Pages: [1]   Go Down

Author Topic: Potential and limitations of new mirrorless digital camera systems  (Read 3221 times)

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600

Another thread has veered off-topic onto discussing the potential and possible limitations of the new shallower and narrower lens mounts for “purely electronic cameras”—ones with electronic sensors and viewfinders. Maybe the best way to avoid a recurrence is to have a thread for it, so here goes.

The Sony E mount when used for 36x24 format is egregiously narrow compared to all the others: Micro Four Thirds, Fujifilm X and G, Canon EOS-M and RF, Leica L, Hasselblad XCD, Nikon F, and even Sony E in its original usage with 24x16 format. Basic optics shows that this has some adverse effects, such as more vignetting with lenses of very low aperture ratios or the need to use a lower exit pupil to mitigate that vignettting — but there is little or no evidence of how much practical effect this will have, especially now that vignetting can to some extent be corrected in software.

My best guess is that there will be at most modest effects with some extremely bright lenses; what do you all think?

And can we avoid disparaging those who disagree with us as “fanboys” or gullible victims of marketing propanda?
« Last Edit: March 12, 2019, 11:42:41 am by BJL »
Logged

Telecaster

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3686
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #1 on: March 12, 2019, 03:52:53 pm »

Sure. A larger diameter mount/throat offers genuine theoretical and possibly also real-world benefits. Leitz and later Leica have needed to "*square the circle" to some degree with their speedier M lenses due to that mount's small-ish size. Sony will have to do the same should they decide to compete in this latest round of the super-speed sweepstakes.

Personally I do think the whole thing is more about marketeering than photography, just as it was in the late '50s–60s with sub-f/1.4 50mms. At least on the part of the camera/lens makers. I own two 50mm SLR mount f/1.2s, an M mount f/1.1 and three m43 f/0.95s, so I guess this makes me a speed lens fanboy.  :D

-Dave-

*The metaphor would work better in this context if it were "circle the square."  :)
« Last Edit: March 12, 2019, 03:59:21 pm by Telecaster »
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #2 on: March 12, 2019, 06:25:26 pm »

I believe that there are two ways to look at this:
- what is Sony not going to be able to do due to their smaller/less shallow mount,
- what will Nikon/Canon/Pana-Leica-Sigma/Fuji/Hasselblad be able to do compared to their historical mounts if any,

The first thing that jumps to the eye is that all the other manufacturers having released mounts after Sony have taken the same design decision to go for wider and shallower (to a lesser extend). And significantly so.

I don't think that any here have the actual expertize in optics to be able to give a fully credible answer to the 2 questions above, but we got elements of answer through marketing materials published mostly by Canon. They indicate that the wider mounts should reduce the optical complexity of designs needed to deliver good image quality in the very corners of the sensor.

In other words you can either design simpler lenses (smaller, lighter,...) to stay at the current level of image quality, or keep the current level of lenses complexity to reach higher levels of image quality in the corners.

Now, that's not the only part to the story. One key aspect of lens design these days is the look of bokeh highlights (presence or absence of aspherical elements grounding marks) that is impacted by technological aspects not directly related to mount dimension (although there may be a need for more aspherical elements to compensate for mount limitation,...). Sony seems to have made great advances on this. We know that Canon hasn't (28-70 f2.0 is pretty awful from that standpoint) and Nikon probably also hasn't (the 58mm f0.95 I saw had the issue to a lesser extend but still visible). This alone may be a good reason for some to prefer some Sony lenses.

Cheers,
Bernard
« Last Edit: March 12, 2019, 06:30:00 pm by BernardLanguillier »
Logged

faberryman

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4851
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #3 on: March 12, 2019, 06:45:09 pm »

The proof of the advantage of a larger throat diameter is in the lenses that are actually offered for sale.
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #4 on: March 12, 2019, 06:51:57 pm »

The proof of the advantage of a larger throat diameter is in the lenses that are actually offered for sale.

Yes, that's very true.

For now we have Canon and Nikon offering high performance f1.2 or brighter designs, Sony hasn't released any such lens nor published any roadmap indicating that they might.

There are third party f1.2 or brighter manual focus lenses for FE mount, but their corner performance is not that great, certainly not up to the standards show so far in the Sony GM series of lenses.

The lack of f1.2 AF design of high quality was the most obvious materialisation of the F mount inferiority relative to the EOS mount.

Cheers,
Bernard
« Last Edit: March 12, 2019, 07:13:48 pm by BernardLanguillier »
Logged

hogloff

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1187
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #5 on: March 12, 2019, 07:19:57 pm »

Yes, that's very true.

For now we have Canon and Nikon offering high performance f1.2 or brighter designs, Sony hasn't released any such lens nor published any roadmap indicating that they might.

There are third party f1.2 or brighter manual focus lenses for FE mount, but their corner performance is not that great, certainly not up to the standards show so far in the Sony GM series of lenses.

The lack of f1.2 AF design of high quality was the most obvious materialisation of the F mount inferiority relative to the EOS mount.

Cheers,
Bernard

All this only matters to the 0.001% that actually make use of these faster, heavier and extremely expensive lenses. The rest just make amazing images with their 1.4 lenses.
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #6 on: March 12, 2019, 07:34:17 pm »

All this only matters to the 0.001% that actually make use of these faster, heavier and extremely expensive lenses. The rest just make amazing images with their 1.4 lenses.

Fair point, but that isn't related to the topic being discussed here.

It may be that the advantage of larger mount has little impact on the needs of a majority of photographers who are happy with good f1.4 lenses.

Cheers,
Bernard

hogloff

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1187
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #7 on: March 12, 2019, 07:37:22 pm »

Fair point, but that isn't related to the topic being discussed here.

It may be that the advantage of larger mount has little impact on the needs of a majority of photographers who are happy with good f1.4 lenses.

Cheers,
Bernard

Yes, same as 20fps...very few really need it.
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #8 on: March 13, 2019, 04:59:40 am »

Yes, same as 20fps...very few really need it.

Now, on the other hand the standard keeps evolving, as is being discussed in another thread.

We used to think that a 28-70mm was a great rangeo for a standard zoom lens, now we want 24-70mm lenses. We used to think that 14mm was super wide, now we want 10mm.

It wouldn't surprised be if f1.2 lenses on FF became the new standard for right prime lenses. Some would argue that it already is among portrait shooters who have been using Canon equipment. The question being how good a f1.2 prime can be. The DSLR version had a great look but were pretty poor performers wide open. The 50mm f1.2 R is very sharp wide open in the center and pretty good in the corners.

How confident are we that great f1.2 primes can be developed for the FE mount?

Cheers,
Bernard
« Last Edit: March 13, 2019, 05:02:42 am by BernardLanguillier »
Logged

Guillermo Luijk

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2005
    • http://www.guillermoluijk.com
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #9 on: March 13, 2019, 08:54:02 am »

Fair point, but that isn't related to the topic being discussed here.

If the topic discussed here wants to be minimally useful to anyone and not just sci-fi irrelevant chattering, the amount of people interested in any kind of lens must be a part of the equation.

Meanwhile real world lenses seem to contradict what some obnoxious users around seem to be very worried about or interested in confirming:

An example here:
https://kenrockwell.com/sony/lenses/24mm-f14.htm

Another example here:
https://m.dpreview.com/articles/1478096987/roger-cicala-tests-new-sony-fe-135mm-f1p8-gm-confirms-insanely-good-mtf-results

Another example here:
https://photorumors.com/2019/03/12/the-new-zenitar-50mm-f-0-95-manual-focus-full-frame-lens-from-zenit-is-rumored-to-be-announced-on-march-15th/#more-107853

This is the kind of stuff 99.9% of users living on this planet could be interested in, and with such beasts appearing in the market all this discusion about system limitations is getting even more irrelevant than it already used to be in the 'past' DSLR times (never heard you complaining of the great limitations suffered from not having f/1,2 primes on your Nikon system Bart).

Think also about the kind of stuff generalist lens designers (Sigma, Tamron, Tokina, Samyang,...) will pay attention to since they are not going to make a new lens design for every FF mirrorless system out there, but design once and then adapt.

Regards
« Last Edit: March 13, 2019, 08:58:21 am by Guillermo Luijk »
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #10 on: March 13, 2019, 12:25:23 pm »

I have never complained about the lack of f1.2 lenses on F mount because the f1.2 lenses that were available on EOS mount weren’t that impressive. In a world were some mounts enable the design of excellent f1.2 lenses the story isn’t the same.

Your assessment that only 0.1% of SLR users are interested in those very much feels like the denial of some Z mount/R mount users about the irrelevance of eye AF... In other words a clunky justification that your platform of choice has no important shortcomings.

I agree about third party lens manufacturers... but I am afraid that the one mount that differs significantly from the others (the FE mount) may end up suffering from its differences. It hasn't been the case till date because FE is both more or less open and was the only game in town for mirrorless... but we pretty much know that Sigma will probably never develop "native" lenses for FE (meaning not just DSLR lenses with a built-in converter) because whetever they will develop will be optimized for L mount. And frankly, Sigma is the only third party lens manufacturer that is consistently worth considering along with Zeiss.

It is indeed pretty unlikely that Zeiss would break their relationship with Sony and develop lenses for R/Z mount that take advantage of the mount.

But, as I am sure you have understood, the very point of this thread was to focus on the actual impacts of the mounts on lens design... not to claim once more that whatever these impacts may be are irrelevant to actual photography.

Cheers,
Bernard
« Last Edit: March 13, 2019, 08:19:37 pm by BernardLanguillier »
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #11 on: March 14, 2019, 07:32:28 pm »

If the topic discussed here wants to be minimally useful to anyone and not just sci-fi irrelevant chattering, the amount of people interested in any kind of lens must be a part of the equation.
Agreed; to put it a bit differently, I am curious first about which lenses (which combinations of focal length, aperture and such) are significantly hampered by the narrower throat, and by how much, in respects like (a) degraded edge and corner performance, or (b) greater bulk and cost needed to mitigate problem (a). Then secondly comes the question of what fraction of photographers are affected by those limitations—or cynically, what fraction of potential gear sales and profits is affected, since it could be that the "photographic 1%" who care generate far more than 1% of profits. Tentatively, I agree that there are probably no problems down to f/1.4 primes, or to anything that Nikon F-mount handles well, and if so, a large proportion of photographers (including me!) would not be affected.

But developments in optics seem to allow good performance in faster lenses like f/1.2 or even f/0.95, and zoom lenses down to f/2 or maybe even f/1.4 (see https://www.canonrumors.com/patent-canon-rf-14-21mm-f-1-4l-yes-f-1-4/), and that new sector might be important, at least for the prestige of a system.
Logged

Paulo Bizarro

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 7393
    • http://www.paulobizarro.com
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #12 on: March 18, 2019, 12:13:28 pm »

Fact:

1. Canon and Nikon are introducing f2 zooms and f1.2 primes that are better than past designs. In the case of Canon, it is not so much the mount diameter (RF is similar to EF), but the smaller flange distance. In the case of Nikon, the Z mount is benefiting from both aspects (relative to F mount).

Speculation:

1. Canon and Nikon have done so for several reasons: to establish a difference relative to the E mount; to make a bold entrance; to call the attention to E mount limitations (according to them).

2. The potential limitation of the E mount regarding narrow diameter may have very little impact on sales and actual usability of the system. Indeed, the advantages (potential) of f2 zooms and f1.2 primes comes at a cost of heavy weight and cost.

Guillermo Luijk

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2005
    • http://www.guillermoluijk.com
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #13 on: March 18, 2019, 12:51:48 pm »

In other words a clunky justification that your platform of choice has no important shortcomings.

My platform of choice? I have absolutely no platform of choice, and that's why I love mirrorless most, because it allowed to introduce the universal body concept in the digital era. I have a 1.500EUR Sony body that today won't sell by half, and decreasing. All my lenses are Canon mount (EF or FD), and can be equally used on any FF mirrorless body, so once my A7 breaks I can change my 'platform of choice' at the same cost as with staying with Sony.

It's you who has a 'platform of choice', heavily investing in gear for it, and somehow needing to make sure your Z system is a better system than any other election (???). Trying to persuade any audience about physical limitations of the Sony system, specially when Nikon 'suffered' much higher constraints for years with zero impact to them designing excellent lenses, sounds really strange.

The future of optical optimisation for image quality is not in the lens design, it's in the sensor dynamic range, in the micro lens design, in software corrections or in new optical paradigms (multi sensor image composites are SOTA in mobile devices). All this story about bulky f2 zooms and f/0,95 primes that weight a ton is just the old fashioned way to show off. I would never think of it as the way to go for any smart photographer.


But developments in optics seem to allow good performance in faster lenses like f/1.2 or even f/0.95, and zoom lenses down to f/2 or maybe even f/1.4 (see https://www.canonrumors.com/patent-canon-rf-14-21mm-f-1-4l-yes-f-1-4/), and that new sector might be important, at least for the prestige of a system.

Totally agree. Prestige is what Leica has been living on since they entered digital. Again, not for smart photographers IMO.

Regards

Telecaster

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3686
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #14 on: March 18, 2019, 04:35:27 pm »

Prestige is what Leica has been living on since they entered digital.

I'd amend this to "…what Leica has had to live on…" No affluent customer base and no marketing to same, no Leica. They don't have the resources to compete with the major players…but for now they've found a sustainable niche.

-Dave-
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #15 on: March 18, 2019, 08:00:54 pm »

It's you who has a 'platform of choice', heavily investing in gear for it, and somehow needing to make sure your Z system is a better system than any other election (???). Trying to persuade any audience about physical limitations of the Sony system, specially when Nikon 'suffered' much higher constraints for years with zero impact to them designing excellent lenses, sounds really strange.

Then why do I include the Canon R and the Leica L in the platforms that have more future proof mounts than Sony does?

Have you considered the possibility that I am trying to avoid making the same mistake twice by investing in a platform that is more limited than others? Understanding obviously that this is just one aspect to take into account. That happens to be the one we are trying to discuss in this thread.

I would disagree that the gap between the EOS mount and F was larger than the gap between Z/R mount and FE mount. The difference of angles to the corners of the sensor is in fact larger now. But history seems to repeat itself... the older mount seems to be the one with less potential. And it shouldn't come as a surprise that Nikon and Canon have paid more attention to this aspect than Sony did. This really related to their DNA and what they think is most important in photography.

I have a high amount of respect for Sony for what they have done with their mirrorless platform and the innovation to continue to unleash in terms of technology, starting with AF. But they don't really have an optical culture.

The future of optical optimisation for image quality is not in the lens design, it's in the sensor dynamic range, in the micro lens design, in software corrections or in new optical paradigms (multi sensor image composites are SOTA in mobile devices). All this story about bulky f2 zooms and f/0,95 primes that weight a ton is just the old fashioned way to show off. I would never think of it as the way to go for any smart photographer.

I don't disagree that there are many avenues for progress. I believe that lenses is one of them though. And a rather important one because photography is in the end all about the way the lens shapes the light before it reaches the sensor.

I am not sure I feel the need to been seen as a smart photographer, I'd rather be seen as someone producing beautiful images. Beauty to which the look of a lens does contributes in many cases. Leica would probably not exist anymore if that were not the case. Many Canon photographers still stick with the brand because of the bokeh of the 85mm f1.2,... example abound. All things that are hard to measure with KPIs.

But I have been stitching and DoF stacking (and to a less extend HDRing) for more than 10 years, I believe that I have demonstrated my interest in computational photography long before the majority of us, not to mention my long term focus on high DR sensors (many still complain about how much I stressed the D3X DR... because I was a Nikon fanboy obviously...)... ;)

So I understand your point and agree that Sony may overall be the best candidate to deliver value to photographers, I am just saying that the optical potential of the mount is important too and that is one
 aspect where Sony isn't as well positioned as Nikon and Canon. I do feel that there are differences. This isn't a lukewarm flat entropic world where everything is the same.

In the end it boils down to what is important for each photographer. But figuring this out starts by assessing objectively the strengths and weakness of each offering, not by denying those.

Cheers,
Bernard
« Last Edit: March 18, 2019, 09:22:51 pm by BernardLanguillier »
Logged

Dan Wells

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1044
Re: Potential and limitations of new mirrorless digital camera systems
« Reply #16 on: March 19, 2019, 01:54:32 am »

Unfortunately, I don't see Sigma producing any dedicated designs for any FF mirrorless mount yet (they have a few inexpensive dedicated lenses for APS-C E-mount and Micro 43. Sigma L-mount lenses are DSLR lenses with built-in converters, just like the FE lenses. The E/Micro 43 lenses are likely to show up in the next batch of L lenses, since they're relevant to APS-C L-mount cameras - but I haven't seen any indication of a single genuinely new lens from Sigma for L or any other mirrorless mount in years.

If Sigma (or someone else) did decide to do a new full-frame mirrorless lens, they'd probably have to make it fit a mount of 20mm depth and 47mm diameter. Any design that needs a wider mount than 47mm loses the majority of the current market because it can't fit Sony FE. A lens that has to have a shorter back distance than 20mm would lose Canon RF and L mount.

The body manufacturers (especially Nikon) have some advantages over third-party lens makers, because they only have to fit their own mount, rather than the combination of FE's narrower throat and L and RF's longer flange distance. Nikon can design right to the edges of 16mm depth and 55mm width, and they're producing some very sharp, compact lenses. The L-mount makers (51.6mm) and especially Canon (54mm) can use broader diameters than third parties trying to fit all four mounts. Sony is constrained by mount diameter, but benefits from a slightly shorter flange distance than L or RF.

Another consideration for a third-party maker is whether to try and pick up one (or both!!!) of the Fujifilm mounts, despite their different sensor sizes. X-mount is right in the middle of the others for flange distance. It's notably narrower (44mm), but it might be possible to use a narrow bayonet that would vignette on full-frame, but covers APS-C just fine. For certain lenses, the focal length would make sense on either APS-C or FF. A fast 50mm would be a nice portrait lens on X-mount where Fuji's only 50mm is f2 (there's also a 56mm f1.2, which is large enough that a FF lens might not be too much bulkier - or it might be so good that the extra bulk was worth it to many photographers), and a very fast 35mm would be a fast normal lens (an average-speed 35mm would be huge in comparison to the very compact Fujinons). If the third-party lens were either a longish telephoto or a macro lens, the crop becomes an advantage - and Fuji is somewhat short on both.

GF mount, of course, requires a lens that has extra coverage on FF - but many longer lenses do, either unmodified or with alterations only to baffling, keeping the optics the same. By the time you reach a fast 300mm lens, it's hard to design one that doesn't cover medium format - I read somewhere that many of them would actually cover 4x5" if you removed the baffles. I wouldn't be surprised if most FF designs over 100 or 150mm would cover GF with few if any alterations. The longer flange distance of GF mount shouldn't affect the longer lenses that are most likely to cover the bigger sensor.
Logged
Pages: [1]   Go Up