My F4s was made around the beginning of 1989; I bought it in February 1999, in absolute mint condition. It looked as if it had never been used, still in its box with an immaculate instruction book.

Due to the fact that it's an early version (about 4 months after the very first off the production line), it has none of the numerous modifications which were made to later bodies. These are only minor and mainly cosmetic, though, so it's never bothered me.

I think the handling is superb. It is chunky and is certainly no lightweight, but it fits really well in my hands and is very comfortable to hold; I especially like the vertical grip with its extra shutter release as I tend to take quite a few pictures in the vertical format. When I prefer to use a lighter or smaller single-digit F camera, I can use my F3 or replace the MB-21 with the MB-20 to turn the F4s into the smaller, standard F4.

Why do I like this camera so much? After all, its technology is old by today's standards and it can be a little awkward to carry around day to day. The answer is that it oozes quality. It is a very solid piece of kit and is a joy to hold and operate. I love the traditionally styled controls: no thumbwheels or fiddly little mode buttons here. To be honest, my initial thoughts were that an F4 (or F4s) may be too big and heavy; thoughts which quickly disappeared the moment I picked one up. Ergonomically the F4s is, for me, close to perfection. 

Although it is an autofocus body, it has been built with manual focus lens owners in mind. Until the release of the F6, in late 2004, this was the only Nikon film body which allowed matrix metering with manual focus lenses. (This is not taking into account the FA with its multi-pattern AMP metering - the precursor to matrix.) All other film bodies with matrix metering require lenses with a CPU for matrix to operate, such as all AF and AI-P lenses, whereas the F4 will happily matrix meter with AI and AI-s lenses. I've found that matrix metering works equally well with both AF and MF lenses. When I want to feel more in control, however, I use spot or centre-weighted metering. The former has proved invaluable in tricky situations whereas the latter can be used in more general situations when one wants to know what the exposure system is doing. It will never be known exactly what the matrix meter is exposing for (unless the reading is compared with spot readings of various elements - unlikely in general use as this would negate the whole point of using matrix) but good old 60/40 centre-weighted metering (cf. 80/20 of the F3) is what I and many others used when first learning photography. I have used the matrix meter a lot, though, and have had no real disappointments; the trick is to try to determine the situation where the matrix meter may have trouble and then switch to either spot or centre-weighted. Even so, I have been surprised at the accuracy of the matrix meter in some situations I deemed tricky, where I also used spot metering.

Incidentally, I've found that the switch used to select the metering pattern is fairly easily moved during normal carrying as the finger catch is rather large and sticks out a fair way, especially when matrix metering is set. This had been addressed in later models which have a smaller, less obtrusive finger catch.

Many people slate the F4 for lack of speed of its autofocus. OK, so it doesn't always snap into focus in the blink of an eye, but it's no slouch as far as I can tell. A friend of mine had an F90x, so I put the AF to the test. Using a 70-210 mm f/4.0 set at around 210 mm, I focussed on a near object then swung between that and another, several feet further away. I didn't use a stopwatch, but for all practical purposes, the F90x seemed to have only a slight speed advantage. What was more obvious though, was that the F4 would sometimes hunt, whereas the F90x didn't at all in my test. I am more than satisfied with the speed of AF on the F4, but care is needed to place something with sufficient contrast within the focus brackets, otherwise it will hunt and/or fail to focus at all. This has only occasionally caught me out, but it is usually very easy to move slightly so that the AF locks on (keeping it locked and re-framing if necessary). The F4 is, of course, a superb camera for manually focussing. The electronic rangefinder for MF assistance will obviously fail under the same conditions that AF fails but, in any case, the screen is bright and easy enough to focus on.

A later note: The above was written some time before I got the F5 and F100. Obviously I no longer need to borrow an F90x in order to compare AF speeds and abilities to lock-on. The F4 is not able to lock-on as well, or at all, to some objects that the F5 and F100 have little trouble with. When talking about AF speed in relation to pure driving speed together with possible delays due to "hunting" then the F4 is undeniably slower than the F5 or F100. But when comparing the rates at which the F4s and, say, the F5 drive a (motorless) lens focusing mechanism, i.e., pure driving speed with no hesitation due to failure to lock-on immediately, there's not much difference. The F100 is slightly slower than the F5 in this respect (though still slightly faster than the F4s), probably due to its 4 AA cells as opposed to the higher torque resulting from the the more powerful focus drive motor and 8 AA cells in the F5. When using my F4s in AF mode in real-world situations, I still don't feel let down by its speed even though I'm used to the F5 and F100. In addition, the F4 is quite happy with AF-S ("silent-wave" motor) lenses, which shows some forward planning, since AF-S lenses didn't exist when the F4 was introduced. This closes the gap still further and I find that I can barely tell the difference in AF performance between the F4 and F5 when an AF-S lens is used, unless conditions become too difficult for the F4's AF sensor and it begins to hunt. Not bad, considering the rate of development of AF hardware and software during the eight years between the introductions of the F4 (1988) and F5 (1996).