Camera Labs

Hi folks,

With 10fps available on pro models and 8fps on semi-pro models I'm tempted to wonder if there are any lenses out there which can't keep up?

Every time the shutter fires the lens iris has to close and then open again quickly enough that the AF tracking can perform its magic prior to the iris closing once more for the next shot. How quickly the iris can perform this feat doesn't appear in the regular lens specifications, so far as I can tell, and it occurs to me that if the iris doesn't reopen quickly enough then AF tracking may be compromised.

Does anyone know the answer to this (should some lenses be avoided with the Canon 7D, for example) or are all mechanically compatible lenses quick enough, even those from third parties such as Sigma or Tokina?

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Does the iris also close when you shoot wide open? I noticed that the DOF preview sound is near-identical to the sound I hear when I release the shutter, when I use the Sigma 70-200.

Also, take into account that burst rate is sometimes also compromised by focusing in between (if you don't have the focus mapped to another button), and finally by shutter speed. You can't shoot 3 frames per second if your shutter speed is half a second.

_________________
I take pictures so quickly, my highschool was "Continuous High".

I don't have the answer to this, but it is a good question.

To people who have been in photography longer than I have, perhaps to film days, how long have speeds like 10fps been available?

As high speed isn't brand specific, I'd have to make the assumption that 3rd party lens makers will also have to design their lenses to cope with the speeds.

On a related note, should wear of the iris mechanism be a concern in a similar way to the camera shutter wearing out? It's mechanical and has to go through repeated fast action during its life.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

It's not the lens that is the determining factor here. It's the shutter speed. So no, the lens you use doesn't make a difference.

_________________
I've upped my Iso, now up yours.
Nikon D700, Nikon D80, Micro Nikkor 105 f2.8 AI-S,Nikon 85mm 1.8D, Nikon 50mm 1.8D, Nikon 35mm F2D, Nikon 16-35 f4 VR, Nikon 24-70mm F2.8 AF-S, Nikon 70-200VRII, Nikon 70-300mm VR, 300mm f2.8 AI IF ED, SB-800, SB-900. Noiseware pro,NIK Dfine, Nik Colour Effex, Nik Sharpener Pro

Good question... Don't know.

- Bjorn -

_________________
Street and documentary photographer | Google+ | Twitter | Blog | Tumblr

Leica M9-P (my article on Camera Labs) | Leica D-Lux 5 | 50mm Summilux

You may be right but yours is such a definitive statement that I'd appreciate it if you could share your sources.

Note that I'm not suggesting that the lens/camera will suddenly stop working here but, as others have agreed, it may be a fair question to ask whether AF tracking performance can be adversely affected of the iris mechanism in some lenses doesn't cycle quickly enough.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

While the shutter can be limiting, I think for the original question you have to assume it isn't set such that it is limiting.

The shutter curtain travel time can be estimated from the fastest flash sync speed, typically as slow as 1/250s. If the camera's controls are perfectly in time, the 1st curtain will pass in 1/250s, the flash goes off, and then the 2nd curtain moves in another 1/250s. Thus the shot is exposed at 1/250s. For a faster exposure without flash, the time will converge to 1/250s, as the 2nd curtain just starts sooner. In practice the shutter is probably faster to allow more tolerance. For slower shutter times, the shutter cycle time could be estimated by the curtain travel time plus wanted exposure time.

If we assume 10 fps, you have 100ms to do everything for one shot.
AF and metering
flip mirror and set aperture
shutter movement
mirror return and aperture reset

1/250s is 4ms. Ok, it will be extended by whatever exposure you set, but even as slow as 1/100s that's around 14ms total shutter time. That doesn't seem significant at all.

Thinking it out like that, the iris has the same window of operation as the mirror movement. Maybe the mirror is the more significant limit.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Never had thought of that....very interesting. This could be a problem for cameras like D3s,that has 11FPS...to be restricted to only pro grade lenses or something,as these lenses could support high burst rates. Who knows...
I have no idea on this matter. It would be a shame for example,for (some)film lenses to not support the FPS abilities of certain cameras.

_________________
Gear: Nikon D80, 18-105mm, 35mm F/1.8, 85mm F/1.8, SB-600
Visit my work here

I'll gladly share my source if you can share yours, that actually told you that such a thing was possible, feasible. I'm simply assuming that you'd have to do a huge battery of tests to actually prove such a thing, let alone get people's attention. Even with advanced 3d tracking and so on, how many photographers do you think would even think that the motion of the iris could even come close to being the reason a frame was out of focus.

The only way you'd be able to even entertain such a theory would be to have a repeatable AF tracking test, and the knowledge of the exact action required to capture an image. Please excuse if I sounded hostile. I just don't see the subject as something you could actually put to the test in real world shooting.

_________________
I've upped my Iso, now up yours.
Nikon D700, Nikon D80, Micro Nikkor 105 f2.8 AI-S,Nikon 85mm 1.8D, Nikon 50mm 1.8D, Nikon 35mm F2D, Nikon 16-35 f4 VR, Nikon 24-70mm F2.8 AF-S, Nikon 70-200VRII, Nikon 70-300mm VR, 300mm f2.8 AI IF ED, SB-800, SB-900. Noiseware pro,NIK Dfine, Nik Colour Effex, Nik Sharpener Pro

Interesting question. I think it should be possible to test this.

Take two cameras on two tripods. One shooting video and one shooting stills. The one shooting video (preferably at a fairly high frame rate) should be set up to look inside the lens of the one shooting stills, slightly from the side so the stills camera can easily focus on a subject other than the other camera, and filming the movement of the iris set to a small aperture in the stills camera. Start the video recording and use some remote/wireless release to let the stills camera fire a burst of your hand which you move to and fro the camera so the stills camera needs to refocus in between shots. Examine the results of both cameras and you might come up with something interesting.

The video camera must be able to keep focus on the iris of course. I don't know how easy/difficult that would be.

Anyone got the gear and the inclination to set this up?

Ben
_________________
When in doubt..... Press the shutter.

Anyone care to agree or reject my earlier theoretical calculations (with reasons) showing shutter speed is not significant unless you choose to make it so?

With the shutter eliminated as a significant contributor, the question then becomes what portion of cycle time is taken by the iris movement. Logically it must occur during the mirror movement, therefore if the iris is "quick enough" it is not limiting, but the mirror is.

Ben's idea did cross my mind. Something like the 1200fps Casio would be ideal, if only I had one.

I've only got up to 60fps on the 7D, which means I'd have to use the 50D to do the shoot. One shot on that would be just under 10 frames which I feel may not be sufficient time resolution to be clear.

I might actually be bored enough to try it as I've not really played with video much yet.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Hi Gregory,

I think your last post amply demonstrates your lack of an official source for your assertion in your first post. I'm not at all sure why you should be getting so worked up over this. I don't need sources in order to ask a question - if I did the question would already have been asked and hopefully answered. But while I agree with you that this hasn't been flagged as being a problem up to now that doesn't actually mean that the problem may not surface. On the Canon side we now have a camera capable of firing off 8fps and it's within reach of the well heeled amateur rather than the pro who is more likely to use top of the range glass. My motive in raising the question is part curiosity and also a desire to avoid a situation in the future where a member has AF tracking issues and we either fail to ask, or even erroneously ask, whether the lens used is up to the job.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

Hi popo,

Thanks for your contributions. I hadn't considered your point about the iris having time to toggle during mirror movement. On the face of it it would seem that the moving mass of the iris would be less than that of the mirror and so given that the mirror has time to cycle so must the iris. But, of course, the electromechanical force applied to the iris would also be tuned to its moving mass so it's still possible that the iris may not have fully opened in time to allow the AF tracking to operate to best advantage.

You are, of course, dead right that the only way to sort this definitively, short of Gordon asking his manufacturer contacts for an answer if he considers the question worthy, is to do an experiment.

Bob.

_________________
OM-D E-M5 + ED 12-50mm, H-PS14042E, H-F007014E, L-RS014150E, M.ZUIKO DIGITAL 45mm 1:1.8.
Gitzo GT1541T, Arca-Swiss Z1 DP ball-head.
Astrophotography: TEC 140 'scope, FLI ML16803 camera, ASA DDM60 Pro mount.

I did it. Here's still frame extracts from 60 fps video of the iris of the 50mm f/1.8 lens going from f/1.8 to f/22 and back again. The time between each frame is therefore about 1/60s or about 16.7ms. The camera used was the 50D set to manual at 1/8000s and f/22.

1
2
3
4
5
6
7
8
9

Note in frame 1 it is clear. In frame 2 you can just see the aperture closing in the lower left. In 5 and 6 you see it remaining closed, returning to fully open status in frame 9. So during a shot it took 8 frames to go open-closed-open, or about 1/7.5s.

With hindsight I should have done a continuous burst so I can see how long it takes to go closed-open-closed too. That would give an indication of the time between each shot which still needs to be counted for a cycle.

Even without that, the claimed shooting rate of the 50D is 6.3 fps, or a cycle time of 1/6.3s. Subtract the iris cycle time from the shot cycle time, we get 25ms (approx 1/40s) of fully open aperture for the camera to do whatever else it wants to. If that is "enough" time is another matter, and would need testing of AF tracking rates. That's far beyond my motivation to create and execute a test for.

And the other part of the assumption: the mirror.

1
2
3
4
5
6
7
8

Frame 1 shows the rest position. Frame 2 shows slight blur as it starts to move. Already in frame 3 it is in the up position where it stays to frame 6. On frame 7 it is returning to the down position which it reaches in frame 8.

It is taking just under 7 frames for the mirror cycle, which is slightly shorter than the iris cycle. If this were a faster camera, then it would be even shorter.

Of course, different lenses may vary their iris at different speeds. The 50mm f/1.8 is a budget lens, so arguably more expensive lenses may have faster operation.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Good question Bob - and great images popo!

Next time I'm with a manufacturer techie, I'll ask them!