Shutter Vibration

[vjbelle]

Test camera vibration files are off to Sony. Camera in vertical position, 90mm Leica, 10s delay. Big difference between the worst and best files. At least they are looking into this.

Victor

 

[cunim]

Just to clarify, the Gaussian blur (shutter vibration) that the A7r exhibits is quite subtle. There is a gross blur as well, that happens if you use a rigid, light and resonant support system but that is not really the camera's fault. Just don't mount that way. In contrast, the Gaussian blur is part of the camera package and is unavoidable without an unusual amount of care.

Vibration is of concern to some because this is a 36 MP camera. People that care about such high pixel density tend to peep. Sadly, in the vibration range the peepers find this Gaussian blur so we squawk. Portraits and street shots won't show it. It will even soften up portraits a bit, which might be nice. Tightly focused heron irises, now those guys might care. Hmmm. Peep, squawk, heron.. too many bird references.

It doesn't matter whether you hand hold or cement the camera to a concrete block. It is not clamps or how steady your hands are. It is a product of a strong shutter in a low mass body with minimal internal damping. Tap the front shell of your A7r. It will ring.

Unlike other high density FF cameras, this one has no mechanism (such as mirror lockup or leaf shutter) to allow you to avoid vibration so as to get that last bit of performance. As Guy points out, you need to use technique to get around the vibration and, with this camera, you need to use a higher level of technique than with a D800. Instead of locking up the mirror for critical shots, stay out of the sensitive range. Or don't if the subtle blur does not matter to you.

I think this is easily fixed if Sony makes it a priority. However, it will need a new internal design. Hey - imagine these electronics in a Leica body! Lust.

 

[MGrayson]

Why would shutter vibration lead to Gaussian blur? I would expect the blur to be parallel to the shutter motion, not nicely anisotropic Gaussian.

--Matt

 

[cunim]

Why would shutter vibration lead to Gaussian blur? I would expect the blur to be parallel to the shutter motion, not nicely anisotropic Gaussian.

--Matt

You'd think, and there are images which clearly show what you suggest - see post #36 in this thread. However, when I look at point sources with a nonresonant mounting system I see a circle of blur. Not sure if it is actually Gaussian, just trying to get across the falloff around a center.

I think the entire CCD package is resonating in relation to the shell. The plane of resonance appears to decouple - at least partly - from its source. Or I could be wrong. I have no testing equipment here. It's great that Sony is now looking into to it (see above posts).

 

[vjbelle]

Just to clarify, the Gaussian blur (shutter vibration) that the A7r exhibits is quite subtle. There is a gross blur as well, that happens if you use a rigid, light and resonant support system but that is not really the camera's fault. Just don't mount that way.

Oddly enough I have Gitzo series 1 tripods and series 3 tripods. Arca Z1 head, and light weight Acra heads. The heavier tripod setup did not result in lesser blur manifestations - or if less very difficult to see with results from a 90mm lens. Longer lenses may produce different results.

I also think that if the first curtain could close at the start of a self timer that this issue could be greatly improved upon.

Victor

 

[sergio lovisolo]

Test camera vibration files are off to Sony. Camera in vertical position, 90mm Leica, 10s delay. Big difference between the worst and best files. At least they are looking into this.

Victor

Victor, could you post an image of your tripod?
With this setup

_DSC3260 by sergio lovisolo, on Flickr

I can't get motion blur at any shutter speed, no trace even with finest details
of test target.(elmarit m 90)

Casual subject at 1/25

_DSC0435 by sergio lovisolo, on Flickr

test target at 1/30 and 100% crop


_DSC0436 by sergio lovisolo, on Flickr

Thanks.

 

[Guy Mancuso]

You'd think, and there are images which clearly show what you suggest - see post #36 in this thread. However, when I look at point sources with a nonresonant mounting system I see a circle of blur. Not sure if it is actually Gaussian, just trying to get across the falloff around a center.

I think the entire CCD package is resonating in relation to the shell. The plane of resonance appears to decouple - at least partly - from its source. Or I could be wrong. I have no testing equipment here. It's great that Sony is now looking into to it (see above posts).

I might call it more of a mirrored ghost in look over a gaussian blur. Either way its there

 

[Ron Pfister]

Sergio, I would guess that the massive, over-sized rubberized/cork-covered mounting plate of your setup does the trick here. And I would add that such a support setup is definitely not what I'd carry into the field along with the A7R…

 

[sergio lovisolo]

Mr. Lovisolo, You and I need a subscription. That is the problem. :ROTFL:

:toocool:

 

[vjbelle]

Holy cow, Sergio - how much does all of that stuff weigh?:shocked: The next step would be to weld that little puppy to an I-Beam. My series 3 and Arca Z1 setup is not enough to calm down the vibration from this little camera. If I have to get beyond that then something surely is amiss.

Victor

 

[cunim]

Victor, could you post an image of your tripod?
With this setup

_DSC3260 by sergio lovisolo, on Flickr

I can't get motion blur at any shutter speed, no trace even with finest details
of test target.(elmarit m 90)

Casual subject at 1/25

_DSC0435 by sergio lovisolo, on Flickr

test target at 1/30 and 100% crop


_DSC0436 by sergio lovisolo, on Flickr

Thanks.

Sergio, my mistake lay in not posting images earlier. Here are two crops. They are a peeper's delight at about 4:1 so the issue becomes fairly obvious. One is at 1/60 and the other at 0.5. I will leave it to you to judge which is which. I have some others using flash in which the difference is clearer.

The point is this camera can deliver wonderful resolution but only under the right conditions. We are deluding ourselves if we think we can get around that.

Conditions: Foba Asaba studio stand with Sinar head, AS clamp on stand and AS plate tight on camera. A7r, iso 200, Apo Summicron M 90 asph. No sharpening applied.

As you can see, you need to look carefully. I would be very interested to see your resolution chart snapped at 1/60 vs 0.5, with a 100% crop.

The best way to lay all this to rest is to derive point-spread functions. I am sure someone here could do that if they wanted to. However, it is not necessary to get technical to decide whether this level of blur matters to you.

 

[Ron Pfister]

However, it is not necessary to get technical to decide whether this level of blur matters to you.

Thanks for posting these, cunim! It seems to me that this level of blur may very well negate the advantage of 36MP over 24MP.

Edit: the same can probably be said about lens quality (i.e. the blurred image may as well have been taken with a far less expensive lens)

 

[Daure]

Sergio, my mistake lay in not posting images earlier. Here are two crops. They are a peeper's delight at about 4:1 so the issue becomes fairly obvious. One is at 1/60 and the other at 0.5. I will leave it to you to judge which is which. I have some others using flash in which the difference is clearer.

The point is this camera can deliver wonderful resolution but only under the right conditions. We are deluding ourselves if we think we can get around that.

Conditions: Foba Asaba studio stand with Sinar head, AS clamp on stand and AS plate tight on camera. A7r, iso 200, Apo Summicron M 90 asph. No sharpening applied.

As you can see, you need to look carefully. I would be very interested to see your resolution chart snapped at 1/60 vs 0.5, with a 100% crop.

The best way to lay all this to rest is to derive point-spread functions. I am sure someone here could do that if they wanted to. However, it is not necessary to get technical to decide whether this level of blur matters to you.

How do you manage the +/- 3 stops between the two pictures ?

 

[jonoslack]

Thanks for posting these, cunim! It seems to me that this level of blur may very well negate the advantage of 36MP over 24MP.

Edit: the same can probably be said about lens quality (i.e. the blurred image may as well have been taken with a far less expensive lens)

Hi Ron
This is the way I'm thinking as well. Together with the reduced shutter lag there seems to be quite a lot going for the A7.

 

[Vivek]

Holy cow, Sergio - how much does all of that stuff weigh?:shocked: The next step would be to weld that little puppy to an I-Beam. My series 3 and Arca Z1 setup is not enough to calm down the vibration from this little camera. If I have to get beyond that then something surely is amiss.

Victor

One of the tripods I have is a Sachtler DA-75 (aluminum) with a Manfrotto fluid head. There is a heavier fluid head as well but that weighs more than this combo. There are carbon fiber versions of the DL-75 which are a lot lighter.

Still the DA-75 + head is not that heavy (excluding the spreader).

 

[cunim]

How do you manage the +/- 3 stops between the two pictures ?

Woops, forgot to specify that. Shots at f3.5 or f4, can't remember. Set a Profoto D4 and soft box to get the 1/60 to work at iso 200. Then just turned down the light for the longer exposures.

 

[Ron Pfister]

Hi Ron
This is the way I'm thinking as well. Together with the reduced shutter lag there seems to be quite a lot going for the A7.

Jono, I agree and I don't. Let's say for now, I'm opting to disagree. I *want* the A7R to deliver as much as my lenses (or its sensor) are capable of, and I'm willing to walk the extra mile to get there. This was the primary reason why I chose this model over the A7. But it may turn out that this goal is not achievable very often in reality. Further use will deliver the answer to this question. For now, I remain positive. I think I can work around the faster than 1/3f or slower than 2s rule when it matters...

 

[vjbelle]

Ron, I have confidence you'll be able to work around this - for now. I'm going to shoot in manual mode with the shutter set at 1/250 and Auto ISO from 100 to 1600. This is strictly for tripod use as I can't imagine hand holding a 90mm setup. A half press of the shutter button shows the ISO in the lower right hand corner - very handy. If I have to I can always attach a 6 stop ND and shoot at 1s or longer with perfect results for the stuff I shoot.

Victor

 

[turtle]

I agree that vibration is evidently a problem under certain circumstances (I have yet to get the adaptor that will let me find out for myself), but I don't think its a show stopper for most users.

• The A7R is still fine hand held, but with the same caveats that apply to any 36MP sensor i.e. keep the speed up.
  
• It is fine with the shorter focal lengths most people will use in day to day use.
  
• It is unlikely to be used for wildlife/birding photography, where a good DSLR remains a far better solution (AF, frame rate etc). It was never designed for this application, I am sure.

• 
  
  For travel and landscape use, it seems the vibration issue will occur only with long lenses, at certain speeds. I notice there was 400% crops being used to show the issue, but IMO, if its not visible at 100% its not ever going to be visible in print in any way that matters. I think for most people, the concern is that it will show up in the 70-200 range, which many use a lot... and therefore the forthcoming 70-200 F4 lens! Will the image stabilisation of that lens help? Personally, I'm not sure that I will ever own or bother to carry around such a lens with this camera. A superwide, the 24-70 Zeiss and the 35 and 55 will be plenty for me.
  

For those who are using long lenses, what is the application and is this a major part of your work? I ask this not to dismiss your application, but because I recognise that my needs are not necessarily yours and its not easy to imagine what other people get up to with their cameras!!!

How many of you were/are expecting to buy the 70-200 F4 G?

 

[Guy Mancuso]

Myself I have a Zeiss 135 in a Nikon mount. I own that lens and on my D800 if I need a little reach for normal work than I can go into crop mode on the d800. I think I could do the same with the Sony. I'll try that today as all my parts will be here today. As far as the 70-200 when I need that I usually rent it twice a year for big runway shows. Even if I switched to Sony I would do exactly the same. But there is a chance if I kept the Sony as my main unit than maybe I would buy it. Also if I decided Sony was my main squeeze than I would add the 7 only version as well for more speed type work. But I love the 135 focal length and I would keep my Zeiss Nikon mount regardless with the adapter, same for my Leica 19r in Nikon mount I would keep as well but also add in Nikon mount the Zeiss 25 F2. So I would have 3 lenses in Nikon mount. Than sell my Sigma 35 1.4 get the Sony 35Fe and 55fe. That would round me out very well. I would also buy the 28-70 zoom with the 7 purchase and replace that later with the 24-70. The real question for me is do I switch entirely. That is a very hard call right now, although I'm freaking nuts and could pull the trigger on a whim as obviously I already thought this all through.LOL I just have to be patient for the moment and see what I am getting but bottom line I was never completely thrilled with Nikon D800 . I need the Mpx no doubt for my high end work which the 7r would cover and the 7 for PR would be maybe better than the D7100 which I thought about getting for the crop factor for one and also faster shooting in AF mode. The 7 would have to be really good at follow focus though. It's for runway with 15k in images in 3 days. That's a load of shooting.

My underlining issue or dissatisfaction is I came from a phase one IQ 160 with tech cam that I DID NOT want to sell but was forced to due to my wife's health which we are still battling even this week with radiation. My issue is I'm still looking for the look of that back in 35mm land the D800e is nice but its work to get close plus it just don't have the look. I'm still searching for that. The nice thing it gets the job done well for clients but Guy is not thrilled with it. At this point in my career I want a thrill. LOL