Review: Leica Noctilux 50mm f1.2

[adan]

Advertisement (gone after registration)

7 hours ago, jonoslack said:

I'm going to do some focus shift tests when I get a minute. I haven't noticed it as a problem, but I would have thought that it was inevitable without a floating element.

Jono, Jono, Jono! You know better than that!

A floating element is for correcting close-up performance (at any aperture) - it compensates for the differing mathematics of parallel (near effective infinity) and diverging (close-up) light rays. It has nothing to do with correcting focus shift.

Focus shift is caused by different bundles of light rays predominating as one stops down to different apertures and eliminates the outer ones. Has nothing to do with a floating element.

Although a myth to that effect has grown up around the 35 Summilux FLE - but it is just a fan-boy myth. What Leica actually says is:

Quote

In comparison with its predecessor, the new LEICA SUMMILUX-M 35 mm f/1.4 ASPH. now incorporates a floating element: the lenses behind the aperture blades are constructed as a floating group that changes its position relative to the front lens group during focusing to ensure that the LEICA SUMMILUX-M 1:1.4/35 mm ASPH. achieves equally outstanding imaging performance at closer focusing distances.

Page one, paragraph one: http://www.summilux.net/m_system/images/Summilux35Asph-2010.pdf

No mention of focus shift whatsoever.

What fixes focus shift is taming spherical aberration by way of different glass or different curvatures (see: C/V Nokton 35mm f/1.4 version II. A new non-floating glass element, but not a floating element in sight ).

Edited February 1, 2021 by adan

[astrostl]

It helps me to remember that focus shift is a spherical aberration and that aspherical elements try to address them.

That isn't to say all aspherical lenses are fully devoid of focus shift, but the situation is usually improved.

Edited February 1, 2021 by astrostl

[Advertisement]

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

[UliWer]

vor 59 Minuten schrieb astrostl:

That isn't to say all aspherical lenses are fully devoid of focus shift, but the situation is usually improved.

I don't think you can generalize this. Aspherical elements are not always helpful for all sorts of optical flaws. The 35mm Summilux asph (pre FLE) had rather modern aspherical elements - but more focus shift than the non-aspherical Summilux.

You can take it as a rule of thumbs that you'll see focus shift  if the optical design aims at high resolution at large opening and hasn't got (or doesn't use) the means to correct the spherical flaws of the optical elements appropriately. The early version of the 1.5/5cm Zeiss Sonnar for the old Contax had - for it's age - tremendous resolution fully opened, but also massive focus shift. They reccomputed it some years later with a slight loss of resolution but the focus shift was tamed. 

Edited February 1, 2021 by UliWer

[adan]

Correct - an ASPH surface is just a tool, that can do various things depending on where it is in the lens. And sometimes a combination of various things - it is just up to the lens designer to fiddle with the power, the glass and the position to solve a particular problem.

Sometimes all it does is reduce the number of required elements (less flare, less weight, less cost - similar performance otherwise). It's been said that one ASPH element can replace two spherical elements, thusly:

Two cemented elements with slightly different indices of refraction. First glass A and then glass B. At the edges the group is mostly glass B, while at the center the group is mostly glass A. The summed power of the lens changes from the edge to the center. (Sometimes called a Merté surface, in microscopy).

()(

Now, a single element with one ASPH surface, which also changes the power of the lens from the edge to the center.

(}

[farnz]

16 minutes ago, adan said:

(}

Aside (and apologies to Andy): does anyone else think this looks like someone wearing a hood?

Pete.

[jonoslack]

15 hours ago, adan said:

Jono, Jono, Jono! You know better than that!

A floating element is for correcting close-up performance (at any aperture) - it compensates for the differing mathematics of parallel (near effective infinity) and diverging (close-up) light rays. It has nothing to do with correcting focus shift.

 

Thank you for the correction Andy! Apparently I didn't. I had simply inferred it from the fact that the FLE had less focus shift than its Asph predecessor but I ought to know better than to conflate the two.

Still, that does make it more worthwhile checking for focus shift on the new Noctilux - When I have a minute I'll give it a go!

Best

Jono

[FDS]

Advertisement (gone after registration)

Regarding post no. 125:

A bit of a Rorschach test?  Rather than a 'hoodie', I see some resemblance to a moustache and and open mouth if viewed from the left side and also the top curvature of a head and a stylised wing-like 'monobrow' if viewed from the right side.  

In 'normal' times, my wife would say that I really must get out more.  

 

 

 

[jonoslack]

On 1/31/2021 at 8:34 AM, elmars said:

One question, Jono: What ist with focus shift? Did You test it? Nearly all older lens constructions, especially the fast ones, suffer from it.

Hi there

Right I’ve done a little focus shit testing at about 2 metres, about 4 metres and about 30 metres. 

Focusing at f1.7 and then stopping down. There may be a little tendency for the point of perfect focus to shift forwards a little, but the original point of focus stays properly sharp at all apertures. Which is, of course, what really matters. It would be interesting to compare it to the original Noctilux 

All the best

Jono

 

[pedaes]

15 minutes ago, jonoslack said:

shit

Nice little Freudian😀

[PhotoEd]

12 minutes ago, jonoslack said:

Hi there

Right I’ve done a little focus shit testing at about 2 metres, about 4 metres and about 30 metres. 

Focusing at f1.7 and then stopping down. There may be a little tendency for the point of perfect focus to shift forwards a little, but the original point of focus stays properly sharp at all apertures. Which is, of course, what really matters. It would be interesting to compare it to the original Noctilux 

All the best

Jono

 

Thanks for testing!

I was reading into the lens a bit more and it looks like the new Noctilux is more different to the original than we may think. In the 'Leica M-lenses' book from Leica-Guru Erwin Puts I found some info about the original lens. The MTF curves look quite different compared to the new edition. Normaly I don't care much about these, because it's all about the looks of the pictures a lens produces. But still, I found it remarkable. The first one with the red lines is the present edition.  

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!

[pedaes]

1 hour ago, PhotoEd said:

The MTF curves look quite

I think, stress think, Leica graphs are calculated and Erwin used a variety of source, including Zeiss test facilities. I am sure glass types and coatings will be different too.

Edited February 2, 2021 by pedaes

[pedaes]

55 minutes ago, pedaes said:

I think, stress think, Leica graphs are calculated and Erwin used a variety of source, including Zeiss test facilities. I am sure glass types and coatings will be different too.

 

Edited February 2, 2021 by pedaes

[NigelG]

Perhaps my understanding of MTF curves is lacking (likely!) but I’m intrigued to see the values for the new lens descend below 0% in the posted charts?

[PhotoEd]

2 hours ago, NigelG said:

Perhaps my understanding of MTF curves is lacking (likely!) but I’m intrigued to see the values for the new lens descend below 0% in the posted charts?

Yes, shows some funny curves, doesn't it? And the messured aperture starts with 1.3 instead of 1.2. Perhaps that proves to be a typo. 

[wattsy]

I think the complicated looking MTF chart for F1.3 essentially translates to "this lens is a bit shit" – at least when judged technically. It's the kind of result that would normally be laughed or sneered at were it not a £6.5k Leica lens.

Edited February 2, 2021 by wattsy

[jonoslack]

3 hours ago, pedaes said:

Nice little Freudian😀

Haha - not Freudian, but quite funny!

[FrozenInTime]

34 minutes ago, wattsy said:

I think the complicated looking MTF chart for F1.3 essentially translates to "this lens is a bit shit" – at least when judged technically. It's the kind of result that would normally be laughed or sneered at were it not a £6.5k Leica lens.

yep, 99% of us would be better served buying the Voigtlander 50/1.2 

[jonoslack]

34 minutes ago, wattsy said:

I think the complicated looking MTF chart for F1.3 essentially translates to "this lens is a bit shit" – at least when judged technically. It's the kind of result that would normally be laughed or sneered at were it not a £6.5k Leica lens.

Well, I think it might be laughed or sneered at anyway, but isn't it all about the look and feel with a lens like this? I like it! These are both shot at f1.2

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!

[jonoslack]

Just now, FrozenInTime said:

yep, 99% of us would be better served buying the Voigtlander 50/1.2 

Of course - if you want a well designed modern lens at a great price . . . . 

But I really don't think that's why you would be buying the 50 f1.2 -

[Advertisement]

Hallo,

Look here: Leica M

[mmradman]

36 minutes ago, wattsy said:

I think the complicated looking MTF chart for F1.3 essentially translates to "this lens is a bit shit" – at least when judged technically. It's the kind of result that would normally be laughed or sneered at were it not a £6.5k Leica lens.

The obvious “Emperors New Clothes” everybody seem to admire.

Any self respecting fast Chinese lens being sold at £300 would die of shame with such “bottom of the coke bottle” quality.