Why you should calibrate Canon EF Lenses with Reikan FoCal

Since the introduction of the Canon EOS 1D Mark III way back in 2007, middle to high-end camera owners have been able to adjust the focus system of their camera to work better with each lens.

Reikan FoCal - Why you should calibrate Canon EF Lenses

 

Reikan FoCal takes the guesswork out of this adjustment by automating the process and using advanced computer analysis to determine the best adjustment for your camera and lens. Over the past 5 years, FoCal users have been uploading their results to our database, and we process this information and feed it back to let them compare how their camera and lens compares with everyone else.

In this post, I wanted to take a look in more detail about the real world need for AF Microadjustment on Canon cameras and specifically Canon EF lenses. In a previous post, we looked at the the results of adjustment across all cameras and lenses (Canon and Nikon) supported by FoCal and saw that on average more than 60% of all cameras and lenses would significantly benefit from calibration.

Spoiler Alert!

This might not come as a massive surprise, but we think it’s a really good idea to calibrate every single lens you own!

But don’t just take our word for it.  I’ve shown below exactly why it’s a great idea to adjust your lenses with your cameras.

Canon Lenses

There have been more than 100 different Canon EF lenses produced since 1987, and our database contains high quality information for nearly 90 of these lenses across the 16 different Canon camera bodies supported by FoCal.

For this post, I’m going to use the results to show how likely it is that any particular lens will benefit from calibration. I want to try and answer the question:

“When I take that shiny new lens out of the box, what are the chances that I’m going to get the best from it without any adjustment?”

How likely is is that a lens needs no adjustment at all?

We’ve taken all the data and worked out the percentage of each lens type that required no calibration at all (i.e. the AF Microadjustment result from FoCal was 0). Here’s that data shown for all the Canon EF lenses (ordered from most-likely to least-likely to need no calibration):

Reikan FoCal and Canon EF Lenses - percentage requiring no calibration
image-16765

(Note: click on the chart to see a large view)

Let’s start at the very left of the chart. The EF 90-300mm lens has a value of just under 25%… so what does that mean? It means that – on average – if you had 4 of these lenses on the bench in front of you, 3 out of 4 would produce sharper photos if you ran them through FoCal. Only 1 of the 4 (25%) would not need any calibration. And that’s for the best on the list!

Move a third of the way along the chart (to the 60mm Macro), and you’re down to less than 10%, i.e. more than 9 out of 10 lenses would benefit from calibration!

And at the far right, the last 2 lens types( the 85/1.2L and 300/2.8L) could do with calibration for every single copy of this lens!

Relax A Little

To be honest, we’ve gone for a pretty stringent requirement above of needing a calibration result of 0. In reality, you’ll usually not notice much of a difference with an adjustment of 2 AF Microadjustment points, so let’s have a look at what happens if you include all the results that need just a tiny amount of calibration:

Reikan FoCal - Canon EF Lenses requiring very little calibration
image-16766

Things are looking a bit better here…

Again, starting at the far left, this time it’s the EF 11-24 f/4L that’s best, and our number has jumped to almost 75%. So now only 1 in 4 lenses will see a significant improvement (remember, the other 3 will probably still see an improvement, just not huge).

But… (there’s always a but!)… only the top 15 lenses have values above 50%. That means that even with this more relaxed requirement, nearly 80% of the lenses are more likely to be significantly better after running through FoCal.

For the half of the lenses on the right side of the chart, you’ve got a 2 in 3 chance that you’ll see a decent improvement using FoCal.

What does this adjustment look like?

I keep mentioning tiny, small and large calibration differences, but what do these actually look like?  The images below show the difference between a well calibrated lens, and an offset of 2 points (barely any noticeable difference), a 10 point error (significantly blurred), and a 20 point error (hugely blurred):

Reikan FoCal - Calibration Example
image-16767

The yellow chart above (Percentage requiring very little calibration) shows the percentage of lenses that fall somewhere between Calibrated and 2 point error.  The remaining lenses – that’s more than 1 in 2 lenses for 80% of the different lens types – are somewhere between 2 point error and 20 point error!

The Scary End!

Finally, let’s take a look at the scary end of the figures – the lenses which needed a huge calibration (more than 10 AF Microadjustment units):

Reikan FoCal and Canon EF Lenses - percentage requiring huge AF Microadjustment calibration
image-16768

This time, the left side of the chart is a bad place to be. If you had 10 copies of the EF 28mm f/1.8 lens on your desk, 4 out of 10 of them would likely get a massive improvement after calibration with FoCal!

Thankfully that number drops quickly, but for around a third of all tested EF lenses you’ve still got a greater than 1 in 10 chance that you’ll need a really big adjustment to get the best from the lens.  And to put that in context, all of those lenses will be producing results between the 10 point error and 20 point error images above!

Conclusion

There’s a pretty simple conclusion here (and it’s great for us at Reikan!): you really should run every lens you have through FoCal!

For most lenses there’s a high chance you’ll get a noticeable improvement in image quality after calibration.

FoCal takes all the guesswork out of adjusting your lenses, so for the sake of a few minutes with each lens it really is worth it to get the best from your kit.

Take a more detailed look at FoCal on the website.

What’s next?

In the next post, we’ll look at Nikon lenses so we can see how things compare.

Fire away in the comments section with any questions you might have about all this data, or any suggestions for other things you’d like to see.

22 comments on “Why you should calibrate Canon EF Lenses with Reikan FoCal

  • Very interesting finding.just shows most first party lenses are not that much different than third party lenses like Sigma and Tamron.There is a general perception that third party lenses need calibration more than the first party lenses like Canon.Though these tests clearly show they are not that accurate as the general consensus is.Alot of people send their lenses back to the manufacturer thinking they have a faulty lens when they get lackluster performance though it is camera body dependant too.just calibrating them TOGETHER can solve alot problems and headaches

    Reply
  • Alternative reading: A vast majority of lenses do not need adjustment. Most people only buy Reikan when they believe they have a problem which means that you are reading from a self-selecting population. You are also ignoring the impact of DoF and perception of sharpness.
    Don’t get me wrong – your product has a use (I have bought one ‘just in case’) but using these charts as evidence of lens error seems over the top marketing.

    Reply
    • Hi Mike,

      I take your point the group is self selecting (we can’t know about camera/lenses not run through FoCal) and there may be some bias, as you say at least some users will look to calibrate when they have personally seen issues. I will say we have a good number of both Canon and Nikon FoCal users and between those two groups the picture with Nikon is significantly poorer overall (see also Why you should AF Fine Tune Nikkor Lenses)

      In terms of what we see within the results above for Nikon, the best lens (Nikkor 18-35mm f/3.5-4.5) – 50% of them require more than +2/-2 AF Fine Tune adjustment – that’s really high. With Canon, the best lens (Canon EF 11-24 f/4L) – 25% of them require more than +2/-2 AF AF microadjustment. So at the best end and actually across the lenses calibrated with FoCal there’s a marked difference between Canon and Nikon (if there were to be some bias it would tend to apply equally across Canon/Nikon users).

      Best Regards,
      Dave

      Reply
    • Actually, I think your only part right.
      I had poor results with one of my old lens when I switched to a new body. I purchased this software to check that lens.
      From that point on, I have ran every lens through the software regardless of my perception of its performance beforehand. I will continue to do this for every new lens I get or rent.
      So you are right that an issue caused me to buy the software, but now they are getting data from lens that I did not think were a problem previously. Some of these do fall into the -2 to +2 range, some did not.

      Reply
  • I calibrated my Tamron 150-600mm G2.I got – 4 at 150mm, +1 at 200mm, – 1 at 300mm, – 1 at 400mm.Reican what difference will it make if I add a – 1 difference into my camera body? Will + 1 enc figures go to +2 enc? What difference will it make if I add a negative figure?Will my positive numbers increase or decrease? My body can unfortunately only take one calibration figure.what number would u recommend Reican?

    Reply
    • Hi Neil,

      Let’s take a simple (made up!) example might explain better, 24-70mm lens with a -5 on the 24mm end and +5 on the 70mm end. If we do nothing (leave/set the camera calibration to 0) then we stick with those results. If we decide the 70mm end is more important (less depth of field at 70mm for similar framing and we might tend to stop down a bit on the wide end). We go ahead and set the camera to +3 as a happy balance. What you now have is a 24mm end that’s effectively -8 ‘out’ and the 70mm end that’s effectively +2 ‘out’. It’s worth checking the How do you calibrate zoom lenses?

      I think given your case it really comes down at least partly to how you typically use the lens. If we ignore to start +1 at 200mm really the overall trend is toward a ‘-‘ / minus adjustment, I would tend to suggest perhaps -2, the overall range of results -4 to +1 is not huge. You will see more consistently well focused shots with a slight minus adjustment but a couple of AF fine tune units is close to the general consistency of what the camera/lens can provide so it’s worthwhile but not going to be massively obvious unless pixel peeping across a large number of images.

      Hope that helps.

      Best Regards,
      Dave

      Reply
  • I’m trying to figure out how your program would work with a Sigma ART series lens, which calibrates focus in 4 different distances using their usb dock. Can you direct me through a detailed video of how this would look? As close as I can gather, I would have to take the lens off of the camera and attach to usb dock each time FoCal asks me to make a microadjustment….. how many times would this typically happen at each of the 4 distances?

    Very interested in your product, but am not sure I’m biting off more than I can handle with the usb dock twist. Thanks!

    Reply
    • Hi Perry,

      Thank you for your interest in FoCal. The Sigma Dock calibration definitely is more involved than a normal AF fine tune / micro adjustment calibration but it is possible. Will include some thoughts below 🙂

      FoCal can be used with the Sigma Dock. Easiest way to think of it is this: the normal camera/lens calibration is a fine tune setting held in the camera and FoCal can be used to work out the best value for this setting. With the Sigma Dock you now have a lot more points of adjustment and they must be set inside the lens rather than inside the camera.

      To get an idea of what is involved I’ve linked to some posts where users describe calibrating using the Sigma Dock with the help of FoCal, http://www.dpreview.com/forums/thread/3809354 and the related thread linked by tektrader where they talk in more detail about the steps. More threads which talk about dock calibration using FoCal are https://www.dpreview.com/forums/thread/4031928 and https://www.dpreview.com/forums/thread/4052613 and another more recent thread at https://www.dpreview.com/forums/thread/4083573

      We don’t have an official written down procedure or tutorial video (yet!) but from what we understand FoCal can be useful for Sigma lens calibration. We’ll need to do our own tests and more research using multiple different Sigma lenses before we’ll have an official procedure that we guarantee will work in all cases.

      Different Sigma lenses require difference calibration distances and are given in the Sigma Dock software. The distance, especially for the close in distances should be fairly accurately measured (measure to the sensor plane usually indicated on the top of the camera).

      FoCal generally expects the target to take up between 20%-90% of the vertical frame. With nearer distances you might find a standard printed target works fine for this, if the target is taking up most or all of the screen you can re-print the target smaller as needed to ensure that it still fits inside the 90% with the closest distance.

      With further distances and indeed with the ‘infinity’ setting it’s unlikely you can print a FoCal target large enough for this calibration. The “infinity question” as I’m calling it is an interesting one(!)

      I think the best way round is to find a natural or environmental target and use that instead. So turn off the function where FoCal expects a target (“Preferences” > “Tests” you can tell FoCal not to look for the specific target design at all by setting “Target Validation” to “No Target Validation”).

      A ‘natural target’ would be anything with high contrast edges in both the vertical and horizontal axis. Perhaps a large road sign with text on it, or a high rise apartment block that has strong defined lines (say window frames).

      This is the approach we’ve come up with as a suggestion for now, one of the things we want to do is provide more information to users and this will be happening soon.

      Essentially the expected process would be the calibrate (using FoCal) the lens at a specific distance (as given by Sigma in their user interface) and then try to dial out that adjustment. The end goal is the camera set AFMA ends up at 0 when the camera/lens is calibrated using FoCal and all the adjustment is entered instead within the Sigma Dock settings and held inside the lens itself.

      It’s hopeful (and there is some evidence to suggest) that once a relationship is found between AFMA / camera fine tune value and Sigma Dock calibration units that relationship holds true for further adjustments on that same lens.

      Best Regards,
      Dave

      PS, there are some videos, not directly related to Sigma Dock but will give you an idea what’s involved at Are there any FoCal calibration videos I can watch?

      Reply
  • As we know the view finder focusing system is different with live view focusing system…
    my canon new 80D has micro-focus adjusting feature, after doing lot of best suggested manual testing/fine tunning with following every settings, recommended distance and patterns, it seems OK only for that focal distance but in real world its fail…
    does this reiken focal system solves this problem or the body it self has problem??
    can i calibirate and registered many lenses with reiken FoCal plus in my 80D with out additional licenses??

    Reply
    • Hi Zakir,

      Thank you for your interest in FoCal.

      The situation you mention is right in the ballpark for what FoCal is designed to do. Aside from the calibration FoCal will provide a lot of information about what might be happening with the camera AF.

      Worth checking out some videos to see what’s involved, FoCal Auto Focus Calibration Videos

      FoCal Plus works really well for calibrations, the limitations with “Plus” is a limit of 400mm on the focal length (ignoring crop factor) and also some functions available in “Pro” such as PDF reports are missing – see FAQ entry Which version of Reikan FoCal?. There’s no limit as such within FoCal on the number of lenses which can be calibrated, the camera itself tends to have a limit but it’s fairly high (40 different lenses on the 80D).

      Best Regards,
      Dave

      Reply
      • Hello Dave,
        thanks for your prompt response, please find the below my lenses for 80D and the work around.

        Sigma 10 – 20MM (i tried shooting lot of photos with whole range of micro adjustment setting that is from – 20 to +20 one by one every day nothing works for me in real world)

        Tamron 70 – 300MM (i tried shooting lot of photos with whole range of micro adjustment setting that is from – 20 to +20 one by one every day but little improvement not perfect)

        canon 50MM (no issue but i think FoCal might improve more)
        canon 18-55MM kit lens (no issue but i think FoCal might improve more)

        here i notice my cannon lenses does not require calibration or very little, the problem is only with my third party lenses.. i purchase third party lenses after lot of research..

        already i tried all range of adjustment but fail so how Reikan Focal is going to solve with in the same range of adjustment?

        is FoCal making any changes in lenses and camera firmware? or only adjusting the range of micro AF by performing the tests?

        Reply
        • Hi Zakir,

          FoCal will show you a lot of what’s happening with auto focus. If you have a lens which doesn’t seem right or strange things are happening FoCal will often show this up or make it easy to spot during a calibration.

          What FoCal can’t do is adjustment outside of the camera range. The adjustments possible are limited to the -20/+20 available in the camera menu, there’s nothing extra FoCal can change beyond those settings.

          Best Regards,
          Dave

          Reply
  • Is it not true to say that it is the lens + camera body together that need calibrating? To say it is just the lens could be interpreted as being somewhat misleading. A camera body requiring a -5 adjustment would be cancelled out when attached to a lens that needs a +5 adjustment, giving an overall adjustment of zero.

    The same lens on a different camera body may give a completely different adjustment value. If this “body” adjustment could somehow be incorporated into your statistics above, a slightly different bar chart distribution may result. Or maybe not 😉.

    Another question, if you have a reliable data set for micro-adjustment values from several lenses acquired on one camera body, could they be transferred to another after compensating for this different “body” micro-adjustment? It would save a lot of work if that were the case. You would of course, still need to find this value initially, perhaps by using the most consistent of those lenses on each body independently i.e.

    Camera 1 + lens 1 gives a value of +11

    Camera 2 + lens 1 gives a value of +3 so if

    Camera 1 + lens 2 gives a value of -2 then

    Camera 2 + lens 2 should be equal to -10 (the body value is -8 in this instance)

    Having said that, my problem is that the light levels are too low in my house at this time of year, so I need to use a laptop & the software outside in daylight.

    That brings it’s own problems of high winds, rain & variable light, not to mention flickering shadows when the sun is behind a tree!! It’s so frustrating when only about 1 day in 7 has the right weather conditions 😰.

    Nonetheless, I’m very impressed with the quality of results & look forward to a High Sierra version soon.

    With regards,

    Rob

    Reply
    • Hi Rob,

      Thanks for your feedback and thoughts!

      Yes, it’s as you say, the calibration is not specifically a lens calibration. Rather it’s an adjustment that accounts for tolerance of both lens and camera.

      What we see in practice is some lenses do (for whatever reason) require more adjustment, with other lenses typically requiring far less adjustment. That’s across multiple cameras and camera models. We could speculate on what might be the cause of this but it’s hard to prove (just as a random theory it could be related to the way the lens is driven and it’s internal design rather than solely related to the mounting offset).

      “if you have a reliable data set for micro-adjustment values from several lenses acquired on one camera body, could they be transferred to another after compensating for this different body micro-adjustment”

      In theory it should work as you say, from a process point of view if you have a lot of lenses and cameras it might pay to make up a table and understand the relationship. Designing a software solution that guides the user through the process that way might cause more confusion than doing each individually (offsetting any potential time saving?!) 🙂

      Glad you’re enjoying using FoCal and thank you again for taking time to share your thoughts 🙂

      Best Regards,
      Dave

      Reply
    • Hi Paul,

      Thanks for your interest in Reikan FoCal.

      Each lens has an internal identifier which the camera is designed to recognise. Whenever the camera sees that particular lens model it will automatically apply the saved adjustments specific to that lens.

      Best Regards,
      Dave

      Reply

Leave a Reply

Your email address will not be published. Required fields are marked *