In any case, the 1/50 result in your recent test is at odds with the previous tests - by 2:1. I can't quite get my head around that one.

Understood. Get a good foundation and then add no more than one variable at a time.

Still, HTP seems like it might be a helpful tool.

Also helpful is knowing what settings don't matter. For instance, knowing that anything over 100 ISO displayed is actually 1/33 means we don't have to sweat getting the light just right so we get 1/50. That can save a lot of time on the set. Just dial the ISO above 100 and shoot.

Holy Cow, I didn't notice that everything above ISO 100 was 1/33. That is really extreme. I went to my earlier data and noticed I had shot everything at 100. I guess they are trying to keep the ISO at a minimum. So they use whatever ISO they have to for correct exposure at 1/33.

I wonder how bad 1/33 looks? It certainly isn't anything we had wished for.

Almost everything that you've seen from the 5D MkII that's not outdoors during daylight has apparently been shot at 1/33.

No wonder people have been complaining about a non-film, soap opera look. It's not just 30p; it's a ~325 degree shutter!!!

Not only do we need fast lenses, we need lights! (or sunshine.)

Or a 360 degree shutter. The test before this last one was giving me a full 1/30 speed. I could hold my finger on the lcd where the band started, advance one frame, and see the band end at exactly the same place. This time there was the small (10%) difference in position when I held my finger there.

That would mean LaForet's video was all 360 shutter. I'm going to go back and look at his blur.

I wonder if the difference between 1/30 and 1/33 was some kind of compression time problem. This last test was of a blanket with a fuzzy pattern and the 1/30 result test was a blank wall.

Once again I am totally confused.

I looked at the Laforet first scene where the guy crosses the screen to kiss the girl. It was a dark scene and no lens could have allowed enough light to get to ISO 100.

If the exposure was 1/30, then I should have seen blur as wide as the distance he moves each frame, right? The blur was much smaller than that, maybe one fourth (90 degrees?). His hand moved 22 pixels and the blur on his sleeve looked like about 6 pixels.

P.S. I don't see how it could make any difference but this latest test was locked at f1.4 and the previous was locked at f5.6. Could the algorithm behave differently when we lock (electronically isolate) the lens?

P.S.S. Looking at the numbers closer (I used excel this time), I found that the strobe lasted about four lines. The smallest number of lines in a fast band was around 300 so that source of error was small.

P.S.S. Could my test be flawed? When a fast strobe can fill the entire screen with light, which happens often, that means it exposes for a long time, close to the 1/30, right? How can this thinking be wrong?

I have an idea. I'll do my flash test and then in the same scene wave my hand in front of the lens or get some other kind of motion in. Then I can compare the blur to the speed.

Mark:

Very interesting data coming out of your experiments. Now, you're probably going to hate me for this, but I assume your tests have been performed with Nikon lenses to take focal length out of the equation. Maybe you should try again with canon lenses and different focal lengths and that might throw very different data your way at iso speeds higher than 100. If that works out to be correct, canon effectively worked out a system to really prevent us users from ever hoping to have any sort of manual control. If you use nikons to control aperture, shutter speed effectively goes out the window at iso's higher than 100. Use canons and shutter becomes predictable but only as a function of focal length and aperture is fixed at certain positions.

The only way we could ever hope to effectively control shutter speed would have to be by tricking the camera into thinking the lens was always fixed at the focal length that gives us the shutter speed we expect. Maybe a custom nikon mount that somehow faked canon lens info into the camera? :)

P.S. Laforet shot usng canon glass.

I have used a Canon 35mm 1.4 for all tests. On the next-to-last test I had it locked by rotation to 5.6. On this latest test it was locked at 1.4.

Are you saying that maybe my results are the way they are because it doesn't know the focal length? I will try a quick run without locking/rotating the lens. If it changes my focal length it won't really matter for the purpose of this test.

No, what I'm saying is that the camera might be trying to adhere to the 1/focal length rule to set your shutter speed. Since your lens is a 35mm lens, the camera might be trying to stick to 1/35 which sounds plausible, given your 1/33 results. The same battery of tests might yield diffrent results with:

a)Unlocked lens
b)Longer focal lengths, like a 50mm, 85mm or a 135mm

Mark employed the lens twist method (as I understand it) so the camera was unaware of what lens he used. As such, the 1/focal length rule should not be a factor.

Mark, I just want to say thanks for doing this work (and Jon before you)... Like many others, I'm following this with great interest. Very very useful!

If you don't untwist, the camera will use 1/focal length for video recording. I rarely shoot less than 50mm, so everything's fine even in low light.

Isn't it true though that sometimes the camera will switch f-stop while shooting if the lens is not locked (rotated)?

P.S. As I said before, all my tests before now had the lens locked (rotated).

This means giving up aperture control, and allowing the camera to dictate shutter - not fine for me at least. Shooting at say 135mm is a joke under these conditions.

After three months switching between EF and Nikkor lenses on the 5D, I find Nikkors are the only way to go for the control I need. I find the numbers posted in this thread invaluable.

It's tricky to get the f-stop where you want it, but once there you can lock it (for one take! then go through the process again). However, the shutter will vary with focal length. I think your work is all about gaining control of the camera through better understanding the shutter (and readouts). I think the way you've gone about it (rotated) is correct. This is the only means (along with using Nikkors) that we can do serious work with the camera. Perhaps the one exception to this rule is when shooting with EF lenses in the 50mm range, where they will shoot 1/50th.

I haven't been able to test this because I'm still waiting for my 100-400 to arrive, but IIRC the camera doesn't go faster than 1/125s at any focal length, which would be a 90° shutter angle. I can live with that. Dodging the 360° shutter is my main interest.

It is a false rumor started by Canon in their manual.

I showed on page 3 of this thread it going to 1/300 and others have shown even higher. Think about it, if can only go down to ISO 100, what can it do if you put more light in? It has to go to a faster speed. It has no choice.