My used 35 lux is back focusing

[Crazy Cat Lady]

Advertisement (gone after registration)

A couple of us posters have asked you to clarify whether the lenses you are referring to are indeed of the pre-aspheric designs. I may have overlooked your response, but I'm still unsure what you've got--and it makes a tremendous difference.

 

--HC

 

 

I have the ASPH version.

[Crazy Cat Lady]

Does the 35 ASPH lux uncoded and the 35 lux non-ASPH, exhibit the same problem? Or is it tied to the lenses being coded?

[tashley]

Does the 35 ASPH lux uncoded and the 35 lux non-ASPH, exhibit the same problem? Or is it tied to the lenses being coded?

 

 

That's one of the things we're all trying to get to the bottom of.

 

The potential variables are:

 

ASPH/Pre ASPH,

Coded at manufacture/Coded later/Uncoded

Current manufacture batch/Older batches

Silver/Black

 

There is a mild possibility that the silver (chrome) ones are better in the backfocus dept.

 

Tim

[ho_co]

Thanks: 35/1.4 ASPH.

 

Forget what I said about the earlier lens not being good enough to worry about. The current lens is recognized as being extremely good--but then, why did the previous owner sell it? Did she have problems as well?

 

Also forget what I said about not using f/1.4. The aspheric lens is very good wide open, so definitely include that aperture in your tests.

 

Do as Tim suggested. Rack the lens to infinity. Point the viewfinder at the moon. You should see a single sharp image in the rangefinder patch. If you don't, then check your other lens.

 

Shoot the moon and check the results. They should be sharp. You might try one more thing: Try shooting first with the moon in the center of the image field, and then with the lens still racked to infinity, shoot another picture by turning the camera to one side and locating the moon toward the edge or corner. Let us know the results.

 

REMEMBER: The moon is not green cheese, but grey rock illuminated by the sun. Therefore, proper exposure for the moon is the same as general daylight exposure here, viz 1/250 ca f/16 at ISO 200 (which is what the M8 actually delivers when set to ISO 160), or 1/8000 ca f/2.8 if I've calculated properly.

 

Progress! Bravo! Progress!

 

As for coding, as Tim said, that's up in the air. Theoretically coding should make no difference, but we're not sure in actuality.

 

--HC

[xrogers]

The cloud test came out blurry. Which I do not know what that means. But this weekend I am going to test it out on other things. So, if the problem is with the lens, then there is no way to fix it, right? Basically I keep it and live with it, at a steep price or return it.

 

::Sigh:: What a bummer

If you do the moon (or other edgy, faraway object) test others are describing, and the lens is blurry at infinity, then it may be simple calibration of the lens focus is all that's required. That isn't so expensive, and Tony at popflash will either pay to have it done, or refund your money (whatever you prefer).

 

My copy of this lens is fantastic, even after six-bit conversion. I might be able to see the results of focus shift if I tested for it, but I'm not going to. I've used it for thousands of M8 photos so far, and have yet to be disappointed by it.

 

--clyde

[cbretteville]

HC thanks for the exposure data.

 

Just a word of caution: Depending on atmospheric conditions the moon might not be a sharp object.

[Jamie Roberts]

Advertisement (gone after registration)

Jennifer--bottom line is you may need your camera or lens adjusted (or both together). That's it. Quite common for many cameras these days, and everyone is looking much more critically at their files these days

 

And as someone who has one of those 'magic' 35 Lux ASPH lenses, I have a couple of observations, though I continue to agree with Sergio (and with William, too, as far as that goes).

 

First, I asked Leica US reps last week directly about whether the 35 Lux was--by design--impossible to focus sharply between f2.8 and f5.6 in the center of the lens's field.

 

They said "of course not"--which at least makes me feel that no magic is going on. But I will also try to get more definitive information from Solms.

 

They also said the lens probably needs adjusting to the body, which is where we left that thread (but I'm still glad your 50 lux focusses more accurately now).

 

Now, could there be a run of completely mis-calibrated 35mm lenses? Yes, of course there could be, and I suspect we're getting close to the answer here.

 

Does the focal field shift as you stop down? Yes. Does the center go out of focus by design? No, I don't think so.

 

No-one is going to buy a blurry 35 lens at f4.

 

FWIW, film is, if anything, harder to get critical focus on because of the media itself. I don't believe for a second that the "depth of the media" has anything to do with this.

 

@ So, Carsten--a slight correction for you: if you blur one channel in photoshop (say, the blue channel), which is more like film, then look at the whole picture in RGB, then you will see more blur, not less.

 

You can actually decrease colour noise (detail) this way, though it works better in LAB by blurring the B channel (when the blue channel is noisy) because there is no luminance information in the AB channels (and there is in the blue in RGB).

 

When you *overlay* (or use any other channel operation) a blurred layer on another photoshop layer, what you increase is the average (thanks blur!) contrast of the image, depending on what channel method you're using and how the blending is set.

 

This looks sharper, because contrast is one of those things that affects perceived sharpness.

 

You don't get more detail, and you're not actually sharpening at all. What you are doing is increasing local contrast (and a better way to do that is sharpen the L channel in LAB mode, or blur a high-pass filter then put that in Overlay).

 

So the analogy about one blurry layer in film making sharper layers doesn't work. Film layers do not make an unsharp mask all by themselves. You need to expose, cut and wedge them together in the enlarger to mask out what isn't sharp Or use PS "unsharp mask" instead.

 

@ Tim--I'm going to post a bunch of 35 Lux shots taken at f4 at night when I get a chance (still crawling out). I'm sorry, but they're all tack sharp in the center FWIW, I picked up the CV 28 1.9 and the 21; they're also sharp. So the only lens in my system--after tweaking--that isn't perfect, is the Canadian 75 Lux wide open. So that lens needs adjusting.

[tashley]

Jennifer--bottom line is you may need your camera or lens adjusted (or both together). That's it. Quite common for many cameras these days, and everyone is looking much more critically at their files these days

 

And as someone who has one of those 'magic' 35 Lux ASPH lenses, I have a couple of observations, though I continue to agree with Sergio (and with William, too, as far as that goes).

 

First, I asked Leica US reps last week directly about whether the 35 Lux was--by design--impossible to focus sharply between f2.8 and f5.6 in the center of the lens's field.

 

They said "of course not"--which at least makes me feel that no magic is going on. But I will also try to get more definitive information from Solms.

 

They also said the lens probably needs adjusting to the body, which is where we left that thread (but I'm still glad your 50 lux focusses more accurately now).

 

Now, could there be a run of completely mis-calibrated 35mm lenses? Yes, of course there could be, and I suspect we're getting close to the answer here.

 

Does the focal field shift as you stop down? Yes. Does the center go out of focus by design? No, I don't think so.

 

No-one is going to buy a blurry 35 lens at f4.

 

FWIW, film is, if anything, harder to get critical focus on because of the media itself. I don't believe for a second that the "depth of the media" has anything to do with this.

 

@ So, Carsten--a slight correction for you: if you blur one channel in photoshop (say, the blue channel), which is more like film, then look at the whole picture in RGB, then you will see more blur, not less.

 

You can actually decrease colour noise (detail) this way, though it works better in LAB by blurring the B channel (when the blue channel is noisy) because there is no luminance information in the AB channels (and there is in the blue in RGB).

 

When you *overlay* (or use any other channel operation) a blurred layer on another photoshop layer, what you increase is the average (thanks blur!) contrast of the image, depending on what channel method you're using and how the blending is set.

 

This looks sharper, because contrast is one of those things that affects perceived sharpness.

 

You don't get more detail, and you're not actually sharpening at all. What you are doing is increasing local contrast (and a better way to do that is sharpen the L channel in LAB mode, or blur a high-pass filter then put that in Overlay).

 

So the analogy about one blurry layer in film making sharper layers doesn't work. Film layers do not make an unsharp mask all by themselves. You need to expose, cut and wedge them together in the enlarger to mask out what isn't sharp Or use PS "unsharp mask" instead.

 

@ Tim--I'm going to post a bunch of 35 Lux shots taken at f4 at night when I get a chance (still crawling out). I'm sorry, but they're all tack sharp in the center FWIW, I picked up the CV 28 1.9 and the 21; they're also sharp. So the only lens in my system--after tweaking--that isn't perfect, is the Canadian 75 Lux wide open. So that lens needs adjusting.

 

OK, I'm curious. I still don't believe you, can't see why Solms would say what they did... but if there's a resolution I'll buy it, obviously...

 

And good to have you back!

 

Tim

[Jamie Roberts]

OK, I'm curious. I still don't believe you, can't see why Solms would say what they did... but if there's a resolution I'll buy it, obviously...

 

And good to have you back!

 

Tim

 

Hey Tim--thanks! I'm trying to understand how this could happen at all, and so I'm trying to get some definitive word here...

 

And it's good to be back, too, though I'm just swamped.

 

By my own stupid fault, I was pretty much Internet-less the last week and a half; in my defence the conference was very packed too (and I spent wayyy too much money on stuff at the tradeshow--without any Leica purchases, no less!).

 

One thing, though, I got to try out was a Noctilux! I shot 4GB of night pix with that thing--at f1.0--both up at my eye and estimating distance (just to see)

 

I was really quite shocked at how easy it was to focus on the M8. Sans filter, too. That's going to be my lens purchase from Solms, I think, though it is a beast compared with the 50 Lux.

 

But pix are coming....I'm still sorting through them

[Jamie Roberts]

OK, just for illustrative purposes, here's one taken with that pesky 35 Lux at f4:

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

I honestly don't think I'd want it any sharper, actually

 

For those interested, this was white balanced with the new Expodisc, then colour corrected in Photoshop. Summilux 35 ASPH f4.0 @ 1/4000s no IR filter. This is also cropped from the out-of-camera 6MP JPEG, not the DNG, which is considerably better

 

Oh--and this is what happens at about 5PM with a good reflector (sorry--different thread!)

[tashley]

OK, just for illustrative purposes, here's one taken with that pesky 35 Lux at f4:

 

[ATTACH]31902[/ATTACH]

 

I honestly don't think I'd want it any sharper, actually

 

For those interested, this was white balanced with the new Expodisc, then colour corrected in Photoshop. Summilux 35 ASPH f4.0 @ 1/4000s no IR filter. This is also cropped from the out-of-camera 6MP JPEG, not the DNG, which is considerably better

 

Oh--and this is what happens at about 5PM with a good reflector (sorry--different thread!)

 

 

Lovely shot but you're cheating: she's not holding five bird boxes...

;-)

T

[ho_co]

I asked Leica US reps last week directly about whether the 35 Lux was--by design--impossible to focus sharply between f2.8 and f5.6 in the center of the lens's field. They said "of course not" ...

Jamie--I was a US rep for Leica for over ten years, and I have great respect for their knowledge and diligence. I know and respect some Leica reps today, but I don't know them all. When I worked for the company, some of us had very little technical knowledge. My position as listed on my business card was "Sales Representative." Although I reveled in technical knowledge, my official job was to sell. If my job is to sell and you ask me such a question, my answer will likely be "No, of course not."

 

I don't believe that the 35 Summilux is designed with curvature of field as bad as that which Tim has shown us, but Solms said it is. As a rep, until I had seen the evidence or received a heads-up from the company (which would never have been made general knowledge when I was employed there), I would also have answered, "No, of course it's not designed that way," both out of surprise at the assumption and due to the definition of my position, viz salesman.

 

So without disrespect to the people you spoke to, their answer is the expected one.

 

But I will also try to get more definitive information from Solms.

Excellent. My advice as well. When I worked with the company, I knew two people from whom I could get the straight poop. The knowledge is there and closer to the surface than you'll find elsewhere. Let us know when all is revealed.

 

They also said the lens probably needs adjusting to the body ...

Does that explanation bother you? It does me: 'Our bodies are made to tight tolerances. Our lenses are made to tight tolerances. But sometimes (?) we need to adjust the tolerances so that a given lens will work on a given body'?

 

Then what happens when I put the lens on my backup body? Or want to sell it?

 

The first time I heard of having to tailor a lens to a body was with the 50/1.2. I didn't understand it then and don't now, but I can sooner understand it on a longer, faster, specialty lens than on the 35 Summilux.

 

I'm going to post a bunch of 35 Lux shots taken at f4 at night when I get a chance .... I'm sorry, but they're all tack sharp in the center

Um, Jamie--

Did you notice that the picture of the couple you posted (very nice shot, sharp as a tack and all that) has the couple along the edge, not at the center? That is, this test is exactly what Tim's tests showed would work just fine with his lens...

 

Seriously, I'm looking forward to a definitive answer.

 

--HC

[ho_co]

Tim, Jamie--

I know this one is off the wall, but do you get the same results both with and without the IR-cut filter?

 

Jamie mentioned that the image above is without the filter; what about your shots, Tim?

 

A filter does refract a bit. The US "Leica School" used to teach that a filter should be avoided except when absolutely necessary, and especially so on wide angle lenses.

 

Could this be a case where the lens is perfect but the filter introduces a problem? If we find that is the case, we have a much easier solution to the problem.

 

And then both of you would be right.

 

Just a crazy idea. Grasping for an explanation. (Till the next run of birdboxes.)

 

--HC

[Jamie Roberts]

{snipped} 'Our bodies are made to tight tolerances. Our lenses are made to tight tolerances. But sometimes (?) we need to adjust the tolerances so that a given lens will work on a given body'?

 

Then what happens when I put the lens on my backup body? Or want to sell it?

 

The first time I heard of having to tailor a lens to a body was with the 50/1.2. I didn't understand it then and don't now, but I can sooner understand it on a longer, faster, specialty lens than on the 35 Summilux.

 

{snipped}

 

Um, Jamie--

Did you notice that the picture of the couple you posted (very nice shot, sharp as a tack and all that) has the couple along the edge, not at the center? That is, this test is exactly what Tim's tests showed would work just fine with his lens...

{snipped}

 

Howard, (and Tim)--

 

This vexes me too a bit, though less than Tim of course...

 

And while I completely understand the whole "we're selling this" perspective I heard from the reps, these people are also Leica shooters with M8s and 35s that work, so I think they were speaking both as representatives and as empirical users

 

I guess when you're talking extremely close alignment, you get to the "I need to tailor the response" place.

 

The parts individually may be made to very high tolerances, but if you're on the outside of both on any given camera and any given lens, you might need an adjustment when they meet...

 

From my experience, it was Canon who first suggested I send both my 1 series bodies and my lenses into them to correct for backfocus. They did, and it was easy to interchange, and easy to sell, too

 

My Canadian 75 Lux that still *front focuses* a bit is also a used lens, though in excellent condition. But still, there's no guarantee the thing doesn't need an adjustment, since it has a history before me!

 

And yeah, I understand too about a longer lens being more prone to slight adjustments. But factor in a quirk of manufacturing on the 35 side and you could easily have the issue Tim is seeing (well, except for that left / right thing)...

 

As for my shot, I know you're kidding--if you look at the glass, it's also tack sharp (and her hand, etc...)

 

BUT--as it turns out, I cropped them from the center anyway

 

On the filter--shouldn't the presence of IR make focus *more* difficult?

 

@ Tim--if I could have persuaded her to hold 5 birdboxes, well, we wouldn't have been taking pictures!

[ho_co]

Jamie--

No argument on any of your points. I do know that most camera company reps (Leica included, myself included) get into the business because they like cameras and photography. But soon we get tired of taking pictures. The constant exposure to the equipment and the constant complaints from dealers and from consumers just wear you down. I know that I would have given you the right answer as I knew it, and I know the people you spoke to did the same. However, when you're dealing with the product at its commodity level, your information may not be the most accurate. Even though I considered myself a more technically aware representative than most of the others at the time, I would still check with our tech people whenever possible--and sometimes find out I didn't have the latest information.

 

As for the filter thing, I just think it should be checked.

 

As you said, your picture is tack sharp and you didn't use the IR filter. And as you said, when there's a fair amount of IR in a picture, the result should be a bit fuzzier.

 

It's grasping at straws, but I think a couple pictures with and without filter could eliminate another variable in this confusing chain.

 

I know I'm not completely crazy, because at the time of the Spotmatic, Pentax offered special curved UVa filters for their normal lenses. The specific reason was different, but the specially designed filters were a solution to a problem: If you used plane filters on those lenses, you got internal reflections; with the curved filters, the reflections disappeared.

 

Damn! I had overlooked the fact that your picture was a center crop! I thought I had finally won one!

 

--HC

[tashley]

@ HC, I will shoot some side by sides today (I hope) with and without filters

 

@ Jamie, I have no left/right problem with the 35 Cron! I have a 35 CV that's a bit smeary on the right and a 24 2.8 elmarit that's a bit smeary on the left.

 

BTW I have two M8 bodies now... so my testing procedure us getting more refined...

 

Both of you: I posted a shot towards the end of The Long Thread and asked people to look at it carefully. Sergio did and thought it was fine. But it really isn't. Here's the link to it again, this was on the 35 Lux and it's a full sized jpeg from DNG.

 

Look at it closely and tell me what you see!

 

http://tashley1.zenfolio.com/img/p399774702.jpg

 

Best

 

Tim

[tashley]

Jamie--

 

 

As for the filter thing, I just think it should be checked.

 

As you said, your picture is tack sharp and you didn't use the IR filter. And as you said, when there's a fair amount of IR in a picture, the result should be a bit fuzzier.

 

--HC

 

An old friend returns: Three centre of frame crops from the 35 Cron all at ISO 160 on a tripod with the same focus and extreme care taken not to shift the focus or the tripod (a huge heavy manfrotto) between shots. First, at F2 with B&W IR filter. Second, F4 with filter. Third, F4 without filter...

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

 

 

 

To my eye the unfiltered shot is hard to call against the filtered but both are less sharp than the F2 shot. The F2 shot is not absolutely critically sharp, not as well as this lens can do when stopped down and correctly misfocussed ;-)

 

Tim

[Jamie Roberts]

Tim--I don't know what to say anymore. Your 35s are obviously messed... the birdboxes show that (and not from IR filtering, either).

 

The Italian shot is interesting, insofar as I don't know what you focused on. I'm guessing it was the flags right in the middle, but I have no idea. They are softer than the edges--I'm guessing that's the point?

[tashley]

Tim--I don't know what to say anymore. Your 35s are obviously messed... the birdboxes show that (and not from IR filtering, either).

 

The Italian shot is interesting, insofar as I don't know what you focused on. I'm guessing it was the flags right in the middle, but I have no idea. They are softer than the edges--I'm guessing that's the point?

 

 

It is indeed! That was on one of the luxes I took back. I shot another shot moments afterwards on the 50 lux and the flags were sharp. (same aperture , 4 or 5.6 from memory). If you look at the 35 lux shot, there are poles in the water which are equidistant to the flagsm but are sharp because they are out of the beckfocus centre spot.

 

What's interesting is that there is a circle in the centre of the frame that look soft - but actually it isn't soft, it's just focussed further out - which is why the 'learn it and shift your focus forwards' technique doesn't work. I stress that this was exactly the same on the two 35 luxes I sent back, which Solms tested and declared normal. Thankfully the 35 Cron doesn't do it as much and in a huge number of shots, the great sharpness of the majority of the frame plus the tendency to focus centrally and recompose subject to a third, means that the effect is lost entirely. I use the lens happily, but not quite as happily as I'd like...

 

That's why I want to get my hands on your lens! Test it, keep it, whatever...

 

best

 

t

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[sergiolov]

@ HC, I will shoot some side by sides today (I hope) with and without filters

 

@ Jamie, I have no left/right problem with the 35 Cron! I have a 35 CV that's a bit smeary on the right and a 24 2.8 elmarit that's a bit smeary on the left.

 

BTW I have two M8 bodies now... so my testing procedure us getting more refined...

 

Both of you: I posted a shot towards the end of The Long Thread and asked people to look at it carefully. Sergio did and thought it was fine. But it really isn't. Here's the link to it again, this was on the 35 Lux and it's a full sized jpeg from DNG.

 

Look at it closely and tell me what you see!

 

http://tashley1.zenfolio.com/img/p399774702.jpg

 

Best

 

Tim

 

Tim,

I evaluated the shot at print size, to judge if it was suitable for publishing on a travel magazine. It is.

Looking at 100% from a technical point of view, the lens shows a slight decentering, as the right side at the focusing distance, probably around 10 meters, is less sharp than the left one. Not easy to see, due to the increasing shadow from left to right.

 

I have one more test for you, if you have time and want to temporarily mess with your rangefinder.

As you know, my theory to justify the fact that my cron 35 stays sharp closing down, and gets better and better doing this, is that it is adjusted to front focus full open, where this means that the focus plane is at the rear of the acceptably sharp area.

 

- Allen wrench in your hands, tripod, focus test chart, F2. Verify where the focus plane is.

- Slightly adjust the rangefinder to have maximum possible sharp area in front of the point where you focused without compromising the sharpness of the point where you focused.

(the rangefinder will be misadjusted at infinity) Approx 2/3 in front, 1/3 rear

- If now your bird box behaves as it should, the fault is in the lens adjustement.

 

Reset the camera at infinity, and start doubting of Leica quality control.

 

If the problem persist, well, this time the lucky man is me, with my strange cron.

(begging for your forgiveness, Jennifer)

 

Sergio