Teleconverter vs Cropping the Picture?

I recently bought the 70-300L to replace my non-L. My plan was to add a 1.4x or 2x Kenko teleconverter before a trip to Yellowstone next summer. But after reading many posts in these forums about loss of autofocus and sharpness, I think I have decided against the teleconverter, and instead cropping my animal pictures in post-processing. Even if I crop my 60D pictures to the equivalent of 600mm (2x, 4.5 megapixels) I should get acceptable results.

Five years ago I visited Yellowstone with a Rebel 300D (6 megapixels) and my non-L lens. I got decent results, although not perfect, when cropping 2x or more. This time I should get better results with my 60D and 70-300L.

I invite any comments you may wish to make.

Forget the 2X extender, but a 1.4X could be very useful. I have a Canon 1.4x version II extender that I sometimes use behind my Canon 70-200 f/2.8L IS lens and I can hardly tell the difference in image quality (sharpness, colors, contrast, etc.) with the extender attached.

Tests I've seen done with the 2X extender show obvious image quality reduction. There may, of course, be the occasional situation where the longer focal length outweighs the image quality reduction.

I've heard the Kenko Pro 300 DGX 1.4X teleconverter has good results with the 70-300L, but more on full frame bodies than crop (or so I've heard - there has been a lot of talk on the photo thread about teleconverters lately).

Personally though, this lens is so sharp and resolves so much detail that I have no issues cropping to and beyond the field of view that a 1.4x converter would have given me natively if the need arises.

Optical solutions for 'magnification' are always better than those that resort to 'cropping pixels' (aka 'digital zoom').

As what Wilt said, cropping is "digital" enlargement, where actual focal length is optical (think of a p&s- digital zoom=crap, vs optical zoom).

While i like to (and often do) crop hard, if you're using a higher ISO (say to get a BIF shot at adequate s/s), cropping will enlarge the noise as well, where a TC won't.

Wilt wrote in post #15575231
Optical solutions for 'magnification' are always better than those that resort to 'cropping pixels' (aka 'digital zoom').

IN camera digital zoom crops and then up-sizes to match the regular resolution. That's why digital zoom looks so bad. It performs a quick and dirty resolution scaling in the camera.

Just cropping is fine, as long as you accept the requisite loss in resolution and maximum print size. Keep in mind that cropping to the equivalent of a 1.4x, leaves you with about half of the original resolution. So your 18MP 60D now becomes about a 10MP camera if you crop to 400mm equivalent. You are also effectively shrinking your sensor to about a 2.2x crop as well, so there will be a little more visible noise. Going to 600mm in a crop gives you a 4.5MP 3.2x crop sensor, but is still OK in a pinch.

When you put a 1.4x on the 70-300 it will become a f5.6-8 lens. The 60D requires a f5.6 lens at worst to autofocus, according to spec. It won't even try to AF at the long. If you tape some pins, you can get the camera to think it is an f5.6 lens and try to AF, but the results are mixed.

Let us assume that a lens manages to capture 2500 line-pairs in the frame height.
Let us assume that a 1.4x convertor loses as much as 20% of the resolution
(photozone.de tests have shown that Canon 70-200mm lenses lost about 10% of resolution with Canon 1.4x teleconvertor, so we are being conservative in assuming 20% loss)

If you do 1.4x crop of pixels, we end up with 2500 / 1.4 = 1785 line pairs
If you do 1.4x optical magnification crop (teleconvertor), we end up with 2500 * 0.8= 2000 line pairs

Wilt wrote in post #15575560
If you do 1.4x crop of pixels, we end up with 2500 / 1.4 = 1785 line pairs
If you do 1.4x optical magnification crop (teleconvertor), we end up with 2500 * 0.8= 2000 line pairs

Is that TC math based on reviews, or where did you get the .8 factor?

I agree a TC is generally better, but only when the lens and camera natively support them. Going out with a taped up TC on a f5.6 lens and hoping the light is good enough for AF to still work is not a recipe for perfect photos, either.

Regardless it would be easy for the OP to see if he likes cropping or not. Go out and find a subject that has some detail. Take a picture at 300mm so the subject nearly fills the frame in the longer dimension. Now step back far enough so that the subject fills about 70% of the length of the frame. Back on the computer, crop it so that shot fills the frame (should be about 3700px on the long dimension). Compare the detail and max viewing size.

You'll find there's not a ton of difference until you try to print big.

tkbslc wrote in post #15575606
Is that TC math based on reviews, or where did you get the .8 factor?

I explained in the earlier post that the Canon 70-200 (excellent lenses) lost about 10% of MTF resolution with teleconvertors. So I then assumed a conservative 20% loss figure, to assume would occur would a less extraordinary lens, like the 70-300mm, for illustrative purposes... it certainly would not cause the 40% loss attributable to pixel losses in cropping.

You bring up a good point about lack of AF; that is yet another perturbation in performance, over and above any differences attributable to the 'crop vs. tele' consideration.

I don't argue with your calculations. If this was a 1D series body that can AF at f8 or a f4 zoom, I'd be recommending a TC hands down. It's just when you get into non-standard configurations, I tend to lean toward the more predictable solution (which is cropping).

In an article in Nature Photographer magazine 2 years ago, bird photography icon Art Morris claimed that a 1.4 extender reduces sharpness by 14%, and the 2x extender reduces sharpness by 28%. I'm not really sure if this means that those percentages are true for use with all lenses and in all situations & conditions, but if Artie says it, I would lean toward believing it.

In my own experiences, using either of the extenders on my 70-200 f4 or my 100-400 . . . did not produce a desirable outcome.

Using either of them on my 400 f2.8 is far, far better than not using them, and cropping. In fact, with the 1.4 on that lens I cannot notice any degradation in image quality whatsoever, and I am quite the pixel-peeper! The 2x usually does result in hair or feather detail that is not as sharply resolved as I like - but will often still result in a marketable image.

It would largely depend on your uses, get a great shot of a bear and you want to blow it up and hang a 20"x30" print on your wall, cropping down to 6mp may not be so good, if they're just going to facebook, it likely won't be an issue.

Also to consider is the quality of the TC, I've got a Sigma that I'm pleased with that has little effect on IQ, then I've got a Quantaray where IQ suffers but my 7D can AF with it at f8 because it doesn't report.

Then there is the fact that some of Canon's lenses are designed with the intention of the end user coupling TCs to the back of them to get the most out of the lens, and IQ is again almost unchanged.

Several places have the Canon MkIII TC's available for rent, if you wanted to try them or have one for the purposes of your trip.

Understand the limitations of each method and choose what will work best for you. The nice thing about photography is there is often times not just one right answer. I personally do tend to favor getting as much of what I want the final image to look like in my head done in camera as I can.

I think the 1.4 teleconverter is the better option but be sure to check focus micro adjustment with the TC in place.

Definitely agree with keeping 2x out.

Then don't forget about the power of sensor. Though glass speak for most of job, its not all that you get by virtue of glass alone.

not sure how long you are going to be there...but renting could be an option...you could rent a 500f4IS from borrowlenses.com for a week for $450...