If some shady lens genie appeared and offered you a choice between these two lenses you'd be mad not to go for the one on the left. Not only could you sell it and buy five 50mm primes, you'd also gain a range of focal lengths and a pretty gold ring. However if lens genies exist they're a well kept secret, and not all of us are free to rush out and spend $1500 on a new piece of glass. I guess what I'm actually arguing here is that the prime produces results comparable to the zoom at 50mm and occasionally outperforms it. I don't want to say that gear doesn't matter, just that it matters less than you might think.
I'm going to show you a series of images taken by both the prime and the zoom. I've used the same settings*, same body and have attempted to replicate the field of view as closely as possible despite the barrel of the zoom lens being almost twice the size of the prime's.
Here's a 1:1 centre crop of a dictionary apparently printed before the invention of digital sensors. Both time the central focus point was placed above the 'ph' of photograph. They seem about equal, but pushed for an answer I'd probably choose the 50mm. Bear in mind the zoom is wide open here whilst the prime is 1.5 stops away from it's maximum aperture of f1.8.
The zoom wins the corner competition hands down achieving not only an appreciably sharper image, but also a more contrast-y image when compared to the prime. I'm not going to go all Kenny R on you and spout off about how little the corners matter - sometimes they do and if that's the case then you're going to have to appreciate how difficult it is for lens manufacturers to maintain corner sharpness and be prepared to stop down your lenses.
Again we're fixed at 2.8 here, so the prime could go blow the background out a stop-and-a-half more, but we're not interested in the sheer background destroying capability of the lens, but the manner in which it does so. The prime will - at f1.8 - blow out more of the background than the zoom - at f2.8/50mm - but the prime won't do it as smooothly, and that can be a problem.
Look at the points in the pictures above where different colour groups intersect and look at the murky tones that are apparent in the right hand shot but not the left. The zoom has 9 aperture blades to the prime's 7, and this might have something to do with it. I'm convinced that no manufacturer knows quite how a lens' bokeh will look before the lens is produced, but there are certainly types of bokeh that are preferable to others. The 50mm loses out to the zoom, but by a shallow margin, and at least we're not dealing with onion rings or double lines.
This incredibly high-tech test will look at the time it takes each lens to hunt from infinity to its minimum focusing distance and back again, and then from infinity to an high contrast object placed about a meter in front of it. You might need headphones.
You might be surprised at how well the prime performed here. The zoom just about wins the focusing but I doubt anyone expected otherwise, the real question is wether or not the zoom is 'too' slow. Stationary subjects won't be a problem, and you might have to defer to the zoom if you're looking to cover fast moving subjects, particularly in low light. Ninety percent of the time the prime will see you through.
What does this mean for owners of either lens? My purpose was to show that the areas in which the zoom wins outright are not always related to image quality but are more to do with usability.
Let's sum up the benefits of each lens whilst assuming comparable image quality:
Nikon AF-S 24-70mm f/2.8 G ED
- Variable focal length
- Durable & weather sealed lens design
- Gold ring
- internal focusing
- commonly used filter thread (77mm)
Nikon AF-S 50mm f/1.8 G
- Costs $1300 less
- Weighs 700g less
- less bulky
- People in the street won't consistently mistake it for a telephoto
- can be replaced if broken
- did I mention the price?
* The Aperture was fixed at 2.8 for both lenses throughout the test, the shutter speeds and iso are constant between each comparison and the picture control was set to Neutral. The images here are OOC jpegs without any distortion control, vignette control or sharpening (either in camera or LR based). The difference in exposure is therefore down to T-stops and fluctuations in natural light coming through the window behind me.