Shutter Shock and the Canon R5
August 27, 2022After having looked at lens sharpness and diffraction blur as well as hand-held exposure times, the next topic related to image sharpness I’d like to look at is shutter shock, which is the vibration caused by the movement of a mechanical shutter and another source of unwanted blur. Shutter shock is nothing new, but again, when trying to make the most out of a high-resolution sensor such as the one in my new Canon R5, the impact of any kind of vibrations is magnified and I’d like to find out how big of an impact shutter shock has on pixel-level sharpness.
I’ll start this post with a short introduction into shutter modes, then continue with different test series showing the impact of shutter shock when shooting hand-held versus from a tripod as well as how the in-body image stabilization (IBIS) deals with shutter shock. If you are short on time, scroll down to the very end to find a brief summary of my findings in the conclusion.
It is important to note that only full mechanical shutter suffers from shutter shock, while the R5’s default shutter mode EFCS and electronic shutter don’t have that issue. The section on shutter modes below explains why you still might to use mechanical shutter in certain situations and therefore would have to care for the effects of shutter shock.
Disclaimer 1: shutter shock is a bit different for each camera model, so my test results are not necessarily applicable to other cameras than the Canon R5.
Disclaimer 2: as in the previous posts, we will again indulge in pixel-peeping to assess sharpness of images coming from a high-resolution sensor. For reasonably-sized screens and prints, viewed from normal viewing distances, the requirements when it comes to resolution and sharpness are much lower than the level we look at here. Only when cropping images or producing very detailed large prints, the additional resolution and higher demands to shooting technique come into play.
Shutter Modes
The shutter basically controls when light falls onto the sensor. The Canon R5 offers three shutter modes: mechanical, electronic first curtain (EFCS) and electronic shutter. With the mechanical shutter, there is a physical first curtain covering the sensor, which is lifted to expose the sensor to light. To ensure that all parts of the sensor are lit for the same amount of time, a second curtain closes in the same direction as the first curtained opened. The electronic shutter does not need the sensor physically covered but uses electronics to control for how long each pixel is exposed to light before it is read. The EFCS uses a combination, starting the exposure electronically, but ending it by closing the mechanical shutter and thus hiding the sensor from light.
Each of these modes has advantages and disadvantages. On the Canon R5, only the mechanical and EFCS will deliver 14-bit images, while electronic shutter will only deliver 12-bit images, a difference that you can feel when doing a lot of post-processing on your images, and then especially in darker areas. Another problem with electronic shutter is the increase of rolling shutter effects, leading to fast-moving subjects being warped. EFCS is also not without its problems, when using wide apertures and fast shutter speeds EFCS can negatively impact the quality and smoothness of the bokeh, or out-of-focus blur (< 1/1000, < F2.0, source). And last but not least, the mechanical shutter suffers from shutter shock, while the EFCS and the electronic shutter don’t.
If you’d like to read more about the different shutter modes, I can recommend you any of the following articles, the first two being general and the third also going into specifics of Canon cameras: photographylife, Photo Review, Canon Snapshot.
I shoot a lot in the dark and appreciate the additional details 14-bit files deliver for my black&white conversions, and also use my primes wide-open a lot. So the mechanical shutter would be my first choice with the R5, with shutter shock being the only downside. So my goal for running tests on shutter shock was also to understand how big of a problem it is for me in practice and what I can do to avoid it.
Overview of Test Setup
For my initial tests I worked with three setups:
- First, to isolate the effect of shutter shock, I put camera and lens on shock absorbers, which would allow the shutter shock to take its effect, but avoid blur from hand shake.
- Second, for a more realistic setup I shot hand-held.
- And finally I shot a series on a tripod to see how well a sturdy mount can avoid blur from shutter shock.
While the tests with shock absorbers and the tripod had consistent outcomes, there was a higher variation in the hand-held results. Therefore after the initial tests I shot another series hand-held, with five images for each setting. And last but not least I wanted to see how the IBIS deals with shutter shock and shot another hand-held test series with a comparison between active and inactive IBIS.
For testing I used the Canon EF 35mm F1.4L and the RF 85mm F1.2L to be able to compare how much more the longer focal length will suffer from the effects from shutter shock. All shots were done with a variable light source to adapt to different shutter speeds with stable ISO and aperture. The image crops were then enlarged to 300% for the visual comparison.
Testing With Shock Absorbers
The tests with the camera rested on shock absorbers show a clear and reproducible impact of shutter shock. When looking at the results with the 35mm and at 1/50 exposure time, the image with electronic shutter is crisp, while the one with mechanical shutter shows noticeable blur. This continues with the short exposure times up to 1/160, with only the image with mechanical shutter at 1/320 being as sharp as the one with electronic shutter.
As is to be expected, the effect can be seen even more with the longer focal length of 85mm. The image at 1/50 with mechanical shutter shows significant blur, which then slowly decreases as the shutter speed gets shorter, with 1/160 exhibiting only a slight blur and the result at 1/1000 being as crisp and sharp as the one with electronic shutter.
Have a look for yourself at the results below. You can also download them in full resolution: Test Results with Shock Absorber.
Hand-Held Testing
Testing hand-held is of course the more realistic setup but subject to additional blur and variations from the shake of the hands. For the first test series shown in this section, IBIS was always active, since this is how most people will use their R5. Have a look further down below for a comparison of results with and without IBIS.
Again, the comparison between mechanical and electronic shutter shows a clear difference. At the 35mm the results with mechanical shutter show a slight blur between 1/50 and 1/160, decreasing with the shorter exposure times. At 1/500 both results are equally crisp and sharp. Again with the 85mm the blur is much more pronounced, especially with exposure times between 1/50 and 1/125.
Have a look for yourself at the results below. You can also download them in full resolution: Hand-Held Test Results.
A Closer Look at Shooting Hand-Held, With and Without IBIS
The hand-held images were less consistent than testing with the shock absorber, therefore I wanted to run another test series with the 85mm, taking five shots per setting to even results out. Additionally I wanted to see if there are any differences between EFCS and electronic shutter, and what impact the IBIS has.
The first series shown below was shot at 1/100 and what we can see from those images is that IBIS cannot really help in counteracting the effect of shutter shock, the results from the mechanical shutter with and without IBIS are equally blurry. However results with electronic shutter are much sharper with IBIS, while without IBIS all three shutter modes show comparable blur. So another takeaway is that at this exposure time only by having the IBIS, the shutter shock becomes relevant, because without IBIS blur from hand shake introduces at least the same amount of blur. Regarding EFCS, the images with IBIS confirmed what was to be expected, EFCS does not suffer from shutter shock anymore than the electronic shutter.
Stopping down by two-thirds to 1/160, the effect of shutter shock in the images with IBIS is already reduced, with only a slight hint of blurriness with the mechanical shutter compared to the electronic shutter. While at 1/100 the images from the mechanical shutter with and without IBIS were equally blurry, at 1/160 we can see that the image with IBIS is sharper, so at this exposure time the impact of shutter shock is already less than that of hand-shake. Images from the electronic shutter without IBIS are as blurry from hand-shake as those from the mechanical, with 1/160 still not sufficient to get maximum sharpness.
Another two-thirds of a stop down at 1/250 we get sharp results across the board, the exposure time is short enough for shutter shock not to have any impact anymore, and also sufficiently short to deliver sharp results without blur from hand-shake, even without IBIS.
Further Testing With and Without IBIS
In some of the images shown in the previous section, the blur from shutter shock is clearly more pronounced vertically than horizontally. This can be seen only in the images with slower exposure times and taken with IBIS, for example here at 1/100:
This raised the question whether this is only from the shutter shock itself, or if the IBIS is somehow exaggerating the vibration instead of dampening it. The previous, hand-held test images cannot really answer this question, since the images without IBIS suffer too much from blur from hand-shake. Therefore I did another test series with the camera rested on shock absorbers with and without IBIS.
The three series shown below shot at 1/60, 1/100 and 1/160 clearly show that results from the mechanical shutter with and without IBIS are equal. On the one hand this underlines what we already saw in our previous tests, the IBIS cannot really compensate the fast vibrations from shutter shock. But on the other hand we now know that the IBIS does not hurt and in no way exaggerates the shutter shock vibrations.
Tripod Testing
Last but not least I wanted to test whether a good and sturdy tripod mount can counter the effects of shutter shock.
When looking at the results shown below I would answer this with a yes. Although when looking at the images from the 85mm, between 1/30 and 1/125 you can tell that with the mechanical shutter there is a hint of additional blur, I don’t think it is strong enough to make a relevant difference. In any case, when using a tripod typically we would neither shoot with very wide apertures and fast shutter speeds, so we could choose EFCS without having to worry about the impact on bokeh quality.
Have a look for yourself at the results below. You can also download them in full resolution: Tripod Test Results.
Conclusion
So, what conclusion to draw from this test?
As any other camera shake, shutter shock has a more pronounced impact with longer focal lengths. At 35mm, the impact of shutter shock is negligible for most applications but still needs to be considered if you are aiming at maximum sharpness with a high-resolution sensor such as the one in the Canon R5. When shooting at 85mm, shutter shock has a much more visible effect, especially since with IBIS we are able to shoot exposure times hand-held that we could not before. And while the IBIS does a good job at compensating hand-shake, it cannot compensate for the higher-frequency vibration caused by shutter shock. On the other hand we could also show IBIS has no negative impact, in no way exaggerating the blur from shutter shock.
How can we avoid shutter shock?
One solution are short exposure times. Blur from shutter shock is more pronounced in slower times from 1/50 to 1/100, gradually decreases till 1/160 and is not discernable anymore from 1/200 onwards. The impact is less pronounced with the wider focal length, and using an exposure time of 1/focal length plus two stops would be enough to avoid the effects of shutter shock.
But the most obvious solution is to use EFCS or electronic shutter. One drawback of EFCS is its unflattering rendering of the bokeh at wide apertures and short exposure times. However at the exposure times where shutter shock is relevant this is not yet an issue, and we could therefore use EFCS without any concerns. Only if we use exposure times shorter than 1/1000 and apertures wider than F2.0 we might want to switch to full mechanical shutter for best bokeh quality.
This also applies to tripod use. While on the tripod the effect of shutter shock is minimal, using EFCS completely eliminates it and should therefore be your first choice if you are looking to maximise sharpness.
As always, I hope you found this post useful and feel free to contact me if you have any questions or comments.