Retina wrote:
...Have you found that cropping with digital has largely replaced their use?
Not at all.
In fact, teleconverters from the film era were mostly much lower quality than TCs today. Over the years I tried a bunch of different ones back in the good/bad old days, but ended up only satisfied with and using one particular 1.5X with one or two specific prime lenses. Most other TC and lens combos were pretty disappointing.
Today I use TCs a lot more often. High quality 1.4X and 2X, mostly with 300mm and 500mm prime lenses. But also sometimes with 135mm and 70-200mm. Recently I got a 100-400mm zoom, too, but I haven't had a chance to try it with 1.4X yet, even though it's said to work pretty well and my current cameras are capable of auto focusing the combo.
Cropping is no substitute for teleconverters (BTW, it was nearly as easy to crop film during the enlarging process, as it is to crop digital images in post-processing). There's much more loss of image quality due to cropping.
Better solutions than cropping are to get closer, or use a teleconverter, or if using a full frame camera you might switch to a "crop sensor" camera, which has some "free teleconverter" effect. By that I mean that when used on a crop camera the lens "acts" longer, but there's no penalty of lost light and/or reduced resolution, the way there is with an actual teleconverter.
Let me give you an example. We did a careful comparison a while back, with a Canon full frame camera (22MP 5D Mark II) and a Canon 1.6X APS-C model (18MP 7D). We took the same shot from the same place with the same lens... then cropped the full frame image down to match the APS-C image. Essentially, that's the same thing as cropping an image to emulate a 1.6X teleconverter, right?
By the time that's done, only about 8MP remained of the the 22MP FF image... while the 7D's 18MP image was still 18MP. The uncropped image from the APS-C camera was clearly superior to the cropped image from the FF camera.
What's happening in this example is that the difference in sensor
area is more greatly reduced than you might initially expect. Same thing will happen if you crop to emulate a teleconverter's effect.
A full frame sensor is 24x36mm or 864 square mm... while a Canon APS-C sensor is about 22x15mm or 330 square mm. So only about 38% of that original FF image remains after cropping to match the smaller format. 38% of 22MP is 8.36MP. Personally, I'd rather have an 18MP image!
Similar would be true cropping any image to emulate the effects of a teleconverter. It would mean a little less loss with a 1.4X TC, slightly more for a 1.7X TC or a lot more cropping to emulate a 2X TC. In fact, cropping to the same image rendered with a 2X would leave little more than 25% of the original image data! Rather than a 22MP image made with that 5D Mark II camera and lens fitted with a 2X TC, cropping the image to the same angle of view you'd end up with an image of less than 6MP! Image quality would be greatly reduced, as a result.
So, in other words, your digital images will probably be a lot better using a teleconverter instead of cropping images
Yes, there's some loss if image quality whenever a TC is used, too. That's just the nature of optics... when you add a bunch of elements, there's going to be something lost from the light that has to pass through them. However, exactly how is lost much varies greatly depending upon the quality of the lens, as well as the strength and quality of the TC, plus how well the two complement each other. There are lots of variables. But in general 1.4X "cost" less image quality than 1.7X or 2X and teleconverters work best on prime lenses, as opposed to zooms. And, almost always, teleconverters work best on telephoto lenses.