M9 shutter lag?

[marcusperkins]

Advertisement (gone after registration)

It's interesting how different people perceive time.

 

There are people who believe the approximate 80ms shutteror lag for the Leica is insignificant, and that the camera fires instantaneously, but at the same time feel that the mirror blackout of approximately 49ms on an equally priced SLR is a big disadvantage in that vital visual information is missed.

 

I also never quite understood the argument that a digital rangefinder has a big advantage over an SLR with regards to the 'decisive moment' because of the mirror. Until the shutter button is released both systems allow the photographer a continuous view of the scene.

 

Once the shutter button has been pressed, there is nothing more either photographer can do to change the point at which the exposure is made, so whether the mirror 'blinks' or not makes no difference to whether you catch the 'decisive moment'.

 

However, what does make a difference is shutter lag, and this puts the digital Leica at a big disadvantage in comparison to film M, and to a lesser extent a pro SLR.

 

There is undoubtedly a reassuring feeling to seeing through an exposure, particularly a slow one. But to my mind that experience is wasted when allied to a delayed or clumsy sounding release.

Edited October 7, 2009 by marcusperkins

[nhabedi]

However, what does make a difference is shutter lag, and this puts the digital Leica at a big disadvantage in comparison to film M, and to a lesser extent a pro SLR.

 

Assuming the numbers quoted here are correct. What I've seen so far are different values from different sources plus values without any sources at all. For me, that isn't enough to base conclusions on. Nor is anecdotal evidence by forum members.

[Advertisement]

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

[marcusperkins]

In the end the numbers mean nothing, it's about the sense of it 'feeling' right. The shutter in my MP feels right, the shutter in my Canon 1 feels right. However, the shutter in my ex M8 always felt wrong. For whatever reason it just did not fit into my way of working, and the M9 (sadly for me) did not feel any different.

 

This is of course a very personal opinion.

Edited October 7, 2009 by marcusperkins

[mat_mcdermott]

Marcus, fair enough. For me the M8 shutter feels no different in practice than my film Ms did -- even if on paper it is less responsive. I'm not missing any more shots because of statistical differences in shutter lag in the slightest and I'd be surprised if having a more responsive shutter really changed that. Again, for me... YMMV

 

When it comes down to it, in all my years of shooting I've found that when shooting with anything other than a point and shoot, shutter lag to be a non-issue in any camera I've used (from manual-wind film SLRs, to the Canon 1 series, film and digital rangefinders). As you've brought up, a more general feel of the camera probably plays more a role in getting the shot than does anything else.

 

As far as always having a clear view goes, this is obviously just down to perception, but I've always found that I get a better sense of if I've gotten the moment, or have to shoot again very quickly, without mirror blackout. Never felt that with SLRs -- even though blackout times are now down to a minimum, there's something about the constant view of the rangefinder that I still just prefer.

[tgray]

Your statements are undocumented as well. "A quick Google will show you ..." documents nothing at all. Give us a precise source than we can relate to.

 

The site you are referring to seems to be a commercial site set up for selling a device aimed at certain people with cameras with a certain shutter lag. But the numbers on this site are completely undocumented and probably unreliable, since they may be set up for the purposes of the site which seem only to be to sell as many devices as possible.

 

Give us honest documentation to reliable sources when making factual statements. If you cannot give such documentation your statements have no credibility.

 

Well, good luck with that attitude. I gave you a link, a document, and you didn't like it. You can reject any document if you want by saying the document itself is also undocumented.

 

As far as the D3 and Canon 1 series go, the numbers are probably NOT unreliable. Some of those cameras shoot 10-11 fps in certain modes. At the very minimum, that gives you 100 ms for each frame (less in the 11fps case). I think its reasonable to believe that it takes about as long for the mirror to return as it does for it to go up, which gives you roughly 50 ms between the time you press the button and the picture is taken. I think the numbers quoted were 55 ms and the Canon 1 series and 37 ms for the Nikons.

 

The Leica 80 ms number comes for a Leica document which is no longer online. You can see it in the html cached version. I also don't know if that's accurate

 

http://m.leica-camera.com/res/dnl/en/LEICA_M8_Folder_engl.pdf

[marcusperkins]

What this thread is increasingly proving is that we photographers get incredibly familiar with equipment, developing an intimate sense of the sounds and timings of our favourite cameras.

 

Most of us can probably sense, and often comment on, the tiniest facets of a camera's operation, so much so that our friends and family probably think we're utterly crazy.

[jonoslack]

Advertisement (gone after registration)

What this thread is increasingly proving is that we photographers get incredibly familiar with equipment, developing an intimate sense of the sounds and timings of our favourite cameras.

 

Most of us can probably sense, and often comment on, the tiniest facets of a camera's operation, so much so that our friends and family probably think we're utterly crazy.

 

HI Marcus

I think you're right - I'd never considered the lag with the M8 / M9 before, but now I look I can tell it's there.

 

But it seems to me that most 'catching the moment' photography is about anticipation and response rather than lightning reactions.

 

For me, what I really hate in a camera is a variable lag - so that you can't be sure when the camera will take the picture, lots of compact cameras do this.

 

What certainly is the case is that I don't feel myself to be missing 'critical moments'. However, I can see that if you were familiar with the action of an instant shutter (an M3), then things would, indeed, be different, and thus irritating.

[ho_co]

... The Leica 80 ms number comes for a Leica document which is no longer online. You can see it in the html cached version. I also don't know if that's accurate

 

http://m.leica-camera.com/res/dnl/en/LEICA_M8_Folder_engl.pdf

 

Tiger, where do we find the "Leica document which is no longer online"?

 

For me, the link doesn't work. Where is the "html cached version"?

 

Thanks.

[ho_co]

As others have mentioned, the advantage of a site like Shooting Digital : Ed Schwartz : Shutter Release Test is that an individual can compare his own times with his own cameras on his own computer, not that an accurate value can be derived.

 

I would guess that although it's a useful test for an individual, just the variation among individuals makes it unsuitable for deriving a 'standard' value.

[diogenis]

no, but it's fun

Guys do you know what ms represent?

btw, that test is a nice reflex test

[tgray]

Tiger, I like that. haha.

 

Paste the link into google and search for it. Then click on "view as html." It's on page 11 I think, or 13. It's not terribly informative.

 

Here's the way I see it. Any test like that one shown linked to that the user can perform himself unfortunately can't account for reaction speed AND anticipation. It might be great for finding out how much to anticipate a shot by, but hard to find the actual shutter speed.

 

If you could trigger all your cameras electronically, you could send them the fire signal at t=0 and shoot a really fast clock, or even another benchmarked source to determine shutter lag. Heck, if you had access to a fast 1000 fps camera, you could just film the shutter and see when it opens up. That way, you'd know all cameras were triggering at t=0 and you could easily find when the shutter opens.

 

However, the M9/M8 doesn't have an electrically triggered option as far as I know. I guess you could rig up a physical and repeatably accurate mechanical 'finger' to push your camera's shutter button at t=0 to give you a consistent start.

 

So we rely, at best in the M case, on manufacturer's claims as to what the shutter lag is. For all we know, Leica might have a different definition of shutter lag than Canon. So even if they both publish a number, it's hard to compare.

 

On the other hand, it seems like Leica is publishing or has published 80 ms for the M8. There's a number floating around on the net that is significantly shorter for film M's. I wonder why there is such a difference in those two values.

[nhabedi]

Tiger, where do we find the "Leica document which is no longer online"?

 

For me, the link doesn't work. Where is the "html cached version"?

 

Just enter the URL into Google and there you have it. And, yes, I checked and it's true that Leica themselves at that time claimed that the shutter lag of the M8 was 80ms.

[nhabedi]

So we rely, at best in the M case, on manufacturer's claims as to what the shutter lag is. For all we know, Leica might have a different definition of shutter lag than Canon. So even if they both publish a number, it's hard to compare.

 

Right, that's what I thought as well. Maybe Canon/Nikon measure the time from the moment the electric circuit is closed while Leica starts measuring when the button is pressed mechanically. In ms, that could already be a difference.

 

On the other hand, it seems like Leica is publishing or has published 80 ms for the M8. There's a number floating around on the net that is significantly shorter for film M's. I wonder why there is such a difference in those two values.

 

The M8 and the M3 certainly have totally different shutters. That's why I asked above if anyone has numbers for the M7. It doesn't have the same shutter as the M8, but at least I think it is more similar to the M8's shutter than to the M3's shutter.

 

Another reason for the shutter lag might be that in a digital camera probably a lot of things have to happen internally before the shutter is actually fired. It wouldn't surprise me if Canon/Nikon where simply better at highly optimizing these tasks using for example custom chipsets. It is hard to deny that there are other areas in the electronics/IT sector where the Japanese manufacturers simply seem to have better solutions. (The speed of zooming into previews on the LCD or RAW write speed come to mind.)

[kipkeston]

Right, that's what I thought as well. Maybe Canon/Nikon measure the time from the moment the electric circuit is closed while Leica starts measuring when the button is pressed mechanically. In ms, that could already be a difference.

 

If that's the case, Leica might want to consider measuring and publishing their shutter lag in a more competitive manner =)

 

Anyway, I can't tell the difference between 15, 40 and 80ms.

[nhabedi]

If that's the case, Leica might want to consider measuring and publishing their shutter lag in a more competitive manner =)

 

Right...

 

Anyway, I can't tell the difference between 15, 40 and 80ms.

 

Subconsciously, you probably can...

[tgray]

Yes, you can tell the difference between small times in certain cases. Take movies versus video. The difference between 24 fps and 30 fps is readily visible, which amounts to 42 ms versus 30 ms per frame. I know its a bit more complicated than that since its motion, but still. Someone else did the calculation earlier of how much of a difference 50 ms can make with objects that are moving at some speed.

 

As far as optimizing electronics, 50 ms is pretty long in terms of processing and times in circuits. I would hope Leica doesn't have that kind of delay due to inexperience. After all, what needs to occur between pushing the button and opening the shutter? If you haven't metered, an exposure calculation needs to take place, but that should be very quick. I can't think of anything else. Intuitively, I would think most of the delays come from the mechanical parts. SLRs traditionally have had long shutter lags most likely because it just takes time to move the mirror out of the way. The cheaper ones have longer lags because the mirrors don't move as fast, etc., but they didn't need to since their frame rate was slower.

 

I would think the M7's shutter is more similar to the other film Ms than it is to the M8/M9. The release is a bit different, but the physical shutter is cloth and moves horizontally, giving it all of the limitations and characteristics of other film M shutters: slow flash sync, noise characteristics, slow top speed, etc.

[diogenis]

I tried the timer and got anything from 0 to 100-150ms: which means that there is no real difference in practice for upto 150-200ms.

Actually, the full cycle of shooting and then recocking is something along 1-1.5secs far more than shutter lag, which in most cases is reflexes+mechanical push button

[nhabedi]

.As far as optimizing electronics, 50 ms is pretty long in terms of processing and times in circuits. I would hope Leica doesn't have that kind of delay due to inexperience. After all, what needs to occur between pushing the button and opening the shutter? If you haven't metered, an exposure calculation needs to take place, but that should be very quick. I can't think of anything else.

 

Resetting the sensor comes to mind. Or "emptying" it - don't know what's the right word in English. Michael Hußmann just explained in the German forum what has to happen before the shutter is opened to let light fall on the sensor. I have no idea how long it would take to empty a full-frame CCD sensor, though, or if resetting a CCD sensor takes longer than resetting a CMOS sensor.

[garygsandhu]

According to a technical representative at Leica, I have been advised that the shutter lag is the same as the M8, which is 80 ms.

 

SHUTTER LAG IS 80 ms.

[Advertisement]

Hallo,

Look here: Leica M

[Overgaard]

...new to digital, finally took the plunge and got the M9 after years of film M use. Intention is to continue shooting film and grow into digital. Initial feeling re: M9 is "wow", but I am a bit concerned about the discernible shutter lag (relative to mechanical Ms) and need to understand if this is normal for digital Ms.

 

Would appreciate input from users that shoot with both film and digital Ms. Thanks.

 

No matter which camea I shoot, M4, R9, D2, M9, I press the shutter to the edge, anticipating the right moment, then press he shutter the last step and it shoots.

 

In my opinion the M9 shoots instantly, or so instantly that whatever micro-delay there might be, is part of your rhythm as a shooter.

 

I don't think you can get that feeling for where the release point, and that rhythm any camera requires for using, in a shop. Not even within the first hours of ownership.

 

I had trouble the first day with the soft and discrete release as the camera fired before I expected. I have an idea I was using my feel and rhythm of the M4, but the M9 fired earlier in the "shutter pressing process."

 

I would say there is no delay in the M9.

 

One could count the "delay" from holding the finger a centimeter above the release button, then measure the delay from pressing down and till the camera fires. Problem is that's not the way to shoot (as little as talking about shutter delay from hanging over the right shoulder to the picture is taken).

 

From you actual release of the shutter, how long is the delay. None, it's Instant. (And then there's an audible delay till the rewind begins; eventually take off the lens and look what the shutter curtain does - the sound of delay is not a picture taking delay but the rewind delay. But if you only listen you might get the idea there's a delay).

 

We all know those p&s where no matter how hard you press, the camera doesn't shoot till 1/2 or 1 second after. But that is certainly not the case with the M9. It fires when activated. But you have to find the point in the "shutter pressing process" where it activates.