Canon lenses calibration

Started 5 months ago | Discussions thread
Zeee Forum Pro • Posts: 21,950
Re: Canon lenses calibration

CESA wrote:

Zeee wrote:

CESA wrote:

Zeee wrote:

J A C S wrote:

Why are you doing this? Do you see AF problems?

One of the mistakes people make is to do the calibration with the target too close. This would create more problems than it would solve - been there, done that.

This is very true. That is why I called that PDF link the most comprehensive document. I had lots of links to various Canon information. Some said 50X, others said at location, etc. Even the manuals say best to MFA at location. Then add all the home made versions, etc. There were just too many options which was one reason I never liked MFA. FoCal which follows the sensor parallel to target method made my life easier. I followed it's minimal distance recommendations.

Just use the lenses you normally would and inspect the shots carefully for AF errors. Do MA only if you see consistent front or back focus.

Hi both.

I have purchased the FoCal software to perform the AFMA to my lenses specially the 85mm f1.2.

I have done several tests including at different distances or in this case at 2.7m and 3.1m.

The AFMA returned AFMA value of +5. Nothing like the +17 that I was talking about. But there is a catch.

After performing this calibration, I went out to take some close shots or in this full body and some half chest shoots (what's that, 1m and 1.5m or so) and pictures were not perfectly sharp.

Changed the AFMA to +17 and bang on at those distances (half-chest and full body).

Followed everything they recommend with exception of using their target, the ones they sell which I am still waiting for delivery. to repeat the tests. I have printed the target at the local supermarket so it is what it is and don't know if it is suitable. But have repeated the test several times and it was more or less consistent in the sense that the AFMA that returned were pretty close +5 and one time said between +4 and +6 but then ended up in +5 after continuing the test.

Actually, before performing the tests I went out and had the AFMA set to +5 and pictures taken at say 2m or 2.5m were not 100% in focus/sharp f1.2 and f2.

Now, next time I am gonna take a walk will change the AFMA to +17 and will do some tests.

Will also perform the test at 1.5m and see if the camera will show the same AFMA that I got by doing real life tests.

That being said I question the efficacy of this software.

Well known Dustin Abbott has mentioned in one of his videos showing how to calibrate lenses that some of the setting FoCal came out after the test he didn't agree.

The testing conditions, lighting specially probably influence a lot the outcome as well.

I am wondering at which extend this software usefull or not. But had to get it to see by myself. Dustin teaches how to perform the test manually using their targets and the LIVE VIEW shot as starting point.

Additionally did the best aperture and also the AF consistency and they were not consistent as well. That is the best aperture for the sharpest aperture, it gave me different values test at different distances from the target (see above) and the AF consistency gave me slight differences but I was expecting to be the same.

Any comments?

Not sure what to say about that part. It has always been bang on for me. I took me 3 days to settle on +13 for my 7D2, 100-400 II and 1.4 @ 560mm. FoCal gave me +13 in 5 minutes. It took me 3 days because I always second guess myself. Maybe they can do a refund I you don't like the product.

On the other hand +17 is a lot IMO. I would not except that. I purchased a 70-200 2.8 II and it needed about +15 at 200mm without a TC on two bodies. I purchased FoCal specifically to verify it. Two Canon techs at New Jersey service told me that with high numbers like that use MFA to get by but it should be serviced. They found a mis-calibrated board.

Back to the 7D2, 100-400 II and 1.4. Without the TC it only needed +3. The TC on primes needed very little as well so it wasn't the TC and the lens was fine. I could only conclude that since the 100-400 II has many moving parts it could be in spec but some components are on one end or the other of said spec. I wasn't sure what Canon could so I decided to leave that particular lens TC combo alone. That was the only exception I have made to date. I know of one more person that had those same results.

Thanks for your valuable comment.

The 85mm was bought from an used shop - don't know if I have referred that in this thread. It has now passed the return date - however it has got a 1 year warranty on it.

Despite this, I called canon and asked to speak to a technician. Cutting the long story short he said to not worry about it. If it required +17 it is fine and that is way the AFMA exists. Another question that I did was that how likely would the existing AFMA range in the cameras be enough after the lens gets used more and more and he just said that this wouldn't happen.

That is not what they told me and I called them twice about it.

Marco, another user in another thread I think he said that sending the lens to canon to get serviced might do more harm than good. If it would be a problem with the board calibration and not related to moving parts or alignment probably that would not be a problem.

I have found Canon service quite good but that is for you to choose. I have my own views on this  which are basically I didn't spend $2,000 + to get a mis-calibrated lens.

In your case, the examples you are giving is related to telephoto lens, like big telephoto lens that in principle are used to shoot very far objects where the DoF is big anyway. So one would expect focal to work properly.

This is true. Perhaps the need for +17 and the shallow DOF may be a factor. I would contact FoCal. I have read they request different test results to analyze them, if you want to go through that.

But in my case what I am doing is calibrating a lens at a given distance that ok, I might use it, but will be likely to be used at smaller distances that the calibrated one.

I will repeat the test again when the targets arrive at the recommended distance of 3.5m and at 1.5m and see the outcome. If the AFMA setting comes out to be around the same +17 value then I would not be suspicious of the software but if it comes out different I will have to rethink if this software is somewhat useful with exception of telephoto lens.

I don't want to give you incorrect information and get to you spend your money on something you aren't happy with. When it first came out it was a little inconsistent but Version 2 (last time I checked) they put a lot of effort to improve accuracy. There are a lot of good reviews about it.  I have not come across too many negative ones.

Ideally it would be nice to test the lens on another body and see if it requires the same amount. Just out of curiosity has this happened to you? With other lens besides the one on the example you gave above? Is this even possible? If it is possible it means that the 'problem' is in the body?

No. My camera bodies and lenses have been pretty consistent only requiring a small adjustment. Fine tuning which is what I believe MFA is for. With the TC and that lens  both my 7D2 and 5D4 needed big moves. I would try it on another body if you can just to rule things out.

How likely would they accept to take in the lens to verify JUST if it is some mis-calibrated board? Provided every lens has got one?

I don't know if it has one or not. I don't know much about lenses.

Thanks in advance.

No problem. If I don't reply for a while we are heading out for supper.

-- hide signature --

I feel even more confident that soon things will have a wonderful conclusion

 Zeee's gear list:Zeee's gear list
Canon EOS R Canon EOS R5 Canon RF 24-105mm F4L IS USM Canon RF 100-500mm F4.5-7.1L IS USM +1 more
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow