Canon 7D mk2 owners thread.

Another weekend out with the 7D MK II at BSB and found it being rather hit and miss with side on pans this outing mainly with the 70-200 and the only thing different is i carried out MA with Focal during the week.Had a lot that it would catch the first image slightly out but then throw a fit the next two or three in the sequence,tried various cases and also spot a/f and surrounding assist points with no improvement.This was even the case with bikes on the slow down lap with shutter speeds way up beyond normal that in all honesty the old 7D would of nailed straight off.

The 1DX hits focus bang on every time in all fairness and maybe im just used to it being so dependable that im expecting a bit much from the 7D MKII.

Have now reset back to original and will try again next outing.

On a positive side when it hit focus bang on the files are superb.
 
Gadget-Guy said:
The 1DX hits focus bang on every time in all fairness and maybe im just used to it being so dependable that im expecting a bit much from the 7D MKII.
Click to expand...

It sounds like there might be something more going on IMHO. I've not had any trouble like that with panning so far, I did when I had my Tamron 70-200VC but that was a lens issue and it was with the 1DX and 5D3.

Any problems with static or head on shots?
 
Toggerman said:
It sounds like there might be something more going on IMHO. I've not had any trouble like that with panning so far, I did when I had my Tamron 70-200VC but that was a lens issue and it was with the 1DX and 5D3.

Any problems with static or head on shots?
Click to expand...

statics are fine and head on shots im getting a lower keeper rate since carrying out calibration but this is even after i turned it off in camera after it started playing around.

Going to have a play again at the weekend but i know for a fact it didnt nail a fair few shots that it should of had no problem locking onto or keeping track of.
 
I had my 7d mk2 since the photoshow it took sometime to get used to it and figure out how to use it. I fined it very sharp with my 70-200 is f4. I do use this for birds and other wild life. What I have done set up the first and second shooting priorities for focus, can't see the point of shooting unless it's focused. I have the af on set for a1 servo I chose which case I want. I have * set for one shot. But so far so good I did down load Canon's pdf on there af set up worth a read
 
I've found that the first 1 or 2 are slightly out sometimes in a sequence of fast moving objects but then its pretty good. In a direct head to head with my 1DX it did pretty well but the 1DX generally had the first 2 frames in focus.

I have noticed that the single AF point seems quite big. When shooting birds you dont need to move it very much for it to miss focus. I tend to shoot spot for single images.
 
Last edited:
wave01 said:
I had my 7d mk2 since the photoshow it took sometime to get used to it and figure out how to use it. I fined it very sharp with my 70-200 is f4. I do use this for birds and other wild life. What I have done set up the first and second shooting priorities for focus, can't see the point of shooting unless it's focused. I have the af on set for a1 servo I chose which case I want. I have * set for one shot. But so far so good I did down load Canon's pdf on there af set up worth a read
Click to expand...

This is exactly how I have set mine up and so far so good... I'll see how it goes this weekend at Harewood Hillclimb.
 
A quick explanation for anyone unsure about AIservo tracking and shutter release.

There's a menu item which lets you prioritise between shutter release or focus. Release will always happen on demand even if you have first image priority set to focus and the only difference is that focus priority (or biased towards focus priority) allows a little more time before release.....ie, AIservo will release the shutter, focused or not. One Shot will not fire until focus has been deemed to have been achieved.

If the 1Dx produces a more consistent batch of initial shots in focus then it's down to either better AF arrays or faster processing and lens drive commands.

Bob
 
I agree Bob. I've back to back tested both bodies with the same lens, same conditions and same AF array and the 1DX has the edge definitely. But the MkII did a decent job and in isolation produced very acceptable results.

I get plenty of OOF shots from my MkII but 99% are down to either me or the conditions.
 
Your mention of "conditions", Gaz (although I'm sure you didn't mean light levels specifically) prompted me to add this....

One Shot has about a 2EV advantage over AIservo....ie, accuracy of AIservo is similar to One Shot when the light is 2EV dimmer. Further degradation (both types) occurs when moving away from the centre of the AF array.

Bob
 
No Bob, I meant general conditions on the day such as the breeze blowing branches into the focus area or even the bird just looking around are enough to put the focus out.
 
Right, I'm going to throw the cat amongst the pigeons!!
I've spent a lot of time this week with my second 7D2 body.
It's a huge amount better than the first body, but I am still getting AF issues, I've found. I've got the new 400 DO mk2 which is an absolute peach of a lens and performs excellently even with converters on the 1 Dx. When I'm using it on the 7D2, with shots that are up to about 40-50 meters, it's good on the 7D2 without converter and with the 1.4, with the 2x it's poor....and it isn't the lens as it's fine with the Dx.....tried static shots (ie eliminating can't camera shake etc) and I realise it's a tall order with an effective focal length of up to >1000mm...it is terrible.

When I try long distance shots, with all combinations it's not acceptable either....and it isn't the lens.......but as you'd expect, in high contrast situations it's "less terrible"...so it points to an af issue under certain conditions.

I've made some enquiries, and I gather that actually the af system on the 7D2 is quite a bit different to the 1 Dx, so watch this space.

Hmm....
 
..Bob, how are you finding the af when you use the 600 at infinity?...that's where my problems are still occurring....but I still believe this will be a great body and with the 100-400 it seems to be fine!
 
I'm typically at 6-10 metres with the 600/4 (small birds) so I'm not overly fussed about infinity, George, but I'll give it a try over the weekend if it clouds over a bit (low level heat haze saps the contrast as the distance increases).

I did do a little controlled testing a few weeks back and decided that the AF sensors simply seem to need more light than the 1Dx/5D3. An f/4 combination on an average day (~EV11 or more) is okay but once the light degrades or the combination is slower then the AIservo performance drops off rapidly (acquisition and tracking) but one shot still looks good though. I'm not being critical of the 7D2 in general, just my impressions of mine.

Bob
 
Agree totally with the low light comments, Bob, be interested in your longer range tests!!..and thanks!
George
 
George said:
Agree totally with the low light comments, Bob, be interested in your longer range tests!!..and thanks!
George
Click to expand...
We've had a couple of showery days and the sun will start evaporating the moisture in the ground...maybe okay over the weekend but the forecast for Monday onwards is the late 20's and anything at distance is out of the question then.

Bob
 
Seems looking back through images of side on pans it was always at the same distance from camera and same spot on track it lost focus regardless of light level yet 50% of the time it wouldnt have a problem.

Static subjects in one shot or Servo it nails every time but the predictive side of things just seems to be a gamble regardless of colour,speed or direction of movement.
 
Gadget-Guy said:
Static subjects in one shot or Servo it nails every time but the predictive side of things just seems to be a gamble regardless of colour,speed or direction of movement.
Click to expand...

I'm not questioning your ability/technique but wondering what the basis for the above statement is. I looked at your Smugmug gallery with the failed shots and it's clear that a good number have missed critical focus. If you're panning and keeping the focus point on the same part of the subject then there's little or no prediction required (based on the shots in the gallery, not all situations) as the change in distance to the subject is minimal. A bike heading towards you, or very obliquely across you, has a reasonable rate of change of distance but that's not happening with your failures.

Bob
 
Canon Bob said:
I'm not questioning your ability/technique but wondering what the basis for the above statement is. I looked at your Smugmug gallery with the failed shots and it's clear that a good number have missed critical focus. If you're panning and keeping the focus point on the same part of the subject then there's little or no prediction required (based on the shots in the gallery, not all situations) as the change in distance to the subject is minimal. A bike heading towards you, or very obliquely across you, has a reasonable rate of change of distance but that's not happening with your failures.

Bob
Click to expand...


Seeing that im not fully side on for a pan or inside a corner the distance to subject still is changing all be it not as much as head on so would expect the cameras A/F to pick this up and adjust not only for the fact that it will recognize that the camera is being panned but also via its own calculations will register the subject distance has itself changed from one frame to the next.

The size of subject from first image to last image in the wheelie shots changes showing that there must also be a change in distance from camera to subject and this can easily be seen in the images without the use of a distance measurement being given and infact regardles of panning skill you can see the front focus shift through prediction on the floor in all shots including the crash sequence where it shifts onto the curb.

The last set are at a very high shutter speed of 1/2000 sec so as to eliminate comments of motion blur from poor panning technique.

This is a quote from the Canon A/F guide all be it for the 1DX as its the closest system i believe to the 7D MKII but i will try and find one specific to the 7D MKII.

With a predictive AF system the camera is continuously recording the position of the subject and predicting where it will be for the next frame based on its motion so far. If the camera fails to detect the subject position in one recording period, the AI Servo AF III algorithm will ignore the negative result and the next focus point is based on the previous accurate results. Equally, the EOS-1D X will ignore the results when the AF distance appears to jump greatly so that it can continue to track a subject even if an obstacle passes between you and your subject. Equally, if there is suddenly a large jump in the focus distance, the camera will not drive the lens to the new distance directly. Instead it will gradually drive the lens focus, based on the previous successful focus distance results.

The increased sensitivity of the focus system has also allowed for faster predictive focus measurements. In previous EOS cameras there was a warm-up period while the AF system began tracking. This has now been reduced so that the EOS-1D X can begin predictive tracking as soon as a subject begins to move.

The complete article can be found here http://cpn.canon-europe.com/content/education/infobank/autofocus/autofocus.do

After 6 years of covering motorsport i know for a fact its not user error i can assure you.
 
Just had a quick look and cant find anything very specific yet on the 7D MKII A/F over the 1DX in the way of tracking subjects or system operation.

And reading my comment above back it seems a little abrupt Bob and isn't meant to be,sorry.
 
Gadget-Guy said:
After 6 years of covering motorsport i know for a fact its not user error i can assure you.
Click to expand...
I wasn't suggesting for a minute that it was user error and tried to make that clear in my post above. I have the same experiences as you....I have a couple of 1Dx's and a 5D3 that perform consistently and a 7D2 that doesn't.

Gadget-Guy said:
Seeing that im not fully side on for a pan or inside a corner the distance to subject still is changing all be it not as much as head on so would expect the cameras A/F to pick this up and adjust not only for the fact that it will recognize that the camera is being panned but also via its own calculations will register the subject distance has itself changed from one frame to the next.
Click to expand...
It won't recognise that the subject is being panned, simply that the region covered by the focus point in use is closer or further away each time it samples.

Gadget-Guy said:
The size of subject from first image to last image in the wheelie shots changes showing that there must also be a change in distance from camera to subject and this can easily be seen in the images without the use of a distance measurement being given and infact regardles of panning skill you can see the front focus shift through prediction on the floor in all shots including the crash sequence where it shifts onto the curb.
Click to expand...
I appreciate that there is a change in distance but it's so minimal (relative to the overall distance) that the change in focus should be equally small (which it appears not to be).

Bob
 
Canon Bob said:
It won't recognise that the subject is being panned, simply that the region covered by the focus point in use is closer or further away each time it samples.
Click to expand...

Im sure that i saw the new 7D MKII features a built-in 2-axis gyro sensor that can detect camera motion both horizontally and vertically when panning the same as the 1DX incorporates unless a super telephoto lens is in use in which case it uses the giro in the lens but i may be wrong.
 
Gadget-Guy said:
Im sure that i saw the new 7D MKII features a built-in 2-axis gyro sensor that can detect camera motion both horizontally and vertically when panning the same as the 1DX incorporates unless a super telephoto lens is in use in which case it uses the giro in the lens but i may be wrong.
Click to expand...
I stand corrected then, I wasn't aware that it had that feature, sorry.

Bob
 
In all honesty Bob it is or should be a mini 1DX with a crop sensor.

As you well know as being in the same position as myself you can just rely on the 1DX to deliver regardless of situation which at this point in time the 7D MKII will not.

Im sure we both picked one up expecting it to behave in a similar way to the 1DX other than possibly not driving the big primes as effective as the 1DX due to a smaller battery capacity.

The complete thing reminds me of the same problems with the 1D MKIII A/F as i stated before but i hope i am wrong.

For now it will be retired away from being a backup when working and be replaced by a 1DMKIV as i just cant depend on it at this point in time.
 
Mine has less than 600 on the shutter in 6 months. I use three bodies for field sports, the main lenses (200/2 and 300/2.8) on the 1Dx's and the third lens (400/5.6 or 135/2) on the 5D3. I thought that the 7D2 would bring something to the party with it's crop factor that the 5D3 didn't offer.

For perched birds in one shot I've got no complaints, so it's found its niche unless the bright, sunny days give it a new lease of life.

Bob
 
Gadget-Guy said:
Will try and update with some better examples soon but a few are here if anyone is interested http://www.smugmug.com/gallery/n-zsrLZW/
Click to expand...

That looks just like an issue I was having last year with a Tamron 70-200VC on my 1DX, first few frames in a burst were fine, it would then lose focus entirely for 3-4 frames and then grab it again. It wasn't very often, perhaps 4-5 times per race day. I was also having a different focussing issue with that lens and a 70D which made me think it was a software based compatibility issue between that lens and the bodies.

It could very well be some weird software bug that can be fixed by a firmware update.
 
Interesting comments guys. I've pretty much put my MkII in the 'use in bright light' box. I've tried it on various subjects including BIF and rallying and whilst I get OOF images I don't expect every one to be pin sharp. There has to be a limit on what the camera can do in tracking AF.

No idea if the above link to canon rumours has any basis to it.

If my body has AF issues then they are reasonably well hidden but I haven't made a concerted effort to test it. I've checked my lenses and that it focuses consistently in one shot and then used it as and when. I've generally no real complaints. Perhaps I haven't come across the right conditions yet. I bought mine for the reach factor for smaller birds and that's what it gets used for most of the time.

My biggest beef with it is the ISO noise, hence the bright light only use. I've seen very few really good images taken in poor light and high ISO from this camera, certainly not birds where, for me its all about detail. I have used it at 2000-2500 ISO for rallying and got stuff I was pretty pleased with.

I'm still not sure i made the right decision in swapping my 5D3 for it, but I've done it now and will persevere and see what happens with Canon and if they have an official response to the issue and is it affects batches or whether they do a Nikon D750 and recall all bodies.
 
George said:
Agree totally with the low light comments, Bob, be interested in your longer range tests!!..and thanks!
George
Click to expand...
Strange things going on here. I had to choose an elevated target to avoid the heat haze (28 degrees and sunny) and one shot looked good at around 350-400 metres. The same target locked with AIservo was less clear despite allowing sufficient extra time. I noticed something strange with the AF point switching in that it was frequently shifting (albeit very briefly) to an off target AF point so I set up something a little more (but not very) scientific.

I set up a series of coloured targets about 20 metres away (rectangular lids off old Tupperware containers..bright and vibrant in the sun). Red, white, yellow, blue and green. These lids covered an array of AF points 3 tall by 2 wide at this distance with the 600/4. All locked precisely in one shot and the focus was good....AFMA has been done using Reikan Focal.
I then switched to AIservo and selected the 25 AF points of the centre group. Shooting the white target resulted in very minor twitching of the lens focus motor as the active point shifted from one to another (case 2 was selected) but it was always constrained to one of the six points covering the target. Same with the red, same with the yellow. However with the blue and green targets it was very obvious that something different was occurring. The target was picked up (as per the other colours) but the AF motor twitches were far more pronounced. I then saw that AF points not covering the target were being highlighted very briefly and the focus position started to move off target before switching back again...the focus never achieved the distance covered by the new AF point but it simply moved the focus enough to soften the target before quickly regaining it. I repeated this numerous times and each test produced the same result....red, yellow and white targets cause the AF point selection to stay within the target boundary but blue and green targets cause the AF point selection to wander off target (for a few 1/10ths sec at least) long enough to cause the lens motor to "spoil" focus but no where near long enough to acquire the distance that was covered by the 'errant' AF point selected. It was always a point adjacent to the six covering the target and I confirmed this by moving the target around the array. The distant backgound colour was a mid-brown.

I'll have a try with the 1Dx and 5D3 tomorrow (adjusting the target distance to replicate the same AF coverage) and see how they cope.

Is the 7D2 colour blind?

Bob
 
Interesting Canon Bob - what Case are you using for the focus for these tests? I know the switching of points varies in different cases.

I have a 5D3 and 7D2 and am really impressed by the speed of the 7D2 in AI Servo, but I am not convinced the focusing is as accurate as the 5D3, so I will watch with interest.
 
Gaz J said:
Interesting comments guys. I've pretty much put my MkII in the 'use in bright light' box. I've tried it on various subjects including BIF and rallying and whilst I get OOF images I don't expect every one to be pin sharp. There has to be a limit on what the camera can do in tracking AF.

No idea if the above link to canon rumours has any basis to it.

If my body has AF issues then they are reasonably well hidden but I haven't made a concerted effort to test it. I've checked my lenses and that it focuses consistently in one shot and then used it as and when. I've generally no real complaints. Perhaps I haven't come across the right conditions yet. I bought mine for the reach factor for smaller birds and that's what it gets used for most of the time.

My biggest beef with it is the ISO noise, hence the bright light only use. I've seen very few really good images taken in poor light and high ISO from this camera, certainly not birds where, for me its all about detail. I have used it at 2000-2500 ISO for rallying and got stuff I was pretty pleased with.

I'm still not sure i made the right decision in swapping my 5D3 for it, but I've done it now and will persevere and see what happens with Canon and if they have an official response to the issue and is it affects batches or whether they do a Nikon D750 and recall all bodies.
Click to expand...
:agree:
It was very rare for me to get a OOF shot with the 5dmk3, it's all very well saying "when it works look WOW" but that's not the point with a £1.5k camera.
 
Bebop said:
Interesting Canon Bob - what Case are you using for the focus for these tests? I know the switching of points varies in different cases.

I have a 5D3 and 7D2 and am really impressed by the speed of the 7D2 in AI Servo, but I am not convinced the focusing is as accurate as the 5D3, so I will watch with interest.
Click to expand...
Case 2 (as mentioned in the text).

To be clear, the accuracy seems fine, it's the tracking (or trying to track when not required ) that I'm finding inconsistencies with.

Bob
 
Last edited:
Canon Bob said:
Strange things going on here. I had to choose an elevated target to avoid the heat haze (28 degrees and sunny) and one shot looked good at around 350-400 metres. The same target locked with AIservo was less clear despite allowing sufficient extra time. I noticed something strange with the AF point switching in that it was frequently shifting (albeit very briefly) to an off target AF point so I set up something a little more (but not very) scientific.

I set up a series of coloured targets about 20 metres away (rectangular lids off old Tupperware containers..bright and vibrant in the sun). Red, white, yellow, blue and green. These lids covered an array of AF points 3 tall by 2 wide at this distance with the 600/4. All locked precisely in one shot and the focus was good....AFMA has been done using Reikan Focal.
I then switched to AIservo and selected the 25 AF points of the centre group. Shooting the white target resulted in very minor twitching of the lens focus motor as the active point shifted from one to another (case 2 was selected) but it was always constrained to one of the six points covering the target. Same with the red, same with the yellow. However with the blue and green targets it was very obvious that something different was occurring. The target was picked up (as per the other colours) but the AF motor twitches were far more pronounced. I then saw that AF points not covering the target were being highlighted very briefly and the focus position started to move off target before switching back again...the focus never achieved the distance covered by the new AF point but it simply moved the focus enough to soften the target before quickly regaining it. I repeated this numerous times and each test produced the same result....red, yellow and white targets cause the AF point selection to stay within the target boundary but blue and green targets cause the AF point selection to wander off target (for a few 1/10ths sec at least) long enough to cause the lens motor to "spoil" focus but no where near long enough to acquire the distance that was covered by the 'errant' AF point selected. It was always a point adjacent to the six covering the target and I confirmed this by moving the target around the array. The distant backgound colour was a mid-brown.

I'll have a try with the 1Dx and 5D3 tomorrow (adjusting the target distance to replicate the same AF coverage) and see how they cope.

Is the 7D2 colour blind?

Bob
Click to expand...

Dont know if it may be worth a try with the EOS ITR turned off Bob to see if it may be that part of the system playing games.

My one was mainly on single A/F point when it missed or didnt gain focus correctly so reading this article points out that it was disabled in my failures by the camera itself.

EOS 7D Mark II: iTR AF and focus point selection
The EOS iTR AF system in the EOS 7D Mark II allows the camera to track moving subjects more accurately by using data from the metering sensor to track both faces and colours. With the EOS iTR AF setting enabled, you may find the camera is not picking up faces or colours to track. This is most likely due to your AF point selection method. To ensure the EOS iTR AF functions, the camera needs to be set to Zone AF, Large Zone AF or 65-point auto selection. In any of the other focus point selection modes, the EOS iTR AF will not function regardless of the menu setting.
 
On another note i have now set my first image and second image so they bias focus lock more than release in the A/F menu to see if this may help as noticed the 1DX has a wider choice of setting to decide if you would prefer image focus over release which i have always had set more for a focus bias on both first and second images on my 1DX for as long as i can remember.
 
Gadget-Guy said:
Dont know if it may be worth a try with the EOS ITR turned off Bob to see if it may be that part of the system playing games.
Click to expand...
I'd actually switched it on figuring that it would be more likely to maintain the AF point over the colour first acquired....good point though, I'll try both options.

Gadget-Guy said:
My one was mainly on single A/F point when it missed or didn't gain focus correctly so reading this article points out that it was disabled in my failures by the camera itself.
Click to expand...
Yes, very different symptoms and setup. You're experience shows that it doesn't track correctly when there's no change of AF point whereas my test shows that the AF point switching is not performing correctly.

Bob
 
You must log in or register to reply here.
Back
Top