Thursday, January 17, 2013

$2000 lens vs. $200 lens: Real World Comparison

This is a follow up post to my post about the depth of field of full frame kit lenses.  In that post, I mentioned that my coauthor MShafik's downsizing (pun not intended) got me to thinking about my gear.  One potential area for savings is my Nikkor 24-70 2.8G.  

The 24-70 is a great lens.  It meets all my needs (other than not having stabilization) and has no significant weaknesses from my point of view.  On the other hand, it is quite expensive (around $1900 new).  

Meanwhile my old Nikkor 28-105 3.5-4.5 AF-D covers a similar range and more, has a not-too-slow aperture, is sharp, and costs about one-tenth as much.  

I'm not really interested in a brick wall comparison.  I'm more interested in their real world usage.  The most significant tradeoff for me in terms of specs is the aperture.  (Incidentally, the 28-105's aperture is the same as that of the new 24-85 3.5-4.5 VR.)  I was most concerned whether the depth of field would be sufficiently shallow.   In addition, I was concerned with whether the lens is usable in low light conditions.  Finally, I wanted to get a high enough shutter speed for a sharp shot without getting too much noise from being forced to use high ISOs.

(Note: for all shots used in this post, you can zoom in on any picture up to 1600pix by clicking on it.)
First, here are my observations regarding my primary concerns:

At longer focal lengths, the 24-70 can get noticeably shallower depth of field than the 28-105.  However, the 28-105 can get what I consider adequate separation from the background.

f/2.8 at 70mm on 24-70

f/4.2 (wide open) at 62mm on 28-105
I like using a high shutter speed to get sharper shots.  The slower aperture of the 28-105 forces me to use a higher ISO than what I would otherwise use with the 24-70.  Nonetheless, because of the D600's superior high ISO performance, the increase in noise is not significant in daylight conditions.

f/4.5, 1/320, ISO 1400

In low light, it's a different story.  With the 24-70 I could get away with not using flash in dim indoor lighting.
The light here was very dim but I was able to take the shot without flash. f/2.8, 1/50, 25,600. 24mm.

With the 28-105, I am pushed to use high ISOs in brighter conditions.

When flash does get used, the 24-70's f/2.8 aperture comes in handy.  At longer focal lengths, it is like having one stop extra flash power compared to the 28-105.
Bounce flash with 24-70.  f/2.8, 1/200, ISO 2800

Bounce flash with 28-105.  f/4.0, 1/60, ISO 12,800

Here are other my observations (some already noted in previous posts) in no particular order:

1.  Chromatic aberration.
The 28-105 has a significant amount of chromatic aberration, whereas the 24-70 has little, if any.
Personally, this doesn't bother me.  First, Lightroom can significantly reduce the appearance of chromatic aberration.  Second, my audience does not care at all about CA.  That includes me.

Check out her legs.  (Did I just say that?)

2.  Bokeh.
The 24-70's bokeh is reasonably smooth and rarely distracting.  The 28-105's bokeh on the other hand can look pretty harsh sometimes.

When the background is not so far, here is how the bokeh looks:

Here is a shot from the 24-70.  Note the rendering of the points of light (sunlight through the leaves).

Here is how points of light are rendered by the 28-105:

4. Flare resistance
First, this is not a very fair comparison because the 28-105's hood is not well designed (it's very wide and doesn't do a good job of blocking light sources).  Besides that, the hood takes up a lot of space because of its width.  I don't even bother to bring mine.  Caveats aside, the 28-105 seemed to be genuinely prone to flare, much more so than the 24-70.  The 28-105 had the occasional blobs of light as well as the kind of flare that reduces the contrast of the image.
Gigantic flare spot

Reduction in contrast caused by flare (plus flare spot)
The 24-70 has much better flare resistance, though is not immune (see yellow/green flare spot on lower right of frame)
5. Autofocus performance.
In daylight I could see no difference between the AF speed and accuracy of these two lenses.  On the D600 they are both very quick (note that the 28-105 doesn't have a built-in motor therefore its AF speed depends on the camera body).

At night, in dim lighting, the 28-105 sometimes had some difficulty locking on unless I used the center AF point, at which point it could lock focus without problems.  I have not yet used the 24-70 at night so I can't comment on that yet.  However, I have used the 24-70 in very dim indoor conditions as shown above and the camera did not hunt for focus.

6. Sharpness.
I have not done a side-by-side sharpness comparison between the two but at my usual viewing sizes, the 28-105 does not lag behind the 24-70 in terms of sharpness.

28-105. 80mm at f/4.5, 1/320, ISO 250.

As noted above the 24-70 is better than the 28-105 in a number of ways.  Whether the 24-70 is worth the extra cost is a personal matter.  As for me, I'm undecided, though with the right opportunity I may sell the 24-70 and keep the 28-105 just because this is only a hobby for me.  Note also that many of the 24-70's advantages are due to its constant f/2.8 aperture.  In that regard, I am curious about the Tamron 28-75, which I used to have but have not yet tried on full frame.

Here are shots from the 24-70:

Here are shots from the 28-105 (note: these have been sharpened in postprocessing):

Grips of steel

A Day With my Precious Ones (and my kids too): D600 + Nikkor 24-70
Cheap, Sharp FX Standard Zoom: Nikkor 28-105 3.5-4.5D
The Truth Behind the Migration

Nikon D600 Resource Page (see under "Lenses Tested")