Mirror Lock Up Test

The image quality was certainly substantially different to.

I often do long exposures but have never bothered to use mirror lock up - I will be doing in future!
 
For long exposures I always do mirror lockup, it just minimises another of the variables
 
I have pondered how effective it is in certain scenarios. I always use it with longer exposures anyway, but with (for instance) a 30 second exposure, how long do the vibrations actually last for, and would a split second of shutter vibration actually make a noticeable difference to such a long exposure?

I can completely understand why it would have quite adverse effects on a 1/10 second exposure, as in the test they did.
 
GrittyShaker said:
I have pondered how effective it is in certain scenarios. I always use it with longer exposures anyway, but with (for instance) a 30 second exposure
Click to expand...

Mirror vibration only happens in the first few moments but that means it's in the first few moments of any and every exposure irrespective of ultimate exposure time.
 
Yes, but a few moments could account for an entire exposure, or a miniscule fraction of an exposure, depending on how long the shutter is open for. That will make a massive difference to the effect of the vibration in the resultant image.
 
GrittyShaker said:
Yes, but a few moments could account for an entire exposure, or a miniscule fraction of an exposure, depending on how long the shutter is open for. That will make a massive difference to the effect of the vibration in the resultant image.
Click to expand...

Any exposure that is longer than the duration of the mirror slap vibration will contain all of it's manifest effects. You'll only mitigate it with a shutter speed than is shorter than it's duration.
 
With a long exposure you reach a point where any mirror vibration will not register on the image. For example, you may well have seen photos of public places with people in them but the resulting image shows none because they've been moving and haven't registered on the long exposure.

I've always believed that there's a cutoff point where mirror lockup has no benefit. It has never bothered me enough to launch an investigation in to what sort of time period this is. In any case, it will no doubt vary to some degree from camera model to camera model.
 
MisterE said:
With a long exposure you reach a point where any mirror vibration will not register on the image. For example, you may well have seen photos of public places with people in them but the resulting image shows none because they've been moving and haven't registered on the long exposure.

I've always believed that there's a cutoff point where mirror lockup has no benefit. It has never bothered me enough to launch an investigation in to what sort of time period this is. In any case, it will no doubt vary to some degree from camera model to camera model.
Click to expand...

That's exactly what I was driving at. Hence why we generally can't see Michael Bosanko in his light paintings.
 
Last edited:
ChrisGilbert said:
If there is any mirror vibration then it registers at the beginning of the exposure because that's where it happens. It doesn't matter how long the exposure actually is.
Click to expand...

And you're missing my point. Yes, mirror vibration happens at the beginning of a long exposure. But just because it happens doesn't mean that the sensor records it and degrades a photo. As I said, that's why people can be made to disappear from long exposures.
 
ChrisGilbert said:
If there is any mirror vibration then it registers at the beginning of the exposure because that's where it happens. It doesn't matter how long the exposure actually is.
Click to expand...

Yes it does. Vibration from mirror slap disappears after around 1 second. So if you use a 1 second exposure all of your image will be blurred. But if you're doing a 5 minute exposure then the blurred image will represent only 0.3% of your total exposure - not enough to be noticeable. In exactly the same way as somebody walking through your 5-minute shot will be invisible, so will the blurred bits.
 
hollis_f said:
But if you're doing a 5 minute exposure then the blurred image will represent only 0.3% of your total exposure - not enough to be noticeable.
Click to expand...

Wrong. A blur is a blur. it doesn't matter how long the exposure is.

The purpose of mirror lock-up is to make sure that the parts of your image that you wish to remain sharp are not adversely affected by the vibration caused by the mirror slap. It doesn't matter how long the shutter is open for unless it's a fraction of the duration of the mirror slap vibrations.
 
Last edited:
Isn't the point of a long exposure to get enough light to record an image. A short exposure wouldn't allow enough light to hit the sensor to record anything. This means that a short period of vibration within a long exposure wouldn't matter as not enough light would be hitting the sensor during that brief period for the blur to get recorded.
 
sxlalan said:
Isn't the point of a long exposure to get enough light to record an image. A short exposure wouldn't allow enough light to hit the sensor to record anything. This means that a short period of vibration within a long exposure wouldn't matter as not enough light would be hitting the sensor during that brief period for the blur to get recorded.
Click to expand...
That's what most of us are agreeing on. Having just done a quick test, I'm actually struggling to deliberately cause any discernible camera shake in exposures ranging from 1/10 to 1sec without using mirror lock up, and that's with my 70-300 at full reach, on a block of text about 1.5 metres away. :shrug:
 
sxlalan said:
Isn't the point of a long exposure to get enough light to record an image. A short exposure wouldn't allow enough light to hit the sensor to record anything. This means that a short period of vibration within a long exposure wouldn't matter as not enough light would be hitting the sensor during that brief period for the blur to get recorded.
Click to expand...

You make long exposures for different reasons and, yes, sometimes it's to get enough light into the camera. Sometimes it's to deliberatley capture motion of somesort. What matters is that you make the choices about how you set the camera up for specific reasons. Mirror lockup is just a tool, albeit an incredibly useful tool when it comes to getting the best possible image out of the camera. It's not something that affects most photographers because they rarely do anything other than look at the image on a computer screen or maybe print A4 but it can be the difference between an image that looks rubbish at A2 and one that look fantastic at A2. Horses for courses.
 
sxlalan said:
Isn't the point of a long exposure to get enough light to record an image. A short exposure wouldn't allow enough light to hit the sensor to record anything. This means that a short period of vibration within a long exposure wouldn't matter as not enough light would be hitting the sensor during that brief period for the blur to get recorded.
Click to expand...

Did you read the article posted by the OP ?
 
Yes I read it but it isn't really relevant to the long exposure question is it, assuming that we are talking several seconds rather than fractions of a second. For a long exposure (whatever the reason it is taken) I would have thought that the percentage of light hitting the sensor during the vibration would be such a small fraction of the overall light recorded that it would be undetectable. I'm brand new to this though and just trying to get my head around it so am happy to be proven wrong.
 
I personally don`t like the way he has done the test using a laser pointer hanging from the back with less than 1% actually in contact with camera this has exagrated the movement imensely.
 
sxlalan said:
Yes I read it but it isn't really relevant to the long exposure question is it
Click to expand...

Yes, it is, because, as I said, if you don't set mirror lock up then the mirror slap vibration is in every shot you take to a greater or lesser degree because it happens at the very beggining of shutter actuation. The actual shutter speed is irrelevant unless it's very short, as stated in the article. One of the arguments being used in this thread is that surely there's not enough light to allow the vibration to manifest itself. Sometimes that may true, sometimes that isn't true. It's up to the photographer to make a judgement call on that but the feature is there for a very good reason. Not using it may limit the uses to which the final image can be put. To most people that wont matter. My issue in this thread is that there appeared to be a misapprehension about what it's purpose is.
 
djinn said:
I personally don`t like the way he has done the test using a laser pointer hanging from the back with less than 1% actually in contact with camera this has exagrated the movement imensely.
Click to expand...

But surely that is the point they are making.... Yes it's exaggerated but it's the same for all the shots.
 
anyway ....... Just use a piece of card in front of the lens... open the shutter and remove the card? ..... No need for all this new fangled stuff :D
 
Thanks Chris. I use mirror up for macro shots which are quick and the effect discernible. For any shot where the vibration time is tiny fraction of the overall exposure time I just can't see how the number of photons hitting the sensor during the brief shake will be sufficient to show up in the final image. If you were getting enough photons in that brief period to be discernible then surley the resultant final image (which collected 1000s of times more light) would be completely burnt out. I confess I come from a scientific background and not a photographic one but am having trouble visualising what you suggest.
 
Thanks! Just starting out but am having a great time so far :-)
 
Nobody had mentioned liveview? If I'm using a tripod I tend to focus in liveview mode and use a 2 second timer. Works for me.

Not sure how that stand up if I was photographing a small line and blowing it up to 1000%. But I don't do that, so I don't care.
 
ChrisGilbert said:
Wrong. A blur is a blur. it doesn't matter how long the exposure is.
Click to expand...

Yes it does. The blur is only happening for a very short time relative to a 5 min exposure. Would you like me to prove it? It's quite simple, just stick a 10-stop ND on, set the ISO to 100, set aperture to f32 and shutter speed to something like 2 minutes in a dim room. Press the shutter button and immediately jog the tripod, then leave it for the remainder of the 2 minute exposure.

Then swap the ND for a 5-stopper, drop the shutter speed to 5 seconds and repeat.

According to you both images will look the same. According to me the second image will show a mass of blur, the first will show no blur. Want to wager on who's correct?
 
Interesting, I forgot about mirror lock-up completely. Used to use it on my old film SLR when I actually took the time to think between each shot!
 
Mirror slap is obviously always present, but it's only noticeable at certain speeds in certain conditions. Namely, between about 1/30sec and 1/2 sec with the camera on a tripod, ironically enough. It's usually worst around 1/15sec to 1/8sec, especially if you have the centre column extended. Hand-holding is very effective at damping vibrations.

At faster speeds, the exposure is too short and is over before any significant movement has time to register, and at longer speeds the camera settles quickly and the wobbly bit of the exposure is too small to notice.

It's easy enough to test, if you provoke it. Camera and long lens on a tripod with centre column fully up, then run through the critical shutter speed range. Also try it with IS on - it's pretty good at cutting down mirror slap.

BTW, the laser test on the link is useless as it is also affected by the mirror banging back down again, largely undamped, when the actual exposure has obviously finished. It makes the point, but has no real bearing on the effect in practice.
 
Frank, sxlalan, Richard Hoppy, and any others.......care to join me and my wall? :bang: :)
 
Last edited:
You must log in or register to reply here.
Back
Top