First and foremost, from the photographic point of view fast 50mm primes are extremely useful tools, and we think they should be of interest to all serious shooters. And just for once it seems that the manufacturers might agree with us, with no fewer than three new F1.4 models announced this year (from Sigma, Nikon and Pentax), suggesting that the fast prime hasn't yet been killed off by the ubiquitous slow kit zoom. One important point here is that the existing reviews give us an essential reference point for testing these new lenses; for example having reviewed Nikon's 50mm F1.4D, we'll be able to make a full comparison with the new 'G' version in a way that would not otherwise be possible. Also the impressive aspheric-design Sigma finally offers an interesting alternative to own-brand optics for all DLR users, but we couldn't demonstrate that properly by reviewing it in isolation.

But there's more to it than just that. We actually have something of an ulterior motive for these tests; clearly the latest round of 14-15Mp APS-C sensors are placing huge demands on lenses, and it's become increasingly apparent to us that the 50mms that we use for reviewing camera bodies are in danger of becoming stretched  (in context a 15Mp APS-C sensor requires the same resolution from the lens as a hypothetical 38 Mp full-frame sensor). So we wanted to have a point of reference in this regard, to see whether we can measure any significant differences between our stock 50mm lenses, and to guage how much further they can be pushed.

To be honest, the results haven't really surprised us. It's not exactly a great secret that 50mm primes are some of the very best lenses available, performing to an extremely high standard in the F4-F8 range, and that's exactly what we've seen. In fact the traditional 7 element/6 group F1.4 designs have all turned out to behave very similarly indeed at the working apertures used for our camera reviews, with the Olympus 50mm F2 macro (which we've always considered an excellent lens) putting in a stellar performance in the studio. We hope that when we've published all of the results, these tests will help to answer the arguments we hear from time to time that we should use the 'same' lens for all camera reviews. It really isn't necessary, as fast standard and short telephoto primes are all excellent performers, and fundamentally these lenses are more similar than they are different.

But having said all that, we still need to consider the general question of how these lenses behave on the latest round of sensors, and it certainly appears from the MTF50 graphs that they are starting to approach their limits (and remember these are some of the sharpest lenses made). This highlights possibly the biggest issue facing manufacturers today; ever-increasing sensor resolutions are simply not being backed up by lenses capable of delivering enough detail, and this begs the question of what all those extra megapixels are supposed to achieve (other than take up storage space and slow the camera down).On the whole, we'd rather see manufacturers directing their R&D resources towards improving sensor efficiency, to give reduced noise and increased dynamic range, rather than continuing the megapixel marketing mania.

Addendum:

One of the bigigest dangers in constructing a quick blog post is that of over-abbreviating your arguments. The 50mm lenses we're testing *are* still capable of providing sufficient resolution to those 14-15Mp sensors (albeit at reduced contrast) but the bigger problem is that many of the consumer-level zooms most commonly used (e.g. 18-55, 50-200, 70-300, and 18-200 class lenses) are looking distinctly stretched. With these optics, those high res sensors simply describe the same slightly soft output with more pixels.