A short time ago, we posted a short article describing a concept for monitoring camera vibration using a laser. Since then, the article and video have made their way around the internet with viral popularity. There seems to be an overwhelming appreciation for Mythbusters style investigations which combine one part curiosity, one part DIY ingenuity, and one part mayhem. And, let’s face it, everything becomes cooler when you attach a laser to it. There have been many questions about the setup conditions and also many suggestions on how to refine the technique for better results. In the spirit of crowd-sourced science, we are revisiting the camera laser for a closer look at the original idea and some new refinements on the technique.
The motivation for the original article was to quantify the amount of vibration induced during the mirror movement of an SLR camera. Most SLR’s have the ability to lock the mirror, but the effect would seem so insignificant to be hardly worth the trouble. One can always take “pixel-peep” comparison images but I was interested in a better, real-time comparison method. So it was with great pleasure in a Dr. Evil eureka moment that I uttered, “Let’s put frickin’ laser beams on their heads!” Note: Remember to use air quotes when you say the word laser.
Here are some of the questions and ideas that I wanted to address in the follow-up study:
Camera: Canon 7D with 28-135 lens attached – Although there are some differences in how different cameras handle mirror vibrations, this investigation is not so much about the 7D as it is about SLR’s in general. Despite all their advantages, one disadvantage of the SLR is the size and vibration of the SLR mirror.
Tripod: Manfrotto 190X tripod with Giottos MH-3300 ball head – Many asked what type of tripod was used with the assumption that a flimsy tripod could amplify any small movements in the camera. While this is not a $1000 Gitzo setup, it is a fairly reliable tripod that I think is typical of what many prosumers might use. The smallest of the tripod leg sections was also completely collapsed to add stability.
Floor Support: Many asked if the tripod was on carpet or bare floor with the assumption that floor vibrations may be contributing to laser instability. This is something we’ve improved upon in the follow-up experiment. Pads were added under the laser camera and video camera tripods to decouple both cameras from any floor vibration.
Laser Mount: Many worried that mounting the laser on one end created a cantilever action that amplified any vibrations. True enough. We’ve improved upon this in the follow-up experiment by mounting the laser to an old flash extension housing which locks firmly to the hot shoe. The laser is now held firmly with it’s midpoint directly over the camera. Believe me, this guy is not moving unless the camera is moving.
Target Grid: There were several ideas for enhancements of the target. One was to use numbered grid lines to quantify laser movement. While this was attempted in the follow-up video, we were really limited by the poor focus of the cheap laser pointer. With such a defocused beam, it is very hard to pick out distinct edges for quantitative measurements. The movements are also too small to be measured with normal graph paper.
Hand-holding vs Manual shutter Press on Tripod: One source of confusion for a few of the quick passersby was that none of the case studies represented true hand holding. The only manual case was conducted with the camera still locked onto a tripod, but with the shutter pressed by hand instead of using a remote. In the follow up video, we have added a hand holding case to demonstrate the worst possible case for camera movement.
Here is a diagram of the final setup:
And so without further ado, here is the follow-up video – Laser Reloaded.
Camera Laser Vibration Detector Revisited from Camera Technica on Vimeo.
The conclusions of the new video are largely the same.
1. The hand holding case was added for amusement to show how hard it is to hold a camera straight without a tripod. It was difficult to even keep the laser in the video field of view.2. With the camera on the tripod, the manual shutter press case clearly shows that human touch is the most prevalent cause of camera movement. All other activities are fruitless until human touch is eliminated. Professionals with a good long lens technique can probably do better, but there is simply no way to touch a camera without inducing significant movement.3. Although the laser stability improvements made a difference, there is still a pronounced bounce in the laser when the shutter is released.4. By examining the case with mirror lock enabled, we can separate how much vibration is occurring due to the mirror-up motion versus the shutter movement/ mirror-down motion. It is clear that a significant portion of the vibration occurs as the mirror goes up. Mirror-up vibrations are the critical ones to assess since they occur just prior to the exposure and may still be reverberating as the exposure is being made. Mirror-down vibrations are also significant, but presumably would not affect image quality since the exposure would have already been made at that point.
Final Analysis – In most cases, I don’t think mirror vibration is worth sweating. We had to look fairly closely using a novel technique at a distance of twenty feet in order to pick up the effect of the vibration. Also, unless you are on a tripod with a shutter remote, enabling mirror lock is clearly pointless. But for critical images with longer exposures, mirror lock up may help you achieve that last bit of sharpness you are looking for. For astro-photographers who take most images with long exposures and long focal lengths, mirror lock is vital.
If you have questions about this experiment or if you have other investigations you would like us to consider, please direct your inquiries to cameratechnica@gmail.com.
Keep on Rolling with These Similar Articles:
The Science of Polarizing Filters
The Science Behind the Red-Eye Effect
How to Create (and Avoid) Starburst Highlights
Or Get More Camera Technica Greatness:
Subscribe to this Feed ◊ Follow us on Twitter ◊ Email Us
Just wondering how the VR/IS/OS technology affect the final result, but it seems impossible to test with laser ?
Anyway, everything is cooler with laser
I’m not sure why long exposures should be particularly affected by mirror lock-up. Surely the exposure that’s going to be most affected is one that is exactly the same time as it takes for the vibrations caused by the mirror to be damped out? Not that I’m suggesting you shouldn’t lock the mirror up because obviously you would if you could, I’m just questioning that statement.
If, for the sake of argument, the vibrations are damped out in 1/10 sec, then an exposure of that length is 100% affected by vibration. For a 100 sec exposure then 0.1% of it is affected.
I agree with your sentiment, John. I was using the phrase “longer exposures” to differentiate from those exposures which happen too quickly to be impacted by the vibration. But I agree that the result will be at its worst exposures ~= the length of the vibration. Thanks for the clarification.
Many times you can use Live View to get the same function as MLU, depending of course on the camera in question, especially for models that make you wade through menus to turn MLU on and off (Canon, I’m looking at you).