Shutter Shock: My ep5 is bad. Would an em5 do better?

Started 8 months ago | Discussions
Anders W
Forum ProPosts: 17,519Gear list
Like?
Re: Shutter Shock: My ep5 is bad. Would an em5 do better?
In reply to Mr Sincere, 8 months ago

Mr Sincere wrote:

gnewylliw wrote:

Mr Sincere wrote

Here's a handheld shot at 1/40, which appears to be under the shutter shock threshold. It's relatively sharp, compared to the 1/250 shot at least.

I would put my money on up-down camera shake at 1/40, being at a 45mm FL.

Well I wouldn't rule that out at 1/40 either, but you'll notice I wasn't complaining about shutter shock in this example. This was supposed to be an example of a photo that wasgood in comparison to the ones showing obvious shutter shock at higher shutter speeds.

If you have a return policy, why not just pick up another e-p5, or if you're inclined to get a e-m5, buy it and test it? If you were buying used, this would be another issue.

I can only afford to buy used. I picked this one up for $700, in like new condition. The seller offered to allow a return, if I paid a small restocking fee, which I'm still debating. I could pick up a used em5 for $650, so it would end up being about a wash.

I'd love to be able to go to my local camera shop to try out another sample, or an em5, but I can't afford to pay full price + tax. And I feel like an a-hole using the local camera shop to demo products, only to hop online later to make the purchase.

Two solutions to the last-mentioned problem:

1. Find some accessory to buy when you visit them so that they still earn something from helping you out with your testing.

2. Tell them the full story. In my experience, many (though not all) would remain helpful, either because they are simply good people or because they realize that if they treat you nicely, you are more likely to do business with them in the future although you won't do it right now.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,519Gear list
Like?
Re: Keep in mind
In reply to DonSC, 8 months ago

DonSC wrote:

that at 1/250 the first and second curtains are moving together so the entire image would not be exposed BEFORE the first curtain closed. If the blurring was caused by the shutter then only the bottom portion of the sensor would show the blurring. Your image has blurring throughout the entire image.

That's incorrect in at least four ways.

First, the E-P5 has a max flash-sync speed of 1/320. This in turn means that the slowest shutter speed at which the first and second curtain move simultaneously (for a fraction, not the whole, of the exposure time) is 1/400.

Second, the first curtain does not close the shutter after exposure. It opens it for exposure. The second curtain then closes it.

Third, the exposure is never finished when the first curtain comes to a halt, regardless of which shutter speed we are talking about. It is finished when the second curtain comes to a halt.

Fourth, your reasoning presumes that the blur is due solely to one specific step in the sequence of shutter action: that when the first curtain comes to a halt. I don't know that we know that. On the contrary, we have a fair amount of evidence suggesting that even shutter movements before the exposure begins, such as that of the first curtain closing the shutter to prepare the sensor for exposure, has an impact. If it didn't, the anti-shock delay would be of no help whatsoever since all it does is to introduce a delay between that first phase of shutter action (the shutter closing to prepare the sensor for exposure) and subsequent phases (those involving the actual exposure).

Most likely it's hand holding technique. Rather than post about it why not put the camera on a tripod and see what you get?

This advice is based on the same "logic" as that of the man who looked for his keys under a street light because it was too dark where he dropped them.

If you want to know how the camera behaves with regard to shutter shock when shooting hand-held you need to test hand-held. If you suspect that hand-holding technique makes a difference, the proper course of action is to test different hand-holding techniques rather than put the camera on a tripod. Putting the camera on a tripod rather than holding it in your hands changes the situation in a large number of potentially important ways and can in no way isolate the effect of the cause you suggest: the particular way the camera is held in your hands.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,519Gear list
Like?
Re: Shutter Shock: My ep5 is bad. Would an em5 do better?
In reply to DonSC, 8 months ago

DonSC wrote:

Anders W wrote:

Just in case you didn't already know, Paulmorgan (alias Alumna Gorp alias Silvernitrate) is the number-one shutter-shock denier on the forum. Judging by the evidence we have this far, no rational argument will make him change his mind.

As the number-one shutter-shock claimer on the forum I'm not sure if you're not throwing rocks while living in a glass house on this one. Surely you can do better than a personal attack.

In case you didn't notice, this post of mine was in reply to the OP, not Paulmorgan. In view of the circumstances, I thought it appropriate to inform him who he was dealing with and what he could consequently expect. What I offered was a factually correct description based on large amounts of consistent evidence.

In this particular case, it's very hard to reconcile the shutter speed of 1/250 with an image which is entirely blurred. At those speeds the second curtain will be closely following the first curtain, so a good part of the sensor (and image) will have been exposed before the first curtain closes. As the Imaging Resource piece which you cite as support for your shutter shock claims makes clear, if the entire image is blurred at 1/250 then the blurring is inconsistent with the theory that the burring is being caused by internal vibrations induced by the shutter/first curtain. Put another way, since non-uniform blurring was the prime piece of evidence cited so support the speculation that the blurring was being caused by the shutter, uniform blurring such as found refutes the idea that the blurring is being caused by the shutter.

The logic of your argument is in error for reasons already spelled out in my earlier reply to you here:

http://www.dpreview.com/forums/post/52992779

On top of that, you additionally overlook here that the test and analysis of Imaging Resource you have in mind, i.e., this

http://www.imaging-resource.com/PRODS/EP1/EP1BLUR.HTM

focuses on the E-P1, which has its max flash sync speed at 1/180 rather than at 1/320 as the E-P5. You also overlook that the mechanism on which IR focuses with regard to the E-P1 (that of the first curtain coming to a halt) is not the only mechanism we have to reckon with when it comes to shutter shock more generally (although it might be of particular importance with the specific gear combination tested and analyzed by Imaging Resource).

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,519Gear list
Like?
Re: Shutter Shock: My ep5 is bad. Would an em5 do better?
In reply to lester11, 8 months ago

lester11 wrote:

Anders W wrote:

Just in case you didn't already know, Paulmorgan (alias Alumna Gorp alias Silvernitrate)

Didn't know, so thanks for that! Alumna Gorp already on my "ignore" list...

The idea of anagrams is of some help when it comes to Paulmorgan and Alumna Gorp. To properly combine Silvernitrate with the other two, you need some additional information.

He seems to be using his three nicks (there could of course be more that I am not yet aware of) interchangeably, switching from one to another in intervals covering a few months. Right now, it is Paulmorgan, which is the oldest of the three (started in 2006). Alumna Gorp (who is registered as living in Albania, the other two are reportedly in the UK) joined in 2012 and Silvernitrate in 2013. Alumna Gorp was last seen active two months ago and Silvernitrate four months ago.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,519Gear list
Like?
Re: More test data with various settings, noticeable improvement.
In reply to lester11, 8 months ago

lester11 wrote:

Steen Bay wrote:

Still find it a bit confusing. Can the first curtain be 'cocked' before it's closed to prepere the exposure? Or maybe it's just the second curtain that's 'cocked' beforehand?

I tried an analysis of shutter action here

http://www.dpreview.com/forums/post/52447334

but am not entirely convinced that it is a complete account.

What is reasonably clear is that the "release short" setting cocks and then holds the shutter open electromagnetically in preparation for the next shot by taking a (high noise high vibration)mechanical action immediately after the previous shot rather than at the start of the current shot. So the current shot can then take place both more quickly and with less vibration.

But what is not clear is why this might reduce shutter shock, since the lower vibrations due to the electromechanical (rather than entirely mechanical) shutter hold are probably finished by the time the next vibration burst occurs, which is the closing of the shutter prior to the release of the first curtain... These vibrations are in action during the entire time that the shutter is supposedly open for the exposure.

I agree that it is not self-evident for what reasons setting release lag-time to short might reduce the blur due to shutter shock. As you suggest, the motion due to charging the shutter (which is moved from the beginning to the end of the shutter sequence when release lag-time is set to short) might be sufficiently distant in time from the actual exposure not to matter even when it occurs immediately before rather than immediately after exposure. What one must keep in mind here, however, is that if the entire camera is displaced in the flesh of your hands, the time it takes for the flesh to resume its original shape might potentially be rather long.

It is easier to see why the 1/8" (or whatever) anti-shock reduces shutter shock. The setting helps by allowing the settling of the shutter closing before the first curtain is released, so that the "only" vibrations affecting the image are due to first curtain release.

-- hide signature --

Lester

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,519Gear list
Like?
Re: More test data with various settings, noticeable improvement.
In reply to Steen Bay, 8 months ago

Steen Bay wrote:

Anders W wrote:

Steen Bay wrote:

Anders W wrote:

Steen Bay wrote:

Anders W wrote:

skyglider wrote:

Anders W wrote:

Mr Sincere wrote:

I quickly ran through some more settings (it's nice out, and I gotta get out and shoot something other than candy bar wrappers ).

Quick summary:

IS 1 vs IS Auto: Makes no difference. Didn't think it would, but thought it couldn't hurt to try.

Short Shutter Release: For some reason, enabling this creates a huge improvement. Does anyone have any idea why this might be? I can hear a click when I enable this, which makes me think it changes something mechanical. This improvement alone could almost be enough to make me keep the camera.

Judging by the evidence presented here

http://www.dpreview.com/forums/post/52306602

the shutter is "cocked" immediately after each exposure when release lag-time is set to short, so that the camera is already prepared to release the shutter again. If it is set to normal, the shutter is instead "cocked" immediately before the exposure. Possibly, the motion associated with the "cocking" operation is sufficiently "violent" to affect the exposure if carried out immediately prior to it.

I put "cock" in quotation marks since the shutters we are talking about here are not spring-loaded (as far as I know). Nevertheless, it seems that the actuator has to be prepared in some way before it can do the job.

Hi Anders W,

Could you expand on your explanation above? I thought that mirrorless cameras have to keep the shutter open in order to display the scene on the screen or in the view finder.

You are perfectly right about that.

If the shutter is cocked immediately after each exposure, then wouldn't the shutter then be blocking the light negating live display on the screen or viewfinder?

By "cocking" I mean "preparing the shutter for action". Even if shutters are no longer spring-loaded but driven by coreless micro motors (if this Wikipedia article is correct), it seems as if some preparatory steps must be taken before the shutter can go into action. Judging by the post I linked to in my reply to the OP, the difference between short and normal "release lag-time" is that with short, this preparation is done immediately after each exposure, before the shutter button is pressed for the next shot, whereas with normal, it is done right after the shutter button is pressed and right before the exposure.

I'm just really interested in learning how the "short shutter release" actually works.

Any downsides to using short release lag time? If not, then I don't quite understand why 'short' isn't the default setting (since it seems to reduce vibrations), and why it's even possible to choose between 'short' and 'normal'.

The downsides are higher power consumption and that the camera may misbehave if subjected to "sharp impacts". Apparently (I haven't verified this with the original source), the E-M1 manual says:

Also make sure that the camera is not subject to sharp impacts while in use. Such impacts may cause the monitor to stop displaying subjects. If this happens, turn the power off and on again.

In other words, take care not to shock the shutter.

What I imagine happens is that the shutter is "charged" (this is the term used on Copal's home page when they describe what coreless micro motors are good for) and held back by some mechanism sensitive to impact. If there is such an impact, the shutter may close accidentally so that live view is interrupted.

Thanks. Guess that sensitive (and power consuming) mechanism could be an electromagnet.

Something like that, yes.

Still find it a bit confusing. Can the first curtain be 'cocked' before it's closed to prepere the exposure? Or maybe it's just the second curtain that's 'cocked' beforehand?

I tried to find some additional literature on how these shutters actually work. What I have found so far is mainly patents (which are a bit time-consuming to read and understand). The one to which I link below, however, suggests that the Wikipedia article on focal-plane shutters that I have previously relied on is partly wrong.

http://www.freepatentsonline.com/6783287.html

It talks about "drive springs", which suggests that even modern shutters are in fact spring-loaded, with the coreless motor used for cocking/loading/charging the springs rather than for directly driving the shutter blades.

As to your questions: Not only can the first curtain be "cocked" before it is closed to prepare the sensor for exposure. If I understand things right, it must be "cocked" in order to move. Whether subsequent phases of shutter action are "cocked" only after the first curtain has made it first move or are somehow "precocked" is not yet clear to me.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,519Gear list
Like?
Re: More test data with various settings, noticeable improvement.
In reply to Mr Sincere, 8 months ago

Mr Sincere wrote:

Jim Salvas wrote:

How about simply trying a 2 second delayed shutter release. If things are better then, it is your technique which is a problem. In other words, camera movement from pressing the shutter button.

Thanks for the suggestion, but it's almost certainly not my technique. At 1/250 and 90mm equivalent focal length, camera shake should be a non-issue, especially with image stabilization. And as I've said before, I shot the same scene with a non-stablized d90/50mm combo, and all my shots were plenty sharp all the way down to 1/100. Same with an LX7 at 90mm.

Even if you know (as I think you do) how to hold a camera optimally from an ordinary hand-shake point of view, that's not necessarily the same as holding it optimally from a shutter-shock point of view. And with IBIS available, the camera is apt to help out with the hand-shake but not with shutter shock. So holding right from a shutter-shock point of view may be what you should concentrate on.

Have a look at the link I already gave you in my first reply to you in this thread for further information on how to accomplish that:

http://www.dpreview.com/forums/post/52603448

For what it's worth, I've found it useful on my E-PM2 to wrap my right hand completely around the camera, so that my thumb is under the camera and my index finger extends all the way around the top. A squeezing motion instead of push sets off the release.

Thanks for the tip, I'll give that a try.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,519Gear list
Like?
Re: More test data with various settings, noticeable improvement.
In reply to lester11, 8 months ago

lester11 wrote:

Jim Salvas wrote:

camera movement from pressing the shutter button.

That's what Andy Westlake at DPR thought as well. I did a test,

http://www.dpreview.com/forums/post/52294229

Using the grip, I took shots with the camera in landscape or portrait, but with my shutter finger always pushing down. The double image shutter shock changed orientation with the camera position, and was never coordinated with my finger position... Shock not due to the shutter button (smile)!

In the course of the discussion about the E-P5 review on the forum, it additionally became clear that DPR's own findings with regard to the distinction between blur due to shutter shock (movement of shutter components) and blur due to shutter-button pressing was inconsistent and/or incomplete. See here:

http://www.dpreview.com/forums/post/52299674

Regrettably, DPR never did what they should have done, i.e., sort out the matter and revise their review. I don't think that anyone who has looked into the matter in some depth still believes DPR's story that it's the shutter-button pressing rather than the subsequent action of the shutter that caused the blur they observed.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Steen Bay
Veteran MemberPosts: 6,188
Like?
Re: More test data with various settings, noticeable improvement.
In reply to Anders W, 8 months ago

Anders W wrote:

Steen Bay wrote:

Anders W wrote:

Steen Bay wrote:

Anders W wrote:

Steen Bay wrote:

Anders W wrote:

skyglider wrote:

Anders W wrote:

Mr Sincere wrote:

I quickly ran through some more settings (it's nice out, and I gotta get out and shoot something other than candy bar wrappers ).

Quick summary:

IS 1 vs IS Auto: Makes no difference. Didn't think it would, but thought it couldn't hurt to try.

Short Shutter Release: For some reason, enabling this creates a huge improvement. Does anyone have any idea why this might be? I can hear a click when I enable this, which makes me think it changes something mechanical. This improvement alone could almost be enough to make me keep the camera.

Judging by the evidence presented here

http://www.dpreview.com/forums/post/52306602

the shutter is "cocked" immediately after each exposure when release lag-time is set to short, so that the camera is already prepared to release the shutter again. If it is set to normal, the shutter is instead "cocked" immediately before the exposure. Possibly, the motion associated with the "cocking" operation is sufficiently "violent" to affect the exposure if carried out immediately prior to it.

I put "cock" in quotation marks since the shutters we are talking about here are not spring-loaded (as far as I know). Nevertheless, it seems that the actuator has to be prepared in some way before it can do the job.

Hi Anders W,

Could you expand on your explanation above? I thought that mirrorless cameras have to keep the shutter open in order to display the scene on the screen or in the view finder.

You are perfectly right about that.

If the shutter is cocked immediately after each exposure, then wouldn't the shutter then be blocking the light negating live display on the screen or viewfinder?

By "cocking" I mean "preparing the shutter for action". Even if shutters are no longer spring-loaded but driven by coreless micro motors (if this Wikipedia article is correct), it seems as if some preparatory steps must be taken before the shutter can go into action. Judging by the post I linked to in my reply to the OP, the difference between short and normal "release lag-time" is that with short, this preparation is done immediately after each exposure, before the shutter button is pressed for the next shot, whereas with normal, it is done right after the shutter button is pressed and right before the exposure.

I'm just really interested in learning how the "short shutter release" actually works.

Any downsides to using short release lag time? If not, then I don't quite understand why 'short' isn't the default setting (since it seems to reduce vibrations), and why it's even possible to choose between 'short' and 'normal'.

The downsides are higher power consumption and that the camera may misbehave if subjected to "sharp impacts". Apparently (I haven't verified this with the original source), the E-M1 manual says:

Also make sure that the camera is not subject to sharp impacts while in use. Such impacts may cause the monitor to stop displaying subjects. If this happens, turn the power off and on again.

In other words, take care not to shock the shutter.

What I imagine happens is that the shutter is "charged" (this is the term used on Copal's home page when they describe what coreless micro motors are good for) and held back by some mechanism sensitive to impact. If there is such an impact, the shutter may close accidentally so that live view is interrupted.

Thanks. Guess that sensitive (and power consuming) mechanism could be an electromagnet.

Something like that, yes.

Still find it a bit confusing. Can the first curtain be 'cocked' before it's closed to prepere the exposure? Or maybe it's just the second curtain that's 'cocked' beforehand?

I tried to find some additional literature on how these shutters actually work. What I have found so far is mainly patents (which are a bit time-consuming to read and understand). The one to which I link below, however, suggests that the Wikipedia article on focal-plane shutters that I have previously relied on is partly wrong.

http://www.freepatentsonline.com/6783287.html

It talks about "drive springs", which suggests that even modern shutters are in fact spring-loaded, with the coreless motor used for cocking/loading/charging the springs rather than for directly driving the shutter blades.

As to your questions: Not only can the first curtain be "cocked" before it is closed to prepare the sensor for exposure. If I understand things right, it must be "cocked" in order to move. Whether subsequent phases of shutter action are "cocked" only after the first curtain has made it first move or are somehow "precocked" is not yet clear to me.

Not clear to me either. Seems that a 'simple' thing like a shutter can be rather complex.

And btw, have you noticed the thread that micksh6 started :

http://www.dpreview.com/forums/post/52984379

Seems that the E-M1 has a sensor that's app. 2.7% larger (linear) than the E-M5/E-PL5 sensor!

Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,519Gear list
Like?
Re: More test data with various settings, noticeable improvement.
In reply to Steen Bay, 8 months ago

Steen Bay wrote:

Anders W wrote:

Steen Bay wrote:

Anders W wrote:

Steen Bay wrote:

Anders W wrote:

Steen Bay wrote:

Anders W wrote:

skyglider wrote:

Anders W wrote:

Mr Sincere wrote:

I quickly ran through some more settings (it's nice out, and I gotta get out and shoot something other than candy bar wrappers ).

Quick summary:

IS 1 vs IS Auto: Makes no difference. Didn't think it would, but thought it couldn't hurt to try.

Short Shutter Release: For some reason, enabling this creates a huge improvement. Does anyone have any idea why this might be? I can hear a click when I enable this, which makes me think it changes something mechanical. This improvement alone could almost be enough to make me keep the camera.

Judging by the evidence presented here

http://www.dpreview.com/forums/post/52306602

the shutter is "cocked" immediately after each exposure when release lag-time is set to short, so that the camera is already prepared to release the shutter again. If it is set to normal, the shutter is instead "cocked" immediately before the exposure. Possibly, the motion associated with the "cocking" operation is sufficiently "violent" to affect the exposure if carried out immediately prior to it.

I put "cock" in quotation marks since the shutters we are talking about here are not spring-loaded (as far as I know). Nevertheless, it seems that the actuator has to be prepared in some way before it can do the job.

Hi Anders W,

Could you expand on your explanation above? I thought that mirrorless cameras have to keep the shutter open in order to display the scene on the screen or in the view finder.

You are perfectly right about that.

If the shutter is cocked immediately after each exposure, then wouldn't the shutter then be blocking the light negating live display on the screen or viewfinder?

By "cocking" I mean "preparing the shutter for action". Even if shutters are no longer spring-loaded but driven by coreless micro motors (if this Wikipedia article is correct), it seems as if some preparatory steps must be taken before the shutter can go into action. Judging by the post I linked to in my reply to the OP, the difference between short and normal "release lag-time" is that with short, this preparation is done immediately after each exposure, before the shutter button is pressed for the next shot, whereas with normal, it is done right after the shutter button is pressed and right before the exposure.

I'm just really interested in learning how the "short shutter release" actually works.

Any downsides to using short release lag time? If not, then I don't quite understand why 'short' isn't the default setting (since it seems to reduce vibrations), and why it's even possible to choose between 'short' and 'normal'.

The downsides are higher power consumption and that the camera may misbehave if subjected to "sharp impacts". Apparently (I haven't verified this with the original source), the E-M1 manual says:

Also make sure that the camera is not subject to sharp impacts while in use. Such impacts may cause the monitor to stop displaying subjects. If this happens, turn the power off and on again.

In other words, take care not to shock the shutter.

What I imagine happens is that the shutter is "charged" (this is the term used on Copal's home page when they describe what coreless micro motors are good for) and held back by some mechanism sensitive to impact. If there is such an impact, the shutter may close accidentally so that live view is interrupted.

Thanks. Guess that sensitive (and power consuming) mechanism could be an electromagnet.

Something like that, yes.

Still find it a bit confusing. Can the first curtain be 'cocked' before it's closed to prepere the exposure? Or maybe it's just the second curtain that's 'cocked' beforehand?

I tried to find some additional literature on how these shutters actually work. What I have found so far is mainly patents (which are a bit time-consuming to read and understand). The one to which I link below, however, suggests that the Wikipedia article on focal-plane shutters that I have previously relied on is partly wrong.

http://www.freepatentsonline.com/6783287.html

It talks about "drive springs", which suggests that even modern shutters are in fact spring-loaded, with the coreless motor used for cocking/loading/charging the springs rather than for directly driving the shutter blades.

As to your questions: Not only can the first curtain be "cocked" before it is closed to prepare the sensor for exposure. If I understand things right, it must be "cocked" in order to move. Whether subsequent phases of shutter action are "cocked" only after the first curtain has made it first move or are somehow "precocked" is not yet clear to me.

Not clear to me either. Seems that a 'simple' thing like a shutter can be rather complex.

It certainly can. It took them a while only to figure out how to make it out of metal (rather than softer material) and make it move vertically rather than horizontally. My first SLR, a Konica Autoreflex T(2) was one of the first to be equipped with the renowned Copal Square shutter (which Konica had a hand in developing). Most other SLRs still had horisontal non-metal shutters (à la Leica) at that point.

And btw, have you noticed the thread that micksh6 started :

http://www.dpreview.com/forums/post/52984379

Seems that the E-M1 has a sensor that's app. 2.7% larger (linear) than the E-M5/E-PL5 sensor!

Yes, I saw that.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
Paulmorgan
Senior MemberPosts: 1,641
Like?
Re: Having a hard time reconciling what you're claiming
In reply to DonSC, 8 months ago

DonSC wrote:

Mr Sincere wrote:

2) I've been shooting for more than 20 years, and I don't even want to mention on how many different cameras. 35mm film cameras, crappy point and shoots, early unstabilized DSLRs, newer DSLRs, you name it. I know how to hold a camera and not move after focusing.

4) I don't even own a tripod.

I've never known an experienced photographer who doesn't own at least one tripod so it's difficult to reconcile these two claims.

Also, as an experienced photographer, you know that every camera and even every lens can require different hand holding techniques, so getting blurred photos when using a new camera or lens would not be entirely unexpected.

This is one thing I have said all along, people need to re learn there handling abilities so that they are more in tune with these new bodies.

Reply   Reply with quote   Complain
tt321
Senior MemberPosts: 3,428Gear list
Like?
Re: Having a hard time reconciling what you're claiming
In reply to DonSC, 8 months ago

DonSC wrote:

Mr Sincere wrote:

2) I've been shooting for more than 20 years, and I don't even want to mention on how many different cameras. 35mm film cameras, crappy point and shoots, early unstabilized DSLRs, newer DSLRs, you name it. I know how to hold a camera and not move after focusing.

4) I don't even own a tripod.

I've never known an experienced photographer who doesn't own at least one tripod so it's difficult to reconcile these two claims.

There are styles of photography where a tripod has no place.

Reply   Reply with quote   Complain
Paulmorgan
Senior MemberPosts: 1,641
Like?
Re: Shutter Shock: My ep5 is bad. Would an em5 do better?
In reply to DonSC, 8 months ago

DonSC wrote:

Anders W wrote:

Just in case you didn't already know, Paulmorgan (alias Alumna Gorp alias Silvernitrate) is the number-one shutter-shock denier on the forum. Judging by the evidence we have this far, no rational argument will make him change his mind.

As the number-one shutter-shock claimer on the forum I'm not sure if you're not throwing rocks while living in a glass house on this one. Surely you can do better than a personal attack.

In this particular case, it's very hard to reconcile the shutter speed of 1/250 with an image which is entirely blurred. At those speeds the second curtain will be closely following the first curtain, so a good part of the sensor (and image) will have been exposed before the first curtain closes. As the Imaging Resource piece which you cite as support for your shutter shock claims makes clear, if the entire image is blurred at 1/250 then the blurring is inconsistent with the theory that the burring is being caused by internal vibrations induced by the shutter/first curtain. Put another way, since non-uniform blurring was the prime piece of evidence cited so support the speculation that the blurring was being caused by the shutter, uniform blurring such as found refutes the idea that the blurring is being caused by the shutter.

As the number-one shutter-shock claimer on the forum I'm not sure if you're not throwing rocks while living in a glass house on this one. Surely you can do better than a personal attack.

And it appears Anders abusive behavior gets ignored by the moderators, some things never change

Reply   Reply with quote   Complain
skyglider
Senior MemberPosts: 2,967Gear list
Like?
Re: More test data with various settings, noticeable improvement.
In reply to Anders W, 8 months ago

Anders W wrote:

I tried to find some additional literature on how these shutters actually work. What I have found so far is mainly patents (which are a bit time-consuming to read and understand). The one to which I link below, however, suggests that the Wikipedia article on focal-plane shutters that I have previously relied on is partly wrong.

http://www.freepatentsonline.com/6783287.html

It talks about "drive springs", which suggests that even modern shutters are in fact spring-loaded, with the coreless motor used for cocking/loading/charging the springs rather than for directly driving the shutter blades.

As to your questions: Not only can the first curtain be "cocked" before it is closed to prepare the sensor for exposure. If I understand things right, it must be "cocked" in order to move. Whether subsequent phases of shutter action are "cocked" only after the first curtain has made it first move or are somehow "precocked" is not yet clear to me.

Hi Anders,

Here's my current hypothesis of how the first curtain shutter works in the E-M1:

The first curtain shutter is made of two parts.  One is a metal piece with substantial mass which is required for an electromagnet to pull it up.  This metal piece is connected to the light first curtain shutter blade with springs.

NORMAL OPERATION:
When the shutter button is pressed, the electromagnet activates which pulls the metal piece up which in turn pulls the first curtain shutter up to the closed position.  Because of the mass of the metal piece, the impact of it hitting the electromagnet causes substantial vibrations in the camera, resulting in substantial shutter shock.

SHORT RELEASE LAG-TIME OPERATION:
When short release lag time is enabled it activates the electromagnet which pulls the metal piece up.  This accounts for the "clunk" sound heard when short release lag-time is enabled.  But the light shutter blade is still held in the down (open) position.  The shutter blade is then "cocked" ready for release. --- When the shutter button is pressed, the first curtain shutter blade is released which allows the springs to pull the shutter blade up closing the shutter. --- Since the shutter blade is light, it can close very quickly and also results in substantially less vibration when it impacts at the closed position.  Result is minimal shutter shock.

This short release lag-time hypothesis would explain why more battery is used and why Olympus warns against bumping the camera which can cause inadvertent release of the shutter blade, closing the shutter and causing the display to be lost on the screen or in the view finder.

What do you think?
Sky

Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,519Gear list
Like?
Re: More test data with various settings, noticeable improvement.
In reply to skyglider, 8 months ago

skyglider wrote:

Anders W wrote:

I tried to find some additional literature on how these shutters actually work. What I have found so far is mainly patents (which are a bit time-consuming to read and understand). The one to which I link below, however, suggests that the Wikipedia article on focal-plane shutters that I have previously relied on is partly wrong.

http://www.freepatentsonline.com/6783287.html

It talks about "drive springs", which suggests that even modern shutters are in fact spring-loaded, with the coreless motor used for cocking/loading/charging the springs rather than for directly driving the shutter blades.

As to your questions: Not only can the first curtain be "cocked" before it is closed to prepare the sensor for exposure. If I understand things right, it must be "cocked" in order to move. Whether subsequent phases of shutter action are "cocked" only after the first curtain has made it first move or are somehow "precocked" is not yet clear to me.

Hi Anders,

Here's my current hypothesis of how the first curtain shutter works in the E-M1:

The first curtain shutter is made of two parts. One is a metal piece with substantial mass which is required for an electromagnet to pull it up. This metal piece is connected to the light first curtain shutter blade with springs.

NORMAL OPERATION:
When the shutter button is pressed, the electromagnet activates which pulls the metal piece up which in turn pulls the first curtain shutter up to the closed position. Because of the mass of the metal piece, the impact of it hitting the electromagnet causes substantial vibrations in the camera, resulting in substantial shutter shock.

SHORT RELEASE LAG-TIME OPERATION:
When short release lag time is enabled it activates the electromagnet which pulls the metal piece up. This accounts for the "clunk" sound heard when short release lag-time is enabled. But the light shutter blade is still held in the down (open) position. The shutter blade is then "cocked" ready for release. --- When the shutter button is pressed, the first curtain shutter blade is released which allows the springs to pull the shutter blade up closing the shutter. --- Since the shutter blade is light, it can close very quickly and also results in substantially less vibration when it impacts at the closed position. Result is minimal shutter shock.

This short release lag-time hypothesis would explain why more battery is used and why Olympus warns against bumping the camera which can cause inadvertent release of the shutter blade, closing the shutter and causing the display to be lost on the screen or in the view finder.

What do you think?
Sky

Hi Sky,

I don't pretend to be an expert on the inner life of shutters. But if you ask me what I think at this point (without having had a whole lot of time to read up on the subject) that's not how I would guess it works. I don't think either of the curtains are operated by electromagnetic force in the manner you suggest or that they include any (relatively) heavy part.

Rather I think both curtains are very light (as light as they can be made without breaking) and that they might be driven by springs charged by one or more coreless micro motors. The shutter curtains of the E-P5 move with an average speed of about 5 m/s, which in turn implies forces large enough to cause problems even if the curtains are very light. Charging the mechanism may also put sufficiently large forces at play, and at least part of this charging apparently takes place after each exposure when the release lag-time is set to short whereas it takes place immediately before the shutter goes into action when the release lag-time is set to normal, thereby increasing the risk that it will affect the exposure.

That's just my provisional account at the moment and it may well be more or less wrong or incomplete.

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
DonSC
Contributing MemberPosts: 899
Like?
Too much religion not any science
In reply to Anders W, 8 months ago

Anders W wrote:

First, the E-P5 has a max flash-sync speed of 1/320. This in turn means that the slowest shutter speed at which the first and second curtain move simultaneously (for a fraction, not the whole, of the exposure time) is 1/400.

Second, the first curtain does not close the shutter after exposure. It opens it for exposure. The second curtain then closes it.

Third, the exposure is never finished when the first curtain comes to a halt, regardless of which shutter speed we are talking about. It is finished when the second curtain comes to a halt.

Fourth, your reasoning presumes that the blur is due solely to one specific step in the sequence of shutter action: that when the first curtain comes to a halt. I don't know that we know that. On the contrary, we have a fair amount of evidence suggesting that even shutter movements before the exposure begins, such as that of the first curtain closing the shutter to prepare the sensor for exposure, has an impact. If it didn't, the anti-shock delay would be of no help whatsoever since all it does is to introduce a delay between that first phase of shutter action (the shutter closing to prepare the sensor for exposure) and subsequent phases (those involving the actual exposure).

This doesn't follow at all from the Imaging Resource findings and in fact flatly contradicts them. Apart from the fact that you've mis-stated -- either intentionally or not -- the X-Sync speed of the E-P5 (it's 1/250 NOT 1/320, the latter higher number being due to the flash pulse not the curtain speed), a reference to the X-Sync speed is meaningless. Imaging Resource found the blur from speeds of 1/100 to 1/200. At all those speeds the blur was more pronounced at the top of the image than at the bottom. In fact the article says that IF the blur was uniform across the image this would be DEFINITIVE PROOF that the blurring was not related to the shutter. This makes sense because at all shutter speeds the bottom of the sensor will be exposed longer to any vibrations caused when the first curtain opens. The article provides:

Accordingly, if we saw equal blur at the top and bottom of the image within the critical shutter speed range, that meant we could safely discount it as user-caused blur, rather than anomalous blur caused by the shutter mechanism.

Given that the X-Sync speed of the EP-1 is 1/180, a speed which is more or less smack in the middle of the shutter speeds at which the blur was observed but well ABOVE most of them, the fact that the E-P5 has a slightly higher X-Sync speed would not affect the basic conclusions, which would be that blurring due to vibrations from the first curtain should be non-uniform and the non-uniform blurring should be evident at all speeds at which blurring occurs. Since the OP never saw anything but uniform blur, according to your own authority the odds of the blur being caused by the shutter mechanism is asymptotically approaching zero.

Most likely it's hand holding technique. Rather than post about it why not put the camera on a tripod and see what you get?

This advice is based on the same "logic" as that of the man who looked for his keys under a street light because it was too dark where he dropped them.

If you want to know how the camera behaves with regard to shutter shock when shooting hand-held you need to test hand-held. If you suspect that hand-holding technique makes a difference, the proper course of action is to test different hand-holding techniques rather than put the camera on a tripod. Putting the camera on a tripod rather than holding it in your hands changes the situation in a large number of potentially important ways and can in no way isolate the effect of the cause you suggest: the particular way the camera is held in your hands.

Whatever. Putting it on a tripod is the simplest way to eliminate hand shake. In fact that's why tripods were invented. But yes, we wouldn't want to eliminate hand shake as the source of the problem because that would might create cognitive dissonance in the minds of the shutter shock faithful. Better to whine about the pressing need for an electronic curtain (be careful what you ask for).

Reply   Reply with quote   Complain
lester11
Regular MemberPosts: 386Gear list
Like?
Re: Too much religion not any science
In reply to DonSC, 8 months ago

"Too much religion not any science"? C'mon! Uncalled-for!

IF the blur was uniform across the image this would be DEFINITIVE PROOF that the blurring was not related to the shutter...

The article provides:

Accordingly, if we saw equal blur at the top and bottom of the image within the critical shutter speed range, that meant we could safely discount it as user-caused blur, rather than anomalous blur caused by the shutter mechanism.

blurring due to vibrations from the first curtain should be non-uniform and the non-uniform blurring should be evident at all speeds at which blurring occurs

Not enough science here. Unequal blur suggests that, whatever is causing the blur, it dissipates pretty quickly in the time interval between the opening of the first curtain and the closing of the second curtain. So yeah, in this case it is unlikely that the blur is caused by the user.

But while we might theorise that first curtain shock dissipates quickly, the evidence from the folks who have wired up microphones and accelerometers is not so clear.  It is not certain that the first curtain vibrations "should" be "non-uniform". In particular, the noise / vibration traces seen here

http://www.dpreview.com/forums/post/52447334

clearly show noise / vibration *throughout* the period of time the shutter is open.  Guy Parsons' trace shows some dissipation towards the end of the 1/50 opening, naturally;  but Polytrophia's trace shows no dissipation during his 1/160 opening.

-- hide signature --

Lester

 lester11's gear list:lester11's gear list
Panasonic Lumix DMC-ZS30 Olympus PEN E-PL3 Olympus OM-D E-M5 Olympus E-M1 Panasonic Leica D Summilux Asph 25mm F1.4 +13 more
Reply   Reply with quote   Complain
Anders W
Forum ProPosts: 17,519Gear list
Like?
Re: Too much religion not any science
In reply to DonSC, 8 months ago

DonSC wrote:

Anders W wrote:

First, the E-P5 has a max flash-sync speed of 1/320. This in turn means that the slowest shutter speed at which the first and second curtain move simultaneously (for a fraction, not the whole, of the exposure time) is 1/400.

Second, the first curtain does not close the shutter after exposure. It opens it for exposure. The second curtain then closes it.

Third, the exposure is never finished when the first curtain comes to a halt, regardless of which shutter speed we are talking about. It is finished when the second curtain comes to a halt.

Fourth, your reasoning presumes that the blur is due solely to one specific step in the sequence of shutter action: that when the first curtain comes to a halt. I don't know that we know that. On the contrary, we have a fair amount of evidence suggesting that even shutter movements before the exposure begins, such as that of the first curtain closing the shutter to prepare the sensor for exposure, has an impact. If it didn't, the anti-shock delay would be of no help whatsoever since all it does is to introduce a delay between that first phase of shutter action (the shutter closing to prepare the sensor for exposure) and subsequent phases (those involving the actual exposure).

This doesn't follow at all from the Imaging Resource findings and in fact flatly contradicts them. Apart from the fact that you've mis-stated -- either intentionally or not -- the X-Sync speed of the E-P5 (it's 1/250 NOT 1/320, the latter higher number being due to the flash pulse not the curtain speed),

You didn't get it. The X-sync speed of the E-P5 is 1/320 when using the internal flash (and 1/250 with other flashes). The higher sync speed compared to earlier bodies was touted as an important new feature when the E-P5 was introduced:

http://www.dpreview.com/news/2013/05/10/Olympus-launches-PEN-E-P5-high-end-Wi-Fi-enabled-Micro-Four-Thirds-model

With a flash capable of multiple-pulse mode (Super FP mode), you can go higher than that with any body supporting that mode. This, however, comes at the expense of weakened flash power (and hence reduced working distance).

a reference to the X-Sync speed is meaningless. Imaging Resource found the blur from speeds of 1/100 to 1/200. At all those speeds the blur was more pronounced at the top of the image than at the bottom. In fact the article says that IF the blur was uniform across the image this would be DEFINITIVE PROOF that the blurring was not related to the shutter. This makes sense because at all shutter speeds the bottom of the sensor will be exposed longer to any vibrations caused when the first curtain opens. The article provides:

Accordingly, if we saw equal blur at the top and bottom of the image within the critical shutter speed range, that meant we could safely discount it as user-caused blur, rather than anomalous blur caused by the shutter mechanism.

Given that the X-Sync speed of the EP-1 is 1/180, a speed which is more or less smack in the middle of the shutter speeds at which the blur was observed but well ABOVE most of them, the fact that the E-P5 has a slightly higher X-Sync speed would not affect the basic conclusions, which would be that blurring due to vibrations from the first curtain should be non-uniform and the non-uniform blurring should be evident at all speeds at which blurring occurs. Since the OP never saw anything but uniform blur, according to your own authority the odds of the blur being caused by the shutter mechanism is asymptotically approaching zero.

You didn't get it. Please reread the fourth of the four points above where I list the errors in your thinking.

On top of that, one might add that the section you quote from IR's analysis represents a logical slip-up on their part, which you apparently failed to recognize. They infer, from the top-to-bottom pattern they observe in their tests shots, that the blur is caused by the first curtain coming to a halt in open position. For a number of reasons (including those already mentioned by lester11 but also others) this doesn't imply that if the blur had been more evenly distributed, it couldn't be caused by the shutter. It would merely imply that it would be less likely to be due solely to the particular phase of shutter action that IR singles out: That of the first curtain coming to a halt in open position.

Most likely it's hand holding technique. Rather than post about it why not put the camera on a tripod and see what you get?

This advice is based on the same "logic" as that of the man who looked for his keys under a street light because it was too dark where he dropped them.

If you want to know how the camera behaves with regard to shutter shock when shooting hand-held you need to test hand-held. If you suspect that hand-holding technique makes a difference, the proper course of action is to test different hand-holding techniques rather than put the camera on a tripod. Putting the camera on a tripod rather than holding it in your hands changes the situation in a large number of potentially important ways and can in no way isolate the effect of the cause you suggest: the particular way the camera is held in your hands.

Whatever. Putting it on a tripod is the simplest way to eliminate hand shake. In fact that's why tripods were invented.

You didn't get it. A tripod eliminates hand shake but it also eliminates a lot of other factors associated with hand-holding the camera. Consequently, it is useless if what you want to know is how shutter-shock affects you in hand-held shooting. If you want to know the latter, the appropriate course of action is that chosen by the OP: Take a sufficiently large sample to distinguish the systematic effect of shutter shock from the random effect of ordinary hand-shake.

But yes, we wouldn't want to eliminate hand shake as the source of the problem because that would might create cognitive dissonance in the minds of the shutter shock faithful.

The psychological problems are all yours

Better to whine about the pressing need for an electronic curtain (be careful what you ask for).

Why would I want to be careful about that?

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +21 more
Reply   Reply with quote   Complain
texinwien
Senior MemberPosts: 2,650Gear list
Like?
Re: Too much religion not any science
In reply to DonSC, 8 months ago

DonSC wrote:

Whatever. Putting it on a tripod is the simplest way to eliminate hand shake. In fact that's why tripods were invented. But yes, we wouldn't want to eliminate hand shake as the source of the problem because that would might create cognitive dissonance in the minds of the shutter shock faithful.

That's rich - the cognitive dissonance is all on your side. As I demonstrated to almost you a month ago, the shutter shock blur doesn't always disappear when the camera is mounted on a tripod.

Here it is again, plain evidence of increased blurriness in the shutter shock danger zone with hand shake eliminated:

Huh. Look at that. The ISO100 (1/40), ISO200 (1/80) and ISO400 (1/160) images are significantly more blurred than the ISO800 (1/320) image.

Source

Assuming DPReview puts each camera on a good tripod when shooting their studio comparison scene (I think that's a safe assumption), how do you explain the above evidence?

Cognitive dissonance much?

 texinwien's gear list:texinwien's gear list
Canon EOS 300D Olympus OM-D E-M5 Canon EF-S 18-55mm f/3.5-5.6 Panasonic Lumix G 20mm F1.7 ASPH Panasonic Lumix G Vario 100-300mm F4-5.6 OIS +14 more
Reply   Reply with quote   Complain
traveler_101
Contributing MemberPosts: 590Gear list
Like?
Re: Shutter Shock: My ep5 is bad. Would an em5 do better?
In reply to Mr Sincere, 8 months ago

Mr Sincere wrote:

gnewylliw wrote:

Mr Sincere wrote

Here's a handheld shot at 1/40, which appears to be under the shutter shock threshold. It's relatively sharp, compared to the 1/250 shot at least.

I would put my money on up-down camera shake at 1/40, being at a 45mm FL.

Well I wouldn't rule that out at 1/40 either, but you'll notice I wasn't complaining about shutter shock in this example. This was supposed to be an example of a photo that wasgood in comparison to the ones showing obvious shutter shock at higher shutter speeds.

If you have a return policy, why not just pick up another e-p5, or if you're inclined to get a e-m5, buy it and test it? If you were buying used, this would be another issue.

I can only afford to buy used. I picked this one up for $700, in like new condition. The seller offered to allow a return, if I paid a small restocking fee, which I'm still debating. I could pick up a used em5 for $650, so it would end up being about a wash.

I'd love to be able to go to my local camera shop to try out another sample, or an em5, but I can't afford to pay full price + tax. And I feel like an a-hole using the local camera shop to demo products, only to hop online later to make the purchase.

Dear Mr. Sincere

Thank you for posting; it has been a tremedously informative and at times entertaining thread.

My advice is to dump this camera as soon as you can and instead of buying a used E-M5 for $650, pick up a brand new E-M10 for $699. Then you ahve a full warranty, plus dpreview has ths to say about it: "The good news is that our preliminary testing suggests it isn't prone to the image shake that can occur with the PEN E-P5."

I don't like to buy used digital cameras--lenses o.k. but cameras no.

 traveler_101's gear list:traveler_101's gear list
Olympus PEN E-PM2 Panasonic Lumix G 20mm F1.7 ASPH Olympus M.Zuiko Digital ED 40-150mm 1:4-5.6 Panasonic Lumix G 14mm F2.5 ASPH Olympus M.Zuiko Digital 25mm F1.8
Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads