SL3 Rumors

[LocalHero1953]

Advertisement (gone after registration)

3 hours ago, SrMi said:

Both Q2’s and SL2’s high ISO images can benefit from AI noise reduction tools like DxO’s DeepPRIME. If you are bothered by IQ at high ISO, I recommend trying it out.

I'm a regular user of Topaz DeNoise and Sharpen, with custom settings, not presets. They are great, but cannot fully solve the limitations of Q2 and SL2 in low light - it's too easy to get artefacts. Nor can they improve low light colour. I'd rather start with a better file from the SL2-S. 

[SrMi]

2 hours ago, Stuart Richardson said:

I think this is primarily about perspective. For you the Achilles heel of the SL2 is its low light performance, for me the Achilles heel of the SL2S is its low resolution. It would be great if we could get both (well...great for me at least. I am sure someone would find a reason to be unhappy), but at the moment it seems to be the tradeoff. The more moderate resolution bodies have better ISO performance and speed and the higher resolution bodies have a lot more sharpness, ability to crop or enlarge, and freedom from things like moiré. I think it is good that Leica has adopted this model. Clearly it makes some financial sense: you please more customers, attract more buyers and presumably there are enough customers who will buy both to even make it likely that they sell more that way too. My hope is that they continue this way. That would mean the SL3 with a high resolution sensor, and then an eventual SL3S taking advantage of a more moderate resolution. As for the Q, however, it seems that Leica has decided that with a fixed wide angle, the high resolution sensor serves more customers than the lower resolution would. As it is, the Q2 gets down to 24mp at around 40mm equivalent, and cropped beyond that it starts to get fairly low in resolution (6mp for 75mm). It is still possible to use 50 or 75mm for the web and prints smaller than A3, but beyond that it is not very practical. Any decent compact camera with a longer lens will outpace it at that point.

For me the Achilles heel of SL2-S is aliasing, which is clearly visible at base ISO. At higher ISOs, the aliasing is hidden by the noise. 
Even discounting resolution, at base ISO I much prefer SL2 to SL2-S.

[Advertisement]

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

[SrMi]

1 hour ago, LocalHero1953 said:

I'm a regular user of Topaz DeNoise and Sharpen, with custom settings, not presets. They are great, but cannot fully solve the limitations of Q2 and SL2 in low light - it's too easy to get artefacts. Nor can they improve low light colour. I'd rather start with a better file from the SL2-S. 

I own Topaz DeNoise AI and found that it produces more artifacts than DxO’s tools. Therefore I mostly use DxO’s tools.

Of course, it is always better to have a noise-free image in the first place.

[Stuart Richardson]

1 hour ago, hansvons said:

Absolutely. As I mentioned, the SL2-S is made for doc and event work, basically journalism, including the occasional low-budget video production. You can, of course, shoot landscapes and portraits with it, but if you're after sharpness and cropping (and there are reasons to do that with landscape photography), 50 MP are better. But for A3 prints, I find sharp 24MP (watch your focus, get the right lens) more than sufficient. As mentioned above, I like texture and don't mind seeing what the image originated (pixels). All of that is a matter of taste, intentions, and the media presenting the images.

 

Agreed. 24mp is overkill for A3. I don't think you will see a substantial difference until you get over 40x60cm (that is around 250 dpi at 24mp, the SL2 is at 350dpi at that point, using more or less the full resolution of printers up to 50x70). This is also assuming you use the entire image. Any cropping will make the differences more visible earlier. But the SL2 will really start to pull away in prints 60x90cm or larger.

[BernardC]

18 hours ago, not12bhere said:

A recent use of an equivalent modern Canon professional camera really challenged my justification as someone who shoots in available light and prefers to have a professional service available to deal with hardware failures.

 

After almost 10 years of SL system, I really hope SL3's Panasonic tech has higher QA and whatever sensor Leica obtains for it has modern low-light performance to match the beautiful and non-clinical results of their glass. 

I guess you mean the R5? I hear it's a great camera. I personally wouldn't switch systems based on a temporary perceived advantage, but I understand why someone might. The thing is, camera/sensor performance evolves in a "leapfrog" fashion. Canon's original competitor to the SL2 was the 5Ds, and that camera was terrible in low light. The R5 is top dog for now (in terms of noise), but it won't be once the SL3 comes out. And then maybe Sony will replace the a1 a year after that, and the R5 mark II come along a few months later...

[Chris Nebard]

43 minutes ago, SrMi said:

For me the Achilles heel of SL2-S is aliasing, which is clearly visible at base ISO. At higher ISOs, the aliasing is hidden by the noise. 
Even discounting resolution, at base ISO I much prefer SL2 to SL2-S.

Yes, on the SL2-s, others have pointed to the aliasing which manifests itself as moiré. I’ve done a number of weddings (where such effects can occur on clothing etc.) but I haven’t yet experienced it. Then again, I don’t use the SL Apo lenses which - again as others have said - are so keen that they out resolve the SL2-s 24mp sensor, and that this is a strong contributory factor. Maybe I’m a bit lucky here in that I do like the M and Sigma lenses. 

Stuart Richardson and Hansvons explain the trade-offs of the SL2-s and the SL2 very well indeed I think, and, for pure resolution, the SL2 would appear to be the way to go. For the low light performance, the noise, shadow and dynamic range handling, I fall in line with Hansvon’s thinking, just because it suits what I do. No wrong answers really - just things to be conscious of when we’re buying. 
 

I have a similar feeling with the original Q versus the Q2. The Q2 is clearly the superior camera in many ways, and certainly so technically, but I didn’t upgrade - scared stiff that I’d lose whatever magic that the Summilux f1.7 and the matched sensor create together. I also continue to get some very acceptable stuff with it in low light - I’m sure it doesn’t compete with my SL2-s but the grain at higher iso can actually be quite appealing. I’m being too subjective now (whereas other contributors on the forum apply the science far better than I do). 
 

One thing is for sure, if a future SL3 bridges the existing characteristics which separate the SL2 and SL2-s, it’ll be a bloody amazing camera. 
 

Great forum this - the depth of knowledge and some of the images I see always make me stop and think. Keep it up 😂👍

[LBJ2]

Advertisement (gone after registration)

From a few comments back...I don't think I'll ever return to a 24MP-only FF camera for photography again. I agree 24MP is more than fine for just about anybody/anything, I just prefer the flexibility of more resolution and since I purchased three SL APOs, I really don't want to pay this kind of price and size premium for top shelf lenses and then use on a 24MP-only FF sensor. But that's me. The M11's Triple Resolution feature pretty much sealed the deal for me as well providind 60/36/18 file size options so I never have to worry about too few/too many MPs again. In fact my hopes as others commented, are the SL3 adopts something similar to the M11's 60MP Sony BSI sensor with PDAF. If I'm pushing my personal day-dream, that sensor will be both BSI and stacked in the SL3. 

The M11's 60 MP sensor seems to measure/chart very close to the low light performance of the 24MP SL2-S ( but I've not compared this for myself) I'm thinking the SL3 with same in-camera DNG processing as the SL2-S might just be THE Leica FF mirrorless camera going forward. However, if Leica holds back and doesn't also improve upon the SL2's buffer size/performance, FPS, current SL2 battery power flaw situation in the SL3 then well...maybe the SL3 continues as the Resolution camera and SL3-S for video and "sports".

Either way, I think it's going be awhile before we see the SL3, probably not this year and then for me, since I'm done paying Leica full price for early adopter status, I'll be waiting some time this go around for when special prices can be had. Unless of course the SL3 turns out to be all that and more...LOL

[hansvons]

2 hours ago, LBJ2 said:

The M11's Triple Resolution feature pretty much sealed the deal for me as well providind 60/36/18 file size options so I never have to worry about too few/too many MPs again.

Without knowing anything below the surface of common knowledge in sensor technology, I could imagine that the M11 sensor's ability to downgrade resolution might be the path to good 4K video as UHD only requires 8.3 MP, and 6K in 16:9 would just fit in with about 18 MP. That doesn't say a thing about read-out speeds (jello) or fps. 

[LBJ2]

4 hours ago, hansvons said:

Without knowing anything below the surface of common knowledge in sensor technology, I could imagine that the M11 sensor's ability to downgrade resolution might be the path to good 4K video as UHD only requires 8.3 MP, and 6K in 16:9 would just fit in with about 18 MP. That doesn't say a thing about read-out speeds (jello) or fps. 

If also a Stacked sensor, then that might support very fast readout speeds which could improve rolling shutter performance ( similar to the Sony A1/Nikon Z9 cameras). A more powerful processor( sure to come)+ BSI+Stacked sensor in the SL3 could be a significant upgrade. 

[SrMi]

11 hours ago, hansvons said:

Without knowing anything below the surface of common knowledge in sensor technology, I could imagine that the M11 sensor's ability to downgrade resolution might be the path to good 4K video as UHD only requires 8.3 MP, and 6K in 16:9 would just fit in with about 18 MP. That doesn't say a thing about read-out speeds (jello) or fps. 

As far as I know, M11's ability to "downgrade resolution" is implemented in firmware (software), it is not part of the sensor's ability.

[FlashGordonPhotography]

Hmmmmm. I’m reading comments about the *poor* low light performance of the SL2 compared to the Canon R5. I have both and I’m just not seeing it. The Canon has mandatory noise reduction applied to higher ISO files so they appear a bit cleaner out of camera. But if you do careful noise reduction in post (or better run a SL2 file through DXO) there’s no real world difference. I did A1 prints to see if there was anything useful. There isn’t. The M11 sensor is slightly better than both. Slightly. Mostly because you’re down sampling to get the same print size, I suspect. Also the M11 doesn’t need to use software to hide the PDAF pixels like the R5 and A7R5 do. If I push my Sony or Canon files really hard I see some artefacts the Leica doesn’t have. But mostly it’s a noise/detail trade off. More of one. Less of the other.

The sensor in the R5 isn’t even new. It’s a derivative of an older sensor from the EOS 90D, which itself didn’t win any noise or DR awards. So about 6 years old. It’s a solid sensor helped a lot by software. But really no better or worse than the SL2 sensor, which is made to preserve slightly more detail and show a bit more noise. There are some processor differences. The Canon does a better job of NR in camera. But I can run my SL2 files through DXO in post. I can’t undo the NR Canon applies.

I have a pretty modern sensor from every manufacturer. And they’re all close. There’s been no major improvements in noise for generations of sensors. I saw the difference between the A7R and A7RII but since then it’s been resolution that’s it. You can go off to DPReview for an hour and do the comparisons. Systems that do NR in camera will be slightly cleaner but will show slightly less detail and vice versa. There’s no big differences.

The SL3 with PDAF will perform in the ballpark of every other 24x36mm high resolution sensor. The processor will be the only real differentiation plus the software that decides how much NR is applied at capture. When I choose between systems the sensor noise is the LAST thing I think about. Mostly it’s about which lenses I want to use. If I want a cleaner image I go up a sensor size as that’s still the only way to improve DR and noise for now. There will be many improvements in an SL3. Likely noise won’t be one of them as it sure hasn’t been for any other manufacturer in the last 5 years.

Gordon

[hdmesa]

3 hours ago, FlashGordonPhotography said:

Hmmmmm. I’m reading comments about the *poor* low light performance of the SL2 compared to the Canon R5. I have both and I’m just not seeing it. The Canon has mandatory noise reduction applied to higher ISO files so they appear a bit cleaner out of camera. But if you do careful noise reduction in post (or better run a SL2 file through DXO) there’s no real world difference. I did A1 prints to see if there was anything useful. There isn’t. The M11 sensor is slightly better than both. Slightly. Mostly because you’re down sampling to get the same print size, I suspect. Also the M11 doesn’t need to use software to hide the PDAF pixels like the R5 and A7R5 do. If I push my Sony or Canon files really hard I see some artefacts the Leica doesn’t have. But mostly it’s a noise/detail trade off. More of one. Less of the other.

The sensor in the R5 isn’t even new. It’s a derivative of an older sensor from the EOS 90D, which itself didn’t win any noise or DR awards. So about 6 years old. It’s a solid sensor helped a lot by software. But really no better or worse than the SL2 sensor, which is made to preserve slightly more detail and show a bit more noise. There are some processor differences. The Canon does a better job of NR in camera. But I can run my SL2 files through DXO in post. I can’t undo the NR Canon applies.

I have a pretty modern sensor from every manufacturer. And they’re all close. There’s been no major improvements in noise for generations of sensors. I saw the difference between the A7R and A7RII but since then it’s been resolution that’s it. You can go off to DPReview for an hour and do the comparisons. Systems that do NR in camera will be slightly cleaner but will show slightly less detail and vice versa. There’s no big differences.

The SL3 with PDAF will perform in the ballpark of every other 24x36mm high resolution sensor. The processor will be the only real differentiation plus the software that decides how much NR is applied at capture. When I choose between systems the sensor noise is the LAST thing I think about. Mostly it’s about which lenses I want to use. If I want a cleaner image I go up a sensor size as that’s still the only way to improve DR and noise for now. There will be many improvements in an SL3. Likely noise won’t be one of them as it sure hasn’t been for any other manufacturer in the last 5 years.

Gordon

R5 applies noise reduction at low ISO, not high. This low ISO noise reduction seems designed to drive up the R5's dynamic range score.

The R5 stops applying noise reduction at ISO 800, so the ISO performance of the R5 at 800 and up is actual performance.

https://photonstophotos.net/Charts/PDR.htm#Canon EOS R5

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 March 1, 2023 by hdmesa
Add chart and link

[hansvons]

9 hours ago, FlashGordonPhotography said:

I have both and I’m just not seeing it. The Canon has mandatory noise reduction applied to higher ISO files so they appear a bit cleaner out of camera.

You see the same in video. The film industry’s workhorse, the Alexa, applies no noise reduction whatsoever. The same can be said about Red’s various cine cameras, and probably the Sony Venice, but I haven't shot with it.

They have their sweet spot around ISO 800, and due to increasing DR, from generation to generation, they become better in higher ISO (ISO 1600-3200). 

All the praised low light heroes (that cost a fraction of the above-mentioned cameras) apply noise reduction that gets hidden in various degrees of lossy video codecs. The obvious gain is a cleaner image. But the downside is less detail and colour in the shadows.

The SL2-S video fidelity isn't better than what the competition offers. But the noise reduction isn't as aggressive (there’s some pleasing texture left), and the colour science is industry-leading in the market of hybrid mirrorless cameras. 

I don't expect the next SL generation to be lowlight heroes. But I expect excellent AF in all situations, probably with machine learning capabilities such as bird recognition, and an articulated screen. It will be a resolution upgrade, but not for the -S, as I expect an even more pronounced differentiation of the two models.

What I would like to see in the -S isn't more resolution, as 6K is plenty enough for a hybrid. I’d like to see better video codecs at full resolution (12 bit 6K raw), a proper ACES integration, and an AF-C that can be adjusted to the particular video needs (non-linear focus pulls). A sturdy, bright and tiltable screen would be a very welcome design change. 

 

[romaing]

New post by Leica Rumors.

SL3 - Late 2023 or 2024.

[Olaf_ZG]

On 3/2/2023 at 8:02 PM, romaing said:

New post by Leica Rumors.

SL3 - Late 2023 or 2024.

 

On 3/2/2023 at 8:02 PM, romaing said:

New post by Leica Rumors.

SL3 - Late 2023 or 2024.

Looks like that site is w/o inspiration…

[MrSMW]

Hmm, what do the SL2 and SL2s have/do that my current S1R/S1H combo can't do and on that basis, what would I like to see in the next gen SL line...

1: Body size/form: no difference other than a TILT screen (a la S1H, or at least the S1R), but not a side-hinged flippy thing which is the devils work.

2: Full frame 4k 50p 10 bit 422 internal, ie, without the S35 crop as with all the Lumix cameras and the SL2s.

3: A higher rez 60mp model for best resolution and a 24mp (or not much higher) lower light model.

Otherwise, there isn't a single thing more I would want. Ultimately, for me it would come down to price vs whatever similar offerings Lumix was punting.

The single biggest issue for me is the lack of any form of tilt screen as I only work from the LCD, stills and video, and usually at chest or waist height or even lower and then if it does not have one (a tiltable screen) it's pretty hard to use.

[FlashGordonPhotography]

3 hours ago, MrSMW said:

Hmm, what do the SL2 and SL2s have/do that my current S1R/S1H combo can't do and on that basis, what would I like to see in the next gen SL line...

1: Body size/form: no difference other than a TILT screen (a la S1H, or at least the S1R), but not a side-hinged flippy thing which is the devils work.

2: Full frame 4k 50p 10 bit 422 internal, ie, without the S35 crop as with all the Lumix cameras and the SL2s.

3: A higher rez 60mp model for best resolution and a 24mp (or not much higher) lower light model.

Otherwise, there isn't a single thing more I would want. Ultimately, for me it would come down to price vs whatever similar offerings Lumix was punting.

The single biggest issue for me is the lack of any form of tilt screen as I only work from the LCD, stills and video, and usually at chest or waist height or even lower and then if it does not have one (a tiltable screen) it's pretty hard to use.

1. Much better menus.

2. Better behaviour with adapted M lenses.

3. *Slightly* better accutance due to the thinner sensor stack.

However the S1R also has significant advantages over the Leica, as you pointed out. Starting with the wonderful tri-flip screen.

The S1R is a criminallly under rated camera. Condemned for one feature by boring lazy youtubers, some of which use the S1H despite bagging the system at every opportunity.

Gordon

[Andricco]

I see a lot of requests for a flip screen. It makes sense but I'm not so sure if they will go down that path because Leica pride themselves on keeping their designs to a more traditional & simplistic look & that's what separates them from other brands and that's what appeals to Leica enthusiasts.I personally think that including a flip screen on a Leica would be like putting a spoiler wing on a Rolls Royce.If Leica does include a flip screen I'm sure it will be designed in a way which will be seamless & unobtrusive   

[Wurlitzer-1015]

8 hours ago, Andricco said:

I see a lot of requests for a flip screen. It makes sense but I'm not so sure if they will go down that path because Leica pride themselves on keeping their designs to a more traditional & simplistic look & that's what separates them from other brands and that's what appeals to Leica enthusiasts.I personally think that including a flip screen on a Leica would be like putting a spoiler wing on a Rolls Royce.If Leica does include a flip screen I'm sure it will be designed in a way which will be seamless & unobtrusive   

I agree, the most I'd like to see is the ability to flip up the eye level finder like the EVFs on the M cameras.

Edited March 22, 2023 by Wurlitzer-1015
Typo

[Advertisement]

Hallo,

Look here: Leica L-Mount

[Simone_DF]

10 hours ago, Andricco said:

I see a lot of requests for a flip screen. It makes sense but I'm not so sure if they will go down that path because Leica pride themselves on keeping their designs to a more traditional & simplistic look & that's what separates them from other brands and that's what appeals to Leica enthusiasts.I personally think that including a flip screen on a Leica would be like putting a spoiler wing on a Rolls Royce.If Leica does include a flip screen I'm sure it will be designed in a way which will be seamless & unobtrusive   

A tilt screen adds the ability to shot at an angle (above the head or below waist level), and it's widely used handheld or on a tripod by a multitude of people. It is not a matter of adding random functionalities and clutter a la Sony, it is a hardware feature that is present on most cameras for good reasons.

The current proposed solution, using the app via live view, is not fit for purpose, and it will never be, because using the smartphone with one hand while holding the camera with the other hand is extremely clumsy and very non-Leica.

Design must also be functional, and it must solve problems, not creating new ones, that's pretty much the first lesson you learn on every UI/UX course. Otherwise you end up with a nice art object with degraded functionality, which is what a fixed screen is in 2023, a nice design for the sake of design that makes things more difficult.

Just look at Apple in the final years of Jony Ive, Macbook became so thin there was no space for ports, and people ended up using dongles, which was a much worse solution than having an extra mm on the chassis. Design for the sake of design.  

PS: most, but not all, cameras have a seamless and unobtrusive screen nowadays. The Fuji XPro3 will even allow you to turn the screen the other way around and have a "M10-D" experience if you want. This is the most Leicaesque thing I've seen in years.

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!