Leica Apo-Summicron-SL2.0/35mm - average performer?

[digitalfx]

Advertisement (gone after registration)

I find it odd that they tested the Leica on a Panasonic body, yet both the Sony and Canon lenses were tested on their respective native bodies. 

 

[Advertisement]

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

[snooper]

I Love it. I come from Cron M 35, and I'm more impressed by the SL version. The only downside being the weight of course.

[Mr.Q]

14 hours ago, jaapv said:

Which only means that such an aggregate score is completely meaningless, and as DXO does not reveal their methods of scoring and criteria, the only place to put the final figure is the rubbish bin.

I agree, I don't pay attention to the scores either.  The interesting bit in there is the comparison of the 35 SL to the Sony and Canon lenses at F5.6 and beyond. The Sony and Canon lenses are tested to have higher PEAK sharpness closed down. That's important to folks that prefer to maximize the performance of their lenses. I've always had a tendency to shoot from F5.6 to F8, whenever I can.

[jaapv]

There we go again (not you, but DXO ) - sharpness is not an optical property thus not expressible by a number or value... Are we talking about resolution, contrast, microcontrast, at what image frequency, etc., all factors that determine the subjective thing that blogs call sharpness?  On top of that the Leica lens was not even mounted on a Leica camera, as opposed to the other lenses, which means that the sensor was not designed to match the lens performance. In other words, the conclusions (if we can even call them that) can not be trusted and are possibly complete rubbish. )

[Mr.Q]

No offense to you either, but perceived sharpness is quite obvious to me when comparing 2 images. The details are either clear or fuzzy. You may call it subjective, but I trust that a trained eye can distinguish a sharp picture from a soft one, like how most people can distinguish a hard fabric from a soft one.

If the Leica lenses do not perform optimally on a Lumix body and vice versa, the L-mount alliance is a complete failure imo.

[Simone_DF]

5 hours ago, jaapv said:On top of that the Leica lens was not even mounted on a Leica camera, as opposed to the other lenses, which means that the sensor was not designed to match the lens performance. In other words, the conclusions (if we can even call them that) can not be trusted and are possibly complete rubbish. )

On which camera should DxO test the Sigma lenses? The FP?

[jaapv]

Advertisement (gone after registration)

2 hours ago, Mr.Q said:

No offense to you either, but perceived sharpness is quite obvious to me when comparing 2 images. The details are either clear or fuzzy. You may call it subjective, but I trust that a trained eye can distinguish a sharp picture from a soft one, like how most people can distinguish a hard fabric from a soft one.

If the Leica lenses do not perform optimally on a Lumix body and vice versa, the L-mount alliance is a complete failure imo.

Lenses in general often perform best on their native bodies. How often have I not read on this forum: "M lenses perform excellently on the SL, but the M10 has the edge nonetheless?" - and Leica designed the SL to handle M lenses as well !  People tend to forget that an imaging system is made up out of three components: lens, sensor, software. All three are optimized to produce the image. 

The L alliance means that three brands share a lens mount and essential software pertaining to the system. But they have not agreed to sell rebadged versions of the same cameras and lenses. Why should they? They could just as well have decided to become one brand in that case.

All components are built to the brand-specific design philosophies. Panasonic lenses will do excellently on SL and future Sigma bodies, but they will still  be matched best to Panasonic bodies. Leica will do best on the SL, but still be an upgrade for a Sigma or Panasonic lens on their respective bodies. Etc., etc..
One key factor will be the microlens, and in Sigma's case, total sensor design.
Lenses have a specific colour transmission as well. It even varies per lens, but much more per brand. The Bayer filter and firmware interpolation and interpretation are matched to the lens colour signature. Lenses perform differently at different wavelengths. It is that subtle.

It is enlightening not only to read the article by Lensrentals linked to above, but even more so to read the discussion in the comments. It will make clear why a simple resolution test and MTF curve are blunt and sometimes misleading instruments for judging lens performance and, incidentally, why images from Leica lenses are often superior, despite similar or sometimes even lesser test results.

It is not for nothing that Sigma did not slap L mounts on their existing lenses but is developing a dedicated L- mount range.
In the past it was different. All brands used the same film, the lens was the only real differentiator.
 

Conclusion of the rant: The L alliance serves to ensure compatibility between three brands, enables the users to mix and match lenses and camera bodies with the guaranty of good results, but does not forbid the participating companies to compete on the the ultimate quality/price performance of their own system. 

[jaapv]

2 hours ago, Simone_DF said:

On which camera should DxO test the Sigma lenses? The FP?

They shouldn't do silly tests like this at all. 

[thighslapper]

3 hours ago, Mr.Q said:

No offense to you either, but perceived sharpness is quite obvious to me when comparing 2 images. The details are either clear or fuzzy. You may call it subjective, but I trust that a trained eye can distinguish a sharp picture from a soft one, like how most people can distinguish a hard fabric from a soft one.

If the Leica lenses do not perform optimally on a Lumix body and vice versa, the L-mount alliance is a complete failure imo.

If we are talking about lenses in this price/quality bracket I am very dubious that anyone can tell them apart in normal usage.

Surely there is a point for a given sensor resolution where additional sharpness or resolving power gives you minimal if any differences in the resulting image. Karbe has been quoted as stating that the current Leica lenses are easily good enough for 100mpx sensors. 

...... and yet again it depends entirely on the medium used to view them..... anyone with a very high resolution monitor can pixel peep the corners and complain about poor resolving power but it will never show up in an A2 print and probably not even on A1 even when viewed closely. All my recent images with the S1R and SL2 with a variety of Leica and Panasonic lenses, zooms included, look perfectly sharp on a 5K iMac. I cannot see how they could be sharper or why I would need them to be sharper. If I use sharpening in processing it is for local emphasis in low contrast areas or with a high level of masking. 47mpx images with most Leica lenses just don't need it. In fact those serious about printing images use 2K monitors deliberately for processing as they more closely depict the degree of sharpness that will appear in prints. 

I have the 75/2, 50/2 and 50/1.4 primes which are as near optical perfection as I will ever need I have every confidence that the 35/2 .... if I decided I needed one....  would perform at least as well. The SL zooms benefit from a little additional LR sharpening. The Lumix 70-200, like the SL primes, needs  no additional sharpening in LR at all. 

There are no doubt a couple of dozen 35mm lenses out there that will give you images on a 40+mpx camera that you be hard pushed to find fault with in normal usage once processed. If you have the money and want complete system compatibility and the very best overall quality regardless, then fair enough get a 35/2, but arguing that it is significantly better or worse than the current competition in terms of the final image is delusional. 

[Chaemono]

vor 15 Stunden schrieb Mr.Q:

I agree, I don't pay attention to the scores either.  The interesting bit in there is the comparison of the 35 SL to the Sony and Canon lenses at F5.6 and beyond. The Sony and Canon lenses are tested to have higher PEAK sharpness closed down. That's important to folks that prefer to maximize the performance of their lenses. I've always had a tendency to shoot from F5.6 to F8, whenever I can.

DxOMark is lying that there is a built-in profile for distortion corrections and they are, therefore, likely lying about the rest.

[caissa]

11 hours ago, Mr.Q said:

No offense to you either, but perceived sharpness is quite obvious to me when comparing 2 images. The details are either clear or fuzzy. You may call it subjective, but I trust that a trained eye can distinguish a sharp picture from a soft one, like how most people can distinguish a hard fabric from a soft one.

If the Leica lenses do not perform optimally on a Lumix body and vice versa, the L-mount alliance is a complete failure imo.

Yes, the L-mount is a complete failure (because of DxO measurements ?!) and btw. the Hassy X1D is the greatest camera of the world.  How come that I expected to hear that again ? Somehow it reminds me of Latin lessons at school and of "ceterum censeo Cathaginem delendam" (Cato).
But their forum must be terrible, why else are all the restless fanboys here instead of there ?     🐿️

Edited January 25, 2020 by caissa

[digitalfx]

12 hours ago, Mr.Q said:

 

If the Leica lenses do not perform optimally on a Lumix body and vice versa, the L-mount alliance is a complete failure imo.

The lens performs quite nicely on a alliance body as this test demonstrates.(is a DXO score of 39 "terrible"?)..the point being they will always perform optimally on their native body. And if you are going to offer up a test...then lets compare apples to apples. That wasn't done here.

Also keep in mind, the L Mount alliance is new, and so is the lens...

[jrp]

There is clearly something awry with the DXO tests.  To rate this lens the same as the Sony / Zeiss 35mm f2.8 (which is not a bad lens) makes no sense, when one compares actual pictures.  The latter is not a bad lens for the £$€ but is nowhere near as well corrected as the Leica (and I have both).  The tester must have been using only the measuring instruments + Excel rather then his/her own eyes.  The Leica lens produces clean, sharp, colourful images.  Faster lenses (from Leica or whoever else) will always produce more aberrations.  The fact that this lens wide open is better than faster lenses stopped down is a feat in itself.  Have Leica sacrificed maximum resolution for better wide open performance?  I don't know, but I do know that with other 35mm lenses, I would often discard pictures that were a bit grungy, one way or another.  Eg, trying to use a fast wide M lens on a Sony body.  But I have still not got used to how reliably usable the images are with this (and the other Summicrons, for that matter).

[Donzo98]

This is on a Panasonic S1 body... @ F2, focus was on her lashes. At 1:1... it's INSANE

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!

Edited January 25, 2020 by Donzo98

[Mr.Q]

@jaapv Again, I don't disagree with what you are saying. But was that true with the MFT alliance as well? Do Panasonic and Olympus lenses perform better on their native bodies? Because I've never read anything like that reported before. And if that's true about the L-mount alliance then it needs to be tested and reported. How much worse do Leica lenses perform on Panasonic/Sigma bodies, and vice versa? That data could sway a lot of purchase decisions. Decisions that weren't common in the MFT world, I might add.

@thighslapper Well, given the price for a slowish F2 lens, and the hype surrounding it, I'm sure most folks thought the 35SL was the new Otus --- the gold standard for reference lenses. When the Otus was released, it was head and shoulders above anything out there. Karbe blatantly said that it was the best Leica ever.  So if the difference between the 35SL and some random Sony/Canon lenses are minimal, I'd consider that a disappointment. 

@Chaemono Lie is a strong word. Any site could be wrong about things with some errors here or there. I consider most of the information provided by DXOMark  to be reliable. Their DR numbers closely resemble the PDR numbers provided by Photons to Photos (the #1 source for sensor performance).  In fact, some of the data on Photons to Photos is derived from DXOMark.

@caissa Uh....wut? I'm here because I shoot with Leica products, including a SL and 24-90. 

@digitalfx Fair enough, but I'd say this test definitely introduced a new question --- "How much worse do the SL lenses perform on Lumix bodies?"  Hypothetically, if the DXO score of the 35SL is a 50 on a SL2, would potential buyers still consider the S1R as an alternative to the SL2?

Edited January 26, 2020 by Mr.Q
typo

[Chaemono]

vor 4 Stunden schrieb Donzo98:

This is on a Panasonic S1 body... @ F2, focus was on her lashes. At 1:1... it's INSANE

...and a waste on 24 MPx.

[Chaemono]

vor 1 Minute schrieb Mr.Q:

Lie is a strong word. Any site could be wrong about things with some errors here or there. I think most of the information provided by DXOMark to be reliable. Their DR numbers closely resemble the PDR numbers provided by Photons to Photos (the #1 source for sensor performance).  In fact, some of the data on Photons to Photos is derived from DXOMark.@caissa

Any site that tests gear and makes up that a lens needs built-in profiles/opcodes for distortion correction is spreading blatant lies.

[rpittal]

18 hours ago, Mr.Q said:

No offense to you either, but perceived sharpness is quite obvious to me when comparing 2 images. The details are either clear or fuzzy. You may call it subjective, but I trust that a trained eye can distinguish a sharp picture from a soft one, like how most people can distinguish a hard fabric from a soft one.

If the Leica lenses do not perform optimally on a Lumix body and vice versa, the L-mount alliance is a complete failure imo.

I am happy with all of my SL lenses (3 zooms, 50lux and 90sum) on my S1R; hopefully I will be able to get an SL2 and see if I am even happier then!

[Advertisement]

Hallo,

Look here: Leica L-Mount

[Mr.Q]

1 minute ago, rpittal said:

I am happy with all of my SL lenses (3 zooms, 50lux and 90sum) on my S1R; hopefully I will be able to get an SL2 and see if I am even happier then!

According to what folks are claiming in this thread, you should experience a substantial jump in IQ 😉

[Donzo98]

2 hours ago, Chaemono said:

...and a waste on 24 MPx.

Yup... I'm sure. It is only works on the SL2 right?? Leica can do no wrong. A total waste of money on the S1. 

I hope Leica is paying you well... you are their best promoter.