1.4x TC + 200mm - filter makes a diff in focus

This is an offshoot of this thread (didn't want to take it over):
https://photography-on-the.net/forum/showthre​ad.php?t=314086

Here's the issue. I shot a few images with my 30d and 200mm f/2.8 with and without the 1.4xTC - just shooting a few birds hanging around in the yard. Granted not a perfect combination but I have nothing longer. When I downloaded the images to my computer I noticed some front focusing but only when using the TC! A bit unnerving, because my 30D just came back from the Canon service center a couple of weeks ago sent there to fix a back focus issue.

Well, I decided to take a test, as "real world" as I could in order to get to the bottom of this. I tried a number of combinations to test out a theory I had (would extra glass affect the focus?). First I tested the 200L f/2.8 with filter, then without. Then I added the 1.4xTC MkII and shot again - with filter, then without.

Below is the image, click on it for full view. Clearly, the addition of the filter on the 200mm lens coupled with the 1.4xTC indicates some focusing issues. I was both surprised and relieved - surprised that a filter could have such an effect and relieved that the camera was not at fault.

Please keep in mind that the image you are about to view is "zoomed" in slightly and does not shot the whole shot, nor is the image post-processed (read sharpened) in any way. I used a screen cap of the ZoombrowserEX image to show the difference.

Judge for yourself (full sized image that pops up may be scaled down, so click again)

Here's the result of the test:
- On the left, 200mm + 1.4xTC + B&W UV F-Pro filter
- On the right - 200mm + 1.4xTC without filter

HTTP response: NOT FOUND | MIME changed to 'image/png'

Here is the result of the complete test:

from left to right:
200L f/2.8 with filter --> 200L f/2.8 + 1.4xTC + filter --> 200L f/2.8 + 1.4xTC no filter

HTTP response: NOT FOUND | MIME changed to 'image/png'

Yes, I had similar problem when using my 70-200 f/4 L with Hayo HMC 67mm filter. It happens with and without TC.

I have changed to B+W 67mm UV MRC, no more such problems.

Is your filter MRC version?

Unfortunately I don't believe it is. Just says B+W UV Haze F-Pro 72mm on it. It's an older (?) filter I suppose.

I ought to try this lens/extender combo with another filter (MRC) and test again.

OK you seem to have a large difference there which is completely at odds with my own tests in this area so far.

I'll just reiterate the points I made in your other post:

a) You will need to refocus after removing/adding the filter.
b) There will be a statistical spread of AF precision , IE it is not perfect. So you can't just compare two shots with and without filter.

My test for this involved taking 10 shots with and 10 shots without filter with a new AF action between each shot.

Additionally I would suggest you use a flat test target pattern as this eliminates the possibility that the AF is locking on to slightly different depths of the tree trunk each time, this concerns me quite a bit with the current method.

Don't get me wrong it is nice to see someone doing controlled tests, but it is best to eliminate as many possible variables as much as possible.

Of course you should be using a sturdy tripod with mirror lockup and a cable release, I am sure you are.

The only other issue is multiple test shots to ensure you eliminate statistical variation due to AF precision.

Once you have those issues checked it might also be worth you trying again with a top end filter like Hoya SHMC Pro1, Heliopan SH-PMC or B+W MRC.

BTW, did you do a with and without filter test with the 200mm f2.8 on it's own out of interest?

I will not get around to it today but I'll see if I can repeat the test with the 200mm f2.8 + 1.4X II using the standard lens filter Hoya SHMC Pro 1.

kolok sorry the offtopic but, what software did you used to see the focus point and those small images?
thank you

Hi,

Thanks for the input guys.

Lester, you're right, at some point here I will have to do a comprehensive multi-shot test to average out the results. I shoot with the 200L very often (always with filter on), and the results are simply stunning (one of the reasons why I purchased the 200L a long time ago - and then went ahead with the addition of the 135L ).

Regardless, either with the filter on or off, the results are visually identical - crisp and clear using the straight 200L. Thing is, the other day, I pulled out the trusty 1.4 MkII and added it to the combo to take some shots of birds (as explained above). That's when I noticed this tempest in a teapot. A few more shots, similar conditions clearly showed some front focus. Hmm, I thought, interesting to see the difference using a varied combination of lens/tele/filter. Preliminary results are what you see up top. No filter = much clearer shots, AF spot on. Filter = fuzz and front focus.

My cam body was recently in the Canon service center where they tweaked the focus and performed an "electrical adjustment". At first sight of the results using that particular lens/tele/filter combo I thought that Canon may have "over-tweaked".. hehe... The removal of the front filter, however, shows a much, much clearer image. Therefore, I'm left wondering if a normal filter will/can have such an effect on focus. Baffling.

Phew.. long winded.. Thanks a lot for your input though, it's much appreciated!

Anycase, as mentioned, my next step will be to test the same using a MRC or SHMC filter and see if there's any difference.

Point of the whole exercise, really, is to clearly indicate to others that focus issues may not always be attributable to either a lens or body.

hello DP86 - I used ZoomBrowserEX and took a screen cap.

Cheers all!

kolok wrote in post #3167193
Point of the whole exercise, really, is to clearly indicate to others that focus issues may not always be attributable to either a lens or body.

Yes I agree, or indeed the filter.

My main reason for testing filters is because we hear so many horror stories. SO far I have found these to be largely baseless.

I'll try and check the 200 + 1.4X tomorrow; it will take a little while to process the results and post.

I'll post this in both threads on this subject for completeness.

OK I have tested the 200mm + the EF 1.4X and not found a significant variation in sharpness due to the filter.

Interestingly there was much more variation in sharpness with the teleconvertor than without. But the variation was, within statistical limits, the same with and without the filter. Also the variation was small enough to not be easily detectable in the 100% crops and probably requires software analysis to spot.

I have updated the writeup on the tests here, this page also includes a link to a page with the 100% crops.

kolok wrote in post #3160467
Clearly, the addition of the filter on the 200mm lens coupled with the 1.4xTC indicates some focusing issues.

This does not surprise me.

jr_senator wrote in post #3171083
This does not surprise me.

It does me, though it is possible that it is because it is a single coat B+W filter.

I think it is more likely the test method in this case, because a tree has been used as the target the curving surface can cause variations in the AF precision. Use of a flat target would eliminate this possible problem.

Careful testing with a good multicoat filter shows no variation due to the filter.

Thanks for taking the time to perform some tests of your own Lester.

I'm in the process of looking for a good filter to see what the difference would be. Should I use a UVa or clear, should I decide to get one?

Other thing, as mentioned in my initial post I used a subject to simulate some real life shooting (in this case it happened to be a tree) to get away from the controlled environment, which may not always mean much if not performed properly.

kolok wrote in post #3171732
Thanks for taking the time to perform some tests of your own Lester.

I'm in the process of looking for a good filter to see what the difference would be. Should I use a UVa or clear, should I decide to get one?

Hi there, you are welcome.

I just use the UV's because they are easy to get but clear will do just fine assuming you don't shoot film. I recommend one of the three types I listed earlier.

kolok wrote in post #3171732
Other thing, as mentioned in my initial post I used a subject to simulate some real life shooting (in this case it happened to be a tree) to get away from the controlled environment, which may not always mean much if not performed properly.

OK, just be aware that the AF lock depth variation may be much larger than any difference you might hope to detect with the filter due to the 3D subject (I must do some tests on that someday). The focus plane probably only needs to move a few mm to make a big difference in sharpness when viewed at 100%, probably a tenth or less than the conventional depth of field.

I suggest checking with a plane target first and then expanding to less controlled environments. Or you could put up a plane target in front of your tree as long as it is 2-3 times the apparent size of the focus point as seen in the viewfinder, that should be enough to minimise confusing signals from other subjects at different distances.

BTW the test I do is quite heavily back-lit (has a window behind the subject so it is many stops brighter than the target) and so also at relatively low light levels. This is to both maximise any potential effect due to the filter and to stress the AF system as much as possible in terms of it's internal noise sources.

Lester Wareham wrote in post #3171948
OK, just be aware that the AF lock depth variation may be much larger than any difference you might hope to detect with the filter due to the 3D subject (I must do some tests on that someday). The focus plane probably only needs to move a few mm to make a big difference in sharpness when viewed at 100%, probably a tenth or less than the conventional depth of field.

Definitely something to remember!

You know, all this has me wondering.

Bearing in mind that the first element on that lens (200 f/2.8L) seems more inset than on other lenses I wonder whether the use of a filter (non MRC) would throw the AF off slightly (at distance) as it may inadvertantly act as an additional lens element (due to the distance between the filter itself and the first element on that particular lens ?) and that the TC already attempts to "tighten" the focal range ... The filter, causing some refraction, is therefore the source of misfocus. If the refraction is negated because of the multi-coated nature of a filter, which simply becomes a "pass-through" then this misfocusing is a non-issue.

Interesting.

kolok wrote in post #3174643
Bearing in mind that the first element on that lens (200 f/2.8L) seems more inset than on other lenses I wonder whether the use of a filter (non MRC) would throw the AF off slightly (at distance) as it may inadvertantly act as an additional lens element (due to the distance between the filter itself and the first element on that particular lens ?) and that the TC already attempts to "tighten" the focal range ... The filter, causing some refraction, is therefore the source of misfocus. If the refraction is negated because of the multi-coated nature of a filter, which simply becomes a "pass-through" then this misfocusing is a non-issue.

Interesting.

Reguardless of where the front element is positioned in relation to the filter, the filter is another piece of glass that can cause problems. Multi-coating may help, I have read that this is not the case (I will try to find the reading), but again, regardless it is another (and I feel usually unnecessary) piece of glass to bounce light around.