DotTune on my 100-400

I have my camera (7d) on a tripod at about 30 feet from this focus chart . I have a 500 watt halogen light pointed at it to try and get enough light on it. My iso is set to 200 and my shutter speed is only 100. I set my micro adjustment to 0 and then use live view to focus on the target using my remote shutter release. I then turn off AF on my lens and start making micro adjustments. First negative and then back to the positive till I find the two furthest points each way and then set it in the middle. I did it 4 different times and got a different result each time. Do I need more light on the focus chart? Also I noticed that the area on the chart that my camera was focusing on in live view was a little bit lower then when I look through the camera. Thanks for any help you can give me.

Yes, try increasing the light you're throwing in your chart. I would increase the shutter speed up from 1/100 a bit too.

Nascar Nut wrote in post #16807712
I then turn off AF on my lens and start making micro adjustments.

One of us is confused. Are you saying that you test the results of changing MFA values with AF switched off?

^ I think that's what you have to do when doing dot-tune. You turn AF off, then keep changing MFA and half-pressing the shutter looking for the focus confirmation light.

OP: it does look your light is not sufficient for reliable AF tuning. I believe I had to get to 1/100s or less at 100 ISO to start seeing convergence of my MFA numbers.

A question: you mentioned that you do remote live view AF - right? I don't think my 50D can do that - perhaps this is something 7D specific? In 50D, live view AF is hard-wired to the back AF button (you cannot use the shutter for that). Perhaps 7D has an option to re-assign it to the shutter button?

Do you mean to say that you cannot use the remote shutter release for focusing in LV? When I do the LV AF on the 70D, I use the remote shutter release button. Perhaps this is what the OP is talking about.

I turned up the iso to 400 and got my shutter speed up to 1/250 and the results were much better. I almost run out of range as I got to +19 on the plus side. As for the remote, I can either use the AF button on the back or the remote shutter release. I think I will wait for a sunny day and do it again at iso 100 if I have enough light to see if it makes any difference.

hollis_f wrote in post #16808895
One of us is confused. Are you saying that you test the results of changing MFA values with AF switched off?

Yes, that is what you do.

How much a spread do you get from a test to a test? Just a couple of notches? Or is something in the 5-10 step range?

Lbsimon wrote in post #16809761
How much a spread do you get from a test to a test? Just a couple of notches? Or is something in the 5-10 step range?

I did it 4 times with the higher shutter speed and I was either +4 or +5 on the low end and +19 or +20 on the high end.

I still think the ol' carpet trick is the best. Throw a quarter out in front of you some distance, enough that the center AF is a bit smaller than the quarter, then take a One Shot focused image, look at the carpet fiber, if the carpet around the quarter isn't in focus adjust, and try again. It takes a few tries, but at some point the carpet will be completely clear around the quarter.

Nascar Nut wrote in post #16809760
Yes, that is what you do.

Yes, so it was me that was confused. My excuse is I've never tried DotTune

Nascar Nut wrote in post #16809771
I did it 4 times with the higher shutter speed and I was either +4 or +5 on the low end and +19 or +20 on the high end.

Please be more specific. Do you mean to say that, let's say, at 400 mm you did the MA four times, and you got the results of +4, +5, +19, +20? Or is it that at 100 mm you got either +4 or +5, and and 400 mm end you got +19 or +20? If the latter, you did everything correctly, there is not much of a difference within 1 point spread. If the former - it is complete screw up.

TeamSpeed wrote in post #16809773
I still think the ol' carpet trick is the best. Throw a quarter out in front of you some distance, enough that the center AF is a bit smaller than the quarter, then take a One Shot focused image, look at the carpet fiber, if the carpet around the quarter isn't in focus adjust, and try again. It takes a few tries, but at some point the carpet will be completely clear around the quarter.

It should work, of course, but the beauty of the DotTune method is that one does not have to take any shots to do the MA.

Lbsimon wrote in post #16809981
It should work, of course, but the beauty of the DotTune method is that one does not have to take any shots to do the MA.

The beauty of doing it my way is that you can do it anywhere on the fly (like on a sideline right before the kickoff) and taking a few test shot either during the tuning, or after to test your tune, is a wash.

It is always good to know a few different ways, because in some situations, one might be better than another.

I do both. MA with DotTune and then grab some test shots with a real-world target to see whether it truly hit the sweet spot. So far, I've been very happy with the agreement between them (as long as I use lighting with the same color temperature).