Problems getting sharp images by 60MP

[lct]

Advertisement (gone after registration)

23 minutes ago, AceVentura1986 said:

Higher resolution cameras are just naturally less forgiving of camera shake, period, full stop. There really is no solution to this problem other than shooting 2x or 3x the normal shutter speed, using a  tripod or image stabilization. Please see this article by Ming Thein as he explains it much better than I can. 

Moreover, there are times when you may want SOME image blur but no camera shake. In cases where you might want to drag the shutter a bit to capture a subject’s movement but still keep the background sharp (a dancer an athlete in motion, for example) the standard technique is to shoot at around 1/30 of a second. That is slow enough to capture the subject’s movement but still keep the background sharp. 

However, to guard against camera shake inherent in higher reso sensors, the standard technique is to shoot at 2x or 3x regular shutter speed and compensate w higher ISO. However, shooting at 2x or 3x to ensure a sharp background (1/60 or 1/90) will tend to freeze the image. Increasing the shutter speed will not work as it will necessarily freeze the subject’s movement, which was the whole point of dragging the shutter in the first place. 

FWIW, I don’t own an M11, but I do own an ME and an M9M (18mpx each) and also own a Canon 5DsR (50mpx). The Canon is usually relegated to landscape shots mounted on a tripod, image stabilized lenses or higher shutter speeds. I use my old Leicas for everything else. 

Moreover, for most people in most circumstances there simply is no benefit or even use for such high pixel counts. I’ve been shooting digital cameras since my original Canon D30 (3mpx!) way back in 2000 and 1D (4mpx) in 2002. I’ve printed 3 and 4 mpx images to 12x18 inches with no issues at all. I’ve printed and displayed photos from my M9M (18 mpx) to 48 inches on the long side, again with no loss in quality. IMHO, and this will likely upset some people, 16 mpx is sufficient for almost everything and anything above 24 mpx is overkill UNLESS the photographer has a specific, articulable need for higher reso, such as advertising photography or landscapes. However, for a Leica M, that is typically shot handheld and under fast-moving conditions, 60 mpx is way overkill. 

Problem is not high res as i understand it but medium to low res on the M11.

Edited February 19, 2022 by lct

[SrMi]

1 hour ago, AceVentura1986 said:

Higher resolution cameras are just naturally less forgiving of camera shake, period, full stop. There really is no solution to this problem other than shooting 2x or 3x the normal shutter speed, using a  tripod or image stabilization. Please see this article by Ming Thein as he explains it much better than I can. 

Moreover, there are times when you may want SOME image blur but no camera shake. In cases where you might want to drag the shutter a bit to capture a subject’s movement but still keep the background sharp (a dancer an athlete in motion, for example) the standard technique is to shoot at around 1/30 of a second. That is slow enough to capture the subject’s movement but still keep the background sharp. 

However, to guard against camera shake inherent in higher reso sensors, the standard technique is to shoot at 2x or 3x regular shutter speed and compensate w higher ISO. However, shooting at 2x or 3x to ensure a sharp background (1/60 or 1/90) will tend to freeze the image. Increasing the shutter speed will not work as it will necessarily freeze the subject’s movement, which was the whole point of dragging the shutter in the first place. 

FWIW, I don’t own an M11, but I do own an ME and an M9M (18mpx each) and also own a Canon 5DsR (50mpx). The Canon is usually relegated to landscape shots mounted on a tripod, image stabilized lenses or higher shutter speeds. I use my old Leicas for everything else. 

Moreover, for most people in most circumstances there simply is no benefit or even use for such high pixel counts. I’ve been shooting digital cameras since my original Canon D30 (3mpx!) way back in 2000 and 1D (4mpx) in 2002. I’ve printed 3 and 4 mpx images to 12x18 inches with no issues at all. I’ve printed and displayed photos from my M9M (18 mpx) to 48 inches on the long side, again with no loss in quality. IMHO, and this will likely upset some people, 16 mpx is sufficient for almost everything and anything above 24 mpx is overkill UNLESS the photographer has a specific, articulable need for higher reso, such as advertising photography or landscapes. However, for a Leica M, that is typically shot handheld and under fast-moving conditions, 60 mpx is way overkill. 

 

If you resize 60MP to 18 MP, in the post or the camera, you will see the same camera shake effect. That is the theory. I believe in the theory until I have seen a comprehensive test that proves or rejects it.
The only serious test of handholding shutter speed that I know of has been done by Jim Kasson: Rules of thumb for handheld shutter speed. It matches my experience with 24MP cameras.

[aristotle]

4 hours ago, AceVentura1986 said:

Higher resolution cameras are just naturally less forgiving of camera shake, period, full stop. 

 

Can I get a comma and ahead quarter speed instead?   Actually, a higher resolution camera is no more, nor no less forgiving of camera shake than a lower resolution camera assuming that the sensors are the same physical size.  Camera shake, or subject motion, does limit the utility of having the higher resolution however.  But it's no worse and no better as a result of the sensor resolution.  A translation of the 35mm sensor by a fixed amount during exposure results in the translation of an object by a fixed amount on the sensor surface, regardless of how many pixels are in the sensor.  The only reason you would need to take the higher resolution sensor into account when selecting a minimum exposure time is if you know at the time of capture that you want to make use of the full resolution of the sensor to either crop into the image or to create a very large print.  The type of subjects where this becomes an issue are things lik birds in flight or sporting events.  For those, sure, image stabilization is helpful, both because of subject motion and camera shake.  Maybe there are folks who like to crop into their street photographs to show a 135mm or greater equivalent field of view, or print them 3' wide and hang them on the wall...  I have to believe that's the exception though.  

Edited February 19, 2022 by aristotle

[M9reno]

5 hours ago, AceVentura1986 said:

Higher resolution cameras are just naturally less forgiving of camera shake, period, full stop. There really is no solution to this problem other than shooting 2x or 3x the normal shutter speed, using a  tripod or image stabilization. Please see this article by Ming Thein as he explains it much better than I can. 

This is what Ming Thein says:

"Empirical observation #1: Cameras with higher resolutions, for a given sensor size, are harder to hand-hold and get a good result than those with a lower resolution. And I define a good result as one which the image is critically sharp at 100% actual-pixels view, with no motion blurring or other artefacts such as slightly missed focus..."

The problem obviously is that a 100% view of a 60MP photo is going to be larger than a 100% view of a 24MP photo.  So he is comparing two differently sized views.

This means that:

If both views have motion blur (for example), then the latter will obviously hide it better than the former.  Magnify the latter 20% more, so that both views are the same size, and you will now see the same motion blur present in both

[FlashGordonPhotography]

You guys are discussing two different things as if they are one. So first you need to define if you're talking about camera shake or shutter shock. Are the perceived differences a result of the increase in resolution or is there something inherent in the design of the M11 shutter actuation that is causing more vibration internally than the M10 cameras.

Camera Shake:

Dependant on the individual user.

Dependant on technique.

Tested handheld.

Measured at a fixed image size not at the pixel level.

Very difficult to empirrically prove due to human error.

 

Shutter Shock:

Not user dependant.

Measured on a tripod.

May only occur at certain shutter speeds or the entire range.

Less dependant on resolution. Usually vied at a pixel level.

Empirracly provable.

 

So decide whether we're more likely to show our own unsteadiness as resolution rises (answer Yes you are.) or whether there is something inherant in the M11 where the CAMERA is affecting the results. Or both. So far, after two weeks of use I see neither with my M11. It's pretty much exactly the same as my M10R and M10M. I shoot a lot at around 1/f but my technique is excellent. Lots of people will need to shoot at 2 or 3 x f to get pixel sharp results at 60MP.

As for the lower resolutions on the M11, down ressing or pixel binning will show similar but not exactly the same results as a (same sized) sensor with a natively lower resolution. It's almost impossible to record blur at a single pixel site. The blur needs to be recorded across more than one pixel (pixels show a single colour dot) for you to see it. The reason we see more blur at higher resolutions is that the blur can travel across more pixels making it more obvoius. The smaller the pixels (photosites) the more likely blur becomes visable (an advantage of larger sensors that can have high resolutions and larger photosites.).

So in theory the M11 down ressing to 18MP would be resampling blur (if it's there) and the M9 wouldn't show it. However in the real world the differences would be incredibly subtle. You might see some differences at 100% but at the output size it's unlikely (unless you're printing at the limits of resolution).

Gordon

p.s. Ming Theins definition of *empirically observable* doesn't make sense. Empirically and observable ar at odds with each other Observations aren't proof. He just made the term up. He did some decent writing. But he also claimed he could accurately calibrate his monitors by eye (no he can't). So like anyone else on the web he's best taken with a grain of salt.

[Mute-on]

19 minutes ago, FlashGordonPhotography said:

You guys are discussing two different things as if they are one. So first you need to define if you're talking about camera shake or shutter shock. Are the perceived differences a result of the increase in resolution or is there something inherent in the design of the M11 shutter actuation that is causing more vibration internally than the M10 cameras.

Camera Shake:

Dependant on the individual user.

Dependant on technique.

Tested handheld.

Measured at a fixed image size not at the pixel level.

Very difficult to empirrically prove due to human error.

 

Shutter Shock:

Not user dependant.

Measured on a tripod.

May only occur at certain shutter speeds or the entire range.

Less dependant on resolution. Usually vied at a pixel level.

Empirracly provable.

 

So decide whether we're more likely to show our own unsteadiness as resolution rises (answer Yes you are.) or whether there is something inherant in the M11 where the CAMERA is affecting the results. Or both. So far, after two weeks of use I see neither with my M11. It's pretty much exactly the same as my M10R and M10M. I shoot a lot at around 1/f but my technique is excellent. Lots of people will need to shoot at 2 or 3 x f to get pixel sharp results at 60MP.

As for the lower resolutions on the M11, down ressing or pixel binning will show similar but not exactly the same results as a (same sized) sensor with a natively lower resolution. It's almost impossible to record blur at a single pixel site. The blur needs to be recorded across more than one pixel (pixels show a single colour dot) for you to see it. The reason we see more blur at higher resolutions is that the blur can travel across more pixels making it more obvoius. The smaller the pixels (photosites) the more likely blur becomes visable (an advantage of larger sensors that can have high resolutions and larger photosites.).

So in theory the M11 down ressing to 18MP would be resampling blur (if it's there) and the M9 wouldn't show it. However in the real world the differences would be incredibly subtle. You might see some differences at 100% but at the output size it's unlikely (unless you're printing at the limits of resolution).

Gordon

p.s. Ming Theins definition of *empirically observable* doesn't make sense. Empirically and observable ar at odds with each other Observations aren't proof. He just made the term up. He did some decent writing. But he also claimed he could accurately calibrate his monitors by eye (no he can't). So like anyone else on the web he's best taken with a grain of salt.

Since I own only Leica film cameras, and have limited experience with digital post-processing, many of the issues in this thread have been challenging to conceptualise. 

OTOH, Gordon’s explanation, and the bolded section in particular, actually makes sense (to me at least).

Cheers  

 

[lct]

Advertisement (gone after registration)

34 minutes ago, FlashGordonPhotography said:

So decide whether we're more likely to show our own unsteadiness as resolution rises (answer Yes you are.) or whether there is something inherant in the M11 where the CAMERA is affecting the results. Or both. So far, after two weeks of use I see neither with my M11. It's pretty much exactly the same as my M10R and M10M. I shoot a lot at around 1/f but my technique is excellent. Lots of people will need to shoot at 2 or 3 x f to get pixel sharp results at 60MP.

Not sure if i asked you already, if so i apologize but those pixel binning things are confusing to me. So... at 33MP and/or 18MP do you feel the need to shoot at 2 x f while  1 x f is enough for you with a "normal" camera (M9, M240, M10 etc.). Sorry to bother you with those low res questions but 60MP sans IBIS is of no interest for me. I realize i could answer in 2 minutes myself if i had an M11 handy but i have none sorry.

[FlashGordonPhotography]

1 hour ago, lct said:

Not sure if i asked you already, if so i apologize but those pixel binning things are confusing to me. So... at 33MP and/or 18MP do you feel the need to shoot at 2 x f while  1 x f is enough for you with a "normal" camera (M9, M240, M10 etc.). Sorry to bother you with those low res questions but 60MP sans IBIS is of no interest for me. I realize i could answer in 2 minutes myself if i had an M11 handy but i have none sorry.

Not for me, no. Personally I'm seeing zero difference between the M11 and another M at equivalent resolutions. For example the M10 (regular) and M11 at 18MP (closest available) I can shoot at about the same speeds (1/2 F normal and 1/4 F if I'm really concentrating) versus 42MP on the M10R or 60MP on the M11 where I'm about 1/F for regular shooting and 1/2F if I'm concentrating.

Currently I'm seeing no shutter shock issues on the M11. So I'm discounting the cameras having any influence. Also the binning seems really well done so someone could shoot the camera at 36 or 18MP with little to no difference to a sensor of those resolutions in the camera.

For camera shake, my personal limitations with a 50mm lens on the M11are currently:

60MP: sharp at 1/60 with normal technique, sharp at 1/30 with thoughtful technique and sharp at 1/15th with very careful technique. This is the same as my M10R give or take.

18MP: sharp at 1/25th with normal technique,  sharp at 1/15th with thoughtful technique and sharp at 1/8th with very careful technique. This is the same as my M10 give or take.

Gordon

[harmen]

To further illustrate what many others have said: motion from not steadily holding the camera will always be recorded.  Two things happen when looking at 100%.  Firstly, a lower MP camera at 100% will cover a much larger image area than a higher MP camera.  in other words, the higher MP camera allows you to look closely at a much smaller area.  By looking at different sizes, motion will seem exaggerated on the higher MP camera.  Secondly, with fewer pixels a sensor doesn't resolve the motion as clearly, in line with it not as clearly resolving the entire image!  I've posted this before, but it's more relevant in this discussion.  See below how even motion of less than 1 pixel width does affect the final image.  Imagine the green object being the edge of a leaf against a white sky.  Now if either the leaf or the camera moves, its projection onto the sensor will travel during exposure.

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!

I see this clearly in my own images: motion blur at an 18MP or 24MP image is a little less obvious than at 60MP, but it's not actually very different.

All in all I'm very happy with a camera where sensor resolution is becoming less of a limiting factor.  I'm happy to be a limiting factor myself, especially when for me it's about the full image and not the 100% view.

 

All that said, I'm still curious why some people report that a majority of their photos is not sharp while it was on their other cameras.  I think we've put hand-held motion blur and shutter-shock to bed.  So what then is it that they are seeing happen with their images?  Is it some setting in the processing, e.g., contrast or sharpening?

Edited February 20, 2022 by harmen

[SrMi]

Josh and David (Red Dot Forum) discuss whether you need higher shutter speed with higher resolution (21:15):

 

 

[lct]

3 hours ago, harmen said:

I'm still curious why some people report that a majority of their photos is not sharp while it was on their other cameras.

Curious indeed. Those ridiculous people were used to get sharp results at normal shutter speeds with their plastic gear and suddenly their pics become blurred with a 9k camera. How is it possible? Go to the gym dudes! Repeat after me: The Real Leica Man doesn't need IBIS .

Edited February 20, 2022 by lct

[mboerma]

4 hours ago, lct said:

Curious indeed. Those ridiculous people were used to get sharp results at normal shutter speeds with their plastic gear and suddenly their pics become blurred with a 9k camera. How is it possible? Go to the gym dudes! Repeat after me: The Real Leica Man doesn't need IBIS .

And, may I add, a nice heavy silver M11 **starts running *** 😂

[52K]

Whew ! there was a bit of Sunday morning reading and the timing was spot on. For a short while I was wondering, should I be worried? I quickly came to the conclusion... 'no'.

Longer version...

Last night I needed a couple of quick 'product' type shots in available light. Weapons of choice, SL2S/50 APO SL, or M11/75APO M. As light was fading shutter speeds were falling. The SL nailed the shot really quickly, pin sharp, high contrast and flawless at well below 1/f, but I liked the 75mm perspective shot on the M11 better, however it was clearly not as sharp. Yes, of course I could have put the 75 APO on the SL, but I wanted to test the high megapixel count for this shot.

I did end up using the M11 (probably out of bloody-mindedness) and took more shots using a tripod. This was a bit of a faff having to first remove the leather protector (which was fiddly) and then set it all up , move stuff to accommodate the tripod etc, in a hurry as the light was fading. But, I got the shot I wanted, which at the end of the day is my gift as a photographer and the challenge I enjoy when taking photos whether for work, or for pleasure. 

Reading this thread this morning over coffee, it promoted me to fire off some hand-held test shots this morning in my living room. Not hugely scientific, I took shots at shutter speeds relative to the focal length where I'd expect to get a clean shot, 1/f or better, then reduced the ISO to where I'd like it, and let the shutter speed fall, at one point 1/30 at 75mm. 

I got a few sharp shots, even at the slow shutter speeds as I'd expect to and of course some were badly blurred. My conclusion is that there is nothing to worry about. It's not hugely different (without the scientific evidence) to any other M I've owned and I'm confident that I'll continue to get great results with it for my photography despite the eye-popping 60MP resolution.

I'm sure this debate will rage on just like the current weather and I'll watch with interest for a meaningful conclusion but confident that 60 Megapixels aren't going to ruin my day. 

[Hanno]

One comment is that shutter shock may be more of an issue at lower shutter speeds, and in that case moving to the electronic shutter will entirely remedy it.

Assuming that at lower shutter speed, presumably one is not shooting fast moving targets.

[Lelmer]

Deleted

Edited February 20, 2022 by Lelmer

[John Smith]

1 hour ago, 52K said:

Last night I needed a couple of quick 'product' type shots in available light. Weapons of choice, SL2S/50 APO SL, or M11/75APO M. As light was fading shutter speeds were falling. The SL nailed the shot really quickly, pin sharp, high contrast and flawless at well below 1/f, but I liked the 75mm perspective shot on the M11 better, however it was clearly not as sharp. Yes, of course I could have put the 75 APO on the SL, but I wanted to test the high megapixel count for this shot..

That pretty much nails it. 

[lct]

59 minutes ago, Hanno said:

One comment is that shutter shock may be more of an issue at lower shutter speeds, and in that case moving to the electronic shutter will entirely remedy it.

The issue is my (your?) hands. You may think that they are steady as i do but they can hardly be as steady as a tripod. There is then a limit (mine is somewhere between 24 and 42MP but i'm an oldie) where motion blur will become significantly visible on screen if not on prints if you don't print billboards. I shoot mostly in e-shutter mode and i can tell you that it can do nothing against that. Solutions are well known: higher shutter speeds or IBIS. Lower resolution too normally but i have no experience with the M11.

Edited February 20, 2022 by lct

[Guest]

23 hours ago, AceVentura1986 said:

Moreover, for most people in most circumstances there simply is no benefit or even use for such high pixel counts. I’ve been shooting digital cameras since my original Canon D30 (3mpx!) way back in 2000 and 1D (4mpx) in 2002. I’ve printed 3 and 4 mpx images to 12x18 inches with no issues at all. I’ve printed and displayed photos from my M9M (18 mpx) to 48 inches on the long side, again with no loss in quality. IMHO, and this will likely upset some people, 16 mpx is sufficient for almost everything and anything above 24 mpx is overkill UNLESS the photographer has a specific, articulable need for higher reso, such as advertising photography or landscapes. However, for a Leica M, that is typically shot handheld and under fast-moving conditions, 60 mpx is way overkill. 

 

+1

Nicely said, couldn't agree more......

[Hanno]

1 hour ago, lct said:

The issue is my (your?) hands. You may think that they are steady as i do but they can hardly be as steady as a tripod. There is then a limit (mine is somewhere between 24 and 42MP but i'm an oldie) where motion blur will become significantly visible on screen if not on prints if you don't print billboards. I shoot mostly in e-shutter mode and i can tell you that it can do nothing against that. Solutions are well known: higher shutter speeds or IBIS. Lower resolution too normally but i have no experience with the M11.

I thought Gordon did a great job in differentiating camera shake (due to one’s hands, to keep your reference) from shutter shock. I was addressing the shutter shock only, so I am not sure why you were commenting on the electronic shutter not helping with your hands. Apologies if I mislead you, but yes you are talking about something I wasn’t.

if camera shake is really an issue, then possibly a mini tripod or something like a Novoflex magic ball on your chest can be of help.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[lct]

33 minutes ago, Hanno said:

I thought Gordon did a great job in differentiating camera shake (due to one’s hands, to keep your reference) from shutter shock. I was addressing the shutter shock only, so I am not sure why you were commenting on the electronic shutter not helping with your hands. Apologies if I mislead you, but yes you are talking about something I wasn’t.

if camera shake is really an issue, then possibly a mini tripod or something like a Novoflex magic ball on your chest can be of help.

I just wanted to say that it is not only a issue for me but for everybody else at various degrees but you know this already obviously.