I don’t put much stock in conventional wisdom. It is not that I make a conscious decision to ignore it, but I just seem to naturally recoil from it. Sometimes that serves me well in that I don’t accept things that are wrong or useless, but other times it causes me to spend a lot of time questioning things that are obvious to other people.
One such area of conventional wisdom that I have questioned for years is the notion that you need to turn off image stabilization when you place your camera on a tripod. The rationale is that the workings of the image stabilization inside the lens will create vibration, which will show up in your image as softness or blur. But that doesn’t make much sense to me, or at least it seems that it shouldn’t be a big deal. The camera is locked down on a tripod. I have never noticed any vibrations or anything from image stabilization. Even the flipping mirror of a DSLR doesn’t have much impact on sharpness, and it seems to me any vibrations from this would pale in comparison to that.
In fact, I frequently shoot from a tripod and I very often I have forgotten to turn off image stabilization when doing so. My pictures still looked sharp, so I didn’t worry about it. After a while, I started just leaving on the image stabilization when shooting from a tripod. I never noticed any lack of sharpness. So this notion of turning off image stabilization seemed to me to be something for the pixel peepers and not something that actually has a meaningful impact on your photography.
But then again, I never tested it. So I really didn’t know. Lately, however, I have received some questions about it, so I decided to put it to the test.
Before I get into telling you all about the testing and the results, let me jump straight to the conclusion. That way you can read the conclusion and carry on with other things, or you can read on and see how I did it and the specifics of the results below.
After conducting my test, I’m forced to admit that the score is Conventional Wisdom 1, Jim 0. Turning off image stabilization does make a difference. Sometimes there is a big difference, and sometimes there is hardly any difference at all. But there is a difference. Here’s what I found:
- If you are shooting at long focal lengths having image stabilization turned off makes a big difference. I was shocked at the difference it made. In fact, when I looked at the shots with the image stabilization turned on there was so much blur I thought that I had bumped the camera or something. But I re-shot them and the results were the same.
- If you are shooting wide angle, it doesn’t make much difference. For digital display, you probably won’t be able to pick up any difference at all. Pixel peepers will notice some difference, but for normal digital display there is hardly difference. Unless you are making a gigantic print you likely won’t see a difference when printing either.
It is quite common for focal length to make an impact in image sharpness (see the Reciprocal Rule for example), and this just appears to be another example of that. I believe the reason I never noticed any difference is that I virtually always shoot wide angle when shooting from a tripod. But now I have learned my lesson and will be turning image stabilization off when shooting from a tripod.
Now let’s talk about how I went about this and the results. First, what was the test? It was nothing fancy. I simply took the exact same pictures from a tripod, one with the image stabilization on, and one with it off. Then I compared them side by side on my monitor.
I have two lenses with image stabilization, one is a 24-105 mm f/4 and the other is a 70-200 mm f/2.8. I did the test with both lenses. I took pictures indoor and outdoors. I took pictures with long and short shutter speeds. I used different apertures. I tried all sorts of things. I’m not going to make you sit through all of them, but I will show you some that demonstrate the effect and what I think is going on here.
Again, my conclusion is that turning off image stabilization when shooting from a tripod can make a difference in sharpness. How much of a difference? Let’s take a look at 3 shots.
Wide Angle Shot
First let’s look at an outdoor shot of part of my backyard. This was taken with my 24-105 mm f/4 lens, with the camera on a tripod. The shot is of nothing, but it has a tree and a fence in it that I thought might give us some good textures to compare. Here’s the two versions of the shot at full size:
I don’t see much difference. But let’s zoom in on a portion of the picture to 200% and see what that looks like.
I still don’t see much of a difference.
This shot is taken during the middle of the day, hence I used a fairly fast shutter speed. Normally, when you are on a tripod, it is because you are using a slow shutter speed. To make this test a little more practical, I went inside to take some shots at a slow shutter speed (measured in seconds). Take my word for it that they weren’t any different. In fact, I took a variety of shots and could not find any difference due to shutter speed.
Photo at 105 mm
When I went inside, to continue taking pictures with the same lens as part of this test, at some point I zoomed in on a subject to take a shot. Suddenly I noticed a difference. It shows up a little bit even at full size (although I confess it is difficult to see in the pictures on this screen):
You can really see the difference when zoomed in to 200% (again I know it is hard to see here, but with the pictures at full size on your monitor I promise it shows up very clearly):
I wasn’t positive that focal length was the determining factor here, but began taking other shots. Ultimately I ruled out other factors and came to the conclusion that focal length was the issue.
Photo at 200 mm
My other lens that has image stabilization is a Canon 70-200 mm f/2.8. I had planned to test it as well, and I did, but now I wanted to see if that issue with focal length showed up on this lens as well.
This lens performed ok at the wider end of its focal length range. However, I zoomed it all the way in to 200 mm to see the effects of focal length. The result was striking:
I didn’t even zoom in on my monitor. These are the full images. As you can see there is a big difference. In fact, there was such a big difference I was worried I had bumped or moved the camera in the first picture. However, I repeated the test several times and the results were the same.
Hopefully you can see the differences in these pictures. The conclusion I draw from this is that turning off image stabilization matters, and the effect is magnified at long focal lengths.
Why does the long focal length matter so much? Well, focal length always matters when considering camera shake. Recall that the Reciprocal Rule, which measures the longest shutter speed that you can use while hand holding your camera without risking camera shake, is a function of the focal length you are using. Long focal lengths tend to magnify small changes.
That’s what I think is going on here. I thought I would pass it along for any of you that – like me – doubted whether you need to turn off image stabilization while your camera is on a tripod. Of course, if you accepted the conventional wisdom all along then none of this is news to you.
I should note that this test is only Canon lenses and it might be different for other brands. If you have alternative theories, or different testing results, be sure to let me know.