shutter speed/tripod reciprocal rule

Sirius Glass said:

Great that it works for you. But for most that is not the case. Come back is 20 years or 50 years and tell us how it works for you then.

There is someone who posts on APUG with a Rollei and claims to handhold it at 1/2 second to 1 second. That is what I would call apocraphal.

Steve

Click to expand...

What the heck does "apocraphal" mean? When I was young I had much steadier hands and could hand-hold a 135 camera at relatively slow shutter speeds and the images easily held up to 11x14 printing. With my health problems today I couldn't hand-hold a corned beef sandwich.

phenix said:

The fact is that the slap of the mirror is the main cause you cannot go for longer exposures than the rule suggests.

Click to expand...

The fact is that the above is fiction of the purest kind.

The movement of even the steadiest hands is way, way worse than anything even a heavy mirror can do with and without proper dampening.

Q.G. said:

The fact is that the above is fiction of the purest kind.

The movement of even the steadiest hands is way, way worse than anything even a heavy mirror can do with and without proper dampening.

Click to expand...

I agree, but it would be nice to see a timeline of events.
I wonder of it is as follows:

1. mirror up
2. mirror dampening (how long?)
2. aperture close
3. shutter open

If so, then mirror dampening (depending on how long that takes) may be no issue whatsoever, because by the time the shutter opens, mirror dampening is finished.

The timeline will be different for different brands and types of cameras.
But it will not be a series of successive events, but one of partly overlapping (in time) events.
For instance: the diaphragm, and leaf shutter in cameras using such a thing, will start to close simultaneously with the mirror swinging up.

The thing to know is how much time there is between the moment the mirror hits its rest, and the moment the shutter starts exposing film.
In a Hasselblad 500 C(...), for instance, it's about 30 ms. But in that camera, there are two baffle doors that take an extra 3 ms afer the mirror has hit the rest to open.
Different cameras, different sequences, different numbers.


But what makes that mirror slap "fact" fiction is not when things happen, but the simple fact that no matter how badly the dampening of the mirror is, hands shake much, much worse.
When handholding, any moment of thought given to mirrors is completely misplaced and wasted. Mirror induced shake is completely drowned, disappears completely, in the much larger hand induced shake.

Oh, and...
Before we get a next "fact" in reply: no. No matter how well trained, and caffeine free, you are, no matter how good your breathing technique, no matter how well tuned into your heart beat cycle you are: hand induced shake remains orders of magnitudes worse than mirror slap.
It's not that you don't have a steady hand. I'm sure you do. But that doesn't change anything. It's just no comparison. You can't beat it.

Q.G. said:

The fact is that the above is fiction of the purest kind.

The movement of even the steadiest hands is way, way worse than anything even a heavy mirror can do with and without proper dampening.

Click to expand...

Didn’t I mention that I’ve got blurred (doubled, in fact) images with a 35mm SLR and a long telelens even when using a heavy (for MF) tripod? Moreover, even when using mirror lock-up? Do you lock the mirror up while shooting handheld? IMHO, mirror lock-up means tripod use too... Mirror’s slap and the movement of the shutter blades, both create vibrations that last until after the exposure ends. Therefore, the fact that the mirror slaps up before the shutter opens is totally irrelevant. And these vibrations cannot be completely absorbed by a tripod. But the hands can do it. They can also shake the camera, but this is a possibility (in my case 1 or 2 blurred shots of 3 handheld), while the vibrations from the mirror and the shutter are certitudes (100% blurred, when using a tripod, even when, in addition, I hold the camera steady with both hands). I repeat, all these are in regard of using a SLR with a long telelens.

As for the diaphragm blades and the leaf shutters, they move in a symmetrical way, self-compensating the movements of their components, and largely of their vibrations. I wouldn’t worry too much about these two mechanisms.

Q.G. said:

Oh, and...
Before we get a next "fact" in reply...

Click to expand...

Here is the next "fact" in reply:
When using the word "fact", I meant from my own experience, not from what I heard, read, seen to others, or simply imagined. Does now make more sense for you?

I see, yes. You mean you believe that your hands are steadier than a tripod.

Would the possibility that you're wrong, that they are not, make sense to you?

Sure: i have had trouble keeping a long and heavy lens still using a tripod.
That shows that some things are pretty hard to keep still.
I still would not have been able to keep it as steady as it was on tripod without one.


The handholding thing is easliy tested: a while ago now, the penny trick video appeared on YouTube. I don't know where to find it in a hurry now, but it is a simple enough test that we don't need to see the video.
Balance a penny, on edge, on a tripod mounted camera, and see if it will fall over when you trip the shutter.
Then try to balance a penny on edge on a handheld camera (never mind the trying to trip the shutter without it falling over part. You won't get that far).


I'll gladly believe that you have produced better pictures handheld than from atop a tripod.
But (after having tried it yourself) give that test some thought, and try to figure out why.

Q.G. said:

The handholding thing is easliy tested: a while ago now, the penny trick video appeared on YouTube. I don't know where to find it in a hurry now, but it is a simple enough test that we don't need to see the video.
Balance a penny, on edge, on a tripod mounted camera, and see if it will fall over when you trip the shutter.
Then try to balance a penny on edge on a handheld camera (never mind the trying to trip the shutter without it falling over part. You won't get that far).

Click to expand...

http://www.youtube.com/watch?v=CkKcbyh2CrA

Steve

The mirror slap is up/down fore/aft, not side-to-side so it would have little affect on the coin in that orientation. The leaf shutter hasn't enough vibration to affect the penny either. I call that test largely bogus.

Mike1234 said:

The mirror slap is up/down fore/aft, not side-to-side so it would have little affect on the coin in that orientation. The leaf shutter hasn't enough vibration to affect the penny either. I call that test largely bogus.

Click to expand...

As are your arguments that 35mm film is useless and should be destroyed.

Steve

So am I right in thinking that if the penny lands heads up, the shake that dislodged it came before the shutter closed, and tails up if it was disturbed after the shutter closed?

Yours in rigorous science,

Lee

Sirius Glass said:

As are your arguments that 35mm film is useless and should be destroyed.

Click to expand...

I don't recall ever saying that but if I did it was in jest. Got the thread and post number?

Mike1234 said:

The mirror slap is up/down fore/aft, not side-to-side so it would have little affect on the coin in that orientation. The leaf shutter hasn't enough vibration to affect the penny either. I call that test largely bogus.

Click to expand...

Nice call.

But instead of thinking about possible reasons to reject it, just try it.
With the coin in any orientation.

If you want it over with quickly, start with the handheld part. Try to balance a coin, in any orientation, on a camera while holding that in your hand. See how steady that is.

I'm a believer in tripods, but the coin trick is bogus and for more than the orientation of the coin. Blur in the image is the product of the angular velocity of the camera multiplied by the exposure time, which means that a slowly rotating camera with a sufficiently short exposure will show no blur, but a similarly short exposure where the camera leaps (e.g. due to massive mirror-slap, like in a 6x7) can show blur. The problem is that what you need to tip the coin is not angular velocity but a certain total angle of tilt. If/when the camera leaps due to mirror slap and crappy tripod, it will move rapidly for a short distance and then return to its original orientation; it is not tilted for long enough to overcome the coin's moment of inertia. Rephrased, the coin test is meaningless because the problem (angular rate) is the derivative of what the test measures (angle wrt gravity).

So: even if you're getting motion blur on a tripod (get a real tripod already and use MLU - it's not that hard and it WILL work perfectly), the coin will not topple. And if you're not getting motion blur with a short handheld shot, you will definitely drop the coin because there's no way you can handhold a camera to the angular accuracy required even though you can hold it still enough.

Edit: and to put it even more specifically though a little more technically, vibration has a spectrum. All tripods get rid of the lowest parts (below about 5Hz, including 0Hz=constant rotation) of the spectrum, meaning that only higher-frequency vibration is left. The problem is that at higher shutter speeds, it's the high-frequency vibration that can show up as blur in the image where the lower frequencies mechanically blocked by the tripod would have no effect. If you want your tripod to be stable up to higher frequencies, it needs to be well damped and have no resonances and such tripods do exist.

Handholding introduces large vibrations but only at very low (<1Hz) frequencies, which means that for all intents and purposes, handholding introduces a constant angular rate to the camera, which is the basis of the t=1/f rule-of-thumb. If hand-motion is the only source of blur, the rule of thumb works perfectly well, modulo the person-to-person variations in wobble speed, so the rule of thumb is very applicable to APS, heavy 35mm SLRs and anything without a mirror slapping about. As soon as you introduce a bigger mirror though, hands will damp the motion of the camera but whether they will damp it well enough to avoid blur is a different issue - I know I can shoot sharply at 110mm and 1/125 on my RZ, often at 1/60 too. But the latter is more reliable if I use mirror pre-fire, which is hardly surprising because I can feel the whole camera bounce in my hands as the mirror hits the stops.

This link may be of interest:

http://markins.com/charlie/report4e6.pdf

That's a nice article, Ralph. I remember seeing that a few years ago but had forgotten about it until you posted it here.

Hope that ends the mirror slap when handholding myth.

That report illustrates what I was saying very nicely. See the graph of 20um p-p oscillation at 40Hz that occurs on tripod and no MLU? That means the peak slew rate of the camera is 400um/s; with a blur criterion of 10um you need a shutter faster than 1/40s otherwise you'll see blur.

Now look at the handheld graph and note the time scale is totally different; while the peak angular deviation is very large, the hands move slowly. The biggest slews I see in the aurhor's data are 250um in 0.25s, i.e. 1000um/s. A bit worse than on the tripod and requiring a 1/100 shutter speed to avoid blur.

Now replace that little APS (it's a D2H with 16x24mm sensor and accordingly tiny mirror) camera with an RZ or Pentax 6x7 and watch the mirror slap increase 10-fold or more, while the hands will still slew slowly: suddenly mirror-slap is the dominant source of motion! I have an APS DSLR and you can't really feel the mirror slap whereas my RZ fairly leaps from my hands. Using the latter on my (admittedly crap) tripod is an improvement for 1/4-1/60s exposures only if I use MLU as well.

While tripod+MLU+cable release will always give the best results, there are many shots that such an approach is not applicable to and for which you can still get very good sharpness handheld. IMHO, a bit of spontaneity and catching the, uh, decisive moment is more important than ultimate sharpness for many genres of photography.

If one peruses forums where d-word cameras featuring stabilization feature heavily, one will discover that it is routine to take perfectly sharp 1/10-1/2s exposures handheld at night with a little technical assistance. One should have and use the right tool for the job. Sometimes that's a tripod, sometimes it's a steady hand.

Oops, I forgot the 2pi when taking the derivative of that 40Hz sine. The slew rate on the tripod without MLU and due to mirror slap is over 2400um/s, which is twice as much as measured while handheld. That means, according to the results in that article, one gets a less blurred image handheld at 1/100 than they do on the tripod.

Of course tripod+MLU+cable is still best, but this is a good illustration of a poorly damped tripod exhibiting high frequency motion (ringing) when driven by a step function (mirror slap) and thereby making sharpness worse.

The handhold rule of thumb = 1/FL was actually invented for rollfilm (medium format), and the world failed to modify it for the newer 'miniature format' 135 when it came out. Since the original rule of thumb had a safety margin built in, it continued to work satisfactorily for 135 film.

The world woke up when digital APS-C format came along, and modified it to = 1/(FL *1.6) for Canon and 1/(FL*1.5) for Nikon. Similarly the Canon APS-H format = 1(FL/1.3).