Is It True That The 24-70 VR is Bad At Close Distances?

Started 2 months ago | Questions thread
OP SmilerGrogan Contributing Member • Posts: 531
Where does the greatness begin???

Wow. Great analysis and thanks for introducing us to the lens designer.

On some of the assignments I shoot there isn’t a lot of time to be changing lenses so I rely on the 24-70 to do everything from product closeups to environmental portraits to outdoor landscapes. I’d love a great macro lens but I have a few priorities higher on the list.

I've owned the VR version since December but don’t have a good sense yet of its best distances. Can you tell us at what distance the lens is at its best? Does it reach greatness at distances farther than 3 feet? 5 feet? 10 feet?

BTW, If you know anyone who wants a sharp copy of the original non-VR 24-70 at a reasonable price send them my way.

anotherMike wrote:

As usual, the internet hype train arrivin the Nikon forum station...

Here's reality:

People tend to (very unfortunately) describe lenses in very binary terms: Amazing or Bad, Sharp as a tack or Soft, etc. The truth, as you might have expected by now, is much more in the middle.

I'll start the discussion with this:

  • NO lens is perfect, they ALL involve trade-offs, which the lens designer navigates to come up with the final product. Remember that sentence, as it critically applies to the 24-70/2.8 VR which I'll get to in a bit.
  • At the same time, in the modern era, there are very few lenses that I'd grade personally as outright "Bad". Even a lens I seriously don't like - the old Nikon 35/2 AFD, is still a usable lens at some apertures and absolutely capable of taking artistically valid imagery. At the same time, there are a lot better. But the lens doesn't grade as "Bad"
  • Each photographer has their own standard for image quality, and what is absolutely fine, "sharp as a tack" for someone might be just "average" for somebody else. Someone printing 30x40" prints for gallery work likely has vastly different standards for image quality and sharpness than someone who mostly prints 8x10" prints, and both would likely have higher standards for image quality than someone for whom a cell phone image is "good enough", and someone printing 5 x 9 foot gallery prints likely has much higher standards for image quality than the other guys in this hypothetical discussion, and likely isn't using a teensy little DSLR to do it either.

So, let's talk about what a 24-70/2.8 class of lens is generally used for. I'd submit that such a lens is more of an "all arounder" as opposed to a specialty use case lens. A wedding shooter would use it for it's flexibility and fast focus, a street shooter might like it's flexibility, a landscape shooter might like flexibility (and quality if it's good enough) to minimize what he/she carries on a hike, a music industry shooter would have one for flexibility, as would a dance/theater shooter, and so forth.

What does the designer do? A lens that is quite wide angle on one end and mid tele is a tough design challeng, one is most absolutely going to give up something to get something, and for sure, it won't be perfect. It's going to be horse trading. So, in the case of the 24-70/2.8E VR, designed by Hiroki Harada, who also happened to be the designer of the earlier 24-70/2.8G (no VR) released way back, likely looked at the earlier lens and thought how to improve upon it. That meant, quite likely, trade-offs were negotiated, and the result of those were realized in different aspects of lens performance.

So, having studied some of his work, he's a man who really likes balance - he's more about designing a lens that "does no wrong" than designing a lens that blows the doors off one particular area, but then s*cks wind at several others.

The end result is that he apparently wanted to make the new 24-70 VR more even in terms of sharpness across the field across a wider range of focal lengths, at longer distances, because this aids the types of shooters who are going to *realize* (remember this word, we'll get back to it) the *potential* resolution in a scene. A guy on a tripod, focused precisely, very clear atmosphere, mirror lock up has the potential to realize more of a lenses resolution than a guy hand-holding (or VR) on a moving subject. And having shot both lenses, I believe he did just that. In the horse trading that occurred, he let "go" some ultimate performance in the closer/moderate distance ranges, likely because he thought that the types of scenarios that occur in these distances for a wide variety of photographers happened to be those that were NOT going to be able to realize the potential resolution of the lens - and hence, while that lens won't blow the doors off the 2D chart test chart sites, which are all done at close distance (the distance range where he let some stuff go in order to get gains elsewhere), at landscape distance, which is what most test sites do NOT test at, it's actually an improvement over the earlier.

Now you add in the internet hype train, or peoples over-reliance on two dimensional test chart results, because they aren't educated in lens performance in real life, and now you have this "thing" that gets you to start a thread asking if the 24-70 VR is "bad" at close distances. The answer, as you might suspect by now, is "no, it's not bad, but it's not the area where the designer put most of the energy into, because he felt most scenarios in the closer ranges are likely to be handheld, and not able to realize the full potential anyway". Once you understand the tradeoffs, and the way so many sites test, now you can make a better purchase decision.

If I shot products and things in the close range and wanted ultimate sharpness, would I get this lens? No, I wouldn't. My use case in this example would be contrary to the design agenda of the lens. Such a use case frankly is better served with a macro lens, every day of the week.

But if I wanted a great all purpose zoom that happened to be good enough to be one of the best at landscape work, to cut how much I had to carry on a trip/hike, this lens would be a great option.

It comes down to matching tool to the task, and knowing enough about the process of lens evaluation and the limitations of both the test process and the design process, to do this matching the best.


PS: The 200-400 "not sharp at long distances" - similar story, although I know far less about it. People who have VERY discerning technique and a lot of experience with long glass have said it's not as "amazing" at long distance as it is closer in. That's totally different than "the lens is bad at such a such".

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow