Question on using Lens Align tool - micro-adjust

Thanks, everyone, for the responses. This was my first try at micro adjusting a lens that only goes to 5.6 (80-400 AF-S), and I think I got it “good enough”.  A real world test afterwards looked good - I photographed a red squirrel in the snow from about 25 feet away, 400mm, f/8, 1/1000 second, hand held, and looking at the squirrel at 100% view in photoshop, the whiskers were nearly tac sharp. The lens isn't in the league of a Canon 400/2.8 for IQ, but the 80-400 is small & light and hand-holdable and can yield perfectly usable images. Not a bad compromise for some photographing. I'm hoping it'll be good for airshow photography too.

E.J. Peiker wrote:

Doug Brown wrote:Here's my take. This notion that you have to go out and microadjust every lens and TC combo that you own is both incorrect and a complete waste of time! I've owned multiple copies of a number of Canon supertelephoto lenses and quite a few different bodies, and there's only one time that I've had to microadjust a camera-lens combination. Reserve microadjustment for times when you're having reproducible problems with focus. Almost every other pro photographer I know feels exactly the same way I do.

You probably own newer Canon gear or if not you may be lucky or you may just not be as critical as some.  The latest Canon gear from the 1Dx on and the new Mk II telephotos rarely need much adjustment at all although usually 1 or 2 points one way or the other does maximize sharpness but granted that's minimal and in real world photography situations doen's make that much difference.  But the generalization that it's a waste of time is not accurate:  Older Canon lenses tend to need quite a bit of adjustment.  Also newer cameras with higher pixel density tend to be more critical of focus in order to get the most out of the sensor (just wait until those 50mp Canon bodies come out).  On the Nikon side, life is nowhere near as good with almost all lens/camera combos needing some adjustment and many needing drastic adjustment.  The 500 f/4 VR with a 1.4x with any Nikon body, for example, in the vast majority of cases is completely incapable of making sharp photos without a huge adjustment, often right to the edge of what can be done in camera and sometimes the combo would still need more adjustment.  Almost all third party lenses need a very significant amount of adjustment to be able to render sharp photos on either the Canon or Nikon platforms.

I've never shot Nikon gear, so I can't speak to the microadjustment requirements of Nikon bodies and lenses. But I've owned both the old and new 600, 2 copies of the old 500 as well as the new 500, the old and new 300 f/2.8, and 2 copies of the 400 f/5.6. I've owned the 20D, 30D, 40D, 50D, 7D, 7D2, 5D3, 1D3, 1D4, and 1Dx, so I think I speak from experience. I suspect that my sharpness standards are extremely high compared to most bird photographers. And I stand by what I said. Let's just agree to disagree.

Well, I can't speak for Canon as I (like you) have never owned or shot with Canon. However, you did say "This notion that you have to go out and microadjust every lens and TC combo that you own is both incorrect and a complete waste of time!"  Sorry but that seems a bit extreme to me to make such a serious blanket statement for all Canon shooters. I know a few myself that are struggling with their Canon gear and the microadjust. Plus, there are many who do not just use Canon lenses, so that statement also cannot apply to them. I am sure EJ has seen and adjusted many times more Canon bodies and lenses than you will see in a lifetime. How you can say to everyone that it is "incorrect" and "a complete waste of time" is just beyond me. There is a reason why Canon started putting that feature in their bodies, and they started it before Nikon.
 

FWIW, here are just a few of the more significant adjustments needed by clients from my database using Canon gear:
-10  EOS 1D4 + EF500 f/4L IS + 2x II
+14  EOS 1Dx + EF180 f/3.5L Macro
+3,+13  EOS 1Dx +EF70-200 f/2.8L IS II + 1.4x III
+17 EOS 5D3 + EF300 f/2.8L II
+16 EOS 1D4 + EF300 f/2.8L
+10 EOS 1D4 + EF800 f/5.6L

Each entry is from a different client so no duplicate lenses or cameras.  However, as I said before, the 1Dx with the series 2 super teles, at least until you start slapping external TCs on the 200-400, rarely need more than a couple of points to be optimized so those are usually plenty good enough straight out of the box.  Also, on average, the EOS 1D Mk 4 requires more adjustment than any other Canon camera that has the facility for adjustment.

Tim Zurowski wrote:Well, I can't speak for Canon as I (like you) have never owned or shot with Canon. However, you did say "This notion that you have to go out and microadjust every lens and TC combo that you own is both incorrect and a complete waste of time!"  Sorry but that seems a bit extreme to me to make such a serious blanket statement for all Canon shooters. I know a few myself that are struggling with their Canon gear and the microadjust. Plus, there are many who do not just use Canon lenses, so that statement also cannot apply to them. I am sure EJ has seen and adjusted many times more Canon bodies and lenses than you will see in a lifetime. How you can say to everyone that it is "incorrect" and "a complete waste of time" is just beyond me. There is a reason why Canon started putting that feature in their bodies, and they started it before Nikon.
 

I stand by my statement Tim. Microadjust if you're having reproducible AF problems. Otherwise don't bother. What is extreme about that statement? The microadjustment setting is there to correct AF problems. If you don't have problems, don't microadjust.

From the Canon 1Dx manual:
"! Normally this adjustment is not required. Do this adjustment only if necessary. Note that performing AF Microadjustment may prevent correct focusing from being achieved."

Doug Brown wrote:From the Canon 1Dx manual:
"! Normally this adjustment is not required. Do this adjustment only if necessary. Note that performing AF Microadjustment may prevent correct focusing from being achieved."

Doesn't mean much. The Nikon manual says basically the same thing. It just means they want you to think their lenses should all work well out of the box. To admit otherwise wouldn't look good, would it?

Joe

SantaFeJoe wrote:

Doug Brown wrote:From the Canon 1Dx manual:
"! Normally this adjustment is not required. Do this adjustment only if necessary. Note that performing AF Microadjustment may prevent correct focusing from being achieved."

Doesn't mean much. The Nikon manual says basically the same thing. It just means they want you to think their lenses should all work well out of the box. To admit otherwise wouldn't look good, would it?

Joe

And their lenses do work well out of the box in my case ( and I shoot Nikon ). I don't see the logic behind "they allow you to micro-adjust the AF yourself = manufacturers no longer make lenses work well out of the box". It just a cynical way of thinking IMO. Of course one can ask: "how well is well?" I think it's the degree of "well" where we differ.

My experience has been that not only do lenses+bodies need microadjustment, but the AF amount is temperature dependent.  I just returned from the Serengeti where I took a 7D2 and 300 f/2.8 version II.  We had about 50 degree F temperature swings.  I generally check microadjustment values periodically throughout the day on targets I am imaging and adjust as needed, and find the value changes, especially with TCs.

Edit: I'll have to check values, but on my 7D2+1.4x, the MA values hovered around 5, +3, -4 (varying with temperature) if my jet lagged brain is remembering correctly.

Roger

DChan wrote:And their lenses do work well out of the box in my case ( and I shoot Nikon ). I don't see the logic behind "they allow you to micro-adjust the AF yourself = manufacturers no longer make lenses work well out of the box". It just a cynical way of thinking IMO. Of course one can ask: "how well is well?" I think it's the degree of "well" where we differ.

It's not logic, it's reality. If you think that it's cynical, you must realize that all machine tools wear as they are cutting metal mating surfaces. There is a certain range of tolerances that are allowable by any manufacturer. A mating surface that is cut with a new tool will vary from one that is the last cut with the same tool. Both may be within the allowable tolerance range, but they are not exactly the same. I do agree with your statement that it's a degree of "well" where we differ, but if you can improve your settings by adjusting them, why not do it? If you don't want to, nobody is saying you have to. That's your choice.

Joe

I apologize for the next statement in advance because I know of no way to sugar coat this:  If you truly believe that lens micro-adjust, or AF fine tune (Canon and Nikon names) is "incorrect and a waste of time" I suspect that you do not understand how off-sensor phase detection works nor how the concept of manufacturing tolerances play out in mechanical and electronics manufacturing.  If you truly learn how it functions, how it is designed and how it is built into camera bodies, you might not make such a statement.

E.J. Peiker wrote:I apologize for the next statement in advance because I know of no way to sugar coat this:  If you truly believe that lens micro-adjust, or AF fine tune (Canon and Nikon names) is "incorrect and a waste of time" I suspect that you do not understand how off-sensor phase detection works nor how the concept of manufacturing tolerances play out in mechanical and electronics manufacturing.  If you truly learn how it functions, how it is designed and how it is built into camera bodies, you might not make such a statement.

+1
body{zoom:150%!important;}



body{zoom:150%!important;}

It would be interesting to see a good statistical test showing how many "ticks" of MFA are typically visible when displaying real world photographs in real world conditions.  It's human nature that we want to fix any error we can measure, especially if we have a bent towards being detailed, perfectionist, obsessive or compulsive etc.  But without a proper sensitivity analysis that really says more about the person than the problem.

SantaFeJoe wrote:It's not logic, it's reality. If you think that it's cynical, you must realize that all machine tools wear as they are cutting metal mating surfaces. There is a certain range of tolerances that are allowable by any manufacturer. A mating surface that is cut with a new tool will vary from one that is the last cut with the same tool. Both may be within the allowable tolerance range, but they are not exactly the same. I do agree with your statement that it's a degree of "well" where we differ, but if you can improve your settings by adjusting them, why not do it? If you don't want to, nobody is saying you have to. That's your choice.

Joe

Well, it's exactly tolerances that I was talking about. Manufacturer's, yours, mine, and many other people's. Sometimes they align, sometimes they don't. If yours and the manufacturer's don't, totally fine for you to fine-tune it to match yours. I don't think anyone here is arguing that nobody should be doing that if that's the case for him or her. Hey, I found I can't tolerate the result - a total disaster - that my 500 + 1.4TC  gave me and I AF fine-tune'd it. So I'm not against AF fine-tune and I don't think anyone here is. It's just that we have different tolerance level and we approach AF-fine tuning differently.

For some perspective here, consider a camera with 4-micron pixels and a very sharp f/2.8 lens (example of real world combo: Canon 7D Mark 2 + 300 f/2.0 L IS II). A focus error of 4 * 2.8 = 11.2 microns enlarges the bur by one pixel. For Americans, that is only 0.00044 inch. Look at the coefficient of expansion of various materials and one sees that sizes of materials are not constant. Thus one would expect some changes with temperature, and with tolerances less than 0.001 inch, manufacturing tolerances will add to the problem. It is not reasonable for lens + camera systems to be perfect out of the box and under all temperature conditions.

There is a saying in science: all precision instruments are thermometers. (Meaning the calibration changes with temperature.)

In my experience, I can see the difference in AF adjustment with my Canon 300 f/2.8 II and 7D2 on the camera LCD with a change of about 3 MA units.

But it is high time for the manufacturers to include a microcode program to do auto-MA calibration:
Photographer pushes the calibrate button and the camera raises the mirror, does a contrast AF with the sensor, lowers the mirror and does a phase detect AF on the same subject and the difference is the microadjustment value. Do this several times and average. It should take about 10 seconds. How to get this info to the manufacturers? (This is not the first time I have mentioned this in these forums.)

Roger

E.J. Peiker wrote:I apologize for the next statement in advance because I know of no way to sugar coat this:  If you truly believe that lens micro-adjust, or AF fine tune (Canon and Nikon names) is "incorrect and a waste of time" I suspect that you do not understand how off-sensor phase detection works.  If you truly learn how it functions, how it is designed and how it is built into camera bodies, you might not make such a statement.

You are misrepresenting what I said EJ. To be clear I said that microadjusting lenses that have no visible AF problems is a waste of time. I freely admit that I had to microadjust my 1Dx and 300 f/2.8 II. But that was because I was having reproducible front- or back-focusing issues (I don't recall which one it was). How off-sensor phase detection works has no real bearing on this discussion. Either your camera and lens focuses properly or it doesn't. If it doesn't, and it's not the result of operator error (which in most cases is the actual cause of AF problems), by all means microadjust your camera body and lens. 

IMO, if you pay $7000 for a camera, and $10,000 for a lens, they should auto adjust for optimum focus.

Somewhere a tour of a camera manufacturer was written up including final quality control. The body was placed on a lens chosen to be a standard. The lens and focus chart were mounted rigid so there would be no question of subject distance. At that point the correction for the body parameters MF was recorded as a starting point in the camera for all lenses.

Now:

How many lenses exactly match the "standard" lens.

As far as MFA on DP Review there was a lot of negative noise when it was found out that there was no MFA on the 60D. It was not that long before the 70D was introduced and that included MFA tables. In other words a lot of users do care.

Interesting read on the subject.

Doug Brown wrote:Interesting read on the subject.

He stated it all in his concluding statement, much as others have here. 

A soft image doesn't mean an AF problem! If you can repeat a focus issue on a measurable scale on a fixed target consistently, then you have an AF problem. Understanding the difference between camera shake, moving subjects, etc., and out of tune AF is crucial in determining if there is a problem. It is obvious if a camera is back or front focusing when you have a rigid mount on the camera/lens and a firmly fixed target with no movement of either.

Joe

SantaFeJoe wrote:

Doug Brown wrote:Interesting read on the subject.

He stated it all in his concluding statement, much as others have here. 

A soft image doesn't mean an AF problem! If you can repeat a focus issue on a measurable scale on a fixed target consistently, then you have an AF problem. Understanding the difference between camera shake, moving subjects, etc., and out of tune AF is crucial in determining if there is a problem. It is obvious if a camera is back or front focusing when you have a rigid mount on the camera/lens and a firmly fixed target with no movement of either.

Joe

That's exactly what I've been saying, despite what some people on the thread have tried to twist my message into.