Call For Caution on "Shutter Shock" Posts

There have been recent posts alleging varying degrees sharpness of photos with OMD-EM1 with differing shutter speeds. These findings have been attributed to "shutter shock" or similar phenomena. However, in at least some of these cases, the "tester" was keeping the exposure constant. Hence, the aperture varied. The result was comparing sharpness at, say f8 vs. f 3.5 and attributing the differences to "shutter shock". Recall that varying the aperature

1. Changes the depth of field, giving appearance of sharpness at smaller apertures

2. Changes the behavior of the lens. Many lenses have "sweet spots"- f8 being a common one.

Hand holding is also a poor way to evaluate this as the results will vary from shot to shot and would require inspection of many images to draw a firm conclusion.

Therefore, I make a plea to be cautious about blaming one variable for changes in sharpness when multiple variables are changed.

Anytime someone gets a photo that is blurry or simply not very good it's deemed to be "shutter shock".

I keep waiting to hear that everything from the Affordable Care Act website debacle, to the sinking of the Titanic, to the Hubble Telescope fiasco, to the fall of Madrid in 1939 was due to shutter shock. Considering the number of new books and TV documentaries showing up regarding the 50th anniversary of the assassination of President Kennedy, I am waiting for some title like: "Secret CIA Documents Show the Rifle that Fired the 'Magic Bullet' had Shutter Shock, Meaning There Really Were Shots Fired from the Grassy Knoll." That's not as weird as it sounds; just saw a new book title that, once again, claims to prove that Lyndon Johnson was behind the murder. And he didn't even have an EM1...

[No message]

Stuart said:

There have been recent posts alleging varying degrees sharpness of photos with OMD-EM1 with differing shutter speeds. These findings have been attributed to "shutter shock" or similar phenomena. However, in at least some of these cases, the "tester" was keeping the exposure constant. Hence, the aperture varied. The result was comparing sharpness at, say f8 vs. f 3.5 and attributing the differences to "shutter shock". Recall that varying the aperature

1. Changes the depth of field, giving appearance of sharpness at smaller apertures

2. Changes the behavior of the lens. Many lenses have "sweet spots"- f8 being a common one.

Hand holding is also a poor way to evaluate this as the results will vary from shot to shot and would require inspection of many images to draw a firm conclusion.

Therefore, I make a plea to be cautious about blaming one variable for changes in sharpness when multiple variables are changed.

Click to expand...

Here's the link to the thread in question:




Note that the bad results were at f/8 (with a 1/100 shutter speed) and the good results were at a larger aperture with a faster shutter.



So, in fact, your arguments about aperture don't hold in this case. Furthermore, the claims that diffraction is the issue at f/8 also don't ring true.




Using the Panasonic 12-35 as a comparable lens, it's clear this lens is only losing about 15 to 20% of its center resolution at f/8 compared to its peak resolution. This wouldn't be sufficient to produce the obvious blurring in the shots. I routinely shoot at f/8 on quality lenses and have a difficult time noticing a difference from peak resolution.

So, in essence, aperture is effectively ruled out as the cause of the blur as far as I'm concerned.

And, unless the OP has very shaky hands, it seems improbable that he'd be getting such significant motion blur at 1/100s with a 12mm lens.

So, what's my point? The OP could very well be onto something, and, even though I agree he needs to do more testing, we shouldn't dismiss what he's shown.

It could also be a defective camera or lens, unless there's evidence that others have the same problem. I'd like to see more testing done to narrow it down to user error, camera problem, or lens problem. Then, if it's one of the latter two, determine if this is a manufacturing defect or a more systemic problem.

Beach Bum said:

Stuart said:

There have been recent posts alleging varying degrees sharpness of photos with OMD-EM1 with differing shutter speeds. These findings have been attributed to "shutter shock" or similar phenomena. However, in at least some of these cases, the "tester" was keeping the exposure constant. Hence, the aperture varied. The result was comparing sharpness at, say f8 vs. f 3.5 and attributing the differences to "shutter shock". Recall that varying the aperature

1. Changes the depth of field, giving appearance of sharpness at smaller apertures

2. Changes the behavior of the lens. Many lenses have "sweet spots"- f8 being a common one.

Hand holding is also a poor way to evaluate this as the results will vary from shot to shot and would require inspection of many images to draw a firm conclusion.

Therefore, I make a plea to be cautious about blaming one variable for changes in sharpness when multiple variables are changed.

Click to expand...

Here's the link to the thread in question:

http://www.dpreview.com/forums/thread/3568609?page=2

Note that the bad results were at f/8 (with a 1/100 shutter speed) and the good results were at a larger aperture with a faster shutter.

So, in fact, your arguments about aperture don't hold in this case. Furthermore, the claims that diffraction is the issue at f/8 also don't ring true.

Click to expand...

Got the same results at larger apertures 4, 5,6 with 1/100 shutter speed, this eliminates the diffraction argument.

Beach Bum said:

http://www.lenstip.com/367.4-Lens_r..._mm_f_2.8_ASPH._P.O.I.S_Image_resolution.html

Using the Panasonic 12-35 as a comparable lens, it's clear this lens is only losing about 15 to 20% of its center resolution at f/8 compared to its peak resolution. This wouldn't be sufficient to produce the obvious blurring in the shots. I routinely shoot at f/8 on quality lenses and have a difficult time noticing a difference from peak resolution.

So, in essence, aperture is effectively ruled out as the cause of the blur as far as I'm concerned.

And, unless the OP has very shaky hands, it seems improbable that he'd be getting such significant motion blur at 1/100s with a 12mm lens.

Click to expand...

It is not a problem of shaky hands, got excellent results at 1/8 sec and you have IBIS

Beach Bum said:

So, what's my point? The OP could very well be onto something, and, even though I agree he needs to do more testing, we shouldn't dismiss what he's shown.

It could also be a defective camera or lens, unless there's evidence that others have the same problem. I'd like to see more testing done to narrow it down to user error, camera problem, or lens problem. Then, if it's one of the latter two, determine if this is a manufacturing defect or a more systemic problem.

Click to expand...

Suspect it is a bad sample. My E P5 does not have the same problem and is tested in the same way.

Call for caution by Stuart

quote

Stuart wrote:

With all due respect,your statements are affecting the decisions of many on this forum .The link you keep referring to does not appear contain sample showing what you are talking about. Blur at 100% is not easy to assess. Perhaps I am missing something here.But if you can't give a direct link to your pics we must conclude that you are planting posts to mislead people.

unquote

My call for caution is not to make such sweeping statements especially in the case that many samples are provided with links to refer to. Yes you are missing something here and hope that posts remain civilized

DP review reported a shutter shock problem with the E P5. Did you ask for evidence, samples ?? They confirmed this problem not me

As i said before i did not have a problem with the E P5 and tested the E M1 in the same way. Posted more the 40 samples. I am now accused of planting post to mislead people.

An apology seems appropriate.

No doubt the E M1 is a good camera but might suffer from sample variation.

Some of the examples have shown double edges rather than blur or the softness I associate with camera movement.

Many of those complaining may simply be guilty of sloppy photography, but I'm sure there is also a real problem.

gotompoes said:

As i said before i did not have a problem with the E P5 and tested the E M1 in the same way. Posted more the 40 samples. I am now accused of planting post to mislead people.

An apology seems appropriate.

No doubt the E M1 is a good camera but might suffer from sample variation.

Click to expand...

No worries. He probably wasn't even referring to you. I also think you've missed his point. He wasn't suggesting that problems weren't real, just that too many people are using the term "shutter shock" to describe every problem. For example, in one of your posts you said that DPREVIEW had found shutter shock to be a problem when testing the E-P5. That's incorrect and the reviewers have said this is incorrect. What they found to be a problem was hand SHAKE. Shake isn't shock, those are two different things. Conflating them just confuses everyone.

From the examples you posted that I looked at, and I only looked at four or five, I didn't see anything that seemed to me to be "shutter shock". The images seemed sharp in some places and not in others. Struck me as more a lens or camera problem.

This in particular is the forum where those with almost no understanding of the principles of how exposure is achieved (but with plenty of cash), meet with those who've plenty of understanding and experience. plenty for us all to learn given the pace of change with m43 cameras. A shame nearly every thread becomes a shouting match. I tend to imagine the physical manifestation of this forum as something akin to the bar in Star Wars.....

DonSC said:

gotompoes said:

As i said before i did not have a problem with the E P5 and tested the E M1 in the same way. Posted more the 40 samples. I am now accused of planting post to mislead people.

An apology seems appropriate.

No doubt the E M1 is a good camera but might suffer from sample variation.

Click to expand...

No worries. He probably wasn't even referring to you. I also think you've missed his point. He wasn't suggesting that problems weren't real, just that too many people are using the term "shutter shock" to describe every problem. For example, in one of your posts you said that DPREVIEW had found shutter shock to be a problem when testing the E-P5. That's incorrect and the reviewers have said this is incorrect. What they found to be a problem was hand SHAKE. Shake isn't shock, those are two different things. Conflating them just confuses everyone.

From the examples you posted that I looked at, and I only looked at four or five, I didn't see anything that seemed to me to be "shutter shock". The images seemed sharp in some places and not in others. Struck me as more a lens or camera problem.

Click to expand...

The reviewers (DPReview included) find blurring at 1/80 to 1/250s with the E-P5 and they somehow conclude that it's not the same problem that affected the Panasonic 45-175x, which also had a problem at those exact same shutter speeds.

Very poor reasoning IMO. The shutter speeds being affected are the key and the commonality.

If you call this "shutter shock" on the 45-175 lens, then it only stands to reason that it should be called "shutter shock" on the E-P5. I have some doubt that it's due to the "improved" 5-axis IBIS (but not much), just like the shake on the Panasonic lens was due to the improved Power OIS.

From the remainder of this thread, I'll simply refer to this issue as "IS shake syndrome", so as to avoid irritating anyone by referring to it as "shutter shock".

It's quite clear to me that certain improvements in the IS system (either IBIS or OIS) can lead to unexpected blurring at certain shutter speeds.

Of note, DPReview did not mention the same issues with the E-M1 that they had with the E-P5. But, IMO, at least some E-P5 samples suffer from the same affliction (whatever you want to call it) that the Panasonic 45-175 had, and I believe the root cause is the same.

artistguy said:

This in particular is the forum where those with almost no understanding of the principles of how exposure is achieved (but with plenty of cash), meet with those who've plenty of understanding and experience.

Click to expand...

There are also those who, like me, have little understanding and little cash.

artistguy said:

plenty for us all to learn given the pace of change with m43 cameras. A shame nearly every thread becomes a shouting match. I tend to imagine the physical manifestation of this forum as something akin to the bar in Star Wars.....

Click to expand...

Brilliant!

I think we had enough people and samples now confirming that it is indeed shutter shock or blur caused by the shutter.

Especially if there are samples using the same lens in the M5 not showing the issue there's not much left to argue about.

Or why else do you think so many threads popping up about this recently (in contrast to other new cameras)?

artistguy said:

I tend to imagine the physical manifestation of this forum as something akin to the bar in Star Wars.....

Click to expand...

and Han's pistol had shutter shock, because he wouldn't have shot first? in other words, "Han shocked first"?

Stuart said:

There have been recent posts alleging varying degrees sharpness of photos with OMD-EM1 with differing shutter speeds. These findings have been attributed to "shutter shock" or similar phenomena. However, in at least some of these cases, the "tester" was keeping the exposure constant. Hence, the aperture varied. The result was comparing sharpness at, say f8 vs. f 3.5 and attributing the differences to "shutter shock". Recall that varying the aperature

1. Changes the depth of field, giving appearance of sharpness at smaller apertures

2. Changes the behavior of the lens. Many lenses have "sweet spots"- f8 being a common one.

Hand holding is also a poor way to evaluate this as the results will vary from shot to shot and would require inspection of many images to draw a firm conclusion.

Therefore, I make a plea to be cautious about blaming one variable for changes in sharpness when multiple variables are changed.

Click to expand...

While there is every reason to warn against conclusions drawn on the basis of naive attempts to test for shutter shock (and I have seen quite a few of those here), the two points you make above are not serious impediments to proper testing. If you use a proper target. like the one shown below (this is a 100 percent crop and the print screen structure is the factor key importance), it is not difficult to tell blur due to camera shake apart from blur due to variations in lens performance or precise choice of focus (where problems with focus can of course be minimized with a bit of care). I would actually recommend keeping exposure constant (and thus vary aperture when you vary shutter speed) since differences in image brightness makes it difficult to evaluate the results.











And while you are certainly right that the results from hand-held shooting will vary on a shot-to-shot basis and therefore require fairly large samples before any clear conclusions can be drawn, the impact of shutter shock in hand-held shooting is certainly something we would want to test for. Although shutter-shock can be a problem even when shooting from a tripod (especially with very long FLs), you cannot generalize the results of tripod-based tests to hand-held shooting. If you want to know what happens when you shoot hand-held, you need to test hand-held however cumbersome that may be (and I know it is since I have done it and surely shot more than 1,000 images in the process).

If you look at my post, it was respectful and civilized. Disagreement is not lack of civility.My point is that there are too many variables to draw any conclusions. In fact,here is another.
This camera does all sorts of things to the raw image when producing a jpeg. Did you get the same problem with raw ?

Anders W said:

Stuart said:

There have been recent posts alleging varying degrees sharpness of photos with OMD-EM1 with differing shutter speeds. These findings have been attributed to "shutter shock" or similar phenomena. However, in at least some of these cases, the "tester" was keeping the exposure constant. Hence, the aperture varied. The result was comparing sharpness at, say f8 vs. f 3.5 and attributing the differences to "shutter shock". Recall that varying the aperature

1. Changes the depth of field, giving appearance of sharpness at smaller apertures

2. Changes the behavior of the lens. Many lenses have "sweet spots"- f8 being a common one.

Hand holding is also a poor way to evaluate this as the results will vary from shot to shot and would require inspection of many images to draw a firm conclusion.

Therefore, I make a plea to be cautious about blaming one variable for changes in sharpness when multiple variables are changed.

Click to expand...

While there is every reason to warn against conclusions drawn on the basis of naive attempts to test for shutter shock (and I have seen quite a few of those here), the two points you make above are not serious impediments to proper testing. If you use a proper target. like the one shown below (this is a 100 percent crop and the print screen structure is the factor key importance), it is not difficult to tell blur due to camera shake apart from blur due to variations in lens performance or precise choice of focus (where problems with focus can of course be minimized with a bit of care). I would actually recommend keeping exposure constant (and thus vary aperture when you vary shutter speed) since differences in image brightness makes it difficult to evaluate the results.

And while you are certainly right that the results from hand-held shooting will vary on a shot-to-shot basis and therefore require fairly large samples before any clear conclusions can be drawn, the impact of shutter shock in hand-held shooting is certainly something we would want to test for. Although shutter-shock can be a problem even when shooting from a tripod (especially with very long FLs), you cannot generalize the results of tripod-based tests to hand-held shooting. If you want to know what happens when you shoot hand-held, you need to test hand-held however cumbersome that may be (and I know it is since I have done it and surely shot more than 1,000 images in the process).

Click to expand...

I agree. However, neither tripod test nor hand held test isolates the shutter's influence from other factors, either in conjunction with shutter motion or independently but contributing in an accumulated fashion.

A proper shutter shock test would be to use a kind of gel structure (the stuff they use to test for ballistic impact on humans for instance) and sit the camera on that - this would simulate a completely stationary hand, i.e. what would happen if the hand is 100% steady. Triggering with wireless remote completes the picture.

Failing that, one might consider suspending the camera on a net of strings in mid air.

tt321 said:

Anders W said:

Stuart said:

There have been recent posts alleging varying degrees sharpness of photos with OMD-EM1 with differing shutter speeds. These findings have been attributed to "shutter shock" or similar phenomena. However, in at least some of these cases, the "tester" was keeping the exposure constant. Hence, the aperture varied. The result was comparing sharpness at, say f8 vs. f 3.5 and attributing the differences to "shutter shock". Recall that varying the aperature

1. Changes the depth of field, giving appearance of sharpness at smaller apertures

2. Changes the behavior of the lens. Many lenses have "sweet spots"- f8 being a common one.

Hand holding is also a poor way to evaluate this as the results will vary from shot to shot and would require inspection of many images to draw a firm conclusion.

Therefore, I make a plea to be cautious about blaming one variable for changes in sharpness when multiple variables are changed.

Click to expand...

While there is every reason to warn against conclusions drawn on the basis of naive attempts to test for shutter shock (and I have seen quite a few of those here), the two points you make above are not serious impediments to proper testing. If you use a proper target. like the one shown below (this is a 100 percent crop and the print screen structure is the factor key importance), it is not difficult to tell blur due to camera shake apart from blur due to variations in lens performance or precise choice of focus (where problems with focus can of course be minimized with a bit of care). I would actually recommend keeping exposure constant (and thus vary aperture when you vary shutter speed) since differences in image brightness makes it difficult to evaluate the results.

And while you are certainly right that the results from hand-held shooting will vary on a shot-to-shot basis and therefore require fairly large samples before any clear conclusions can be drawn, the impact of shutter shock in hand-held shooting is certainly something we would want to test for. Although shutter-shock can be a problem even when shooting from a tripod (especially with very long FLs), you cannot generalize the results of tripod-based tests to hand-held shooting. If you want to know what happens when you shoot hand-held, you need to test hand-held however cumbersome that may be (and I know it is since I have done it and surely shot more than 1,000 images in the process).

Click to expand...

I agree. However, neither tripod test nor hand held test isolates the shutter's influence from other factors, either in conjunction with shutter motion or independently but contributing in an accumulated fashion.

A proper shutter shock test would be to use a kind of gel structure (the stuff they use to test for ballistic impact on humans for instance) and sit the camera on that - this would simulate a completely stationary hand, i.e. what would happen if the hand is 100% steady. Triggering with wireless remote completes the picture.

Click to expand...

While it would be nice to have a good substitute for actual hand-held shooting, I doubt that even something like what you suggest would be good enough. So I am afraid we just have to do what I have done: Shoot and shoot and shoot until you have a sufficiently large sample to come to grips with random error (shot-to-shot variations due to other factors). Of course, the conclusions I can draw on that basis are not necessarily valid for anyone else but me, since there may be systematic variations between my hand-holding and that of others. But again, I am afraid there is little to do about that. Anyone who really wants to know how it impacts his/her own shooting simply has to test it for him-/herself.

tt321 said:

Failing that, one might consider suspending the camera on a net of strings in mid air.

Click to expand...

Anders W said:

Stuart said:

There have been recent posts alleging varying degrees sharpness of photos with OMD-EM1 with differing shutter speeds. These findings have been attributed to "shutter shock" or similar phenomena. However, in at least some of these cases, the "tester" was keeping the exposure constant. Hence, the aperture varied. The result was comparing sharpness at, say f8 vs. f 3.5 and attributing the differences to "shutter shock". Recall that varying the aperature

1. Changes the depth of field, giving appearance of sharpness at smaller apertures

2. Changes the behavior of the lens. Many lenses have "sweet spots"- f8 being a common one.

Hand holding is also a poor way to evaluate this as the results will vary from shot to shot and would require inspection of many images to draw a firm conclusion.

Therefore, I make a plea to be cautious about blaming one variable for changes in sharpness when multiple variables are changed.

Click to expand...

While there is every reason to warn against conclusions drawn on the basis of naive attempts to test for shutter shock (and I have seen quite a few of those here), the two points you make above are not serious impediments to proper testing. If you use a proper target. like the one shown below (this is a 100 percent crop and the print screen structure is the factor key importance), it is not difficult to tell blur due to camera shake apart from blur due to variations in lens performance or precise choice of focus (where problems with focus can of course be minimized with a bit of care). I would actually recommend keeping exposure constant (and thus vary aperture when you vary shutter speed) since differences in image brightness makes it difficult to evaluate the results.

Click to expand...

I forgot to mention here that you can of course vary the light level rather than the f-stop if you want. But I wouldn't say that's of much importance. Both methods will work if you know what you are doing.

Anders W said:

And while you are certainly right that the results from hand-held shooting will vary on a shot-to-shot basis and therefore require fairly large samples before any clear conclusions can be drawn, the impact of shutter shock in hand-held shooting is certainly something we would want to test for. Although shutter-shock can be a problem even when shooting from a tripod (especially with very long FLs), you cannot generalize the results of tripod-based tests to hand-held shooting. If you want to know what happens when you shoot hand-held, you need to test hand-held however cumbersome that may be (and I know it is since I have done it and surely shot more than 1,000 images in the process).

Click to expand...

tt321 said:

Anders W said:

Stuart said:

There have been recent posts alleging varying degrees sharpness of photos with OMD-EM1 with differing shutter speeds. These findings have been attributed to "shutter shock" or similar phenomena. However, in at least some of these cases, the "tester" was keeping the exposure constant. Hence, the aperture varied. The result was comparing sharpness at, say f8 vs. f 3.5 and attributing the differences to "shutter shock". Recall that varying the aperature

1. Changes the depth of field, giving appearance of sharpness at smaller apertures

2. Changes the behavior of the lens. Many lenses have "sweet spots"- f8 being a common one.

Hand holding is also a poor way to evaluate this as the results will vary from shot to shot and would require inspection of many images to draw a firm conclusion.

Therefore, I make a plea to be cautious about blaming one variable for changes in sharpness when multiple variables are changed.

Click to expand...

While there is every reason to warn against conclusions drawn on the basis of naive attempts to test for shutter shock (and I have seen quite a few of those here), the two points you make above are not serious impediments to proper testing. If you use a proper target. like the one shown below (this is a 100 percent crop and the print screen structure is the factor key importance), it is not difficult to tell blur due to camera shake apart from blur due to variations in lens performance or precise choice of focus (where problems with focus can of course be minimized with a bit of care). I would actually recommend keeping exposure constant (and thus vary aperture when you vary shutter speed) since differences in image brightness makes it difficult to evaluate the results.

And while you are certainly right that the results from hand-held shooting will vary on a shot-to-shot basis and therefore require fairly large samples before any clear conclusions can be drawn, the impact of shutter shock in hand-held shooting is certainly something we would want to test for. Although shutter-shock can be a problem even when shooting from a tripod (especially with very long FLs), you cannot generalize the results of tripod-based tests to hand-held shooting. If you want to know what happens when you shoot hand-held, you need to test hand-held however cumbersome that may be (and I know it is since I have done it and surely shot more than 1,000 images in the process).

Click to expand...

I agree. However, neither tripod test nor hand held test isolates the shutter's influence from other factors, either in conjunction with shutter motion or independently but contributing in an accumulated fashion.

A proper shutter shock test would be to use a kind of gel structure (the stuff they use to test for ballistic impact on humans for instance) and sit the camera on that - this would simulate a completely stationary hand, i.e. what would happen if the hand is 100% steady. Triggering with wireless remote completes the picture.

Failing that, one might consider suspending the camera on a net of strings in mid air.

Click to expand...

Good thinking, now go execute it....

I´m 100% sure this will show even more shuttershock then hand held.

But the thing is that it is not important if there is user interaction. If you have much more unsharp shots in a certain shutter range then expected, it is important to know. And what else then shutter shock can it be, mirror slap? Either avoid the range or find better holding techniques. In fact if keepers are important it is good to know your statistics, for every lens and camera combo you own.

When I went from a 45-200 to a 40-150 I had better sharpness results handheld, on a tripod results were equal. Why? No idea, maybe weight? But it is important to know.