Thanks Dennis.

Kel, that's a really informative comment. Thankyou. There is nothing like hearing from someone who owns and uses gear for much the same things you use it for yourself. Your view is particularly valuable to me because you also have a 500/4 to provide a common baseline. The way I'm thinking at present, I'd rather my second big lens was either much lighter (300/2.8; 400 DO), or else did f/2.8 for dark places (300; 400/2.8). Either that or sell the 500 and start everything from scratch. So I'm particularly interested in how and why you use the 200-400 and the 500/4 in combination. They strike me as lenses broadly similar in purpose: both f/4, both quite long, both very heavy. But unless I miss my guess there will be a method in your madness.

Oh, and those are lovely pictures. I think I'd like to marry your night heron.

_______________________________________________________________________________________________________________________________

Kel's question on reach is a good one.

If we are going to be consistent, surely we should treat all lenses the same way. I think it's perfectly reasonable to treat the 200-400 as a 560, but only if we do the same with the others - treat the 500/4 as a 700, the 400 DO as a 560, and so on. Does it really matter whether the teleconverter is internal or external? Not much. Granted, the made-for-exact-purpose internal TC of the 200-400 is said to be a particularly good match to the lens - Kel's comments reinforce that view, which I have also heard elsewhere - and granted also that there is a significant convenience factor: messing about with plugging and unplugging lenses in the field is always an awkward business. But where do we draw the line? Why not treat the 600/4 as a 1200/8 given that we can use a 2.0 converter? Or the 70-300 as a 600/11? I think the answer has to be based on what we would actually use in practice and get good results with.

Probably we will all have different answers. For me, the effective maximum length of a lens is primarily determined by aperture. (Sharpness counts too, of course, but can largely be accepted as a given in this company. We are not considering cheap 75-300s which go blurry anywhere over 200mm.)

I've owned the 500/4 for 11 years now. For the first year or two I mostly used it with a 1.4 converter (700mm f/5.6) and sometimes with a 2.0 converter (1000mm f/8) but as I gained experience I slowly moved away from that habit. I stopped using the 2.0 entirely, having decided that I got equal or better results with a 50D or 7D at 700/5.6 as compared to the 1D III at 1000/8, and eventually sold it. And, little by little over the years, I find that I'm also using the 1.4 less and less. These days, I probably use 1D IV and 500/4 (bare lens) something like 60% of the time, with the remainder split between the 100-400 and 500 with 1.4. (I also used to switch to 7D and bare-lens 500/4 for reach sometimes instead of mounting a converter on the 1D. Now that I have a 7D II I may do that more often - too early to say yet.)

Anyway, when it comes to reach, I draw three lines in the sand.

  • The ideal line, or the "this is exactly what I want" line: f/4. Regardless of the shooting aperture (which might be anything between f/4 and f/11), an f/4 lens is just so nice to use. Focus is fast and accurate; the viewfinder is bright and clear; and even when shooting at f/8 or f/11 you know there is spare aperture ready to hand anytime it's needed. I love using an f/4 lens!
  • The acceptable line, or the "it will do at a pinch" line: f/5.6. It's not nearly as nice as f/4, but in most circumstances it's still perfectly OK if the light is good.
  • I seldom go to what we might call the desperation line, f/8, though now that I have a 1D IV (instead of a 1D III) and a 7D II, that may or may not change.



Throwing in a few extra examples to my previous list of lenses, in IDEAL REACH terms we get:

  • 600mm: 600/4L IS II
  • 560mm: *400/2.8L IS II
  • 500mm: 500/4L IS II; 500/4L IS
  • 400mm: 400/4L IS DO II; 200-400/4L IS
  • 420mm: * 300/2.8L IS II
  • n/a: both 100-400s, 70-300, the various third-party zooms
  • n/a: 800/5.6


A couple of points to note.

First, while I'm happy to own and use f/5.6 lenses like the 100-400, I wouldn't dream of paying close to $10,000 for anything that wasn't f/4 or better.

Second, two of the lenses listed above (400/2.8 and 300/2.8) require 1.4 converters for this reach. Converters always degrade image quality (but usually not enough to be too concerned about) and more importantly, degrade AF speed. But by how much? Does a 400/2.8 with a 1.4 focus as quickly and accurately as (say) a bare 500/4 or 600/4?

I don't know.

Searching on the web, I get vague and contradictory answers. Even the peerless Bryan Carnathan doesn't have much to say on this. Being reduced to guesswork, I have assumed pro tem that the f/2.8s with 1.4 converters will focus slower than a Mark I 500/4 bare lens, but faster than the 500/4 at 700mm, making the 400/2.8 with 1.4 around about as useful in practice as a 500/4 - we gain 60mm but lose something from the converter - and the 300/2.8 at 420mm a bit less useful than either of the true 400/4s.


In ACCEPTABLE REACH terms we get this list:

  • 840mm: * 600/4L
  • 800mm: 800/5.6
  • 800mm: ** 400/2.8
  • 700mm: * 500/4 II; * 500/4 I
  • 600mm: ** 300/2.8
  • 560mm: * 400 DO; * 200-400
  • 400mm: both 100-400s.
  • 300mm: 70-300
  • (The various third-party zooms all seem to be f/6.3 designs. They come very close to our line in the sand at f/5.6, so we could be generous and count them at 600mm if we wanted to.)


There are several caveats to note here. The only big lens (over 500mm) not using any form of converter for this reach is the 800/5.6. There would be no possible reason for making it at all if it didn't out-perform 600/4s and 400/2.8s with converters, let alone spending more than it costs to buy a new small car to buy one. I imagine that sport photographers and paparazzi use them, I really can't see it as a birding lens.

That said, I believe people often overrate the image quality difference between bare lens and converter: to my eye, the main difference in the final picture is a slight to moderate loss of contrast and a (mostly very minor) increase in CA, flare, and blur. This is only to be expected. Despite the best high-tech lens coatings in the world, every extra bit of glass in the image path costs colour and contrast, and converters add quite a few.

  • 5 extra elements: 1.4 II
  • 7 extra elements: 1.4 III and 2.0 II
  • 8 extra elements: 200-400 (the built in converter)
  • 9 extra elements: 2.0 III


It might be worth remembering here that one of the improvements Canon made to the Series II big IS primes over the Series I models was removal of the protective meniscus glass at the front of the lens. (Presumably this wasn't done only to improve contrast; they have a new, tougher coating now, making the protective glass less necessary than it was, and they also seem to have been very keen to save weight wherever possible.)

With all this in mind, let's look again at the list of "acceptable reach" focal lengths above. I've marked the two f/2.8 lenses for special consideration as they are now using 2.0x converters. I have been unable to get reliable information as to how they perform at 2..0x when compared to something like a 500/4 with a 1.4. My guess is that they are pretty good but somewhat inferior. However I have no firm knowledge of this. Test shot samples show a modest and (to my mind) perfectly acceptable loss of sharpness. What the test shots don't show is:

  • (a) How long they take to auto-focus.
  • (b) How much difference the TC makes to focus accuracy (if any).
  • (c) Whether AF copes as beautifully with difficult conditions (backlighting, confused subjects, low light) as it does through (say) a 500/4 bare lens or with a 1.4.
  • (d) Whether the lens/converter combination is as amazingly good at overcoming difficult conditions as a bare-lens prime. (That is, setting AF aside, how much do images taken with a 2.0x suffer from flare, backlighting, or dull light?)


For all of these things, pending authoritative advice or hands-on experience, I have to make an educated guess based mostly on experience with a 500/4 and Mark II teleconverters. I have assumed that a quality 2.8 with 2.0 x converter will be slower to focus than a 500/4 at 700mm but much faster than the 500/4 was at 1000mm (back when I used to use the 2.0x and 1D III); much the same as regards accuracy; marginally less capable of AF in difficult conditions; and significantly less capable of resisting difficult lighting challenges.

(Putting this last thing a different way, one of the things that sets the very best lenses apart is how forgiving they are. In beautiful light, a cheap lens is often perfectly OK and seems "just as good" as a quality one. But the very best lenses keep on delivering good results even in remarkably bad conditions. That's why you spend the extra money. I'm doubtful that even the finest of lenses will consistently give its best through a 2.0x converter.)

The 2.8s aside, there are two other lenses in the "acceptable reach" list that need special consideration: these are the two 400mm f/4s.

In theory, zooms are never quite the same as primes, and should be discounted a bit from that point of view. In the case of the 200-400 (as with the Nikkor 200-400/4) we can pretty much ignore that. Every report I've ever read says that they give nothing away in IQ or focus speed to the equivalent primes. (Kel's testimonial in this thread just adds weight to that claim.) Secondly, although I disagree with Kel's notion that the 200-400 should be regarded as a 200-560 because the teleconverter is built-in, we should recognise that the 200-400's built-in 1.4 converter was designed and built especially to suit that exact lens, and as such could be made without multi-purpose compromises. Kel regards it as superior to an ordinary converter, and I see no reason to doubt that. In short, the 200-400 at 560mm f/5.6 should be regarded as punching a little above its weight.

Actually, no pun intended, weight is a big factor in my assessment of this lens. It weighs more than a 500/4 II and almost as much as a 500/4 I, and I'm not getting any younger. If I buy a birding lens next week, I'll expect to get 10-15 years use out of it, at which time I'll be over 70. Plus it's particularly expensive even in this company and - Kel's view notwithstanding - a bit short for a 3.6kg lens. If I was starting from scratch, it would nevertheless be right up there near the top of my short list - zoom is a wonderful thing to have when birding, and the excellent minimum focus distance would be more than handy.

The other 400/4 is the DO II. The original Mark 1 400/4 DO was not a success. There were many reports that, although a miracle of advanced, lightweight technology, it was nevertheless a bit NQR. It was regarded as odd, as a bit difficult, as producing unacceptable artifacts.

The Mark II has changed all that. Canon must have put a tremendous amount of hard work and money into R&D for it. As a company, Canon has a lot of pride and really, really hates having egg on its face. After the humiliating 1D III AF system fiasco, remember, Canon spared no expense to rush out the all-new and superb 1D IV AF system; put a genuine pro-grade overkill AF system into the consumer-level 7D just to show that they really were good at AF after all; did wonderful things with the AF systems for the 5D III and 1D X; and invented the world-leading dual-pixel AF mechanism for Live View and movies.

My take on the 400/4 DO II is that they did something similar, probably spending far more for the sake of wounded pride than the accountants could possibly justify on a low-volume specialist lens. (Not without long-term benefit though: there is a 600/4 DO due out soon, using the same technology. It won't weigh much more than 3kg, will be half the length of the conventional 600/4 but much fatter, and should be a huge success. Expect it to cost 20 or 25 thousand dollars. That one's not on my radar, alas.)

All those improvemernts notwithstanding, I don't believe the 400/4 DO II is quite as robust with regard to lighting conditions as a top-class orthodox lens. If I buy one, I expect to use it in good-to-middling light and go back to the 500/4 when the going gets tough.


As for the DESPERATION LIST, why bother? It's not as if anyone is going to spend many thousands on a lens with the primary aim of using it at f/8 minimum. Let's regard performance at f/8 as a possibly welcome bonus, not the main game. In any case, such a list would be much the same as the previous one, differing only by the addition of a bigger teleconverter.

(I'll write something on the third-party lenses a little later.)