imagemeister wrote:
...( can anyone actually see 5% degradation ?? - IF it exists)...
It exists. And if you can't see it, you aren't looking close enough:
http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=822&Camera=614&Sample=0&FLI=4&API=0&LensComp=822&CameraComp=614&SampleComp=0&FLIComp=5&APIComp=1In this case the differences are rather subtle, but some small loss of contrast and sharpness can best be illustrated with test shots like those at the link above.... One can expect that all images taken with the lens vs lens + TC will show similar differences, though it may be even harder to see.
5%/15%/25% loss is a just generally accepted guideline or rule of thumb. There are so many different possible lens/teleconverter combinations that may perform a little better or worse. Maybe or maybe not you can find someone using the exact combination you're considering... But even if you do, the range of variation between copies of specific lenses and teleconverters, and the way people use them are bound to vary, making it a bit of an educated guess how well the teleconverter you purchase will work with your particular lens.
Even test shots like those at the above link only give us limited information. We know what model lens was tested and the camera model it was used upon... But we don't know if they both were "good copies" that were as perfectly calibrated as possible. Also we don't know the specific brand or model of 1.4X that was used or if it too was perfectly calibrated. And we can only hope that the testing was done using optimal technique and have to accept that it's only being done at the specific distances dictated by framing the target used.
All this makes it difficult to give recommendations and predict how a particular combination will work. But it's the best information we've got.
In general... weaker teleconverters cost less loss of image quality, while stronger ones will cost more. It's
estimated that a 1.4X will "cost" about 5%, a 1.7X will reduce IQ roughly 15%, and that IQ will take approx. a 25% hit with a 2X. They are much less common, but if you look hard enough you even can find 3X... though I'd estimate the loss of IQ with them is close to 50% and there I don't know anyone who would use them.
In general... teleconverters work better with primes than with zooms.
In general... the higher quality the lens, the better it tolerates being used with a teleconverter or the less noticeable IQ loss will be.
In general... teleconverters work best with telephoto lenses.
In general... it will nearly always give better image quality to use a longer focal length lens than to use a shorter one with a teleconverter to achieve similar focal length.
It's easy to say "check out Tamron and Kenko teleconverters" or "only use OEM teleconverters". But that still leaves a lot up to chance because there are different qualities and models of TCs from all manufacturers. For example, Kenko currently makes two models of 1.4X: a more expensive Teleplus 1.4X Pro 300 DGX and a slightly lower priced Teleplus 1.4X MC-4 DGX. Both are considered very good, but the MC-4 is a little sharper in the center, while the Pro 300 tends to be sharper across the full frame, edge-to-edge and corner-to-corner. Someone using it on a crop (DX) camera might be happy with the MC-4, while a full frame (FX) user probably would want to use the Pro 300 version. I don't know about Nikon, but to date Canon has made three versions each of their own 1.4X and 2X teleconverters. (Canon actually calls theirs "Extenders"... which has made for some confusion with macro extension rings at times.)
Image quality aside, adding teleconverters also effects auto focus performance and the loss of light makes for a dimmer viewfinder. So the native aperture of the lens the TC is being used upon can be important, too.
With the effectively increased focal length, getting a steady shot is also made more difficult. Image stabilization can help, just as it does when not using a TC.
Another nice thing about TCs is the closest focusing distance of the lens doesn't change, so the magnification potential of the lens is increased by the TC's factor, too.
Ultimately, all you can do is ask for opinions about very specific lens/teleconverter combos, search out online reviews and look at any test chart examples you can find. Then - so long as none of those have scared you away - take a bit of a risk, buy the TC and try it on your lens, to see for yourself if the results are satisfactory for your particular uses.
Personally, I use Canon gear, including their 1.4X Mark II and 2X Mark II. I use both on 300/2.8 IS and 500/4 IS lenses (both "Mark I" ), without much concern. I also use the 1.4X on 300/4 IS fairly often and more occasionally on 135/2L. I've very infrequently used the 1.4X on 70-200/2.8 IS ("Mark I" version)... IQ and AF performance are both okay, I just don't have much need for it because I usually have a 300mm lens with me. I also have 70-200/4 IS, but have only had it for a couple years and can't recall ever trying it with 1.4X.
Canon uses fluorite elements in a lot of their highest quality telephoto lenses, and those typically tolerate teleconverters the best. For example, the latest 70-200/2.8 IS Mark II now uses fluorite and gets very good reviews with most 1.4X, and even quite good reviews with the 2X Mark III specifically. My older version of this lens doesn't have a fluorite element and the loss of IQ to any 2X I've tried has been unacceptable to me. Among Canon lenses, the 70-200/4 IS, the new 100-400 Mark II and 200-400/4-1.4X zooms and most of the super telephoto primes use fluorite.