I purchased an inexpensive 2x teleconverter, and put it to the test. As with my first teleconverter test, I've read online that using a TC is no better than taking a normal photos and resizing it to make the same increase in apparent size. This is usually accomplished using a paint program, although there are advanced software programs that do a better job. My goal was to see if, in general, the claim holds true for the method that most people would use (bicubic interpolation).
I took photos with and without the teleconverter. For the below photos, I found two that were roughly equivalent, and I processed them similarly (but not identically) with a bit of noise reduction. Then, I increased the resolution of the non-TC photo to match the 2 times increase in size of the photo with the TC, and cropped both for web viewing.
First, I'll show both at "50%", where this would be the normal crop for the non-TC photo. The first photo is without the TC, the next one with.
(If your browser renders it correctly, the first photo should be very close to what the original non-TC photo looks like.)
Next, the same two photos at full size. If your monitor is not large enough to hold them side-by side, I'm afraid they'll land one after the other.
You can see that with resizing the first photo, it also enlarges the noise. I could increase the noise reduction, but the point of the exercise is to preserve as much detail as possible, not necessarily compete on the basis of noise. However, it is an interesting effect to have ISO 200 look noisier than ISO 400; if this turns out to be a consistent effect, it means you can more easily justify the use of higher ISO when using the TC (using the reasoning that you'd otherwise increase the noise anyway, as you cropped and enlarged to attempt to get the same view). Although, I should have used ISO 100 for the non-TC photo, which would have reduced the noise. The point is that ISO 400 isn't necessarily the liability that you would assume it would be.
In practice, most of the teleconverter photos simply didn't turn out well. Most of the photos where blurred, probably due to two factors:
Another negative is the increase in "purple fringe" in the highlights.
While viewing at 100% pixels, the teleconverter photos are flawed. And yet, they manage to extract more detail, and have an advantage over cropping and resizing a non-TC photo. You can't look at this as a high-quality solution, but as a work-around for not paying for a 600mm lens and making further use of your 300mm lens instead. A true 600mm lens would offer greater quality and detail.
Last change: June 1, 2009