MFA'in my 7d mark ii

So i got my camera out today and swept the dust off and said to myself.. I got to figure some things out..

1. figure out how to use the Neewer 2.4ghz remote for the grip - Done...
2. Google and learn about MFA

well I am on 2. now and been reading up on it. Started out with Dot-Tune and didnt actually try it. I then got to this video
https://www.youtube.co​m/watch?v=m2tKs1zQ58E

and downloaded his chart. What a easy way. I am at work and decided to print the chart and give a handheld test. I know tripod - will do at home...
I then decided to give 24mm and 105mm a try.

So still trying to get a grip on this but would you say...

1. @24mm - Way off
2. @105mm - pretty close

FYI handheld method for quick view took less than 5 minutes

Nope, I wouldn't say that at all.

The problem with using a target that's not parallel to the image sensor, is you don't know what the camera is actually focusing on. The focus points are actually bigger than the boxes you see in the viewfinder, but regardless, when you put that point over an angled surface, it's actually covering a range of focus distances- and the camera has to pick what distance it thinks you want.

MFA works best when you give the camera an unambiguous target to focus on. Look at this target: http://www.bhphotovide​o.com …ap=y&m=Y&is=REG​&A=details

Note that you focus on the big white cross on the left side- which is parallel to the sensor plane. You also get the front-to-back scale in the photo, so you can tell whether it's front- or back-focused, but you've now given the camera something very distinct to focus on.

I've basically made my own target by making that front-to-back scale like you did, and then setting a vertical object (like a wine bottle) so that the front of it is in line with 0 on the target. Does that make sense? I can take a pic of the setup later if you need, but I'm at work now.

Thanks for your explanation and yes it makes total sense.

A pic would be great!!!! hmm wine bottle... does it have to be empty first? please say yes!!!!! hah...

well regardless to the true fine tuning.. would you say that my lens does need adjustment on the wide side?

once again appreciate your feedback...

You really need to test it with a tripod and maybe a bit of better light but you are on the right tracks.

Oh yeah, tripod and a lot of light are also important things. Good catch.

Wine bottle doesn't *have* to be empty, but if you were to pour a glass out to drink while testing, it would work just as well, so there's that...

Impossible to say if you need to MFA for the wide end. Or the tele end- without a more reliable test, it's impossible to say if you just got lucky this time.

Worth considering: does the 7d2 store an MFA value for the wide and and the tele end of zooms? I know the 7d does not. If you can store a value for each end of the lens, then great. If not, you have to pick one that best approximates both ends. This can be tricky, but I usually go with what the tele end needs, since that will have a thinner depth of field and be more focus-critical.

yes there is a wide and tele setting.

I will give a go later tonight with lots of light, tripod, wine bottle or bottle of sampling choice!

Thx

My typical target for checking focussing accuracy is three beer cans on the bench side by side with either one each side being half an inch in front and the other behind.

Using good daylight focus on the middle can on a high contrast area. Low contrast is no good. Pop your image file into the PC and enlarge to 100% - middle can should be sharp and ones on either side slightly less sharp, depending on your aperture.

This is from about 5 feet away for a short or medium length lens 24/35/50 mil, and 10 feet away for a 85/135/200 mil using a tripod as suggested already.

I tried this on a 24mil bought new on the grill of my car once and got strange results. Tried it again on matt high contrast area later and it was perfect. Conclusion? Must have been the reflections thrown the focussing on the car grill.

I MFA'd that lens then the next day tested it again and got different results. reset it back to zero and it's still fine to this day!

I did my 7d2 with a Sigma 150-500 that had a front focusing issue with several charts and never was totally satisfied. Last thing I tried was the dot-tune and am fairly happy with the results. As others have mentioned, tripod and good light really help. Also, kill image stabilization. Some lens are okay on a tripod with it still on, but my Sigma does better with it off.

The method that you are trying to use proved to be unreliable unless it employs single malt scotch not less than 18 years old.

A better target is a cereal box. It is flat, and it usually has all kinds of text and pictures of all kinds. However, I still prefer DotTune method. It only seems intimidating at first try, but in reality I was able to MFA all my lenses on two cameras in less than an hour. Please remember that good lighting, a tripod, and remote shutter release of any kind are essential.

You need something that has 2 planes in use, a vertical plane for a target that is parallel to your sensor, and a metering plane that shows you where your DOF lies.

This is something I have offered for years, but anything you can create that emulates this should work, even if it is a DVD/movie case standing upright and a ruler next to it at an angle.

Having had a good read up on Dot Tune that is the option I will be using to set up my Sigma 150-600 after christmas when I have my USB dock. The advantage of Dot Tune is that it removes the shot to shot variability of other methods. Other methods require that you take multiple shots of the target and try to average out the variability. I also have Magic Lantern on my camera, and that allows you to run Dot Tune automatically. With the Sigma Dock allowing for 16 variations in zoom and focus distance, it will make it much easier, as all one has to do is run Dot Tune for each setting combination that the dock allows, and use the dock setting to adjust the lens until each is at zero. Just setting the MFA at the one setting the my 50D allows, using a target at 400m requires a lot of shots to average out where the setting should be.

Alan

I've tried a few methods, but broke down and bought FoCal when I bought a 7d ii, because it was a great excuse to try it out if nothing else. It's well worth the price if you've got more than a lens or two to tune. The down side of upgrading from a body that has MFA to another body that has MFA is you have to reMFA all your lenses. However, with FoCal I could get more confident results than I ever got with any other method and I could do it in just a couple of minutes per lens. I think if I had to do reMFA all my lenses without FoCal, I'd simply never find the time or energy to mess with it all.

Using the proper chart as described earlier and understanding how the adjustment works, you should be able to take 4-5 shots at most to get a good MFA value.

Keep in mind, though, that while Dot Tune will give you a great and repeatable value, it will be at the distance where you tested it, with the lighting you used. It is possible that your setup might front or back focus at different distances; that it might have a focus shift when focusing on moving subjects; and that the lighting may have an effect on the camera's focus. It will depend on the lens and your shooting technique.

What I usually recommend is to set up your camera in a consistent manner, using known distances and lighting, and a stable mount. The environment should be as close as possible to what you're expecting during a normal shoot, especially distance and lighting. Then, try it out on an actual event and check to see if there are any discrepancies. If so, make corrections and try again.

Sometimes my static tests give me perfect results at a shoot; sometimes there's a small offset. It usually depends on the lens more than anything else, for me. At least I know when I've tested it beforehand, whether there's an obvious defect in the camera or lens. After that's it's just tweaking the settings with confidence.

GregDunn wrote in post #17821611
Keep in mind, though, that while Dot Tune will give you a great and repeatable value, it will be at the distance where you tested it, with the lighting you used.

ALL of the methods discussed above have that limitation. Your other comments are in line with Canon's recommendation to do on-site MFA tweaking as necessary. Take a couple test shots when you get where you're going and if the focus plane appears to be behind your subject, add some MFA.

I had limited success with dot tune. The Focus genie / lens align methods are also easy but it's not always easy to see those little rulers at any kind of distance.

You can test your results when you're done - live view focus on an object the rough distance that you plan on shooting and then half-press the shutter button. if the lens moves, you probably didn't get it right.