7D Mark II - Focus Discussions

habenero wrote in post #17443635
I've been following this thread for a while. Even contemplated a 7d2 earlier on but after this thread and others around the net, glad I did not. It amazes me that this discussion continues to grow and with new contributors.

Maybe it's just me, but a $1800 camera should focus right out of the box. MFA is for "fine" adjustments/tweaking. Like I dropped $40k on my new beamer but I have to go home and set the fuel curve for it to work properly...

Good luck to all with this body.

As I posted earlier, there may be some bodies with issues but we're only seeing the posts from those that have issues since that is what this thread is about. I've seen lots of people posting great photos and expressing plenty of satisfaction on several forums. In my experience, those that have problems will be a lot more vocal about it than those that are satisfied. While this thread has had several people posting their focus issues (I'm not willing to count how many over the 127 pages!) my bet is that it is still a small percentage of the total number that have purchased a 7D Mark II.

Yes, my 7D Mark II required MFA for front focusing and has been sent to Canon for adjustment but the photos I got with it (after MFA) were fine. These photos were all shot at high ISO (ranging from 2500 to 8,000), AI Servo, have had some noise reduction added and a small amount of unsharp masking to make up for the compression of posting on the Web. The only thing I really haven't been able to test is the AI Servo tracking for someone or something moving quickly as I only shoot sports on occasion and I won't have another dance program to shoot until April.

A couple of more recent photos with my 7D II.

rejay14 wrote in post #17443480
So I just conducted the best test I could think of at four o'clock in the morning. I MFA'd all of my lenses again using yet another technique, I changed a few settings by 1 or 2, one by 5. Some stayed the same. (The 50mm 1.4 is a nightmare.)

Then I got out my trusty tripod and took ten photos of my croquinole pieces. Please don't make fun of me... croquinole is fun after too many drinks.

Sorry, I digress:

Remote shutter control, mirror lockup, 1 second between lockup and trigger, tripod mounted, 1/200th ( I keep forgetting that it syncs @1/250th), flash on ( to help freeze the motion of the wooden playing pieces while they are sitting there, stacked up).

Several fails with one-shot, several fails with servo. I heard the lens refocusing between shots. I wondered, and still do, why.

I got 1/2 keepers with One Shot, less with Servo. On a tripod. WTF. **(Welcome To Facebook Moderators...)

I also tested the Canon CPS loaner combo (7D2 + 70-200 IS II 2.8). I guess they don't MFA the gear they send out, but in their defence, it would be a lot of work. Theirs needed T+13, W -2. (Please, nobody tell them that their lenses are out of spec and need this. They'll accuse us of a conspiracy.) With so many post contributors adding their valuable insight and suggestions, I was honestly going to work on my technique. I thought that it maybe really was me. (It still could be).

It's been a while since I played sniper poker and I probably shake more since then too. If I'm wrong or find something that I did incorrectly, I'll say it. This forum runs on assumed honesty. We need to figure out what we screwed up and pass the word on to save other members valuable time.

But on a tripod? This kinda P1$$3d me off a bit.

So, for the misses, for a given lens, where was sharpest focus, front back, both, none?
If none, IS off or on (since was a tripod)?
Was there good contrast and was single focus point enveloped by the target since the actual Af point is larger than the Vf indication square?

I agree its not everyone having issues. If there were, this camera would not have 4.5+ stars on every site that sells it.

I have seen some amazing shots come out of this camera.

I noticed some one saying something about the camera acting like its still tying to focus after the shutter button is hit, and I have heard my camera doing that as well. Even on the new body. I turned the beep sound on, because I thought maybe I was just trying to "snap photos" too quick, but even if thats the case, what a shame that my camera needs me to be still and practically hold me breath for a good shot. I want to try this MFA thing and see if I get better results, but I need to be able to pinpoint if its FF, BF or just inconsistent. My other camera was strictly inconsistent.

Weather is finally starting to break here, somewhat at least. Went out yesterday afternoon to a mostly clear sky. Finally managed to find something alive to shoot. The pin point of sunlight in the birds eye revealed the small amount of camera shake like a moving flashlight in the dark. No real detail to the birds feathers due to the sunlight. I did get the AF second guessing itself and had to put it in one shot.

Went out again today, light overcast so diffused sunlight. Due to some posts here I disabled MFA and ran a set of shots from 150-600 at 50mm increments (all shots with tamron). I then enabled MFA and did it again. All shots with MFA enabled were better than the shots with it off excluding one and that shot was sharper than all the others regardless of MFA status.

The shots below won't win any awards (taken today) but it is a worse case scenario for AF: white on white. Willow ptarmigan. The real trick will be to get one on the wing.

Both shots in 'M' mode, F7.1, auto ISO (158) and 1/800 sec. SOOC jpgs that were just resized. Must of bumped the main dial as I had set it to 1/1600. Oh well. Single bird at 600mm, one shot focus; 3 birds at 480mm and AI servo (aimed at middle bird)

magicmikey wrote in post #17442346
@ScPhotoMom - One more thing. Because the sensor in the 7D Mark II is so much higher in resolution than in your 40D, it will show lens imperfections more. That means a lens that isn't razor sharp is going to look softer when used on the 7D Mark II than on the 40D. That's why people end up having to buy more expensive lenses to keep up with their camera's resolution. (Also, that's one of the reasons why Canon made newer versions of the 24-70 f/2.8L and 70-200 f/2.8L IS lenses. The higher resolution camera sensors were showing the flaws in the older versions.)

I'm not sure I follow the argument that the 7D2's high pixel density would make shots soft when compared to a body with lower density. At 100% crop I completely agree the 7D2 is going to have a higher chance of looking soft. A given number of pixels on a 7D2 take up less space then the same number of pixels on a body with a lower resolution. Essentially one would be zooming in further on the 7D2 image at that point.

I don't think pixel density should have a negative effect on the average, not aggressively cropped image. If I take a the same shot with both a 7D2 and a 40D and then look at them side by side at the same relative zoom (so both images appear the same), I would expect, at a minimum, that the 7D2 be on par with the 40D. As long as the lens is projecting a focused image on the sensor, all those extra pixels shouldn't matter. Perhaps there is some part of the science behind this I am not grasping.

In my mind, moving to these higher density sensors simply affords the shooter the opportunity to crop even tighter or produce larger images without losing detail. This is where I think the new lenses come into play. We couldn't employ all those extra pixels if the lens simply cannot resolve that level of detail properly.

It's not that the high pixel density of the 7D II makes photos look softer than a lesser pixel count, it's that it reveals flaws in lenses, focus and camera movement more because of the ability to enlarge it more. It also shows the camera movement (camera shake) more. I didn't explain it well in my post above.

MNiceGuy wrote in post #17444721
In my mind, moving to these higher density sensors simply affords the shooter the opportunity to crop even tighter or produce larger images without losing detail...

Exactly, and it is because of the tighter crops and larger images that it is easier to notice flaws.

magicmikey wrote in post #17443700
I think you would be a lot better off doing MFA with good light and not using flash. Using flash adds another element to the focusing system if it is using the focus-assist beam. When I did MFA with my 7D II, I used the DotTune method and used lots of light. I was in my garage so I added a 150 watt quartz light to my focusing target.

I already MFA'd my lenses, for the fourth time now, before the test. Regular overhead room light used for the test, flash to see what it might help. If this thing can't focus consistently on a dark object in a lit room, that would throw a lot of real world shooting out the window. I'll do it again in bright daylight.

habenero wrote in post #17443635
I've been following this thread for a while. Even contemplated a 7d2 earlier on but after this thread and others around the net, glad I did not. It amazes me that this discussion continues to grow and with new contributors.

Too bad. You're missing out on a great camera. Second time out with my 7D2 and I'm much much happier. It's as suspected, just takes some practice and familiarity with the camera to learn how to use it properly. I filled a 32GB card shooting perched and ground feeding songbirds from 30-50 feet and waterfowl in flight from far too far away, but that's life. Using canon 100-400 I. Handheld, IS turned on (I) and set to mode 2 (for panning). Occasionally I'd set the lens to look for objects beginning at 1.8m, but mostly was set on 6.5m to infinity. Camera set to auto ISO, Tv, Case 1 (primarily, but I did try out other Cases), centerpoint AF when shooting songbirds and full 65-point AF (open brackets { }) when shooting birds in flight (BIF). Shutter speed ranged from 1,600-8000. Back-button AF (a first for me). Full sun when starting out on the songbirds, but clouds rolled in later. Regardless, I continued to shoot because the ducks kept flying.

I was amazed by the 65-point AF, not only did it lock on BIFs that I'm sure I would have missed with my 60D, but it allowed me to frame the shot rather than forcing me to keep the focus at the center. (Most BIF people use single center AF point when shooting BIFs.) I have to admit, I would like a faster method of switching from centerpoint to full 65-point AF. Flipping through all the options takes too long.

Here's two samples; Singlepoint AF nailed the birds eye hiding behind the limb and 65-AF point locked on the canvasbacks and allowed me to frame them. Neither image is cropped or altered except maybe a slight exposure adjustment.

8612images wrote in post #17432644
I'd like to give a shout to Adorama. My 7d2 was a bit quirky on AF and I had dropped them anote about 2 weeks in on the body. Tried to muddle thru it then the LCD scrambled. I was about 10 days past the return period. Spoke to one of their reps and they said send it back.

Delighted to hear you were looked after, well

rejay14 wrote in post #17445136
I already MFA'd my lenses, for the fourth time now, before the test. Regular overhead room light used for the test, flash to see what it might help. If this thing can't focus consistently on a dark object in a lit room, that would throw a lot of real world shooting out the window. I'll do it again in bright daylight.

You have to do MA with a brightly lit target either with a LED or daylight is best. You are not testing low light performance at this point. You want to accurately calibrate your lens and camera. Well lit target no shadows and daylight or artificial lighting that supplies daylight color is best. Also, absolute proper alignment of the focal plane is necessary for accurate results. I like LensAlign with Focus Tune the best out of any method I have tried because done correctly it takes all of these factors into account.

Well I have to say, my first 7D2 definitely had some AF issues in servo mode. Tried all the different settings, and having had 2 x 1DMKIVs there's no real mysteries there.

It went back and go swapped for another, and the difference is chalk and cheese. The new body is dynamite. I've had it since about 2 weeks after launch, so quite a while now. I have to say this is a great camera.
Really pleased with it.

Just wanted to say that, as so many people seem to have returned it and then gone elsewhere, and I think that's a mistake. One swallow a summer does not make, as they say.

butterfly2937 wrote in post #17445659
You have to do MA with a brightly lit target either with a LED or daylight is best. You are not testing low light performance at this point. You want to accurately calibrate your lens and camera. Well lit target no shadows and daylight or artificial lighting that supplies daylight color is best. Also, absolute proper alignment of the focal plane is necessary for accurate results. I like LensAlign with Focus Tune the best out of any method I have tried because done correctly it takes all of these factors into account.

This is good advice. Don't invent your own procedure for MFA. Follow a proven method.

I followed the Dot Tune method and it worked great.

habenero wrote in post #17443635
...

Maybe it's just me, but a $1800 camera should focus right out of the box. MFA is for "fine" adjustments/tweaking. Like I dropped $40k on my new beamer but I have to go home and set the fuel curve for it to work properly...

Good luck to all with this body.

Just my opinion of course, but if all it took to get a 7D2 to work as well as they do for some, I'd suggest it's worth taking time to MFA.

At least try it with your main lens and see if that solves it. I'm not saying it does, but just that if it did, then go for it.

And as for t5he $40K beamer, maybe that's the wrong car. Maybe if it was an Alfa romeo, then everyone would understand that some wrenching is required to get to work perfectly. Hopefully the 7D2 stays working longer than an Alfa!

Seriously it sounds like the troubles are well beyond a simple MFA fix.

P.S. one of the fixes that was recommended for the 1D3 was MFA.
It did not help then either.