dpreview lens tests are great! thanks phil & andy

Started Oct 22, 2008 | Discussions thread
studio5photo Contributing Member • Posts: 770
Re: Focusing

Thanks for the reply. I do enjoy seeing the reviews on lenses (from both places). I find it interesting that we spend so much time on cameras, but if you ask any photographer, they will all say the same thing, make sure you have good glass (or spend your money on good glass first). One reason I asked about focus speed is the Nikon 85mm 1.8 lens (which I just purchased on ebay). Some people say it's slow, others say its fast. I don't have the assets to go out and buy lenses to see how fast they are, I have to rely on other's opinions, but in many cases these opinions can be all over the place and it would be nice to have a more scientific approach with scientific results. I did it for the lenses I owed, but that was for my camera and my lenses only, under one set of lighting conditions. It was very time consuming since I had to average several trials to get a good indication, so I do understand the difficulty. By the way, the results were suprising to me. Lenses people said were fast (AF-S ones), really weren't and some of the older lenses like the Nikon 70-210 F4.0 weren't nearly as slow as people thought, in some cases less than 10% slower.

Dave Etchells wrote:
Hi Andy -

Congrats on your own testing work, I know how much work is involved
in it!

Andy Westlake wrote:

Strangely enough I have read this article in great detail, and much
as I applaud Dave for providing this level of information, I'd also
point out that their use of DxO gives them a different kind of
problem to us.

The DxO test chart is essentially a grid of dots, with no other
features, which makes it remarkably difficult to focus on properly
(either in auto or manual). This means that SLRGear are essentially
forced to work in the way they describe by the very design of their
test (not to mention that fact that they started well before live
view was a common feature).

Thanks for pointing out the need for clarification in my Fallibility
of Focus article: In actual fact, when I was running the
focus-accuracy tests, I was using a target with elements in it
designed to allow very fine focus discrimination; a USAF target chip
at the center of the frame. The progression of very fine lines on
this target provide the best possible target for manual focusing
(that I can think of, at least), as you can discriminate focus very
finely based on how well the clusters of lines are resolved. Even
with this absolutely optimal target, though, and even with a
magnified Live View display, the level of error was still more than
we were willing to tolerate. - And I even used about a 3x magnifying
glass to inspect the rear-panel LCDs on the Live View cameras I
tested. (It is true, though, that a magnified Live View display did
better than any other manual method.) This is why we went with the
"focus sweep" method, not because the target wasn't conducive to
accurate focusing: There was no point adding elements to the target
(which we found we could in fact do with Analyzer), because any sort
of manual focusing still left our results more prone to error than we
were comfortable with.

Thanks for pointing out the unclarity; I'll try to get back to that
article in the next day or two, to make it clear what I was using to
focus the cameras on.

I also wish they would/could test
focus speeds. I know that's related to the camera as well, but if
there was some kind of metric they could come up with that would be

Coming up with a metric is easy.

Coming up with a metric which helps in any way is a whole different

I agree with Andy here: There are such differences between camera
platforms, even within a single manufacturer's line, that it's very
hard to come up with an indicator of performance that would be
applicable across the board.

We do, however, commonly report on how long it takes the lens to slew
from infinity to closest focus, as that's an indication of how fast
the lens elements can travel. We do this on a pro-level body, so the
transition time is about as good as it can get. Beyond that, YMMV (by
a lot) depending on the camera body you're using, and how the lens
interacts with your body's AF system. - But the measure of time to
transition from infinity to closest focus does give a pretty good
general idea of how fast a lens' AF drive is.

  • Dave Etchells

-- hide signature --


Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow