Peter Karbe Speech on Friday - Questions?

[michali]

Advertisement (gone after registration)

8 hours ago, Jeff S said:

Yes, OIS provides an extra half stop (from 5.5 to 6) when combined with IBIS.

https://www.reddotforum.com/content/2019/11/leica-sl2-announced-47mp-ibis-cine4k60-and-more/

Jeff

Whether the combination of OIS & IBIS actually gives you this in a real world situation, is something I've been trying to work out after my 4 month honeymoon with the VE 24-90 L & VE 90-280 L on the S1R was cut short by far too many blurry images in the field for my liking. User error?  Possibly....  For other reasons as well, I sold the S1R and went back to my SL, the problem went away.

With my SL2 expected to arrive in the next 10 days, I'm revisiting the issue.

Ming Thein's written one of the best articles  I've come across regarding this subject of OIS & IBS:

Stabilisation is good…but only up to a point. 

An excerpt from the article that deals with OIS & IBIS:

"On the subject of dual IS systems – i.e. moving optical elements moving sensor – potentially, capable of greater stabilisation effectiveness because your possible maximum displacement is increased; however, you may well see that whilst the overall structure of the image is maintained out to slower shutter speeds, the micro contrast is not – we now have two sets of moving parts to keep perfectly parallel! Of course, below a certain resolution – dependent on your hardware – it may not be visible.

When stabilisation fails, the obvious problem is we start to see motion blur in our images at shutter speed/ focal length combinations you don’t expect – usually when you would otherwise experience a crisp image even handheld. However, it can also take the form of slight double imaging when the moving elements ‘jump’ into position."

 

 

 

 

Edited December 14, 2019 by michali

[Advertisement]

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

[SrMi]

6 minutes ago, michali said:

Whether the combination of OIS & IBIS is actually gives you this, is something I've been trying to work out after my 4 month honeymoon with the VE 24-90 L & VE 90-280 L on the S1R was cut short by far too many blurry images for my liking. User error?  Possibly....  Sold the S1R and went back to my SL, the problem went away.

<snip>

 

S1R cannot use combined OIS & BIS with Leica zooms. Instead one needs to specify to use either OIS or BIS. When specifying OIS, one should get similar results as with SL (did not test it, though). I would think that OIS helps more for focal lengths above 150mm. IMO, Dual IS not only helps increase the stabilization, but also adds convenience of managing  OIS vs BIS.

[michali]

Just now, SrMi said:

S1R cannot use combined OIS & BIS with Leica zooms. Instead one needs to specify to use either OIS or BIS. When specifying OIS, one should get similar results as with SL (did not test it, though). I would think that OIS helps more for focal lengths above 150mm. IMO, Dual IS not only helps increase the stabilization, but also adds convenience of managing  OIS vs BIS.

Thanks,  I had tried different combinations and eventually settled on IBIS in camera OFF and OIS ON. 

[jrp]

And nothing for CA correction?

[verwackelt]

And was there a opportunity to ask him some of the questions that have been collected here in this thread?
Any answers ;-)?

[thighslapper]

7 hours ago, michali said:

....... however, you may well see that whilst the overall structure of the image is maintained out to slower shutter speeds, the micro contrast is not –

..... ah ..... the mysterious 'micro contrast' ...... that mythical component that appears to be invisible as far as I can see but adds 'magic' to a photograph. 

IBIS/OIS either yields an 'acceptably sharp' image or it doesn't. 

Multiple factors contribute to effectiveness and usability ..... and like every tool you need to learn its limitations and how to use it to best advantage. 

[BernardC]

Advertisement (gone after registration)

12 minutes ago, thighslapper said:

.... ah ..... the mysterious 'micro contrast' ..

Nothing mysterious about it. It's just high-frequency (fine detail) contrast, as opposed to overall contrast (the type that is controlled by paper grades in the darkroom).

It's easy to control the overall contrast in post-processing (wet or dry), but it's hard to increase micro-contrast without getting that brittle, over-sharpened look that screams "digital!" That's why many people prefer to use lenses that have that quality baked-in, at least for some types of scenes.

[scott kirkpatrick]

16 hours ago, mediafotografie said:

...today there wasn't any news - beside the lens design of all 7 Summicron-SL-Lenses

But Peter Karbe showed the quality and philosophy of the Summicron-SL-line very impressive; a wonderful event

 

 

 

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!

No news??   That's a pretty informative slide, showing how much repeated use there has already been in the SL-SC line.  That's why they don't cost $12-13K, like the M 75 and 90 Noctiluxes.

Also the commonalities seen in the 28, 24, 21 series makes me think that the 28 and 24 might appear closer together in time, with the 21 lagging because its greater differences.  The last set of wide angle bleeding edge lenses that Karbe's team designed were the 28, 24, and 21 M-Summiluxes, which appeared over a period of several years.  They look similar but are not the same size. In that triad, the 24 and 21 appeared together and are more similar.  I'd love to hear him talk about what could be obtained then and now with the current design paradigm.  (The 16-35 SL zoom may have involved a Far Eastern partner, as its patent drawings appeared there under the name of another company.)  I notice that the focus frame seems to be common across all seven lenses, even though the moving glass is not.  So this means that more things discovered later will probably be reduced to firmware corrections. 

Edited December 14, 2019 by scott kirkpatrick

[SrMi]

23 minutes ago, scott kirkpatrick said:

No news??   That's a pretty informative slide, showing how much repeated use there has already been in the SL-SC line.  That's why they don't cost $12-13K, like the M 75 and 90 Noctiluxes.

The slide compares only Summicrons. I wonder how they compare with Summilux lenses.

[Chaemono]

vor 2 Stunden schrieb jrp:

And nothing for CA correction?

No, not in the opcodes, at least.  Take a shot with the S1/R with either of the Summicron-SL lenses mounted on it of a scene that should show CA/PF.  Make s duplicate of the file. Remove the opcodes in the duplicate.  Open both files in LR.  Neither will show any CA/PF.  

[lik]

vor 2 Stunden schrieb verwackelt:

And was there a opportunity to ask him some of the questions that have been collected here in this thread?
Any answers ;-)?

I made a summary of all your questions and passed it as a written list via email.  He told me in a little personal talk that he really read it. But in terms of lenses in future he could not give any comments. 
But I also took some notes of his speech which I found interesting. I will type them down and post here. Maybe I’ll manage tomorrow - please give me some time. 

[scott kirkpatrick]

1 hour ago, SrMi said:

The slide compares only Summicrons. I wonder how they compare with Summilux lenses.

The SL-Summilux was designed in a different environment, as the only announced prime in the SLlineup.  I suspect it has the style of the M-Summilux 50 -- rendering very precisely at f/2.8 and below, but exuberantly at f/1.4, with falloff around the central region of the image.  But I have never shot with one.  The SL-Summicron line really does seem to have been planned as a single project, and was probably not green-lighted completely until Leica knew that the SL was a success, perhaps around the time that a lot of effort went into firmware 3.x.

[scott kirkpatrick]

30 minutes ago, lik said:

I made a summary of all your questions and passed it as a written list via email.  He told me in a little personal talk that he really read it. But in terms of lenses in future he could not give any comments. 
But I also took some notes of his speech which I found interesting. I will type them down and post here. Maybe I’ll manage tomorrow - please give me some time. 

I'm not surprised on the first point -- future products are seldom discussed except in the vaguest terms.  But did you get a response on the question of where IBIS and OIS make a difference?

[jrp]

I seem to remember a video of a couple of Leica guys from when the SL was being introduced that had the suggesting that OIS was switched off by default because it had an adverse effect on image quality. As usual, they offered no specifics. 

[SrMi]

10 minutes ago, jrp said:

I seem to remember a video of a couple of Leica guys from when the SL was being introduced that had the suggesting that OIS was switched off by default because it had an adverse effect on image quality. As usual, they offered no specifics. 

I remember that being said for Q, not for SL.

[jrp]

I thought that it was the SL, but why would it be one and not the other?

[SrMi]

21 minutes ago, jrp said:

I thought that it was the SL, but why would it be one and not the other?

I don't know, I am just guessing:

Q's lens is more compact and requires a lot of software distortion correction. Its corners are more 'fragile' ;-). Personally, I have not noticed any image quality degradation caused by OIS in Leica Q.

[lik]

An Evening with Peter Karbe in Wetzlar. Two hours of knowledge and messages.

A very personal report.

I noticed the event with Peter Karbe too late. It was fully booked, and only thanks to the wonderful Bodo Philipp of Leica Store Berlin and his friendly colleague Benedikt Hartmann of Store Wetzlar, I have slipped in. He was also the one who sent my summary questions from the forum to Peter Karbe and even introduced me to him in the store. He, in turn, had read everything and immediately clarified that he would refuse any answer to future planning. It became not that bad in the end.

Karbe is an incredibly nice person. One who lives his work in harmony with his own thinking. He doesn't need a sales approach. Even when he says, "You've never gotten as much value as with these new lenses." It comes as a sober statement. And when he explains that when polishing aspheric lenses, the human skills factor is crucial, because unclean polishing produces onion rings and the whole thing is a quality but also price driver, then we stop whispering.

So I try to write down everything I can catch and that I think is interesting in order to be able to reproduce it as close as possible to the spoken word later on, or at least as a sense. He announces hard boring stuff. Expectation management.

The digital evolutionary

In the first few minutes, he clears up a cliché known to Leica enthusiasts (like me): "We never slept through the digital revolution, but our priority was the Leica M. We adapted the digital technology (above all sensors) to the M and not the other way around. It just took time." It is apparently more the electronics than the optics that are the technology drivers of the current progress. And he constantly „evolutionese“ them. In other words, we could forget the finest optics that we now get for the SL without mastering all the associated technologies: "The complexity of the entire process chain was like driving backwards with a number of trailers. To do this, we had to lay new rails. That's why it took so long." Something astonishing to our ears when he says that the lenses should have become less sensitive to a higher offset of the individual lenses due to AF. Centering, closing, controlling, reopening, correcting – for example, the process with M-lenses, that would no longer be possible if everything is adjusted and glued. When asked by the audience about the quality of service with so much complexity: "Don't worry about that." You don't do it if you listen to him.

The Optics Challenger

Most Leica photographers know his statement that the new generation of lenses can easily manage 100 million pixels. Few people know the derivation except for some of the connoisseurs gathered here. I finally understood it as plausible, though not so deeply that I could explain it in detail. So for all those who, like me, feel good with just some more plausibility: "With the SL summicrons (also the Summilux SL 50 and also other lenses like the Q lens) we have increased our target by 50%(!) up to 60 lp per mm at 50% contrast. Because it used to be 40 lp before." Some manufacturers still work with 30 lp and one (no, I do not name it here) internally with 50. He casually notes that it was the work on the APSC sensor which has already begun to raise the benchmark. When this was finally successful, it was clear that there must be more for the SL system. So that's how it was.

Then the explanation for the 100 million pixel statement comes up, the Nyquist frequency for the various pixel values that are now possible. It is 172 lp per mm for 100 MP sensor at 100% contrast. Thus 50% contrast (1/ 2 Nyquist) would be 86 lp. Karbe: “This is not any problem to our lenses, even in the corners.” And then the superlative: „Our Summicron SL 35 is also our best lens because we had a lot room for 35 mm in that tube.” It handles 204 lp / mm at 50% contrast. Silence in the room. For some people, maybe not so new all this. To me it was quite new – hard but not boring at all.

There are some comments on color defects, CA etc. The lenses are corrected so that the color failure to be corrected on the computer is practical 0. Even I can see that.

Finally (thanks from me as an unfortunate S-user to the questioner), the question of S-lenses remains not unanswered. Indeed, they were expected to be 50% for 40 Lp, but: “There you have less additional magnification. The final picture result is much better at the S.“ Thank God.

By the way he says, they do only as much as really need to make software corrections of the lenses. As the Leica-JPEG's contrast are not so much too high exaggerated, some tests in magazines are dropping them. I don't care.

The sharpness-blur juggler.

He also wanted to clarify his view on the sharpness / blur reaction of these SL Crons. The sharp loss of the new SL summicron is about the optical impression of the difference of sharpness and blur. But if you look at how thick the spots are in the background shooting wide open, then a Summilux is still a difference. After so many comparisons here and elsewhere, it's good to have heard it out of his mouth. And I continue to use my M-Summilux lenses. It’s the hour of the wide open shooter: Peter Karbe wants a monitor that shows a message when stopping down: “Are you really sure?” All lens performance ambitions are wide open. "Whoever photographs with 8 should take a smaller sensor. APSC is also the better sensor if you need DOF images. And more: “Instead of stopping down, a ND filter can do in full format”. Laughter.

Alan Greenspan of optics.

Finally, the Q&A time. And it's a pleasure to see how smart the audience is asking. Nobody asks directly for future plans. And so the event becomes valuable, even if the answers are like the ones that once Alan Greenspan gave as market outlook. The question of general product strategy and whether Leica leaves the very fast lenses to other manufacturers like the one with the „S“: “When we make some faster lenses I cannot say. Now we have this project.” To ask if that would mean that they finally would come once - no answer.

The question of cooperation with the L-mount – he makes clear that this is not cooperation, but the opening up of Leica’s own mount and thus opening the market, in order to enable other good companies making good L-lenses.

No question of whether extenders come of course – but whether they are technically feasible at this high level. The answer makes everything clear to me. “With an extender, CA will also be magnified. An optical extender would be less good than a digital crop today. So, actually, we don't need it anymore.“ And: “Extender wouldn't be my ... [Smile and silence] … but I would never say we never make one.“ Understand?

The Tilt-Shift question is next: “I may not and cannot answer that.” The number of these lenses  by other producers should lead to a realistic view here.

A long telephoto lens? Of course he would like to make a long tele, he says, but if at all, and if so, when – he really doesn't know. Then it comes: “Why don't you use the 90-280 and crop it?“ And Sigma is a good company as well.

The question of whether there’s no demand for macro lenses anymore - here he makes a purchase recommendation: “probably there is further demand. I also recommend the EIpro-lens. Thank you anyway for that point. I'll pass it on.” He's showing his own macro photos. I think he'd enjoy developing a macro. That's all I could get out.

This friendly modest anti-guru makes us astonished again: He's just doing JPEG because he's too lazy to work on LR he says. Culture clash for a notorious DNG nerd like me. And besides, he's still using an M8. Among others I guess.

That’s what I noted. Unfortunately I’ve got nothing about OIS and IBIS – sorry for that. Maybe because I’m not that focussed on that topic and other participants remember more about it.

Anyway – listening to Peter Karbe is informative and enjoyable. Actually, Leica should clone him and thus save a lot of glory marketing expenses.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[nicci78]

Thanks for your transcript. 
I guess that APO-Extender-1.5x name is the CL or the SL2 in crop mode ! 
 

[papimuzo]

Thanks a lot for all the work you made for us at LUF. Impressive.

Happy picts .