Problems getting sharp images by 60MP

[LocalHero1953]

Advertisement (gone after registration)

13 hours ago, lct said:

Optimistic version: We can still use our dirty old hands at 33MP or 18MP with e-shutter mode and a faster Viso. "What's not to like" as they say .  

Das Wesentliche?

[harmen]

Once there is an M11-S we should be able to get someone make even easier to interpret comparisons 😋

[M11 for me]

vor 1 Stunde schrieb harmen:

Once there is an M11-S we should be able to get someone make even easier to interpret comparisons 😋

The M11 is an M11-S at the same time. And even an M11-SS when you choose 18MPix. What should be different in the M11-S that you expect?

The method is called pixel binning: When one uses the sensor at 36Mpix then the blur behaves like with a full frame 36Mpix camera. And such cameras are on the market since at least 5 years without any complaints.

Edited February 21, 2022 by M11 for me

[harmen]

19 minutes ago, M11 for me said:

The M11 is an M11-S at the same time. And even an M11-SS when you choose 18MPix. What should be different in the M11-S that you expect?

I don’t expect any differences.  However there seem to be many perceived differences and quality concerns compared with previous Ms such that:

• there may be a market for an -S version
• the comparison we will be able to do then will show that there actually is no reason for an -S, except for cost

hence the 😋

Edited February 21, 2022 by harmen

[rramesh]

20 hours ago, raizans said:

The Real Leica Man prefers his rangefinder camera without IBIS.

The Real Leica Man is perfect and perfection stops at 10.

Edited February 21, 2022 by rramesh

[lct]

2 hours ago, M11 for me said:

The M11 is an M11-S at the same time. And even an M11-SS when you choose 18MPix. What should be different in the M11-S that you expect?

The method is called pixel binning: When one uses the sensor at 36Mpix then the blur behaves like with a full frame 36Mpix camera. And such cameras are on the market since at least 5 years without any complaints.

Wouldn't be the same sensors obviously so who knows? Would be larger pixels vs pixel binning so slight differences are to be expected. If they are slight as between my A7s and A7r2 and i don't have to go to the gym i'm all in . 

[M11 for me]

Advertisement (gone after registration)

 

When you set your M11 to 36Mpix you have a 36Mpix camera. Yes there is a difference between this and that. There is always a difference between 2 things unless they are clones. I wonder what exactly your problem is as you have no M11 🤣. 

[lct]

14 minutes ago, M11 for me said:

When you set your M11 to 36Mpix you have a 36Mpix camera. Yes there is a difference between this and that. There is always a difference between 2 things unless they are clones. I wonder what exactly your problem is as you have no M11 🤣. 

I respect theories a lot, especially those confirming my experience , but i have zero experience with those pixel binning things. Otherwise i have no problem at all as you can see it given that i am the most talkative poster in this thread . I just wonder if i will order an M11, wait for a possible M11-S or wait for an hypothetical EVF-M. Differences between this and that are indeed what might interest me then.

[aristotle]

There are theories and then there are theories.  I have a theory that the silver M11 will provide the same ISO performance as the black  M11.  I suppose I could look for test results that confirm my theory, but I'm OK with that working theory given that it's hard to imagine why there would be a difference.  Sure, it's possible....but seems pretty darn unlikely based on the fact that "in theory" at least, the exterior color seems unrelated to ISO performance   To me at least, pixel binning and camera motion blur specifically are similarly unrelated.

Edited February 21, 2022 by aristotle

[lct]

36 minutes ago, aristotle said:

To me at least, pixel binning and camera motion blur specifically are similarly unrelated.

Haha a clear theory would be to say "pixel binning and larger pixels don't make a difference". Is there any good theorist here sure enough of his science to state that? .

[SrMi]

Why are we talking about pixel binning? AFAIK, M11 does not use pixel binning to implement triple resolution (I can't see how it would be possible).

[lct]

1 minute ago, SrMi said:

Why are we talking about pixel binning? AFAIK, M11 does not use pixel binning to implement triple resolution (I can't see how it would be possible).

OK so let's assume we have two cameras: 
• Camera one (let's call it M11) doing pixel binning and triple resolution including 36MP
• Camera two (let's call it M11-S) not doing pixel binning with a single resolution of 36MP
Now can you or anybody else here state hand on heart that there won't be any difference at 36MP between them?
(waiting patiently)

[SrMi]

12 minutes ago, lct said:

OK so let's assume we have two cameras: 
• Camera one (let's call it M11) doing pixel binning and triple resolution including 36MP
• Camera two (let's call it M11-S) not doing pixel binning with a single resolution of 36MP
Now can you or anybody else here state hand on heart that there won't be any difference at 36MP between them?
(waiting patiently)

A camera that uses pixel binning and triple resolution would need to upsize the pixel-binned 15MP image (60/4). Therefore, it should produce inferior results than the camera that reduces the 60MP image in firmware.

A camera that reduces 60MP image in firmware may create better results than a 36MP camera.

P.S.: Sean Reid did some relevant comparisons.

Edited February 21, 2022 by SrMi

[aristotle]

1 hour ago, lct said:

Haha a clear theory would be to say "pixel binning and larger pixels don't make a difference". Is there any good theorist here sure enough of his science to state that? .

I was just responding to your initial question specifically about motion blur and higher resolution sensors where you were concerned that there was somehow a difference between the two specifically having to do with blurry images and down-sampling (through whatever technique...pixel binning or otherwise).  If this question has expanded to how well an M11 image, down-sampled to say 24mp, compares to an M10 in terms of noise performance for example), that's a different story.  All indications are that they are similar, but at least there one can imagine a reason for them to be different.

That wasn't the concern that you originally stated.  

Edited February 21, 2022 by aristotle

[aristotle]

53 minutes ago, SrMi said:

Why are we talking about pixel binning? AFAIK, M11 does not use pixel binning to implement triple resolution (I can't see how it would be possible).

Correct Sr!  Should have said downsampling (though whatever technique).

[pgk]

1 hour ago, lct said:

Now can you or anybody else here state hand on heart that there won't be any difference at 36MP between them?
(waiting patiently)

Unless the systems are identical then there will inevitably be some difference. Whether this will translate into a visible difference is, of course, yet another question. But I don't understand the basic problem. If I bought a 60MPixel camera (I have 43 and 24) I would do so in order to use all the 60MPixels, As I see little point in using 60MPixels (43 is usually excessive enough as it is) I won't be buying one, but if I did, I wouldn't use it except for at 60MPixels - I can always change the dimensions in Photoshop easily enough later on..

[FlashGordonPhotography]

1 hour ago, aristotle said:

Correct Sr!  Should have said downsampling (though whatever technique).

Because Leica stated that it was pixel binning. That's why. They're touting their new way of doing it.

I thought it was interpolation as well until someone actually found Leicas statement and posted it for me.

Gordon

p.s. does anyone know whether Canon's mRAW and sRAW are binned or downsampled.

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!

[FlashGordonPhotography]

2 hours ago, lct said:

OK so let's assume we have two cameras: 
• Camera one (let's call it M11) doing pixel binning and triple resolution including 36MP
• Camera two (let's call it M11-S) not doing pixel binning with a single resolution of 36MP
Now can you or anybody else here state hand on heart that there won't be any difference at 36MP between them?
(waiting patiently)

Difference as in IQ or difference as in the *look*?

For the first there won't (for a similar tech level) and for the second there will.

I would expect if I pulled out my old A7R (36MP) that the newer M11 at 36MP would look better.

Gordon

[IkarusJohn]

On 2/21/2022 at 9:39 AM, FlashGordonPhotography said:

I'd summarise it as follows, based on my usage.

1. The M11 does not have a shutter shock issue. (I've done a lot of testing in the 1/15 to 1/125 range.)

2. The M11 rewards excellent hand holding technique at higher resolutions so practice and get better.

3. Lower resolutions allow hand holding at slightly lower speeds.

4. If you're not prepared to improve your technique then the M11 is not for you. If you are the M11 is a heap of fun to shoot.

5. Get a Thumbie. Or the grip. Or both if you want at least a stop of handheld improvement. But definitely a Thumbie.

6. Nothing has changed from previous M's at *equivalent* resolutions so buying a M10R isn't going to help much.

Gordon

Great summary, Gordon.

I doubt Leica would release a camera with shutter shock. That said, with live view as standard (I’m using shorthand for the shutter opening when the camera is turned on), a lot more is happening than in traditional mechanical shutter mode - shutter open->close->open->close->open - for a single exposure. That said, the same happens with other cameras without issue (the X1D springs to mind).

What interests me more is the impact of increased MP. I don’t need it, but I don’t mind it, if there is no cost.

Take a comparison of pixel pitch of the cameras I’ve used (there is a flaw in the comparison in that there will be a difference in the efficiency of coverage of the pixel sites on the sensor surface area):

• 18 MP M9 (Monochrom) - 6.9 microns
• 24 MP M10-D - 6.0 microns
• 50 MP X1D - 5.3 microns
• 24 MP TL2 (APS-C) - 3.9 microns

I don’t have an M11, but at 60MP, it’s pixel pitch is 3.76 microns.  I haven’t included the Nikon D800E as the issue there was shutter shock.

Surprisingly, motion blur isn’t an apparent problem with the TL2, though I do take care.  My conclusion is that shutter shock hopefully isn’t a problem, and that there are diminishing returns with increased MP for a given sensor size.  The 35mm format is ideal for handheld photography, and anything which detracts from that is not a positive development.  Similarly, medium format would not be my first choice for most handheld photography.  The X1D II was the exception to this - it’s wonderful body form and fabulous sensor made it a great camera for most photography.  My only reservations were the unavoidable lens size and weight (for given focal length and speed), and heat management.  If I walked around with the camera turned on, and AF on, the camera quite quickly cooked up.

So, I see no benefit in high MP on its own.  The M is a compact, stripped back to the essentials camera of the highest quality primarily for handheld use.  If the sensor gives better IQ and dynamic range, then that’s great.  I suspect Leica could have achieved that without the potential complications of 60MP and permanent live view - I see these as negatives, with no positive gain from what I already have.

Let’s see what comes next.  100MP?  Don’t really care one way or the other.  IBIS?  Not particularly interested, provided it is effectively invisible.  Neither of these are virtues in themselves.

 

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[aristotle]

1 hour ago, FlashGordonPhotography said:

Because Leica stated that it was pixel binning. That's why. They're touting their new way of doing it.

I thought it was interpolation as well until someone actually found Leicas statement and posted it for me.

Gordon

p.s. does anyone know whether Canon's mRAW and sRAW are binned or downsampled.

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!

SrMi posted a link to a Leica Tech Talk video in another thread where a Leica developer said unambiguously something to the effect that "there is no magic to the lower resolution...you get the same result if you start with the high resolution DNG and lower the resolution yourself."  I was surprised, but there wasn't a bunch of room for interpretation.  Maybe SrMi can find that?