Image stabilization: Sigma 100-400 vs Nikon 200-500

Started 3 weeks ago | Discussions thread
ForumParentFirstPrevious
Flat view
lokatz
lokatz Contributing Member • Posts: 559
Image stabilization: Sigma 100-400 vs Nikon 200-500
2

In another thread , I shared my sharpness testing results between the Sigma and Tamron 100-400 versions, concluding after testing several copies of both lenses that the Sigma is quite a bit sharper.

One weak spot that is often quoted for the Sigma, however, is that its OS/VR/.. (whatever you prefer to call it) is poor. Actually, I stated that myself, in spite of not having evaluated this in detail. Since I wasn’t comfortable with that omission, here is now the result of more extensive testing of these lenses' image stabilization performance that I did to either prove or disprove the point. Since I no longer had the Tamron on hand, I instead pegged the Sigma against my Nikon 200-500, which in my view is a more interesting comparison anyway.

The Test:

I took a 'perfect' picture of a target that is about 65ft/20m away, the way it should be done: on tripod, OS/VR switched off, etc. Next, I shot a series of 50 hand-held images each at seven different shutter speeds (yes, that's 350 test shots per lens), from 1/20s to 1/1250s, all from the very same position. All of these were shot at the maximum length (400 resp. 500mm), with the lens wide open (f/6.3 resp. f/5.6) and ISO set to Auto. Both lenses were tested on a Nikon D500, which has a crop factor of 1.5. For comparison purposes, I also conducted the same tests with the Sigma 100-400 on a Nikon 1 J5, which with a crop factor of 2.7 gives it an effective focal length of 270-1080mm. (The J5 does not work with the Nikon 200-500.)

For each lens/camera combination, I analyzed each of the 350 shots and put them in one of four categories:

  • Excellent: Except for differences in ISO, the shot is virtually non-distinguishable from the tripod shot and is equally sharp. Between different 'excellent' shots, there is very little variance, if any.
  • Good: The shot is sharp and usable, just maybe a quarter-notch below perfect in terms of sharpness when compared to the tripod shot. While small variances exist between different 'good' shots, all of them are still keepers.
  • Fair: The target (a sticker with different-sized lines of text on it) remains easily legible and traces of motion blur are still minimal, the shot is not super-sharp. Variances between different 'fair' shots exist, though none are huge. Unless this was a once-in-a-lifetime shot, you probably would not keep it.
  • Poor: The text on the target becomes hard to read and motion blur indicates that the camera has been moved in one direction quite a bit while shooting. Variances between different 'poor' shots are significant. This is definitely not a keeper.

Here is the graphical summary. The X axis always shows the shutter speed, the Y axis the number of shots in the four categories, in total 50 per shutter speed setting.

Conclusions:

  1. On the D500, the Sigma stabilization works well, with 80+ percent of the shots being green = keepers at around 1/125s or faster. According to the old rule of thumb (inverse of focal length), this equals a stabilization of a bit more than 2 steps (1/125s vs. 1/600s). The Sigma’s results on the J5 indicate the very same number. Here, the 80+ percent limit is at about 1/250s, which again means a stabilization of slightly more than 2 steps (1/250s vs. 1/1080s). Shooting at slower shutter speeds becomes a lottery.
  2. The Nikon stabilizes more effectively on the D500, with 80+ percent keepers at 1/80s or faster. In its case, the stabilization offers a little more than 3 steps (1/80s vs 1/750s). In other words, it is about one full step better than the Sigma. What I found even more impressive is that at 1/20s = more than 5 steps, it already gave me a better than 50 percent keeper rate.
  3. At 1/320s and above, I’d say both lenses stabilize equally well on the D500.
  4. A personal observation: it seems to me that the Sigma does a better job in actually stabilizing the shot than it does during the focusing phase, where the viewfinder image is far more jittery than with the Nikon lens and led me to believe the difference between the two was bigger than it actually is.
  5. Relying on my different evaluations of the Tamron 100-400 without having tested its stabilization in detail, I'm nevertheless comfortable to say I think its VR performance is about half-way between the Sigma and the Nikon, maybe a little closer to the latter.

Results will vary some across different tests. I am sure some of you have a steadier, or vice versa, less steady hand than I do. Since I was taking so many shots, fatigue certainly also influenced this, so your keeper rate will probably be a little higher if you take less shots. Nevertheless, the difference between the lenses would likely remain about the same in different tests.

As always, interested in your thoughts and observations.

.

Side note: The above graphs confirm that the method I used to do my sharpness testing of the Sigma and Tamron (see thread link at the beginning of this post) was perfectly valid, in spite of drawing lots of flak there.

 lokatz's gear list:lokatz's gear list
Panasonic ZS100 Nikon D700 Nikon D7100 Nikon 1 J5 Nikon D500 +17 more
Nikon 1 J5 Nikon 200-500mm F5.6E ED VR Nikon D500 Sigma 100-400mm F5-6.3
If you believe there are incorrect tags, please send us this post using our feedback form.
ForumParentFirstPrevious
Flat view
Post (hide subjects) Posted by
ForumParentFirstPrevious
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow