5D-3 AF microadjustment ?

[ST-EOS]

I hope someone can help me out here, before I send my camera and lens to Canon Elstree for a Pro-AF service.

I have posted some images here and received feedback that some images appeared soft. That started me to examine other images of mine to see if I could determine what I was doing or the camera was doing that caused the softness. I'm not convinced that it's the equipment Im using which is the problem but more likely that it's the operator.

When I check some images in DPP and view the AF Points. The area under the focus point does look soft but...other parts of the image are in focus. My thoughts are that at the time of focus that part of the image would be sharp but when the image was made the subject had moved (subjects were bees, flowers etc) this isn't always the case in fact in some instances the next image of the same subject is sharper but not tack sharp. This is why I'm considering the Canon Service option.

Anyway I decided to try and check the autofocus on the 5D-3 with a 100mml-IS lens, which should be a good combination and that combination was what I had used to capture the images which were being seen as soft.

To check the auto focus I set the camera on a tripod with a remote release the test target was on a level surface in front of the camera with the camera at approx 45degrees to the target, live view enabled. The target filled the frame. I used manual mode, zoomed live view to 10 focused and made the image. For this exercise I used a number of micro adjustments (-10, +10, +5, -5 and zero) making notes of the micro adjustment settings for each of the series of images. When finished I reviewed the set of images and decided that IMO the +5 seemed to be the most accurate. So that was the setting set the lens to and registered the lens to the body.

This morning I decided to look again at the test images, I'm from Yorkshire remember so even when only contemplating paying Canon to do the service my wallet was having funny turns

However this morning I also thought I was having funny turns! as when I was reviewing the test images using DPP. The image information for the series of test shots didn't show any lens micro adjustments!

I have just experimented with a couple of quick images using both Live View and normal shooting. Well the experiment proved I wasn't having a senior moment. The Live View images don't show any AF adjustment settings. The non live view images do show the +5 setting.

So my first question is this the norm with images taken using Live View and DPP i.e. DPP doesn't show the lens adjustment setting?.

My second question, given that I not convinced that part soft images are my fault all of the time, has anybody used Elstree or other Canon Repair Centres for the Pro-AF Service which is available to CPS members.

Sorry for the rather rambling post I hope it all makes sense.

[Muscat]

Re: 5D-3 AF microadjustment ?

I had a problem when I purchased my 400mm lens in as much as I couldn't get it to focus correctly with my 5D3 so I took the camera and lens to Eltree where they made some adjustments to I believe both the camera and lens. I must say that when I got them back all my other lenses where sharper but I still needed to go through a micro-adjust sequence with the 400mm which resulted in a +12 setting being added. Later in the year I fell and dropped my 5D3 with the 400mm attached so had to send the camera back to Eltree again where they made some sensor and AF adjustments, the camera having had it back is performing better than when I got it new, it has less noise at higher ISO and the images are much sharper than previously. Yes I rate the service from Elstree.

as to why you are not seeing AF adjustment data in some images I have no idea as to be honest I haven't looked, my guess is that if you are using live view the camera would expect that you will be manually focussing which would nullify the need for the Auto-focus system and hence the need to use any micro-adjustment associated with auto-focus.

[ST-EOS]

Re: 5D-3 AF microadjustment ?

Thanks for your reply Ian, much appreciated. Like you I have had trauma with my camera. At the beginning of May whilst on a trip to Derbyshire my tripod toppled over with the 5D3 attached. At the time I thought I had got off very lightly for my stupidity, with just a scrape on the camera body and a smashed Lee ND grad. However having read your comments I suspect that my focus problems may all stem back to that Derbyshire pre dawn tumble.
I'm nearly convinced that I should just bite the bullet and send the 5D3 plus the 100mm lens off to Elstree for the Pro-AF service. At least then I should have confidence that the equipment is in spec, just the operator to blame for soft images. My camera is still in warranty but the lens isn't. So hopefully if the camera needs parts the warranty should cover it.
Thanks once again Ian. I'm off to package up the kit now

[ST-EOS]

Re: 5D-3 AF microadjustment ?

Well the camera and lens are boxed up, all paperwork completed. I will give Canon Elstree a call in the morning just to confirm their requirements and ask about the lack of visibility in DPP of the AF micro adjustment detail when using Live View.

Then its off to the Post Office with it. I will let you know how I get on when the camera and lens are returned. Looks like I may have to bring the 50D out of hibernation

[DrJon]

Re: 5D-3 AF microadjustment ?

Yes, I'd expect the Live View images not to use AF micro-adjust. The adjustment is allowing for the difference between what the Phase-detect sensor sees and the image on the main sensor, which are physically separate things. With Live View the focusing is usually via contrast detect using the main imaging sensor, so you don't want an adjustment as it's picking the sharpest setting for you actually at the image sensor (rather than the separate phase-detect sensor).

If you use phase-detect focus for live view, which is an option on my 5DmkII and presumably many/most/all others (where it does the mirror flip when you hit the AF button) then I'd expect it to use the micro-adjust value.

[ST-EOS]

Re: 5D-3 AF microadjustment ?

Thanks for taking time to comment John I appreciate your knowledge and experience in all of this technology, some may say smoke and mirrors.

I spoke with the Canon service centre this morning and the camera and macro lens are now in transit. I hope they find something wrong with either the 5D-3 or the lens, preferably the camera as that's still in warranty. I didn't ask about the Live View/Micro adjust issue as I was more concerned about getting to the root cause of the soft images I and others were seeing.

Thanks once again John

[DrJon]

Re: 5D-3 AF microadjustment ?

Good luck!

[ST-EOS]

Re: 5D-3 AF microadjustment ?

Well the camera and lens were returned here this afternoon from Canon Elstree.

So I thought I would post one of my test images here to see what you thought of it from a focus perspective. The camera has AF microadjustment set at zero.

_V1A8956-web-Edit.jpg

[pelliott1954]

Re: 5D-3 AF microadjustment ?

Hi Peter,

Not sharp at all, not that I know anything about Macro lenses, but have you got the camera on a Tripod, using remote release and Mirror lock up, if you do a test shot like that, then you will know there are no other factors affecting the focusing such as camera shake at too low a speed.

What speed and Aperture did you use for this test shot?

Regards Paul

[DrJon]

Re: 5D-3 AF microadjustment ?

What should the depth of field be at the distance/aperture/etc and where was it focused?

[ST-EOS]

Re: 5D-3 AF microadjustment ?

Hi Peter,

What speed and Aperture did you use for this test shot?

Regards Paul

What should the depth of field be at the distance/aperture/etc and where was it focused?

Paul ISO 100, 100mm FL, f2.8, 1/500sec

I have uploaded the same image this time using Photobucket and to cover DrJon's question included one from DPP showing the AF Point used.

original but upload via PB




Same image but showing AF Point, (NB. adifferent crop and no PP on this version)

[MX5]

Re: 5D-3 AF microadjustment ?

Are you wondering why the image isn't sharp (the whole creature)? With a 5D3 and 100mm;

At f/11 and distance of 1' ... DoF is 0.01', but at f/2.8 it'll be less than a bee's whisker. So you can see even the slightest movement has a dramatic effect. [I#ve been playing with an extension tube recently and getting decent pictures is quite hard ... and I have admiration for those that persevere (and have the skill) and get 'pin sharp' images].

[ST-EOS]

Re: 5D-3 AF microadjustment ?

Are you wondering why the image isn't sharp (the whole creature)? With a 5D3 and 100mm;

At f/11 and distance of 1' ... DoF is 0.01', but at f/2.8 it'll be less than a bee's whisker. So you can see even the slightest movement has a dramatic effect. [I#ve been playing with an extension tube recently and getting decent pictures is quite hard ... and I have admiration for those that persevere (and have the skill) and get 'pin sharp' images].

Thanks for commenting MX5, to be honest I think I've been barking up a number of wrong trees WRT the sharpness of images with this body and lens combination as previously detailed in this thread.

The issue started when some images that I had posted online were deemed to be soft (tree 1)
Then I started to wonder if the issue was with the equipment or user (tree 2)
I tried a number of lens AF tests using the AF microadjustments, using various test targets (tree 3)
Then I remembered that DPP could show the AF point used, I don't use DPP to any great extent but...from this it seemed that the image at AF point used seemed softer than other areas of that image (tree 4)
I did have abit of a trauma when my tripod when A over T and thought that may have had a bearing on the issue (tree 5)

I could go on documenting the issues I "thought" I had (I'm sure you will remember what thought, thought ) but having had the lens and body in for a Pro AF service with Canon Elstree I have to say that my conclusion is that :-

(i) the operator could have been the main problem i.e. barking up the wrong tree.
(ii) the softness noted may not have been caused by the equipment but...maybe the Post Processing of the images plus the method used to post the images on a forum.

Thanks once again and finally I have to agree with your thoughts on the impact that movement can have on image sharpness and I share your admiration of those who can actually achieve images that are "pin sharp".
There are people on this and other forums who do achieve pin sharp images. I shall persevere with this macro lens and hopefully get images that I'm happy with.

[MX5]

Re: 5D-3 AF microadjustment ?

There are so many things that contribute to a sharp image. From the operator, the kit, how the kit is deployed )tripod, remote release, mirror lockup, ...), how still the object is (and the air), the exposure (all three variables, aperture, shutter speed and ISO), the post processing ... even with all these optimal I've seen hosting sites (or indirections to hosting sites) screw everything up. I tend to mitigate against each ... but alas the operator is rumoured to human.

(I was recently fortunate to get into moth, butterfly, newt, lizard, frog and natterjack toad country with a licensed to pick up listed species ranger and a pro nature photographer ... sort of inspired me again into macro ...)