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.