The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

New 35 Lux announced

PeterA

Well-known member
Hi Peter

Hmm - I think Tim Ashley did some 'definitive' tests, and Leica have certainly acknowledged it as a feature of the lens. Never having owned the lens I can't remember the tests, but presumably you focus at around 2 metres at f1.4 on a point on a receding scale, and then change the aperture successively downwards and look at the point of correct focus - I remember seeing examples where the focus point at f1.4 was well out of focus at f2 until around f8 (where the dof rescued it again).

But maybe you have a magic lens?

all the best
Hi Jono,

hmmmm so the focus shifts along a line away from a point either front or back accordign to aperture used?

well that isnt a definitive test for me at all - in fact from my naive perspective it is a very silly test that says nothing much about anything much useful or interesting to me..- let me explain before being accused of being a bore...

My 'definitive test is about how I make a photograph using any camera and any lens. First I look at the light and decide what is possible - then I consider how much DOF I wish to use and how much separation I wish to use - that pretty much delivers the aperture and speed combo required - then I focus and ..go click

in this world the only focus shift that can occur is from a poorly aligned rangefinder or shaky hands - I have experienced both.

Now if someone can point to me the relevance of the focus shift test - that would be interesting...is my aspherical ( with focus shift I can never see) a poor lens? really? In what away would this focus shift - have ANY baring on making ANY photo ever?

Pete
 

Terry

New member
Peter,
I have read that the chrome versions for whatever reason were less prone to focus shift than the regular versions.
 

Paratom

Well-known member
I had a Noctilux/1.0 and a 35/1.4asph - both were send to Leica together with my M8 for calibration of focus and tested by Leica.
Still at closer distance They would not reliable focus at shorter distances wide open, but focus would be spot on when stopped down 1-2 stops.
So IMO focus shift can be an issue for real world photography -maybe not for some people, maybe there is also sample variation, and maybe there are also some people who believe they havent focused accuratly (user error) but it was just focus shift.
For me it meant I would get less keepers than I would get with some other lenses and it lead to frustration - even though it has been only a slight effect.
I really liked the Noctilux rendering, but I could not handle that lens reliably.


Hi Jono,

hmmmm so the focus shifts along a line away from a point either front or back accordign to aperture used?

well that isnt a definitive test for me at all - in fact from my naive perspective it is a very silly test that says nothing much about anything much useful or interesting to me..- let me explain before being accused of being a bore...

My 'definitive test is about how I make a photograph using any camera and any lens. First I look at the light and decide what is possible - then I consider how much DOF I wish to use and how much separation I wish to use - that pretty much delivers the aperture and speed combo required - then I focus and ..go click

in this world the only focus shift that can occur is from a poorly aligned rangefinder or shaky hands - I have experienced both.

Now if someone can point to me the relevance of the focus shift test - that would be interesting...is my aspherical ( with focus shift I can never see) a poor lens? really? In what away would this focus shift - have ANY baring on making ANY photo ever?

Pete
 

jonoslack

Active member
Hi Jono,


Now if someone can point to me the relevance of the focus shift test - that would be interesting...is my aspherical ( with focus shift I can never see) a poor lens? really? In what away would this focus shift - have ANY baring on making ANY photo ever?

Pete
Well, it certainly had a bearing for Tim, where he got a bunch of pictures out of focus. The point being (presumably) that whereas the lens can focus spot on at f1.4, then that spot may be out of focus at f5.6 (which is certainly a 'real world' problem). His discovered the issue shooting at the Venice shindig.

Daniel's shot is interesting if you compare between f4 (where the bar in front (to the left) of the upright is largely in focus, but at f5.6 it is not.

But hey - I'm only the messenger (shoot me if you like), I can honestly say I've never touched one of these lenses - but I've talked to enough people who believe it to be the case, and I've certainly seen cases where it is the case.

I must say, I can see how it would vary theoretically between lenses - if, for instance, your lens is set up with your rangefinder to be spot on at f1.4, or worse, if it slightly forward focuses at f1.4 - then you may get problems by f4, however, if your lens is set up to be perfect at f1.8, then perhaps the increasing dof sorts out the whole range?

Certainly, my Noctilux f1 could easily be shown to suffer, but I can't say it mattered to me that much.

What is unquestionably true is that the older 35 1.4 'lux is a fine lens, and if you ain't getting focus shift issues (and I believe you Peter and Daniel - don't think for a minute that I don't!) , then it hardly matters whether it's there or not!

all the best
 
Last edited:

scott kirkpatrick

Well-known member
First part of Puts' review of the new lens:

http://www.imx.nl/photo/leica/lenses/page171/SX35FLE1.html

"focus shift is substantially removed and that the performance wide open and stopped down has improved in the close range domain from 1 meter to 3 meters"
If you read Puts' article and inspect his numbers carefully, the numbers do not quite agree with the words. You have to ignore that he does not define the "quality factor" numbers which he reports, and the typo in which he reports an accuracy of +- 0.01 mm but means +- 1 cm. His experiment is to focus carefully and then move the camera back and forth 1 cm per step and record the "quality" of the test target image, taken on an M9. It's not clear whether a + movement which improves the image is moving the camera closer to the target, so that this uncovers front focus, or the reverse. I'll assume that + in his table means that the camera moves closer to the target. The new lens at f/1.4 then shows sharp focus at an indicated 135 cm actually occurs from 1.5 cm in front of the target to 3.5 cm in back of the target, with the sharpest plane just over 1 cm in back of his target. At f/2.8, the sharpest focus appears to occur at 4.5 cm in back of the target. While Puts correctly points out that the shift is about 3% of the focus distance, 3 cm is more than the depth of a nose -- the sort of focus error that drives some of us nuts.

He compares with the previous 35/1.4-asph, for which the plane of sharpest focus shifts from 2.5 cm front-focused wide open to at least 5 cm back-focused at f/2.8 in this experiment (his table shows "quality" still increasing at 5 cm). So the focus shift has been more than halved, but not eliminated.

The more important news is that pixel-peepers will see improved image clarity at useful distances, like 1-4 m, at least if his "quality factor" scale is meaningful.

scott

PS, if it turns out that Puts' + table entries mean that the camera was moved away from the target, then simply switch the words "back" and "front" everywhere in what I wrote above.
 

Paratom

Well-known member
.....
Certainly, my Noctilux f1 could easily be shown to suffer, but I can't say it mattered to me that much.

What is unquestionably true is that the older 35 1.4 'lux is a fine lens, and if you ain't getting focus shift issues (and I believe you Peter and Daniel - don't think for a minute that I don't!) , then it hardly matters whether it's there or not!

all the best
I totally agree-as long as one doesnt see or feel there are any issues than I wouldnt search for an issue.
 

glenerrolrd

Workshop Member
Focus shift on this lens is an issue and it does affect the sharpness of your images. Its just darn hard to identify. The fence post is a good example. First look very carefully at the absolute focus point....the easiest way to find it is to examine the leading and trailing detail . I used to use the old 40/60 approximation ....slightly more depth of field behind the focus point. I now understand that this changes with focal length..so use 50/50 for a 35mm lens. So now you feel pretty good that you know the exact point of focus..a spot on the post. Look through the images and you can see the focus point shift. The focus point moves back as you stop down.

A good away around this is to optimize the focus at around 1.8/2.0 allowing f1.4 to be just a little front focused. This keeps the focus shift just about within the range of acceptable sharpness. Where most photographers would give you credit for nailing the focus. This is how my 35/1.4asph is calibrated. I did this with the technician at Leica NJ and I was in the shop doing the testing with him. This places the focus point at 1.4 about 1cm front at 2M ..its that close.

Two problems most tests don t show.(1) Focusing at intermediate distances ....the small acceptable errors are magnified into clean misses. How exactly can anyone say ..I never had a problem? You never missed focus..was it you or the lens...how would you know unless you tested for this. (2) Focusing on a parallel target....here any slight focus issue (calibration,photographer variable,focus shift) just results in less resolution. How would you know?

The 2nd is what Puts test is designed to show ....you can still get excellent results with a little focus shift but its not the best the lens is capable of. This is why Sean Reid takes a half a dozen captures and picks the best ..he is trying to eliminate the loss of resolution due to a focusing variable. I think what Puts is saying is that by minimizing focus shift ....you actually can improve the lens performance by a material amount .

Typically my street shooting is done in week long dedicated efforts . I shoot about 2000 captures over a week hope for 20-24 worthy images. Granted there is a lot of movement in street photography...the subject, me or both of us. Loss of sharpness due to misfocus is the number one reason to reject an image. I do think you could argue that most problems are calibration verse the focus shift inherent in the lens design...but this is difficult to test.
 

JimBuchanan

New member
Just as I said, I've never had an image (out of thousands) suffer from OOF because of focus shift shooting from 1.4 to 8. I even tested mine and found very tiny degree of shift that was covered by increasing DOF stopping down. I know this is not always true, but there are many happy users of this excellent lens who have reported the same experience as me.
I don't deny it!

The focus shift moves back as one stops down. My 35 Summicron ASPH has it, but at wide f/2.0 is right on focus. Stopping down just does cover the focus shift.

The 35/1.4 ASPH is worse. In fact, I would presume that there could be a spot at f/2.0 that could be out of focus that was in-focus at f/1.4. Add to that the variability of the focusing mechanism, and that is what all the talk is about.

Surely, the probability of focus, with all the varibles, is in play somehow, here.
 

JimBuchanan

New member
Peter,
I have read that the chrome versions for whatever reason were less prone to focus shift than the regular versions.
Excuse me, but that is ridiculous.

The chrome versions, being made of more brass than the black aluminum versions, could have better feel/action, but thats it.
 

JimBuchanan

New member
Now if someone can point to me the relevance of the focus shift test - that would be interesting...is my aspherical ( with focus shift I can never see) a poor lens? really? In what away would this focus shift - have ANY baring on making ANY photo ever?
Pete
Does anyone here know how to measure focus shift?

M8/9 on a tripod with focus target at 1 meter or so. Could be a page of text at a 30 degree angle.

Adjust the focus point to a line of text at wide open aperture, by exposure/inspect on LCD, etc.

Then take exposures stopping down one stop at a time.

view at 100%, and you will have your answer.
 

D&A

Well-known member
In general, I've found most Leica 35mm Lux ASPH lenses are adjusted one of two ways.

1. The first is basically as Roger described, where the lens is adjusted to slightly front focus at f1.4 and possibly just minisule amount at f2. Upon stopping down any further, the growing size of depth of field will generally cover any rearward focusing shift through f8 etc...keeping the subject weel within this zone of focus.

2. The second way I've seen these lenses set up is as Jono described, where the lens focus is adjusted to be "dead on" at f1.4, putting the subject well withing the depth of field at this aperture. Upon stopping down, the focus point shifts rearward and the subject quickly exists (is outside) the depth of field by f2 and continues to do so till around f6.7-f8.0. The larger depth of field that results upon stopping down the lens isn't large enough while shifting rearward to keep the subject in the zone of focus (until almost f6/7-f8).

Most shooting film, don't notice focus shifts with this lens since they are not examining at high maginification. Likewise, many of those shooting digital also don't always notice the focus shifts unless they are examining images closely, espcially 100% crops.

There are of course variations of the senarios I described, but of all the 35mm Lux ASPh's I've tested, most fall generally into these two catagories, when tested very carefully in carefully set-up and controlled tests.

Dave (D&A)
 

glenerrolrd

Workshop Member
Excuse me, but that is ridiculous.

The chrome versions, being made of more brass than the black aluminum versions, could have better feel/action, but thats it.
Keep in in that most users can t separate focus shift from calibration issues. If the lens focus point is slightly back focus and then focus shift kicks in you have a back focus variance that can be greater than your perceived DOF.

The tolerances considered(by Leica) as "in specification" allowed for a more forgiving film assumption. Many of the summiluxes tended to be at one end of this range (tending toward back focus). To stay within this range the summiluxes were often individually shimmed. This caused Leica fits when they started to change out mounts in order to roll out the 6 bit coding. A summicron by contrast could exchange a mount with little impact on the calibration.

I am not certain exactly why ....but pulling the focus point forward is difficult on many of the summiluxes. Those with brass mounts seem to have been easier to calibrate toward the middle of the range.
 

PeterA

Well-known member
Jono - I am not in the business of shooting anyone and am quite relaxed really about the whole topic. I think i have sold two Leica M lenses a few years ago and t this day regret it..and certainly not defending the poor old sick 35mm lux for personal gain or profit reasons..

TEB - some say teh chrome version is better - I dont know - I just happen to have a chrome version..which I buy in foal lengths and type whnever I can because the just look sexy

I do know how to do a focus test - just liek I know ( theoretically how to jump off cliffs , and ride unbroken horse the thing is as you get older - and I am officially middle aged now passing 40..you get to be more particular about use of time

the whole premise of the focus test is so funny and I mean o hilarious to me I just larf

lemme see know I use a tripod focus on a page with letters or a ruler with lines ( pity no one builds sloping brick walls for focus test huh) and then shoot the same shot @ various apertures..


then..pt files into computer and check m out @ 100% or 200% or whatever..and wala!

ahhahahhahahah

so funny..

versus my hand hold shoot at aperture that you wish to use given the varables and outcomes that you need to consider ..focus and click.

one thing has got somethign to do with photography - the other thing is just ..ummm

funny.

:ROTFL::ROTFL:
 

jonoslack

Active member
Jono - I am not in the business of shooting anyone and am quite relaxed really about the whole topic. I think i have sold two Leica M lenses a few years ago and t this day regret it..and certainly not defending the poor old sick 35mm lux for personal gain or profit reasons..

the whole premise of the focus test is so funny and I mean o hilarious to me I just larf

lemme see know I use a tripod focus on a page with letters or a ruler with lines ( pity no one builds sloping brick walls for focus test huh) and then shoot the same shot @ various apertures..


then..pt files into computer and check m out @ 100% or 200% or whatever..and wala!

ahhahahhahahah

so funny..

versus my hand hold shoot at aperture that you wish to use given the varables and outcomes that you need to consider ..focus and click.

one thing has got somethign to do with photography - the other thing is just ..ummm

funny.

:ROTFL::ROTFL:
Wellllllll I rather agree with you (until we get tilting brick walls):)
However, if you aren't going to test for it, I don't think you can really say you haven't got it?. You can certainly say
it isn't important, or
it doesn't affect me, or
only sad bastard pixel peepers would care
(come to think of it, I think you just said that :ROTFL:)
 

glenerrolrd

Workshop Member
Wow this thread sure degenerated into total BS. Most photographers would prefer to have something sharp in their images(some exceptions of course). We buy ultra expensive M glass because of its high performance and would reasonably like to get the best out of it. In general its reasonable to seek a high level of IQ . Focus shift affects all three. How is that funny or did you mean to say "a stupid waste of time" ?

The most significant controllable variable in an M systems IQ is focus point. Understanding your equipment and how it performs is an important element in being able to control your focus point. Having your equipment calibrated to perform to its specification, optimizing your vision thru diopters/magnifiers/contacts/glasses so that you can see to focus and practicing/or shooting enough so that your technique is a routine are the three cornerstones of achieving accurate focus.

The thread was about the new 35 lux. One of the benefits that Leica is touting is the lack of focus shift ...an improvement over the existing 35 lux .

E. Puts does a nice job of showing you to what extent focus shift at different apertures impacts IQ.

If you subscribe to Reid reports you will see that he struggles to test M glass. The differences in performance can frequently be thrown off by even minor focus differences. The zeiss lens is a dog not because of its design but because its focus point was off by the smallest amount.

Diglloyd another subscriber test site also speaks to the issue of focus shift in certain lenses and the impact on his lens tests.

So what it doesn t seem to show up much? Placing a relative value on minimizing focus shift depends ,of course , on your photography requirements but I think its fair to say that its an improvement that will be appreciated by many.
 

JimBuchanan

New member
Keep in in that most users can t separate focus shift from calibration issues. If the lens focus point is slightly back focus and then focus shift kicks in you have a back focus variance that can be greater than your perceived DOF.

The tolerances considered(by Leica) as "in specification" allowed for a more forgiving film assumption. Many of the summiluxes tended to be at one end of this range (tending toward back focus). To stay within this range the summiluxes were often individually shimmed. This caused Leica fits when they started to change out mounts in order to roll out the 6 bit coding. A summicron by contrast could exchange a mount with little impact on the calibration.

I am not certain exactly why ....but pulling the focus point forward is difficult on many of the summiluxes. Those with brass mounts seem to have been easier to calibrate toward the middle of the range.
So, the chrome versions are easier to adjust focus than the black versions? OK, but the focus shift is still there. The referenced link to this topic dosen't reveal any meaningful info on the subject.

I'm OK with moderate focus shift, as long as 1 or 2 stops down, I don't loose the point of sharp focus (the widening of DOF covering the shift). The 35 Lux apparently has more, where one does loose the point of sharp focus (no firsthand experience). The common workaround is to adjust the point of sharp focus forward a bit to compensate.
 

jonoslack

Active member
HI Roger
So what it doesn't seem to show up much? Placing a relative value on minimizing focus shift depends ,of course , on your photography requirements but I think its fair to say that its an improvement that will be appreciated by many.
I'm not sure that it's BS (although the quality of typing does leave a little to be desired :). I'm sure that most of us would agree with Peter that the pictures are more important than the tech . . . and it took quite a long time to realise that the lens did exhibit focus shift.

I learned to deal with the focus shift on the Zeiss 50 sonnar, but now I'm using a 50 'lux asph it's very nice not to have to think about it.

Seems to me to be a worthy upgrade, and one I'm saving my pennies for.

all the best
 
Top