Sample Variation in Lenses

[Rick in CO]

Advertisement (gone after registration)

" An OEM Schneider is no better or worse than the Linhof branded version, it is the same lens, otherwise how could Schneider get away with marketing an inferior product that is identical to a Linhof badged product? And then there is the experience of using them, in which, there is no difference, just the smoke and mirrors of marketing name plates. Prove me wrong."

 

Well, the marketplace proves you wrong. Linhof-badged lenses go for a premium price. You might say its hype, but I have not observed Linhof customers complaining about the issues pertaining to this particular thread. To be clear, its not to say that any Linhof lens is better than any non-Linhof lens. Not every lens from a particular manufacturer gets sent to them for testing. What is not clear is what happens to the % of lenses that don't meet testing spec's and get sent back to the manufacturer by Linhof & Sinar. The additional factor that hasn't been mentioned yet is that the Linhof lenses for the Technika rangefinder cameras are matched with custom cut interchangeable focusing cams. Similar to the Leica IC with non-standardized interchangeable lens mount. Aside from manufacturing defects (like de-centering or element spacing) this custom adjustment solves the up to 5% variation in focal length that is typically seen for lenses out of the box. For the prices paid, Leica QC should catch the manufacturing defects, however the variation in focal length is more difficult to remedy. Leica will adjust a lens to your camera - it worked wonders for my second-hand 135 Elmarit. The bottom line to all this discussion is that any new lens should be tested just as if it were previously used.

[gpwhite]

I believe the issue that Ming Thien addressed, the last time I discussed this with him via e-mail, was the compound effect of lens variation itself with rangefinder calibration. Either side of that equation could be quite nominal in tolerances (i.e., "spot on specification"), yet the images taken from the system will be thrown off by variation from tolerances in the other component. Ming said he had enough, and other professional reviewers have published similar findings.

 

Is there quite noticeable sample variation in Leica lens performance? For sure. Is there quite noticeable sample variation in Leica M series rangefinder calibration? For sure. So it takes some fiddling by either testing multiple copies or learning how to shoot your particular lens-bodies systems with a fine eye to where in the RF patch one aims for perfect correspondence. Of course, I you live near Wetzlar, you can have it your way

 

One thing struck me about Ming's opinion on this topic. Even acknowledging that he is an accomplished professional, I found his strong support of the D800E focus system surprising. The OVF on that body is just terrible, as the system is designed for AF. Perhaps Ming is doing his food work with a tripod, LV and a Zacuto?

[pgk]

Well, the marketplace proves you wrong. Linhof-badged lenses go for a premium price. You might say its hype, but I have not observed Linhof customers complaining about the issues pertaining to this particular thread.

The marketplace is not a very scientific methodology for determining such things. Unless Linhof actually publish data as to what they require in a lens then there is very little comment to make. My guess would be that if anything, Linhof specify performance/technical specifications which the lenses supplied to them must meet and that said, I'd guess that such performance/technical characteristics are well within Schneider's lens build tolerances anyway;). I can't see any manufacturer supplying 'premium' equipment for another which could suggest that its own products were somehow potentially 'inferior' even if that's only in QC terms.

 

Leica M lenses have very different requirements and the number of variables is high so any problem of many can reduce performance. As I said in my last post, the hand-built product system which Leica uses for its M lenses does at least allow for economically viable adjustment rather than simply replacement.

[pico]

About what? Linhof testing lenses, or OEM being as good? I'm not a mind reader. It would help if you have a 'theory' you interjected with it instead of asking everybody else to prove your idea for you.

 

It was you, not me who asserted:

"It is true that Linhof test lenses, but a 'good' Linhof branded lens is no better or worse than the ones supplied by the OEM, it is just a marketing ploy to justify an increase in price."

 

In other words, you wrote (in an ambiguous sentence) that Linhof's marketing is simply a lie. That kind of statement deserves evidence, at least a credible citation.

 

FWIW, I have and use a few Linhof lenses, all for 4x5.

[250swb]

In other words, you wrote (in an ambiguous sentence) that Linhof's marketing is simply a lie. That kind of statement deserves evidence, at least a credible citation.

 

 

Schneider don't have a separate line to build 'better' Linhof lenses than the ones they market themselves. Linhof cherry pick, they probably even have Schneider do it for them, but the best Schneider lens (assuming some sample variation is weeded out in extra checking) isn't worse than the same lens made with a Linhof logo. So Linhof charge for double checking the lens and putting their own logo on it. I guess only you can decide if the price difference is worth it, but as it is a considerable difference the general assumption might be the customer is getting something optically better, rather than getting something simply 'quality controlled more'.

 

It would be the same if Hermès marketed editions of Leica lenses at a much higher price, and that had been quality controlled to the nth degree. So there would be absolutely no Friday afternoon lemons. But the optical design is exactly the same, so it can't be better than the best Leica branded version. You wouldn't need to send a Hermès lens back, and that is the convenience you'd pay for (plus a badge). Of course as with Linhof branding it wouldn't take long before Hermès owners would start to think they had something extra special, and just like 'King of Bokeh' another myth would be born.

 

So where are we? I haven't offered proof, as in company documents etc. Which means you must have proof to the contrary and you are just waiting to deliver it? Or is it I'm required to prove something where your lower standard is just to disagree?

 

Steve

[Rick in CO]

Well, we have deviated from the original query of this post. The facts remain that no manufacturer on the planet can make a lens that is perfect (except perhaps by fortuitous happenstance). Lens elements that deviate from absolute vertical, from perfect spatial alignment with each other, lens groups that deviate from the specified focal plane, yet alone defects in glass manufacture, are all subject to variation. There is a +5% tolerance which the industry (supposedly) follows when listing the specifications of a product. Your new "50mm" lens can actually be 47.5mm or 52.5mm, either of which is within industry "standards". The latter is particularly important with rangefinder cameras and is a separate issue from the "focus-shift" observed with some high aperture lenses - that is a consequence of lens design. If you wish to read more, go to Roger Cicala's LensRental website and search his older posts. Erwin Puts has cleaned up his posts on the subject so you will have to dig it out of his Leica Compendium which has a whole chapter exhaustively discussing lens design and manufacture. Leica goes to extraordinary effort to minimize these variations in manufacture, that is why their lenses are expensive compared to the competition. The bottom line, again, is that one should test every lens purchased, new or second-hand, by taking pictures with the camera on which it is intended to be used.

[Jeff S]

Advertisement (gone after registration)

Well, we have deviated from the original query of this post. The facts remain that no manufacturer on the planet can make a lens that is perfect

 

That's because we covered this in the first couple of posts, including the link here.

 

Jeff

[pico]

Steve, we're good. I would search the Large Format Information forum looking got 'linhof select bob salomon'. I have learned to trust Bob over the decades.

 

 

 

 

Sent from my Etcha-sketch.

[pgk]

Steve, we're good. I would search the Large Format Information forum looking got 'linhof select bob salomon'. I have learned to trust Bob over the decades.

I tried that pico, and (as someone who used to test lenses for a similar purpose myself;)) my original query remains in force. Exactly what constitutes a 'better' lens? - I can't see that the original manufacturer would have lower QC requirements. I'm not a statistician, but my betting is that QC failures are easily statistically shown because they will fall in the lowest % grouping, and allowing cherry picking of the 'best' - those in the highest grouping - is to me, a very dodgy idea as it leaves the rest of the manufacturer's lenses as demonstrably poorer lenses.

 

I'd suggest that 'cherry picking' is more likely to be designed to pick up a poorly QC'd lens once in a while (this is how things tend to work in my experience), and is more likely to allow the vast, vast majority of lenses to pass - because they will! They will still have been subjected to further QC but I'd be doubtful that many would fail - without actual technical specifications of pass/fail requirements and/or specific % rejected back to the manufacturer, I'm very dubious about all this 'better' lenses mythology, although I could understand a manufacturer setting up rangefinder cams individually or suchlike, which would optimise the specific lens in use for the task in hand, or perhaps very careful individual alignment of aero lenses at infinity or similar.

 

In my days of lens testing (many years ago now), the vast, vast majority of lenses were fine and were sold and had no problems. The very few lemons should have failed initial QC tests, but crept through until spotted by a further test - as they would have been later, in actual use.

[Rick in CO]

"In my days of lens testing (many years ago now), the vast, vast majority of lenses were fine and were sold and had no problems. The very few lemons should have failed initial QC tests, but crept through until spotted by a further test - as they would have been later, in actual use."

 

Or, spotted by a further test like in Linhofs QC lab.

 

"That's because we covered this in the first couple of posts, including the link here."

 

Sample variation is not the only real world issue. I wonder how many rangefinder camera lenses do not perform satisfactory on their owners cameras (ie. Mr Ming) and are sold on eBay (or returned) to someone else who discovers its the sharpest lens they've ever used on their camera?

[pgk]

Rick

 

The problem I see is that the manufacturer would have to take back as returns, lenses in the lowest % which could not (according to the link) be adjusted and/or accept that there manufacturing and QC were somehow lacking and that they could make lenses 'better' by adjusting them better themselves. I can't see this being viable and my assertion remains - that the vast majority of lenses would pass all QC checks as they stand. Myths are harder to counter than objective technical explanations.....

[Jeff S]

Sample variation is not the only real world issue. I wonder how many rangefinder camera lenses do not perform satisfactory on their owners cameras (ie. Mr Ming) and are sold on eBay (or returned) to someone else who discovers its the sharpest lens they've ever used on their camera?

 

Camera/lens connection was covered in the other link I provided, in the first response to the OP….and this one in that some post.

 

Anything else you'd like to revisit?

 

Jeff

[250swb]

I would search the Large Format Information forum looking got 'linhof select bob salomon'. I have learned to trust Bob over the decades.

 

Every link and conversation I find concerning the entirely unbiased Bob Salomon ()seems to still confirm the same point, the Linhof lenses get checked more, that's it. In the case of Schneider it does mean the end user has to be clever enough recognise the fault, which of course they mostly are, because they could spot the catch of paying a lot more for a Linhof lens in the first place. As far as Mr Bob Salomon is concerned I suspect that if he earned his living selling Schneider as well he would be singing in a clearer voice from a different hymn sheet rather than encouraging the cult of Linhof.

 

Steve

[pgk]

.....the Linhof lenses get checked more, that's it......

I'd be surprised if Linhof would be able to 'fine tune/adjust' lenses better than the manufacturer. Individual fitment to optimise performance is one thing, but to actually adjust is another and would, I suspect, significantly increase the cost if such work was to be carried out (if it was thoroughly economic, the manufacturer would do it, obviously). I'd suggest its simply an extra QC check and individual fitment where possible (ie, rangefindercam adjustment).

[250swb]

Yes checked, not adjusted. As regards the rangefinder cam that is something any lens sold for a rangefinder camera may have (it's not essential for all types of photography), but is irrelevant if a Linhof lens is bought second hand and not still matched with the camera. For many lenses this is now a redundant cost, built into the price for perfectly matching the lens and panel to a camera/cam, but separated from the camera is just a fee, and the optical arrangement wasn't touched anyway.

 

Steve