Lag test - M11 in rangefinder vs liveview mode

[digitalfx]

Advertisement (gone after registration)

1 hour ago, dkmoore said:

I get this delay is very very small but would it still impact focus if you are not perfectly still or using a tripod?

for instance, you may sway slightly after clicking the shutter release which could move the focus point. I think this would only matter with 1.4 and faster lenses and shooting portraits and focusing on the eyes. 
 

Is the lag far too short a time for the above to be an issue?

1/1000th of a second slower than M10, is that enough to really make a difference? I’d think your body movement (from moment of focus to actual shutter release) has a far larger impact of rf focus. We are literally splitting hairs here, but unless you are shooting while jumping on a trampoline I cant see how 10ms would make any appreciable difference. Human reactions and natural movements have far a larger impact on rangefinder focusing (both photographers and subjects). 

[SrMi]

1 minute ago, mmradman said:

Thanks for clarifying.

Shutter lag regardless of reason is never good thing.  If Electronic shutter is laggy (if there is such word) than it would be good for static & semi-static subjects only.  No capturing decisive moments there.

More importantly, how responsive is mechanical shutter considering that M11 works with  shutter curtain open at all times?  My only current digital M is 246 has mechanical shutter only and when shooting using OVF only it is as good as instant while shooting in LiveView/EVF (shutter again only mechanical) is well known to be painfully slow - about 1 second to take a shot. 

Nobody said that the electronic shutter in M11 is laggy. Leica said that the lag is higher than with the mechanical shutter. The new mechanical shutter is, according to Leica, 10ms slower than the old-style shutter. That is not noticeable. LV in M11 is much better than in any previous M-s. The main "issue" with LV (in all cameras) is the lag from live view (30ms, AFAIK).

[mmradman]

27 minutes ago, SrMi said:

Nobody said that the electronic shutter in M11 is laggy. Leica said that the lag is higher than with the mechanical shutter. The new mechanical shutter is, according to Leica, 10ms slower than the old-style shutter. That is not noticeable. LV in M11 is much better than in any previous M-s. The main "issue" with LV (in all cameras) is the lag from live view (30ms, AFAIK).

Maybe I got it all wrong but going back to Leica own video Leica own representative don’t make strong case for own product when it comes to shutter. 

If the shutter latency be it mechanical or electrical is generally in keeping with industry average than there no issue.  It would be in Leica own interest to be clear what to expect in terms of new camera in relation to previous models and in respect to competition.  The fact that this is being debated may leave impression that M11 is a case of latest sensor with clear improvements on past models installed in M10 shell but under developed electronics 🤷‍♂️. 

[digitalfx]

13 minutes ago, mmradman said:

Maybe I got it all wrong but going back to Leica own video Leica own representative don’t make strong case for own product when it comes to shutter. 

If the shutter latency be it mechanical or electrical is generally in keeping with industry average than there no issue.  It would be in Leica own interest to be clear what to expect in terms of new camera in relation to previous models and in respect to competition.  The fact that this is being debated may leave impression that M11 is a case of latest sensor with clear improvements on past models installed in M10 shell but under developed electronics 🤷‍♂️. 

How do you get that from the video?  he says "the shutter lag as compared to the M10 is quite the same, its 10ms slower, but in practice you dont feel any difference..."

Its being debated mostly by those without the camera in hand. IMO its not an issue and I have shot with every digital M. 

[Daedalus2000]

1 hour ago, mmradman said:

Thanks for clarifying.

Shutter lag regardless of reason is never good thing.  If Electronic shutter is laggy (if there is such word) than it would be good for static & semi-static subjects only.  No capturing decisive moments there.

More importantly, how responsive is mechanical shutter considering that M11 works with  shutter curtain open at all times?  My only current digital M is 246 has mechanical shutter only and when shooting using OVF only it is as good as instant while shooting in LiveView/EVF (shutter again only mechanical) is well known to be painfully slow - about 1 second to take a shot. 

I did some experiments here to give you an idea, but I do not have another rangefinder to compare with. Maybe you can do the same experiment and see what you get.

 

[AandW]

I wonder if the reason electronic shutter has more lag is turning on all pixels has to mimic the reading at the end of exposure to ensure it is uniform.   I understand a sensor’s pixels can be turned on very fast (hence EFCS) so with mechanical shutter the camera closes shutter, turns all pixels on as fast as possible, and then opens. With electronic shutter it clears pixels and then turns on row by row at the same pace they will be read to ensure each row has the same exposure time.  If a mechanical shutter opens or closes in 5 ms and an electronic sensor read takes 30 - 50 ms then shutter lag would be longer with an electronic shutter depending how it is measured.  If lag was based on the last pixel turned on then much longer, medium pixel, less longer.   This is conjecture my part, I could not find it discussed in my brief search of the web.

[mmradman]

Advertisement (gone after registration)

Two fundamental issues with modern cameras equipped with electronic shutter:-

First, latency to start recording once shutter release button is pressed, as I understand it delay is typically in tens of milliseconds and it probably not much worse than some purely mechanical cameras. Probably cameras with first electronic curtain are slightly faster.

Second, sensor readout time, slow sensors typically 1/10 sec. including M11 record moving objects distorted, I.e. subject is not frozen although exposure time can be much faster, up to 1/16000 sec.

Interesting illustration here from well known YouTube reviewer...

 

 

 

 

[SrMi]

Here are shutter lag measurements for Sony a7rIV.

- Shutter lag is shorter if holding the shutter button half-pressed. I wonder if that is the case with Digital M-s.

- EFCS on is faster than off.

- Electronic shutter is way slower.

[digitalfx]

3 hours ago, SrMi said:

Here are shutter lag measurements for Sony a7rIV.

- Shutter lag is shorter if holding the shutter button half-pressed. I wonder if that is the case with Digital M-s.

- EFCS on is faster than off.

- Electronic shutter is way slower.

Its pre-focusing when holding the shutter...unfortunately the M11 doesnt have auto focus

Sony has 125x more lag in MF vs M11

 

[SrMi]

37 minutes ago, digitalfx said:

Its pre-focusing when holding the shutter...unfortunately the M11 doesnt have auto focus

Sony has 125x more lag in MF vs M11

 

Prefocused means holding the shutter half pressed. With Sony that means focus and metering locked. 

[Ba Erv]

When I did my rotary wing training the first thing the instructor told me was you have to learn to anticipate the aircraft...so, just start thinking about your index finger 10ms sooner😀 

[SrMi]

The shutter lag of M9 is a "horrible" 0.125 sec (link). Nobody noticed, nobody complained, AFAIK. Sean Reid measured M11's shutter lag to be 0.064sec.

[hdmesa]

On 1/24/2022 at 12:04 AM, adan said:

...I think it requires a "stacked" BSI sensor, with its own dedicated processor and buffer memory attached right to the back of the sensor (for processing and storage speed), and electronic shutter - usually global (all pixels read at once - no rolling shutter)...

Stacked sensor like in the Z9 is not a global shutter and does not expose the sensor all at once – the readout still cascades across the sensor. It's just so fast that the speed of reading the sensor is now equal to the speed of a mechanical shutter – hence no need for a mechanical shutter in the Z9. A global shutter is an entirely different tech that is yet to make its way into mass production.

[orcinus]

Scenario 1: OVF

t=0ms - Event you are photographing occurs. Your eyes see it.

t=150ms - Your brain has received the input from the eyes, and made a decision.

t=200ms - Your finger finally moves and presses the shutter release.

t=264ms - Shutter is activated. Event has been captured with a 264ms delay.

 

Scenario 2: EVF

t=0ms - Event you are photographing occurs. Sensor sees it.

t=50ms - Image of the event shows up in the electronic viewfinder screen.

t=200ms - Your brain has received the input from the eyes, and made a decision.

t=250ms - Your finger finally moves and presses the shutter release.

t=314ms - Shutter is activated. Event has been captured with a 314ms delay.

Could somebody please explain to me how is this extra EVF delay significant compared to the bulk of total reaction time of the system (camera+operator), because i’m obviously still missing something?

Edited January 26, 2022 by orcinus

[adan]

33 minutes ago, orcinus said:

Could somebody please explain to me how is this extra EVF delay significant compared to the bulk of total reaction time of the system (camera+operator), because i’m obviously still missing something?

Every little bit helps - or hurts.

 

Edited January 26, 2022 by adan

[Daedalus2000]

30 minutes ago, orcinus said:

Scenario 1: OVF

t=0ms - Event you are photographing occurs. Your eyes see it.

t=150ms - Your brain has received the input from the eyes, and made a decision.

t=200ms - Your finger finally moves and presses the shutter release.

t=264ms - Shutter is activated. Event has been captured with a 264ms delay.

 

Scenario 2: EVF

t=0ms - Event you are photographing occurs. Sensor sees it.

t=50ms - Image of the event shows up in the electronic viewfinder screen.

t=200ms - Your brain has received the input from the eyes, and made a decision.

t=250ms - Your finger finally moves and presses the shutter release.

t=314ms - Shutter is activated. Event has been captured with a 314ms delay.

Could somebody please explain to me how is this extra EVF delay significant compared to the bulk of total reaction time of the system (camera+operator), because i’m obviously still missing something?

First of all we need to identify 4 different cases for the M11 vs previous rangefinders comparison:

1) OVF mode in old rangefinders where the sensor is not used for metering (e.g. M10 )

2) OVF and NO live view in M11

3) OVF and live view in M11

4) EVF (and live view of course) in M11

I have no clue what the extra lag between (1) and (2) is, maybe this is what Leica says is 10ms.

My personal estimates are that the lag between (2) and (3) are 100ms, and between (3) and (4) another 100ms. They do sound big so I am waiting for someone else to verify or negate this.

To your question... I guess the human brain focuses on differences between different options and takes into account even small differences. I thought the same that 10-20ms cannot really be perceivable, but then I realized the shutter lag of the film camera M3 is less than 20ms... So as adan says above, every little helps. But, my own crude calculations show that having live view on can be an extra 100ms or even more, depending if you use the EVF/LCD or not.

 

 

 

 

[hdmesa]

38 minutes ago, orcinus said:

Could somebody please explain to me how is this extra EVF delay significant compared to the bulk of total reaction time of the system (camera+operator), because i’m obviously still missing something?

Capturing a "derp" facial expression from the subject instead of the one you wanted? 🥴

The human ability to recognize a meaningful change in facial expression is about 500 ms, so in either scenario, we're capturing the image with a delay that impacts the outcome.

That being said, the perfect moment to press the shutter has always been slightly in advance of an optimal moment, so with time we attune ourselves to the delay. Anticipation has always been the key to being a good photographer of anything in motion.

Edited January 26, 2022 by hdmesa

[SrMi]

44 minutes ago, hdmesa said:

Capturing a "derp" facial expression from the subject instead of the one you wanted? 🥴

The human ability to recognize a meaningful change in facial expression is about 500 ms, so in either scenario, we're capturing the image with a delay that impacts the outcome.

That being said, the perfect moment to press the shutter has always been slightly in advance of an optimal moment, so with time we attune ourselves to the delay. Anticipation has always been the key to being a good photographer of anything in motion.

Hear! Hear!

[Ba Erv]

I suppose I could see if I can get reimbursed for PET, EEG, and Nerve Conduction Studies to help you guys out here…I’m just not sure which ICD-10 Code is “M11 Shooter Lag Syndrome”.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[hdmesa]

4 minutes ago, Ba Erv said:

I suppose I could see if I can get reimbursed for PET, EEG, and Nerve Conduction Studies to help you guys out here…I’m just not sure which ICD-10 Code is “M11 Shooter Lag Syndrome”.

F42, R46.4