Motion Blur and the Panasonic Lumix DMC-FZ7

“We think our camera is broken and are wondering if we should take it back.” said a colleague at work, describing her family’s new camera.
“What’s making you think it’s broken? I asked.
“Almost all the pictures are blurry,” she said.
I immediately suspect camera motion and ask a few diagnostic questions. “Are you using the camera indoors or outdoors? Are you zooming in for tight close-ups? Are you taking action photos or staged portraits”
Basically, she answers, “Yes.”
“It’s probably camera-shake.” I offer.
“But it has image stabilization! Do you think you could take a look at it?”

I’m no expert, but in this situation, given that we both flounder at the low end of the clueless scale, I figured two heads would be better than one. I agreed to take a look at the camera, the camera manual, and some examples of the problem photos.

Figure 1 F3.2 Shutter Speed 1/4 ISO 200

A presumably stationary Christmas ornament. This is an example of motion blur caused by camera movement.

The Panasonic Lumix DMC-FZ7 seems to fall into the really-complicated-point-and-shoot-digital-camera category. The specifications for the camera seem pretty impressive. It’s a 6 megapixel camera but the thing that seemed the most amazing to me was the camera’s lens specification (Optical 12X zoom, f = 6 mm to 72 mm). To be honest, that last part didn’t seem so impressive to me because I didn’t know exactly what was being measured until I went on to read that f = 6 mm to 72 mm is equivalent to 36 mm to 432 mm in the world of 35mm film cameras.

I compared the lens specification for the Lumix to a lens I was familiar with. Sigma has a telephoto lens for 35mm cameras that’s supposed to be 18-200 mm. If I were to use it with my Canon digital Rebel, I’d have to take into account the camera’s sensor size and its placement in relation to the lens. I’ve learned that to get 35 mm camera equivalencies for the Rebel, I need to multiply the specified focal length of the lens by 1.6. The 35 mm camera equivalent for the Sigma lens paired with the Rebel works out to 28.8 – 320 mm.

So far, we’ve taken into account how ‘long’ the Lumix lens is, which I guess is another way to describe its powers of magnification. Sticking to the 35 mm camera equivalencies, we can say that the Lumix, at 432 mm is 112 mm longer than the Sigma lens. In other words, if I want to take surveillance photos of the anarchist in the parade, the Sigma lens will allow me to zoom in on the anarchist’s nose while the Lumix camera will allow me to count the nose hairs.

But there is kind of a trade-off associated with long lenses and it has to do with how much light can get into the lens to reach the sensor (formerly film). This is measured with another set of specifications that have to do with how wide a diaphragm at the back of the lens can be opened to let light in. Looking at the Lumix lens, I can see the aperture numbers written as 1: 2.8 - 3.3. The manual doesn’t really explain this nomenclature, but I think it means that at the short end of the lens’s focal length, the aperture is F2.8 (Where ‘F’ stands for aperture) and at the longer end it drops to F3.3. Suffice it to say, when considering aperture numbers, smaller is better (faster) and each larger increment on the aperture scale lets half as much light in. The Sigma lens is labeled with aperture data that looks like this: 1: 3.5 - 6.3. It looks to me, based on these numbers that the Lumix lens should be ‘faster’ even though it is ‘longer’.

One of the nice things about today’s digital cameras is their built in ability to keep track of the camera’s settings and embed this information into the picture’s digital file. I reviewed the settings for the pictures that were blurry, and in every case, the blurry photos in question showed aperture values wide open at F2.8, but the shutter speeds were still in the range of 1/5th to 1/20th of a second. Despite the relatively fast aperture (compared to the Sigma lens), the camera still had to choose long exposure times to allow enough light in to make pictures.

To get a practical idea of how shutter speed relates to stopping motion, I used the Lumix camera, a ferocious-windup-walking-Spark-Dino, a halogen work-lamp and a turntable to create the following sequence of pictures.

Figure 2 F6.3 Shutter Speed 1/5 ISO 100

Figure 3 F 6.3 Shutter Speed 1/8 ISO 100

The turntable is set at thirty-three and one third revolutions per minute. In retrospect, I probably should have set it slower. The sparking dino is towards the outside edge of the turntable, but he is not wound-up and consequently is not walking or sparking. In all cases, I attempt to trip the shutter when the sparking dino is closest to the camera lens. What seems clear at these shutter speeds is that I am nowhere close to stopping motion. Also, since the aperture is recorded as F 6.3, my analysis of the meaning of the cryptic markings on camera lenses above is flawed.

Figure 4 F 3.6 Shutter Speed 1/20 ISO 100

Figure 5 F 3.3 Shutter Speed 1/25 ISO 100

To get the shutter speed to be shorter, I made the aperture wider (a smaller number = faster), however, I soon reach the limits of the lens and actually end up underexposing figure four and five to achieve 1/25th of a second.

Figure 6 F 3.3 Shutter Speed 1/60 ISO 100

Figure 7 F 3.3 Shutter Speed 1/250 ISO 100

To achieve even higher shutter speeds, I ended up having to hold the halogen work-lamp increasingly ever closer to the subject.

I topped out at 1/250th of a second, but even then, I was unable to completely stop the motion produced by the sparking dino’s placement on the turntable.

Figure 8 F 3.3 Shutter Speed 1/250 ISO 100

Evidently, thirty-three or so revolutions per minute is conducive to movement that is generally faster than most of the motion one would encounter in a low- light portraiture situation. Generally, I feel pretty comfortable if my camera is showing me shutter speeds at 1/60th of a second or faster/shorter. Even so, I usually make a point of asking my subjects to hold still. Not only will there be motion from the subject, but unless one is using a tripod, there will likely be some movement from the camera operator pressing the shutter button.

Zooming in tight on the subject requires much more light (slower shutter speeds). Since the lens we are looking at is already stressed by low light situations, a tripod is essential. Also, giant clamps to hold the subjects head and limbs firmly in place should be seriously considered.

Finally, I checked the Lumix camera manual to find out about Optical Image Stabilization. This is what it said:

The stabilizer function may not be effective in the following cases.
- When there is a lot of jitter.
- When the zoom magnification is high.
- When taking pictures while following a moving subject.
- When the shutter speed becomes slower to take pictures indoors or in dark places.

Is it just me, or are those all the conditions that occur when you might want something like image stabilization to work? Presumably image stabilization works better when the subject is motionless and outside on a bright, sunny, cloudless day.

Finally, one other thing that I failed to account for when I took the sequence of experimental photos was the ISO setting (a measure of the camera’s sensitivity to light). It turns out that the ISO was evidently set at 100 for the manual mode. The camera itself is capable of being adjusted to ISO 400. Some modes don’t allow the user to choose the ISO, but where possible, it would be a good idea to bump it up to 400.

Summary:
The camera isn’t broken.
Each lens has built in limitations.
To reduce blurred images:
Increase the ISO setting to 400 if possible
Use shutter preferred mode (instead of SIMPLE MODE) and set the shutter speed to 1/125 of a second or even a shorter fraction of a second if possible.
Use a tripod or brace your elbow(s) on a table or something that doesn’t move.
Open window shades, turn on lights, use the flash, or relocate to a brighter space.

Unanswered questions:
What does 1: 2.8 - 3.3 really mean?
What is the relationship between the focal length of a camera’s lens and the size of the camera’s sensor (film)? Based on the difference in the physical size of a lens’s diameter, aperture must be more than the physical measure of the lens’s diaphragm. What is aperture?

Pondering 'Depth of Field'

I was monkey-ing around with depth of field the other night.


FIGURE 1. In this image, the aperture value = 29, and the shutter speed , = 30 seconds under these dim lighting conditions.



FIGURE 2. Here, the aperture value = 4.5, and the shutter speed = 1.3 seconds.


There seems to be three main factors that affect the depth of field: The focal length of the lens, the aperture, and the distance of the subject in relation to the background and the lens.

Here I tried to limit the variables to just the aperture. For these images the global values are: An 18-55mm zoom lens (in this case the focal length was set at 35mm). I also had the camera set to aperture priority.

The sock monkey was sitting about 3 feet from the front of the lens, and approximately 4 feet beyond the monkey sits the monitor. The depth of field in Fig. 2 is quite narrow. The material in the chair immediately behind the monkey is already starting to blur (compare to Fig. 1)

The monitor sets up a backlighting situation which I tried to balance with my hallway light. I used my spot meter to expose for the monkey’s brown skin tone. It seems that metering with the camera’s spot meter is somewhat hit or miss.


To really increase the amount of fuzziness in the background, I had to adjust the Zoom lens so that its focal length was at the maximum 55mm. In this image, the aperture value is 5.6, the widest value possible for this lens at 55mm. I haven’t really changed my distance from the monkey, though the longer focal length has the effect of bringing one closer to one’s subject matter. Somehow, it seems that the magnification of the subject has a correlation to the compression of the depth of field, but I still can’t visualize the physics.

Even though I'm becoming aware of the variables involved with setting up depth of field shots, I still can't control it. In this picture of a shell, the sand grains show that my point of focus was a little bit too much towards the front of the shell. This is a case where the subject isn't really engaging in a lot of activity, and yet I still wasn't able to get the whole shell in focus (Sadly this is the best of three attempts). In the olden days, lenses used to have distance/aperature markings on the focus ring that worked to show what the depth of field was for a given distance and aperature setting. These markings don't appear on the Canon EFS 18-35mm kit lens that comes with the Rebel.

While the instruction manual details a depth of field preview button, the only thing that appears to happen to me when I use it is that in some cases, the image in the view finder becomes darker, which, instead of helping me determine the range of acceptable focus, just makes it harder to see. I know there must be some kind of a correlation there, but so far, it hasn't been immediately obvious.