Focusing Problem with 600mm f/4.0L IS II with 1.4X Extender

[wearle]

To all,

I've been having a difficult time getting sharp images with the Canon EF 600mm f/4.0L IS II USM and the Canon 1.4x EF Extender II.  Basically, I shoot from my Toyota Tacoma with the lens/camera combination resting on a BLUBB (bean bag).  I turn the vehicle off while taking images.  I've tried where the focusing ring is on the bean bag and not on the bean bag.  Will the movement of the focus ring during AF impact the overall image quality?  That is, is the focus ring turned off during AF?  I've tried placing my hand over the tripod foot, over the hood and under the hood for added stability.  While looking through the view finder I find the had over the hood appears to work best.  I see very little if any movement during this phase.  I took around 600 images of the eagle, and saved only 17 marginally sharp images.  This is a terrible hit to miss ratio.  I have attached several images to show the common out of focus images.  What's really interesting is the strange catch light pattern in the eagle.

I've tried every AF combination and AF pattern available on the Canon 1Dx to no avail.

It is possible that the problem lies with me (i.e. poor technique); however, I've read several articles relating to  proper super-telephoto techniques and try following them as best I can.  If anybody lives near Pendleton, Oregon and is an expert at proper super-telephoto techniques, I would enjoy your company for an afternoon along the Foothills of the Blue Mountains.

Thanks,

Wade

[Mt Spokane Photography]

The lens has full time manual AF.  If the AF ring moves, it overrides the autofocus.  Resting the lens on the AF ring  is going to cause a problem.

[wearle]

I appreciate the clarification.  I figured it was likely degrading the focus so I made certain during these images that the focus ring was not resting on the bean bag.  Nevertheless, there may be times where it did make its way back onto the bag.

I also forgot to mention I used FoCal for microadjustments and this combination came out to -1; however, I'm beginning to wonder if it was a valid result.  In addition, IS was off the entire photo shoot.

Wade

[wearle]

To all,

One final note, I believe the catch eye in the second image points to a potential problem.  There appear to be at least five individual catch eye spots.  They all appear to be tightly focused (i.e. near pinpoint).  Typically, if it is user error, the catch eye will drag across the eye or become every blotted. It almost makes me believe there is something happening with the motors inside the lens despite IS being off.  I can't see how you can have five or more focused catch eye spots within a single 1/3200s exposure.  Any thoughts?

Wade

[Don Haines]

If you say that some shots are in focus, that means that the lens/teleconverter combination CAN focus, but isn't doing it properly... hopefully that means that the combo is optically fine.

What happens if you put the camera on a super-solid tripod and aim at a fixed target (a bird like stuffed animal in a tree)? If you manually focus on the object, then try again with autofocus on, does it change focus?

[jasonsim]

Do you have another camera to try it with.  I tried my 600 II out for the first time on Sunday with a 5D III.  I too was getting fuzzy results.  Come to find out...the MA was set at -8 automagically.  I never set it for that lens.  So I entered into MA and use the INFO button (I think twice) to enter in the lens serial number and set the MA to 0.  That seemed to correct the problem.  Have you tried disabling MA all together?

Here is a test shot from Monday night shot at 840mm (1.4x III was used) at ISO 800 and f/5.6 with 0.3 second exposure (IS program 1 was on):

[East Wind Photography]

I am no longer a big fan of FoCal.  Particularly when it comes to AFMA adjustment of long tele's.  There are too many variables that will skew the results...the largest being vibration.  The shutter activation even with Mirror lockup can affect the image.

The best way is to just do it manually.  You can start by taping a newspaper to a wall and running through AFMA adjustments until you get the best image.  Then go out and shoot a duck on the water to dial in the adjustment.  I also discovered that you have to bring the image onto your computer to accurately see where the focus point is.  Not accurate enough to just view on the LCD...so that means perhaps several trips to the shooting site.

With the 600mm and a 1.4 extender, your DOF will be around 12" or less.  an AFMA adustment of + or - 1 will shift your focus point several inches.  My guess is about 6 based on my set up at around 30 meters or so. 

So it has to be dead nuts on or you will get an OF image.

I appreciate the clarification.  I figured it was likely degrading the focus so I made certain during these images that the focus ring was not resting on the bean bag.  Nevertheless, there may be times where it did make its way back onto the bag.

I also forgot to mention I used FoCal for microadjustments and this combination came out to -1; however, I'm beginning to wonder if it was a valid result.  In addition, IS was off the entire photo shoot.

Wade

[Recoil]

Looks like thermal distortions to me.  It happens to me when I shoot from my vehicle, especially in winter (with 500mm).  Have to be careful with heat escaping the window.  It can happen in other situations also, It even happened at close to mfd once (hot trail totally messed up my shots...). 

[neuroanatomist]

The best way is to just do it manually.  You can start by taping a newspaper to a wall and running through AFMA adjustments until you get the best image.  Then go out and shoot a duck on the water to dial in the adjustment. 

Makes sense...and that's pretty much what I do, although still using FoCal.  Basically, I trust an image analysis program to determine which is the 'best image' more than I trust my own eyes (and I should point out that my day job for many years was analysis of microscopic images, both 'by eye' and with digital image analysis algorithms). 

When I do a FoCal calibration, I use lots of light (my 600 II tests were in the EV 14-15 range) and a solid support (RRS TVC-33) on a stable surface (concrete or asphalt).  I don't tether the camera and let FoCal do the iterative adjustments...rather, I take 83 shots per 'test' - 2 shots at even-numbered AFMA values from |20| to |12| with one starting focused at ∞ and the other starting at the MFD, then three shots at every AFMA value from +10 to -10 (inclusive), one from ∞, one from the MFD, and then one without moving the focus ring.  I load the 83 images into FoCal in manual mode, and there are ample data to drive a curve fit.  I run a test at 25x focal length and 50x focal length, and for zoom lenses, at each end and 1-3 intermediate focal lengths.

I then take the value(s) from FoCal and do real-world testing with the lens with AFMA applied, to make sure I'm getting sharp images.

[East Wind Photography]

My point was that with the 600 every error is magnified and if you dont have the "ideal" conditions to conduct a focal test you are just wasting time and shutter cycles.  Most people dont have a stable enough set up to conduct a focal test at 600mm much less with an extender attached.  I recall you needed a football field to conduct the calibration.

Short of making a recommendation to someone to fix their set up, the next logical course is to do it manually.  I find it quite easy to find the focal point when shooting ducks on the water and easy to see if you are 6" off in focus.  Seeing the difference between one AFMA is tough but if you shoot a range in the ball park you can pretty much tell which one is the best even if you find the ones that get fuzzy first of each end and extrapolate in.

2nd, I dont trust Focal after the 1.4 release.  The algorithm changed and I became less confident in the results.  It became more of a mathematical model rather than looking at the real sharpness of the target.  The argument has been beat to death in other forums.  So lets suffice it to say that if you did not get the desired result with focal your only recourse is to do it manually or send the entire setup back to Canon for calibration.

Regardless of wether you use focal or not, you should still validate the setting by taking shots through the +-5 AFMA range to validate that it actually calculated the right AFMA.

The best way is to just do it manually.  You can start by taping a newspaper to a wall and running through AFMA adjustments until you get the best image.  Then go out and shoot a duck on the water to dial in the adjustment. 

Makes sense...and that's pretty much what I do, although still using FoCal.  Basically, I trust an image analysis program to determine which is the 'best image' more than I trust my own eyes (and I should point out that my day job for many years was analysis of microscopic images, both 'by eye' and with digital image analysis algorithms). 

When I do a FoCal calibration, I use lots of light (my 600 II tests were in the EV 14-15 range) and a solid support (RRS TVC-33) on a stable surface (concrete or asphalt).  I don't tether the camera and let FoCal do the iterative adjustments...rather, I take 83 shots per 'test' - 2 shots at even-numbered AFMA values from |20| to |12| with one starting focused at ∞ and the other starting at the MFD, then three shots at every AFMA value from +10 to -10 (inclusive), one from ∞, one from the MFD, and then one without moving the focus ring.  I load the 83 images into FoCal in manual mode, and there are ample data to drive a curve fit.  I run a test at 25x focal length and 50x focal length, and for zoom lenses, at each end and 1-3 intermediate focal lengths.

I then take the value(s) from FoCal and do real-world testing with the lens with AFMA applied, to make sure I'm getting sharp images.

[neuroanatomist]

FoCall is still using sharpness to quantify the images (I'm running the latest Mac beta).  Where their 'modeling' comes in is to reduce the number of data points this software requires to make a call on the AFMA value.  This is pretty evident when I load in >80 data points and look at the curve fit.  Sometimes the fit is 'fair' or even 'poor' when the same data loaded into a bona fide data analysis program yields a much better fit, and a resulting value which matches what I see from visual inspection of the plotted data, but is sometimes off by 1-2 units from FoCal's selected value. Basically, I think FoCal tries to fit a few data points to a pre-specified curve, rather than fitting a curve to the data.

RE what the data points actually are, I have done what FoCal does (measure sharpness and peak contrast) with a MATLAB script and gotten the same rank-ordering of image sets. FoCal is just easier to use, as I don't need to batch-convert to TIFs, etc. 

I guess the key point is that there are LOTS of wrong ways to do an AFMA, and a few right ways.  As long as you find a right way that works for you, and gives you sharp images, you're fine.

[wickidwombat]

The best way is to just do it manually.  You can start by taping a newspaper to a wall and running through AFMA adjustments until you get the best image.  Then go out and shoot a duck on the water to dial in the adjustment. 

Makes sense...and that's pretty much what I do, although still using FoCal.  Basically, I trust an image analysis program to determine which is the 'best image' more than I trust my own eyes (and I should point out that my day job for many years was analysis of microscopic images, both 'by eye' and with digital image analysis algorithms). 

When I do a FoCal calibration, I use lots of light (my 600 II tests were in the EV 14-15 range) and a solid support (RRS TVC-33) on a stable surface (concrete or asphalt).  I don't tether the camera and let FoCal do the iterative adjustments...rather, I take 83 shots per 'test' - 2 shots at even-numbered AFMA values from |20| to |12| with one starting focused at ∞ and the other starting at the MFD, then three shots at every AFMA value from +10 to -10 (inclusive), one from ∞, one from the MFD, and then one without moving the focus ring.  I load the 83 images into FoCal in manual mode, and there are ample data to drive a curve fit.  I run a test at 25x focal length and 50x focal length, and for zoom lenses, at each end and 1-3 intermediate focal lengths.

I then take the value(s) from FoCal and do real-world testing with the lens with AFMA applied, to make sure I'm getting sharp images.

Man how long does that take to do 1 lens 

[neuroanatomist]

Once the target is set up, taking the 83 shots takes 10-12 min, so a prime takes me <25 min and a zoom usually <1 hr. Manual analysis in FoCal takes just a few minutes per image set.

[wearle]

To all,

Thanks for your suggestions.  I'll redo the AFMA, perhaps something similar to neuro's.  At this time, I have it set to 0 which is pretty close to the original calculation from FoCal of -1.  Is it possible to set the AFMA through Canon's software?  It would be ideal to be able to do this as any touching of the camera during the calibration process is going effect the results (i.e. slight changes in distance to target).

I have used the lens/camera assembly attached to a Gitzo GT5541LS/Wimberley Gimbal Tripod Head II in manual focus.  I tried LiveView at 10X, but touching the focus ring resulted in too much movement.  4X seems to work pretty good though.  What experiences have you all had at an 840mm focal length?  The manual focusing results were not as sharp as I would expect from this lens; however, they are better than what I'm getting with AF.

I'm not sure if the problem was heat currents.  Towards the end, the cabin of the truck had pretty much acclimated to the outside temperature; nevertheless, I'll keep this in mind the next time I go out.

I'm still a little concerned about the multiple catch-eye signatures in the second photo.  Individually, they are tack sharp.  There's no way I can move fast enough (i.e. operator error) in a 1/3200 second of exposure to produce five or more catch-eyes on the target.  It's a very strange, but interesting phenomenon.

Wade

[wearle]

To all,

Here's an image of the Moon I took a few moments ago.  I was in manual focus using LiveView at 4X.  It was still bouncing around a little which made precise focusing a little difficult. It appears the lens in manual focus is fine.  It's AF that's a little whacked. 

Wade