Electronic Shutter Readout & Hand Shake (dumb question no. 1)

[AndrewLT]

Advertisement (gone after registration)

I’m new here, so …. Hello, everybody!

 

Thinking about buying a colour digital M for the first time, I have what may be a dumb question about the electronic shutter.  (I expect to have other dumb questions in due course.)

 

Background: I am happy with my Sony A9 for wildlife (particularly BIF) but, increasingly, I prefer my Leica MP and Monochrom type 246 for anything else.  Hence being tempted by the idea of either an M10-R or M11.

 

Viewing @Overgaard ’s insightful M11 mechanical / electronic shutter video has caused me to wonder about the effect of hand shake when using an electronic shutter.

 

Suppose the readout time is 1/10s.  In that time, my hands may well move slightly — I would not handhold with a 1/10s exposure time — so I am wondering if, irrespective of the exposure time, there would be an amount of hand shake that corresponds to the sensor’s readout time.

 

Is this a valid concern?  

 

Now, it may be that, because the amount of time taken to read adjacent rows on the sensor is quite small, the overall effect will be somewhat mitigated (and be less than with an exposure time of 1/10s), but it is hard to see why there would not be some degree of distortion.  If that is indeed the case, it does perhaps diminish one of the M11’s supposed advantages.

 

(I should also try hard not to think too much about the relative merits of the M10-R and M11.  Life and enjoyment is not a spreadsheet; so long as I convince myself that I will have enough time to use it this summer, I just need to go ahead and try one.)

 

Many thanks in advance for any thoughts you may have.

 

Andrew

Edited May 5, 2022 by AndrewLT
To simplify explanation of my question.

[jaapv]

Electronic shutter is used to mitigated shutter slap which has nothing to do with hand shake, which occurs irrespective of shutter method. I would say you are overthinking this. The readout speed of the shutter reams is the same irrespective of shutter or no shutter, so that will not make a difference either. At worst it would create some invisible distortion from hand shake but no unsharpness . 

[Advertisement]

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

[SrMi]

M11's readout speed is the same as a7IV's readout speed. So any potential issue you would observe with one camera, you would observe with the other. 
With a long lens, it is possible to see the effect of slow readout on a camera like X1D (250ms readout speed): a vertical line is not entirely vertical. However, M11's readout speed is likely fast enough that the distortion cannot be observed in most cases. 

 

Edited May 6, 2022 by SrMi

[Kwesi]

Here is my non scientific answer: you may or may not have a problem. 
on the M11, if you set the shutter to hybrid, the electronic shutter only kicks in at shutter speeds over 1/4000sec.  So no problem with your 1/10 scenario .

on the other hand if you wanted to shoot birds in flight with a shallow depth of field you may very well be shooting faster than 1/4000sec which because the bird is moving could be a problem. 

[Tailwagger]

As Kwesi suggests, the primary use case for the e-shutter is to expand the range of usable shutter speeds beyond that of the mechanical shutter, ostensibly for shooting fast lenses wide open in strong light. As Jaapv implies, a second use case is using the e-shutter at lower speeds to mitigate any effects of shutter slap. The third case, which tends to be ignored in these discussions is the ability to rig for silent running.  ie. you can avoid annoying an audience when shooting a play or wedding ceremony. 

Beyond the drawback of motion distortion, the other issue you might encounter is uneven exposure or banding artifacts when shooting under LED or florescent lighting.

6 hours ago, SrMi said:

M11's readout speed is the same as a7IV's readout speed. So any potential issue you would observe with one camera, you would observe with the other. 

Assuming, of course, you have IBIS turned off. 

[SrMi]

42 minutes ago, Tailwagger said:

<snip>

Assuming, of course, you have IBIS turned off. 

Good point!
Notably, the shutter lag with an electronic shutter is longer than with a mechanical shutter (RDF: 22ms vs. 13ms). However, in practice, one may not see a difference.

[AndrewLT]

Advertisement (gone after registration)

Many thanks indeed for all the considered and helpful replies.  Much appreciated.  If I may respond to a few of the points made:

18 hours ago, jaapv said:

Electronic shutter is used to mitigated shutter slap which has nothing to do with hand shake, which occurs irrespective of shutter method. I would say you are overthinking this. The readout speed of the shutter reams is the same irrespective of shutter or no shutter, so that will not make a difference either. At worst it would create some invisible distortion from hand shake but no unsharpness . 

Re. the sentence I have highlighted: that's precisely what I tend to do!  😀  Well said!

Having said that, and if you'll pardon a little more overthinking ....

I appreciate that hand shake will occur regardless, but the essence of my concern (which may be completely misplaced) is that, whilst with the mechanical shutter, the period when hand shake could occur is purely the exposure time, whereas with the electronic shutter, hand shake could be having an effect during the period of the readout (i.e. 1/10s).

Am I wrong to think this?  I may be betraying a lack of knowledge of the intricacies of shutter operation.  My assumption is that, with the mechanical shutter, the sensor is only exposed for the exposure time (even though the readout will then take 1/10s), but that with the electronic shutter, it will remain exposed for 1/10s (even though each individual line of the sensor is only collecting light for the exposure time).  If this is incorrect, I then don't understand why the electronic shutter -- when used with somewhat longer exposures -- has a problem with moving targets ("rolling shutter") in a way that the mechanical shutter does not (with the same exposure time).

It may be, though, that 1/10sec's worth of hand shake is typically less than the effect of shutter shock?

 

10 hours ago, SrMi said:

M11's readout speed is the same as a7IV's readout speed. So any potential issue you would observe with one camera, you would observe with the other. 
With a long lens, it is possible to see the effect of slow readout on a camera like X1D (250ms readout speed): a vertical line is not entirely vertical. However, M11's readout speed is likely fast enough that the distortion cannot be observed in most cases. 

 

Good point.  This is absolutely not specific to Leica, exactly the same would apply to any camera with a relatively long readout time.

In fact, I have now done a quick and nasty test using my wife's A7R IV (110mm lens, 1/250s exposure, IBIS turned off).  My tentative impression is that there is no obvious difference between the two shutters, but I will try to make a more methodical test in better light when I get a chance.

 

6 hours ago, Kwesi said:

Here is my non scientific answer: you may or may not have a problem. 
on the M11, if you set the shutter to hybrid, the electronic shutter only kicks in at shutter speeds over 1/4000sec.  So no problem with your 1/10 scenario .

on the other hand if you wanted to shoot birds in flight with a shallow depth of field you may very well be shooting faster than 1/4000sec which because the bird is moving could be a problem. 

Yes, you're quite right to make the point about hybrid mode.  However, it's not a 1/10s exposure time (or anything longer than 1/4000s) that I'm thinking about, but the fact that, with the electronic shutter, hand shake could be having an effect for the 1/10s readout period -- irrespective of exposure time -- in a way that can't happen with the mechanical shutter.

Interestingly, in his excellent video that discusses the mechanical and electronic shutters, @Overgaard recommends using the electronic shutter for all situations other than under artificial lighting and with moving targets.  That makes good sense to me, so long as my concern about hand shake during the readout period is not, in practice, a material problem.  

 

Thanks again to all who replied.

Andrew

[jaapv]

Camera movement will not create unsharpness but minimal distortion. You will not see it at all under normal circumstances.

But with speeding cars and a slow focal plane shutter, you do:

 

 

[jaapv]

Camera movement will not create unsharpness but minimal distortion. You will not see it at all under normal circumstances.

Here you have a typical effect,  one 1/1600 classic shutter and 1/12000 electronic shutter. No appreciable difference in sharpness (such as there is here  ) In fact, the classic shutter shows motion blur on the blade tips, the electronic shutter not.

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!

 

[AndrewLT]

Thanks, Jaap.  I'm grateful for your taking the time to post these examples.

Here is another quick illustration, taken on my wife's Sony A7R IV, 110mm lens.  ISO 400, 1/80s, f/5.6.  Hand held.  Artificial light (!!).  Imported into Lightroom.  Cropped to approx. 1/9th of the whole image.  Sharpening reduced to zero, exposure +0.5, no other changes.  Exported without sharpening.  I took six each with mechanical and electronic shutter, these are what I judge to be the best two.

(This is the first time I have posted photos, so apologies in advance if it goes wrong or if I have not done this the best way.  I'm not very good at this social media stuff.)

The first one (3917) is with the mechanical shutter, the second one electronic.  If anything, the electronic one looks marginally better.  QED?  (Although I should probably repeat this little test in stronger daylight.)

Andrew

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!

[Kwesi]

5 hours ago, AndrewLT said:

Interestingly, in his excellent video that discusses the mechanical and electronic shutters, @Overgaard recommends using the electronic shutter for all situations other than under artificial lighting and with moving targets.  That makes good sense to me, so long as my concern about hand shake during the readout period is not, in practice, a material problem.

Does he explain why? I admire Thorsten’s photographic style and can see why he would find being able to dispense of ND filters in broad daylight a godsend but does he give a technical reason why an e-shutter would produce a better image than a mechanical one at an exposure of let’s say 1/500 @  5.6 ?

 

[jaapv]

Because it avoids mechanical vibrations from the shutter.

[Overgaard]

The reason to use the electronic shutter would be to avoid the (confusing) sounds of the mechanical shutter. That was the initial response to the first firmware shortcomings where the mechanical shutter was also found to not release at first shot and/or several of the shots taken in a series. After a while of use now, my experience is that the electronic shutter shows to create warped images of faces, buildings, etc even at very fast shutter speeds (like 1/1000, because the readout of the sensor is 1/10th second which basically mean the sensor records movements in that period, which is impossible to avoid in any handheld situation).

So, more about those problems and possible ways to work around it in later episodes of the magicoflight.tv videos and overgaard.dk articles. 

 

[SrMi]

4 hours ago, jaapv said:

Because it avoids mechanical vibrations from the shutter.

My tests found barely visible mechanical vibrations with long lenses only (e.g., R180).

 

 

[jaapv]

It is indeed something that is more applicable to longer lenses. See the L-mount forums

[AndrewLT]

Blimey!

6 hours ago, Overgaard said:

The reason to use the electronic shutter would be to avoid the (confusing) sounds of the mechanical shutter. That was the initial response to the first firmware shortcomings where the mechanical shutter was also found to not release at first shot and/or several of the shots taken in a series. After a while of use now, my experience is that the electronic shutter shows to create warped images of faces, buildings, etc even at very fast shutter speeds (like 1/1000, because the readout of the sensor is 1/10th second which basically mean the sensor records movements in that period, which is impossible to avoid in any handheld situation).

So, more about those problems and possible ways to work around it in later episodes of the magicoflight.tv videos and overgaard.dk articles. 

 

This is fascinating -- and is, of course, precisely the concern that I tried to articulate yesterday: that, irrespective of exposure time, in electronic shutter mode, the sensor will record any movement, such as hand shake, during the readout period (even though individual lines in the sensor are only recorded for an exposure time's duration).

That being the case, it is hard to see how this could not, at least sometimes, have a material effect.

Now, my quick and nasty test yesterday evening was almost certainly not adequate.  It is sunny here this morning, so I have made a few quick shots (in much better light than yesterday).  Rather amazingly, I can see a warping effect, sometimes quite pronounced, roughly half of the time when using the electronic shutter.  I will hold off posting examples until I have convinced myself that it was not a fluke, and I certainly do not wish to conclude anything prematurely, but this is getting interesting ....

Thank you once again for the continuing replies.

Thank you also to @Overgaardfor stating this so clearly.  If you happen to see this post, Mr Overgaard: are you able to give any indication of when you may release any further videos or articles?  (I am close to convincing myself that I should buy an M10-R or M11, and this is one of the factors that will sway my decision, so any further perspectives that you are able to give will certainly be helpful.)

Andrew

[AndrewLT]

Fascinating.  Having taken a few more shots, I am seriously wondering whether this effect really does exist.  The word “warping” prompted me to think: why not try shooting something with obvious lines?  (And, yes, I know that the door needs painting!)

I should preface this by saying that there may well be a different explanation.  In the two images I have posted below, it is entirely possible that I have done something stupid or that something else is happening. I am not regarding this as conclusive proof.  Nonetheless, I am finding it increasingly hard to see how there could not be some degree of hand shake with an electronic shutter and a readout time of 1/10s.  It may be more pronounced with longer lenses — but, equally, I would never normally use a 1/10s exposure time with a 50 or 35mm lens.

These two images were taken, once again, with my wife’s A7R IV.  110mm lens, ISO 100, 1/500s, f/8.0.  Lightroom export with no sharpening, having reduced sharpening to zero, no other adjustments.  I have cropped each image to show the top two thirds of the central part of the image (i.e. two ninths of the total image), simply to reduce the file size.  You can probably guess which one is the electronic shutter.  These are the two images that show the effect (whatever it may be) most clearly.

I did wonder if I could somehow be twisting the camera with some of the shots I did.  But the “warping” effect was discernible in four of eight electronic shutter images, but none of eight mechanical ones.

You might argue that I should not worry my pretty little head about this, but I have to confess that I am finding it quite interesting, in a geeky sort of way.  And I find myself beginning to think that, if this really is a characteristic of an electronic shutter (when combined with a slowish readout time), there is an argument — for me, at least — for plumping for the traditional “open / close” operation of the M10-R’s mechanical shutter.  Although some of the other features of the M11 are quite tempting!

(I hope it is okay to be posting images from a Sony camera here.  Obviously, the point is to illustrate use of mechanical vs. electronic shutters with a sensor that has a comparable readout time.  I do not posses a Leica M with an electronic shutter.)

Andrew

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!

[LocalHero1953]

10 hours ago, Overgaard said:

the readout of the sensor is 1/10th second

I don't have the M11, but I'm surprised at this - it is very slow compared to other Leicas and other brands, which are typically 1/20s at the slowest, and often faster than 1/30s.

A few years ago I did a Q&D test of the SL by panning across a vertical line on a wall, including it in the frame over about 1 sec. Measuring the amount of 'bend' across a 36 wide frame gave me an estimate of about 1/30s.

It would be disappointing if 5 years later the M11 was worse than this.

[SrMi]

37 minutes ago, LocalHero1953 said:

I don't have the M11, but I'm surprised at this - it is very slow compared to other Leicas and other brands, which are typically 1/20s at the slowest, and often faster than 1/30s.

A few years ago I did a Q&D test of the SL by panning across a vertical line on a wall, including it in the frame over about 1 sec. Measuring the amount of 'bend' across a 36 wide frame gave me an estimate of about 1/30s.

It would be disappointing if 5 years later the M11 was worse than this.

The readout speed is faster with lower resolutions and when using 12 instead of 14 bits.

Every camera using Sony’s 60MP sensor has the same readout speed.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[frame-it]

7 minutes ago, SrMi said:

Every camera using Sony’s 60MP sensor has the same readout speed.

interesting..how do you know its a sony sensor?