35mm Summilux - do I dare?

[pemayeux]

Advertisement (gone after registration)

I wouldn't necessarily worry about a lens not working properly. My M8 has had the 15 CV, 21 asph, 28/2.8 asph, 35 lux asph, MATE, 50 lux asph, nocitilux and 75 lux. One lens - the 50 lux, had a problem, but it also was evident on film. My 35 lux does not show the focus shift I keep reading about. It's my favorite lens film or digital. I wouldn't hesitate to buy a new Leica lens.

[carstenw]

So, if I send in an out of warranty 35mm Summilux asph, they'll adjust it properly...and for free?

 

I don't think that Leica will automatically do so, no. Two years is long enough to notice a problem. However, why don't you start a conversation with them; it can't hurt. Perhaps they will be able to help you somehow. Note that not all 35 Lux Asph lenses which have gone back to Leica have returned fixed. Some (Tim's, at least) have come back with the note that everything is within spec. It seems that this lens may be too tight in the tolerances to guarantee proper functioning with the M8, sadly.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[guy]

My 35 lux does not show the focus shift I keep reading about. It's my favorite lens film or digital. I wouldn't hesitate to buy a new Leica lens.

 

I'm 90% decided that I'm going to do just that. I'm mentally factoring in sending it off on its Grand Tour to Solms, so that if there is an issue I won't be too disappointed. I'm also thinking of trying to get the Voigtlander fixed, now that Wilson and others have pointed out that it ought to work. All of this, of course, dependent on finding some cash down the back of the sofa.

 

Thank you again, everyone, for help and advice and reassurance. I've been hugely gratified by how the whole Leica / rangefinder experience works for me when the kit is on spec – I certainly never thought I'd get this confident with it this fast after years of hypersonic autofocus – and the part of me that isn't nervy about tolerances and focusing issues is filled with childlike excitement at the idea of new toys that make pretty pictures. So, one way and another, gradually it's all working out for the best!

[Bill W]

I bought a used mint lux that was probably 3 or 4 years old. I tested it at all apertures and it appeared to be fairly sharp. It just needed to be coded. Hopefully it comes back with no adjusting and just the new mounting ring.

[ho_co]

... The methodology was antiquated but should have been adequate. It was proved to be inadequate, so either it was incorrect, sloppy procedures or worn equipment. Given the ethos at Leica, I would plump for the latter.

Wilson, I'm prejudiced by having worked for Leica; and I disagree: I think they would have noticed and replaced worn equipment, because (as the hideous phrase has it) their "core competency" is lensmaking. People involved in those procedures would have noticed.

 

There is a very interesting and balanced article on the Summilux 35 asph and focus shift in the newest LFI. ... [T]he only solution will be a new design with a floating element. The M8 is simply too critical.

Jaap--Thank goodness they've finally published the article, and done it on the topic of the 35 Summilux! Your last sentence seems to me to be the key: Older test methods are no longer adequate.

 

... The correct way to size an outside sourced (if it is) component like this on a modern production line is with a laser against an electronic template but that sort of technology does not come cheap.

Good observation. Some errors may have crept in with new out-of-house sources. My feeling is that for Leica no cost would have been excessive when it comes to keeping lens construction to spec. The problem with my logic is, of course, that some recognized needs might go unmet when you're broke.

 

... Note that not all 35 Lux Asph lenses which have gone back to Leica have returned fixed. Some (Tim's, at least) have come back with the note that everything is within spec...

Carsten, your description is correct. Both of Tim's 35 Summiluxes received the same diagnosis. But as I understood his response to a later query of mine (after the adjustment of his 35/2), he agreed that it is likely that that diagnosis was shortsighted, and that the lenses could have been adjusted if Leica had understood at that time what they discovered in approaching the Summicron.

 

 

Guy Mancuso described the projection of special test patterns in testing lenses--a guarantee of optical performance, but a procedure that clearly failed to detect failure of focus register. In other words, I feel there was in place a procedure that did not meet the requirements of the digital standard, and there was no recognition that the procedure needed to be changed.

 

Some people may have seen the problem on film, but it was when the M8 appeared that the mis-adjustment became clearly and immediately noticeable. I think it was an understandable oversight on Leica's part, because the initial argument is that "sensor and film are just two different media, but the lens is the same." It took them a while to recognize that the plane surface of the sensor and the immediate ability to blow up each image to 100% had changed the equation.

 

Mark Norton was right when he said the 35 Summilux design was inadequate to the M8. But that could simply mean that series production of the lens is too expensive when attempting to keep the lens to the tolerances of the M8's world.

 

Again, Guy B: The current 35/1.4 was considered Leica's best lens when it came out, and its performance was not surpassed until the aspheric 50 Summilux appeared. If I were thinking of the 35 Summilux, the contents of the LFI article would not retard but spur my purchase, before any replacement shows up.

 

Good luck!

 

--HC

[sean_reid]

Sean, that's really interesting. I really liked the Nokton on paper: right aperture, right length, single-coated etc. It sounded perfect for me. When I quoted the guy at Robert White, I was just passing on what I'd been told, which seemed to suggest to me that I'd have to scrap the Nokton and go for a Leica 35. Frankly I'd be much happier to have a working Nokton at the moment, for a number of reasons, not least the gaping price difference.

 

I still have the Nokton since I'm abroad at the moment and can't return it to the shop til I'm back in the UK. They've very generously agreed to allow me to hang on to it in the meantime as long as I keep it in its box and don't use it. It really does backfocus very obviously with the lens wide open, which is where I'd want to use it most. But given my experience with Leica UK themselves "testing" lenses on M8 bodies which turned out to have the same misadjusted rangefinder as mine, do you think it's worth my instead swapping this Nokton – the thinking being that perhaps it's a dud lens but that the others in the shop might be better *despite* what the salesman found when he tested them (ie, that his M8 might be faulty)?

 

Further, does anyone know of a reasonably economical way to get a Voigtlander lens adjusted in the UK? Robert White suggested this simply wouldn't be cost-effective, but you guys may know differently. If I could get the Nokton working I'd be a very happy bunny indeed, and could put my Leica 35 savings towards a 28 Lux instead...

 

Many thanks, again, for everyone's help with this.

 

Step One (and some seem to skip this) is to be absolutely sure that your M8 rangefinder is adjusted correctly. As of now, which lenses do and do not focus correctly on your M8? It's important to isolate which variable you need to deal with and, as a rule, cameras can go out of adjustment much more easily than lenses.

 

Cheers,

 

Sean

[sean_reid]

Advertisement (gone after registration)

There is a very interesting and balanced article on the Summilux 35 asph and focus shift in the newest LFI. Basically they say the problem cannot be resolved within the current design without compromising quality at maximum aperture (the way Zeiss seems to have done) and the only solution will be a new design with a floating element. The M8 is simply too critical.

 

Hi Jaap,

 

That seems quite possible. What causes confusion is the fact that all Summilux 35 ASPH lenses probably show minor focus shift but some examples seem to show quite noticeable focus shift (which is *not* inherent in the design).

 

Cheers,

 

Sean

[sean_reid]

Jaap,

 

I too have just put down the LFI article. I think, I am afraid, it makes a terribly good argument for the Biogon 35 rather than the Summicron, which like the Summilux also suffers quite bad aperture shift and maybe also for the CV 35/1.2 Nokton rather than a Summilux. How often do I use F2 on the Biogon - probably less than 20% of the time and have I noticed any softness - only once or twice. Does it make sense to compromise 80% of your images for a marginal improvement on 20% - not in my book. If I need to go faster, I always have my Nokton 35/1.2, which draws very differently from the B35 and in general is better on people that the B35, which seems to find blemishes on the best complexion.

 

As a take anywhere pair, when I don't want to carry a Billingham, for me it's the Biogon 35 and Elmar-M 50.

 

Wilson

 

Also...the CV 35/1.7 and the CV 35/2.5. You know, of course, that I looked at all of these lenses in a recent review.

 

Cheers,

 

Sean

[sean_reid]

pretty much the reason i sold the lux in favor of a new cron. My Lux had a focus shift that i decided was not worth fighting the problem and bought a brand new Cron which does perform very well.

 

Hi Guy,

 

The 35 Cron shows some focus shift as well but it seems like your sample of the Cron was better than your sample of the Lux.

 

Cheers,

 

Sean

[Guest guy_mancuso]

exactly Sean and the cron I really liked at f2. As you know I can get the lux repaired very quickly, I really just did not want to hassle Leica with it. They have been extremely good to me and many members I sent for help. so going to the cron

made sense

[guy]

Step One (and some seem to skip this) is to be absolutely sure that your M8 rangefinder is adjusted correctly. As of now, which lenses do and do not focus correctly on your M8? It's important to isolate which variable you need to deal with and, as a rule, cameras can go out of adjustment much more easily than lenses.

 

As of now, I only own the 21mm Elmarit and the 40mm Voigtlander Nokton. Both lenses and the M8 went to Solms; the rangefinder was way out, and was adjusted, and the Elmarit was apparently off spec also, and that too was fixed. They didn't touch the Voigtlander for obvious reasons, but they did test it, and said that it exhibited some backfocus – as indeed it continues to do, though it's improved by about 50% since they sorted the rangefinder on the camera. It's still there, though. The 21mm is spot-on accurate. As I said earlier, I don't own any more lenses: I'm new to this!

 

So my fervent hope is that the camera body is now sorted, which should give me a decent tabula rasa from which to start adding lenses.

 

As a point of interest, the shop I bought the camera and lens from (Classic Camera in London) agreed that there may well be a rangefinder issue, but swore blind that there would absolutely NOT be an issue with the lens – that the construction tolerances for the 21mm were at the bleeding edge of the industry's ability. The chap I spoke to had visited Solms and seen the 21mm in production there himself. No way, he said, would the lens be off. Leica tested it and it was. Not sure what the moral there is, except perhaps, as several others here have pointed out, that the M8 changes things.

[wlaidlaw]

If these Leica lenses are discovered to be "off" when returned to the factory, why is this not being discovered at the final production testing stage? Is the procedure different for production testing to that for repair testing? If so, does this make sense?

 

When customers like me have to send a lens back for a second time, Leica totally ignores requests for refund of postage. In my case it cost me £50 in post and insurance to return my Noctilux for a second service. It cost me £25 to send my Elmarit 90 only for them to fail to code it. I think the very least Leica could do when they mess up, is to return the hopefully correct item with vouchers spendable against Leica goods to the value of the post and insurance costs incurred by the customer. Why should we pay for Leica's mistakes? I hope to visit the factory at the end of September and I intend to bring this up. In law, if Leica have erred, then they are obliged to put the customer back in the same position as if they had not erred. They are just banking on the fact that nobody is going to be bothered to sue them for £50.

 

When I had two new faulty HP B9180 printers one after the other. HP sent the third one to me by taxi with a set of cartridges as an apology. They must have blown their profit many times over by this gesture but boy did they have a happy customer, who has recommended the B9180 and HP in general, to all and sundry.

 

Wilson

[guy]

If these Leica lenses are discovered to be "off" when returned to the factory, why is this not being discovered at the final production testing stage? Is the procedure different for production testing to that for repair testing? If so, does this make sense?

Wilson

 

It's a worrying question. You'd have thought that getting things right as the product goes out the door would save them time and resources in the long run. I suppose if a lens comes back they know they have to look hard for the problem (which suggests that quality control at the production stage has slipped somehow). It's interesting to hear of people having older lenses that work just fine while brand-new ones don't.

 

It's certainly true that the instant review function of the M8 allows the user to test and evaluate a lens's performance quickly and simply. Without the turnaround of development and printing you KNOW what you shot and how you shot it, and there's no margin for wondering after the event if maybe your focus was a tiny bit off. This isn't "pixel-peeping" (a phrase I loathe: since when has fastidiousness over image quality been something we denigrate?) – this is just careful evaluation, and you can do it at home, easily. And it leads inevitably to the question of why, if it's so easy, that evaluation hasn't been completed comprehensively at the factory.

 

Works great when it works, though!

[sean_reid]

As of now, I only own the 21mm Elmarit and the 40mm Voigtlander Nokton. Both lenses and the M8 went to Solms; the rangefinder was way out, and was adjusted, and the Elmarit was apparently off spec also, and that too was fixed. They didn't touch the Voigtlander for obvious reasons, but they did test it, and said that it exhibited some backfocus – as indeed it continues to do, though it's improved by about 50% since they sorted the rangefinder on the camera. It's still there, though. The 21mm is spot-on accurate. As I said earlier, I don't own any more lenses: I'm new to this!

 

So my fervent hope is that the camera body is now sorted, which should give me a decent tabula rasa from which to start adding lenses.

 

As a point of interest, the shop I bought the camera and lens from (Classic Camera in London) agreed that there may well be a rangefinder issue, but swore blind that there would absolutely NOT be an issue with the lens – that the construction tolerances for the 21mm were at the bleeding edge of the industry's ability. The chap I spoke to had visited Solms and seen the 21mm in production there himself. No way, he said, would the lens be off. Leica tested it and it was. Not sure what the moral there is, except perhaps, as several others here have pointed out, that the M8 changes things.

 

The 21 has a lot more DOF than the 40. Try the camera, maybe at a dealer, with some 50s, 90s, etc. to confirm that the RF is properly adjusted. If it is, then I would exchange the 40 or get it repaired under warranty. It should be dead on at F/1.4 and then shift a bit as it is stopped down. Have you read my review of it on the M8?

 

Cheers,

 

Sean

[guy]

The 21 has a lot more DOF than the 40. Try the camera, maybe at a dealer, with some 50s, 90s, etc. to confirm that the RF is properly adjusted. If it is, then I would exchange the 40 or get it repaired under warranty. It should be dead on at F/1.4 and then shift a bit as it is stopped down.

 

Good thoughts, Sean. Once I'm back in the UK I'll take a trip to London and do just that.

 

Thanks,

 

Guy

[pklein]

There is a very interesting and balanced article on the Summilux 35 asph and focus shift in the newest LFI. Basically they say the problem cannot be resolved within the current design without compromising quality at maximum aperture (the way Zeiss seems to have done) and the only solution will be a new design with a floating element. The M8 is simply too critical.

 

Jaap (or anybody?): What month(s) edition is this issue of LFI, and what is the article title? I just called our international newstand, and they have the June/July issue of LFI, but do not see any article with the words "focus shift" in the contents.

 

I'm finding in practice that the focus shift is not so big an issue. When using f/4 and f/5.6, I just focus on something a little closer than what I want, e.g. the nose intead of the eyes if I'm close-in, or the nearest person in a group if I'm farther away. It's a bit of a nuisance, but the lens is so good otherwise that I'm willing to put up with it. So far.

 

--Peter

[carstenw]

Peter, it is 6/2007 (August), and it just came out, and hasn't reached everywhere yet. Give it a day or two.

[wlaidlaw]

Peter, it is 6/2007 (August), and it just came out, and hasn't reached everywhere yet. Give it a day or two.

 

Mine arrived yesterday morning.

 

Wilson

[carstenw]

Subscriber copies apparently get sent out first by LFI. It is sadly rare to see a company which takes care of its loyal customers like that.

[Advertisement]

Hallo,

Look here: Leica M

[wstotler]

My Summicron 35 asph chrome has no preceptible focus shift in daily use. It is a darn sharp and well-focussing lens at all apertures

 

Ditto on this observation. Same here.