Fine tuning a 70-200 f2.8L IS II USM lens

[hombrevg]

All, 

I am trying to dial in my 70-200 f2.8L IS II USM lens that I bought. I am not sure I am doing the right steps to get it razor sharp and may have some high expectations of lens/glass. 

Here is my current workflow and details. Mark 5D III body.

I put the camera in P mode, I set to one shot and set spot focus on a point and do not move the camera/lens. I have lense at 200mm. I have done at 185mm as well. I change the ISO setting scrolling thru 100, 150, 200, 300, 400, 600 and take a photo. About 20 yards away (currently indoors overcast), I have a few tape measures set up and focus directly on the center one (one to left and one to right). I also do with IS on and off. 

I then look at photo in RAW on computer and zoom in 500% to see how sharp the picture is...

Questions:

Is 500% too much to grade quality of image/details? If so, what should I be doing?

Any other suggestions? Thanks!

I bought the lens used and just tryting to determine if it is functionally correctly with the correct details/razor sharpness.

[rs-eos]

I use a Spyder LensCal from DataColor to see if I need to make any microadjustments.  You can create your own as well which is effectively a ruler at a 45º angle and a target to focus on.   The target is right at the ruler's zero mark and you can see where focus is actually at in terms of the target: spot on, back-focused, or front-focused.

I think things are definitely trickier with zoom lenses due to the multiple focal lengths.  I've only ever had to fine tune one lens so far (50mm f/1.2) and that was only with an EOS 6D.  I didn't need to do any microadjustments with the same lens on my 5D IV.

[hombrevg]

Thanks I have played around with this approach and dis my own with ruler on ground with 45 degree camera on tripod. 
My bigger question is testing quality of glass and fine details based on a object about 30. -35 feet away

i set up a few tape measures vertically in picture at this distance. I zoomed to 200m and took pics with above variables with the same spot focus. 

I then looked at CR2 outputs on a 5K monitor and zoomed to 500% and when I do this, I lose all of the detail on picture no matter the setting on body.

 
I am wondering if 500% is a fair ask as it is 5X on a 22 megapixel sensor. 

the reason I ask this is I currently have a Mark 5D III and considering migrating to mirrorless 6D Mark II with adapter from EF to RF to use all lens and move to this as my primary body/rig. 

so lots of questions before investment as well as validating lens I have is solid. 

happy to do 1:1 detailed discussion as well. 

[shadowsports]

Greetings,

The 6D2 is a DSLR, not mirrorless.  

So if you want to go mirrorless.

FF options: R, RP, R6/mkII, R5/C, R3

APS-C: R7,R10

I think the R6mkII is going to be a top seller this holiday season.  

You'll need a EF to RF adapter.  Get Canon brand with or without a control ring.  

[hombrevg]

Yes mis typed R6 Mark II. And yes will probably get the Canon adapter with control ring. Will not get filter one. 

i will post all the lenses I have but I have the Canon 85mm 1.8 that said it conss to life with R body. 

still get the impression that my 70-200mm os not as sharp as it could be. Wondering if I should get serviced. 

[jrhoffman75]

A DSLR has two focusing systems - one when using the. Optical viewfinder and one when using Live View. If you are running tests to determine lens sharpness you need to be using Live View since that is showing what the actual camera sensor sees. 

Focusing through the optical viewfinder is an inferred approach. A sensor determines sharp focus; camera calibration relates that result to the sensor. That is why many bodies have AFMA to adjust for differences. 

Enlarging an image greater that 1:1 will not give an accurate image - as you have found it falls apart. 

[shadowsports]

Good Morning John 😄

@hombrevg

John's advice is always stellar.  Canon does offer a lens maintenance service at a very reasonable price.

Canon Support | Canon Maintenance Service | Canon U.S.A., Inc.

You can use the link above for more information.  No worries about the typo.  I believe you will really enjoy the R62.  I have a the 85 1.8 and believe it to be one of the sharpest lenses I own.  Its certainly a bargain. The newly announced RF135 will also be a great portrait lens, and be a lot less wieldy than a EF 70~200.

[ebiggs1]

Also keep in mind something that most people don't realize, Micro adjustment does not make any lens sharper. The IQ of a lens is fixed at manufacture and cannot be changed unless physical parts are altered, replaced and or damaged.

Micro adjustment simply moves the critical focus point either forward or backward.

[jrhoffman75]

---

@ebiggs1 wrote:

Also keep in mind something that most people don't realize, Micro adjustment does not make any lens sharper. The IQ of a lens is fixed at manufacture and cannot be changed unless physical parts are altered, replaced and or damaged.

Micro adjustment simply moves the critical focus point either forward or backward.

---

And to add to Ernie’s response, the adjustment is for the camera, not the lens. 

[Waddizzle]

Performing AFMA is just as much a test of the photographer as it is the gear.  Here are two things to consider before performing AFMA.   

One, performing AFMA should be a last resort, not a first option.  I recommend not attempting AFMA without cause to do so.  “Just Checking” is a rabbit hole to avoid.

Two, performing AFMA is not as simple and straightforward as it might first seem.  Do not assume that you will get it right on the first attempt, because you won’t.  You won’t get it right the first time, almost no one does.  There are too many things that can trip you up that are easily overlooked.  This is the main reason for first piece of advice. 

Take a look at this popular video.  He covers a lot, but still misses a few points.  The video is mainly about a simple technique.  It only covers the basics without taking a deep dive into common mistakes.

Dot Tune: Autofocus Fine Tuning in under 5 Minutes. 

A couple things you are doing are already causing your AFMA to go off the rails.  These are commonly overlooked details that easily cause an inaccurate calibration.

Number 1.  Lighting.  Perform the AFMA under shooting conditions that are as close as possible to the actual shooting conditions when you will be using the lens.  If you will be using the lens outdoors, then calibrate the lens outdoor sunlight.  This is the most commonly overlooked detail.

Number 2.  Distance.  The video recommends performing the AFMA at a distance that is 50x the focal being tested.  For a zoom lens, this means that you need to perform tests at the wide and long ends of the zoom range at the very minimum.  If you are testing an APS-C body, then you need to allow for crop factor, and multiply those test distances by the crop factor.  The 50x distance figure is just a starting point, a generic distance.  If you know the actual distance to your subject for your shooting scenario with the lens, then calibrate at that shooting distance.

Number 3.  Sampling.  You should collect several test shots and average the result, instead of relying on one shot.  Defocus the lens between sample shots, too.  Also, capturing a several samples can also be a measure of your technique.  Your samples should consistently be within a limited window of results.  If your sample measurements are all over the map , then your testing technique needs to be tightened up.

Number 4.  Targets.  Your test target should mostly fill the frame at 50x the focal length.  Most test targets in the commercially available calibration kits work well with lenses up to around 200mm.  If you are trying to calibrate a super-telephoto lens, then you may want something a little larger than 8x11 sheet of paper.  I print targets on 24x36 paper and use a yardstick set at angle to judge front or back focusing.

Number 5.  Capturing.  Using a tripod is highly recommended.  You want to make sure that the camera is level to the horizon, both left-to-right and front-to-back.  Set your center column to its lowest position.  Using a shutter delay and mirror lockup is highly recommend.