This post is a continuation of my earlier tripod-test posts:

1. Considerations for Building and Using a Tripod
2. Tripod Stability Tests, Part I
3. Tripod Stability Tests, Part II
4. Tripod Stability Tests: on Shutter Speed

Part II compared different equipment setups; in this post, I'll look at the same results, but with an eye toward differing photo-taking techniques.

All images on this post were taken at f/8, ISO 100, with a 1-second shutter speed, with a pro-level 70-200 f/2.8 zoom at 200mm.

These are all from the same test session as those presented in earlier posts, so all the same caveats apply.

We'll start with my reference setup, which is a pro-calibar set of Gitzo legs without any of the leg sections extended, no ballhead, and no center column. (The specifics of all my components are detailed here.)

Here are the details on each button that can be mouseovered above:

• Ref — my standard reference image, taken with the most stable setup and technique.
• Mup (10 sec) — image taken with a remote shutter release, and mirror lock up left to settle for 10 seconds between the mirror being put up and the shutter being released. (Because the set of tests in this particular box are with my most stable tripod setup, and because “Mup (10 sec)” is my best technique, in this box only, “Mup (10 sec)” is the same image as “Ref”.)
• Mup (1 sec) — only one second pause between putting the mirror up and releasing the shutter, after at least a 10-second pause since last touching the camera.
• no Mup — simple remote shutter release without mirror lock-up, having waited at least 10 seconds since touching the camera before firing the shot.
• Bonk — simple remote shutter release, but I gave the lens a bonk with my hand on second before releasing the shutter. (I'm not really sure what this test is intended to show. Certainly, there is quite a bit of variance in how I bonked the lens each time.)
• Hand (Careful) — snapping the picture without a remote shutter release, being as careful and gentle as I could.
• Hand (Not careful) — snapping the picture without a remote shutter release, not paying too much attention to being soft and gentle.

With the reference (most stable) setup shown in the images above, it seems that you could use a hammer to release the shutter and the results are still excellent.

Let's move to a more realistic setup: 3 of the 4 leg sections extended (of the very tall Gitzo 3540XLS, described here), Really Right Stuff's best ballhead (BH-55), and no center column....

There doesn't seem to be much benefit to mirror lock up, but there is clearly benefit to not bonking the lens, and to being careful when shooting by hand.

Let's add a center column and raise it 10cm (about 4 inches)...

Again, not much benefit to using mirror lock-up, but the benefits of using remote-release are more clear (no Mup vs. Hand/Careful).

Let's now be crazy and raise the center column to its maximum height of 32cm (about a foot)...

The results seem comparable to the 10cm-height tests.

Let's get rid of the center column, and extend all the tripod's leg sections to put it at its maximum almost-eight-feet height...

These are more clear across the board, as we'd expect from the tests presented in Part II, which showed little degradation in adding a center column, but notable degradation when raising it 10cm. I'd still not recommend bonking the lens, though.

Perhaps the most real-world tests, here are the same set of techniques on the small, flimsy, cheap tripod described in Part I....

This shows that the need for remote shutter release goes way up as the quality of the tripod goes down.

You can really read a lot into these results, but let me again stress the caveats (some listed here) of these tests. In some ways they're more brutal than real-world usage (that is, less forgiving of vibration) which may well use much faster shutter speeds, but in other ways they're isolated in the studio away from real-world problems like wind.

(Note: this long-winded, painfully verbose post will be of interest only to serious camera geeks, if anyone)

After posting the first parts of my tripod stability tests (Part I and Part II), some discussion ensued about my choice of a one-second shutter speed for the tests, both in the comments added to the posts (comments for Part I and Part II), and in the forums at Digital Photography Review (starting from this post).

Well-presented comments from a number of different people suggested that a one-second exposure was too long, and by keeping the shutter open longer than the vibration vibrated, the added time would tend to “wash out” the more subtle effects of the vibration, rending the test results misleading. For example, if there was vibration for only 1/10^th of a second in a 1-second shot, its effects would be seen only 10% compared to a 1/10^th-second shot.

I understand the point they're trying to make, but I believe that it does not apply to how I did my tests. As I said before, I'm not an expert at this stuff, so perhaps I'm wrong. Hopefully, this post will make things more clear, one way or the other.

One important mistake I made in my earlier presentations was that I didn't make it clear how my tests are different from normal tests. My defense is that I didn't really know how normal tests were done, which is probably a good thing because knowing could have limited my thinking while developing my own approach.

“Normal” Test Charts

It seems that normal tripod testing is done by pointing the camera at a paper with a test pattern printed on it, such as this:

(That's a crop from a low-res version of a video test pattern first made in 1956, available as a high-resolution vector PDF from a site with a lot of test patterns.)

My Test Charts

I created a digital test chart, color-on-black, and displayed it on my laptop's LCD screen....

The strong points of this test pattern, in my mind, are:

1. The background is black, and as such, makes almost no impact on the test image.
2. The “railroad track” lines use no anti-aliasing, so the LCD's pixels are sharp (either 100% on, or 100% off).
3. Each of the three primary colors appear individually, in an attempt to create the most clear image on common image sensors (which generally have pixels that respond to the primary colors). (This is only a half-hearted attempt, because my laptop LCD's idea of primary colors are likely a chromatic mile away from the camera sensor's, but it's a start.)

Its the first point — a black background — that's important for this post....

The Fundamental Difference

I'll explain in detail further on in this post, but in short, the fundamental difference is that the normal “black ink on white paper” method involves taking a photograph, of which shutter speed is an integral and important part of a correct exposure. On the other hand, my method uses the camera along the lines of a device to detect photons, where the shutter speed merely indicates the duration over which photon detection occurs.

Perfect Situation: No Vibration

Let's consider the perfect case of absolutely no camera vibration while taking a test shot. In that case, the shutter speed doesn't matter for the background areas of the test chart, because fast or slow, no photons will arrive from it. (I use “no photons” quite loosely here, but you know what I mean.)

So, in the perfect case, the shutter speed has an impact on only how bright the lighted portions of the test chart show up. Too fast a shutter speed and nothing will show up, and too long an exposure and electrons will start to “bleed” to adjoining pixels. That last situation is well beyond simply “blowing out the highlights,” and requires a really long exposure, relatively speaking.

So, as long as the shutter speed used for the test is between the “too long” and “too short” speeds, the result should have meaning. At the ISO and aperture I used (100 and f/9), I could see the test image well enough at 1/250th of a second, so we'll call beyond that “too short.” I didn't test a shutter speed longer than 10 seconds, but they were usable, so we'll call anything beyond that “too long.” Their highlights were likely “blown out,” but that doesn't really matter to vibration testing unless electrons start to bleed to adjacent photosites.

So, considering that the range I just presented (1/250th to 10 seconds) represents about 12 stops, and the D200 sensor has almost that much dynamic range to begin with (as described here), I think most any speed within that range would properly reflect the perfection of our utopian case.

Imperfect Situation: Vibration

So, if there is vibration, how will that show up in the test image? If even the slightest vibration causes the lit portions of my test chart to shine onto the area of the sensor where the background normally appears, those photons will be like a flashlight in the dark, and show up in the result.

Vibration is often a back-and-forth thing, so if a shot's vibration lasts for, say, 1/10th of a second, it'll cause photons to collect on either side of a bright area (in an area that should be black) for half that time, or 1/20th of a second. That's well long enough to show up.

This is all very different than what happens with a normal test chart, where the background is white and is flooding the sensor with photons almost all the time. In fact, the only way the vibration shows up in such a situation is because the vibration caused a reduction in photons, a reduction that indeed can get “washed out” as the exposure time approaches a length intended to properly record the whiteness of the paper.

This difference in the two approaches is doubly noteworthy because of the nature of human perception of light. When a light is dim, as the background of my test chart is, a 10% difference in light intensity (number of photons) results in a much larger percentage change in brightness (human perception of light). Thus, a little vibration resulting in a little extra photons can be noticed.

On the other hand, when something is perceived as bright (like a properly-exposed sheet of white paper), a 10% difference in the intensity is perhaps not even noticeable. (It might come as a surprise, but you may not even notice half of the sun being blocked during an eclipse unless you happen to first notice the changing shape. You have to cut light by about 82% to cut the perceived brightness by half. What remains after 82% is cut — 18% of the light — is where a standard 18% gray card finds its origins.)

An Analogy

I set my camera up in a dark room, and took pictures of myself making a circle with a flashlight. In each shot, I made about one revolution with the light (sometimes a bit more, sometimes a bit less), and took about three-quarters of a second to do it.

Here are four of those exposures, with shutter times ranging from 1/2 second to 15 seconds:

What I intend for this to show is that a short shutter time can cut off some of the “vibration” (my flashlight circles), but a long one does not reduce the impact of what was recorded. A lot of extra nothing is still nothing (as the entirety of my blog so poignantly illustrates ).

Even if the circle-spinning went on for a long time, you'd see a blown-out circle, but you'd still be able to tell that there wasn't a big X through the center, or whatever. It would still tell you where the flashlight was and wasn't present.

Thus, for my tripod tests, I chose the relatively long shutter speed of one second so that they'd be more brutal: I'd watch for vibration for much longer than a normal photographic exposure.

Relevance?

What I've explained in this post is why my 1-second tests are not irrelevant for showing camera vibration. What I've not explained is whether these tests are actually relevant to how a tripod will perform in the real world.

I know that the tests I've already presented (and those I've done but haven't yet presented) are only a starting point. In one sense, they're supposed to be brutal in how they test the equipment, looking for the most minor vibration. Yet, at the same time, I know that I've tested them only under the best conditions (no wind, photographer not rushed, etc.), so in another sense I've gotten a “best case” reading.

A theoretical “best case” might be applicable in some studio situations, but in the real world, wind, rumbling trucks, etc., can have a strong impact, so the more important goal is to find out how a tripod degrades or holds up under those conditions. As Alexander Kiel notes in one of his comments, “it would be good to do some outside tests in a windy environment. I think that then the cheap tripod won't look so good even with the best technique.”

Indeed, I hope to do those tests sometime...

Looking at my server logs lately, it's clear that people are looking forward to fall, as I can see the popularity of posts like the following bubble higher:

• The Color of Kyoto — Desktop Background
• Kyoto Fall Foliage Desktops
• Photoshop's “Darken” Blend Mode (and more fall-foliage desktops)

Those three are particularly popular at the moment, mostly via people coming from the search engines' text and image search. As fall gets closer, more and more posts from my Japan, Fall Colors category will become more popular.

Fall hits Kyoto a month or more later than it seems to hit elsewhere in the world, so interest will wane by the time the colors actually arrive this year, so to help satisfy the current colorful-foliage needs of the Internet, I've dipped into my image archives to show a few rejects from last year....

The photo above was taken on the same day as these colorful shots, from a small street leading up into the mountains on one of the routes to Mt. Daimonji, taken very near where the “Pausing to be Silly” shot from this post was taken.

The image below is from the garden at the most-excellent Enkouji Temple (north-east Kyoto). With the focus at the far background and most things out of focus in the front, this might not “work” for most people, but the colors are wonderful.

Photos from my one visit to Enkouji last year have ended up in four other posts:

• Revisiting Kyoto Temples in Autumn
• Relaxing Amid the Fall Colors
• Kyoto Fall Foliage Desktops
• Fall 2006 at Enkouji Temple

The next shot shows the temple complex nestled in the bamboo-covered mountain, with northern Kyoto in the background.

Kawabata St, which runs parallel to the big river in eastern Kyoto, is one of the most pretty main thoroughfares of the city. It helps that there's a river and park all along one side (as opposed to big ugly office buildings, parking lots, and convenience stores), and also because it's planted with so many kinds of, er, plants.

The whole length is an explosion of color in the fall (the yellow shown above only one of the many fall colors), and has an even more beautiful and varied outburst in the spring (such as shown in the first pic on this colors of spring post).

The yellow trees above (seen closer in the first pic on this post) are ginkgo trees (in Japanese: イチョウ), which are beautiful, but when, unfortunately, once its smelly seeds start to fall, it makes the whole area smell like dog poop.

With that pleasant thought in mind, finally, here's another one from the same area as the lead photo on this post....

I introduced my tripod stability tests in the previous post, and I'll continue in this post with more details.

Before I continue, I must note that there are some serious caveats that limit how much one can really read into the result I'm about to present. I don't list the caveats up front because no one will pay attention to them anyway /-:, so I detail them later in this post.

In this post I'll focus on equipment setup. In another, I focus more on technique.

Equipment Tests: Basics

To test the effects of changing equipment, I used a photo-taking technique that I hoped would give me the best results: I used a remote release so that I didn't actually touch the camera, and I used mirror lock up to raise the camera's mirror, and let it sit for at least 10 seconds before actually tripping the shutter, so as to allow any vibration from the mirror movement to fade away. It's extra work, but should give the best results possible for any given setup.

Here are the results of using that technique with the pro-caliber Gitzo 3540XLS carbon-fiber legs (described here) set up in various ways, with one final “crappy tripod” example using the cheap, flimsy tripod described yesterday.

Test Description

• The “Ref” image above will be the first image presented in all the sets, and is what I would consider a reference image — the best I could produce with a one-second exposure. It uses the tripod legs without any of the sections extended, and without any center column, ballhead, or other movable camera mount. It's a fairly impractical setup for real-world use due to the lack of a movable mount, but since that lack removes much potential for instability, it should be ideal as a reference setup.

• The “+ Ballhead” version adds a pro-caliber ballhead (the Really Right Stuff BH-55, with the Really Right Stuff PLC-1 quick-release clamp, also described here).

• For the “+ 3 Legs” version, I extended three of the four leg sections, resulting in a camera about eye height for me. (I'm 6'4“)

• For the ”+ 4 Legs" version, I extended all the leg sections. This was much taller than I could normally use, with the camera almost touching the 8-foot ceiling in my living-room.

• And again, as an opposite reference, there's the “Crappy Tripod” version.

Test Analysis

The two extremes, “Ref” and “Crappy Tripod” are certainly different, but the Crappy Tripod one is not that bad, indicating that good technique can make up for a lot.

Oddly, the quality seems to get a bit better when moving from “+Ballhead” to “+3 Legs” (that is, by lengthening some of the tripod legs), which is counter-intuitive. This is an excellent example of one of the caveats that I point out later: I was perhaps a bit more sloppy in my tests that I should have been, and because I took only one image at each setup rather than taking several, I lose a bit to a wider margin of error. The difference between these two pictures is pretty small, and so falls into that margin of error, I think.

All in all, there's little practical difference from the “Ref” through “+4 Legs”, which leads me to conclude that good technique and good equipment under ideal circumstances can lead to good results. Under ideal circumstances, you don't lose much stability even when raising the camera up 8 feet on quality legs and a quality ballhead.

Note that the results shown above don't indicate anything about less-than-ideal conditions (e.g. wind, trucks rumbling by) or technique (not using a remote shutter release, for example). The next set of tests attempts to address half of that.

This set has the same equipment configurations, but the picture-taking technique changed from using a remote shutter release, to my pressing the camera's shutter-release button by hand. For these tests, I did the pressing very carefully, controlling my breathing and trying to be as soft and gentle as possible....

Test Analysis

The value of a quality tripod makes itself clear in the “crappy tripod” version, which shows a lot of movement despite my having been very careful.

Relatively speaking, the others look much better and are all in the realm of usable. Still, they've all lost noticeable sharpness from the reference image.

Oddly, the same counter-intuitive difference in the move from “+Ballhead” to “+3 Legs” is also present here. I can think of three possible explanations:

1. The “margin for error” just coincidentally manifested itself in the same spot. (Time to play the lottery!)
2. Something was wrong when I did the “+Ballhead” tests. (Maybe my refrigerator running full tilt?)
3. These results accurately reflect reality, and for some reason, stability is increased when extending the legs. (Seems highly unlikely, but who knows?)

That aside, the results make a lot of sense, getting progressively worse as you move across. The conclusion supports common sense: good technique become more and more important as your equipment gets taller / more complex / lower quality.

Equipment Tests: Center Column

A center column can be really convenient, allowing you to make quick and fine-grained adjustments of camera height. Without a center column, you have to futz with the length of the tripod legs to change he height, which believe me, is not fun. However, it's clear that adding an extra movable component to the support system will weaken it, but does it really matter enough to, well, matter?

I did some tests with my tripod at its three-leg-section setup, first without a center column, then with the center column installed but completely down, and then at various heights.

Test Description

1. Ref — standard reference image
2. No Column — three of four leg sections extended, ballhead installed, but no center column.
3. Column Down — like above, but with a center column installed and positioned all the way down (its most stable).
4. 10cm Up — like above, but with the column extended 10cm (about 4 inches), a likely height to use it at.
5. 32cm Up — the center column extended to its maximum of over a foot, its most unstable. This is an extreme height to use a center column at.

Test Analysis

All the configurations yield images that are in the realm of being usable, but there's a clear penalty to raising the center column. Simply having the column installed seems to yield no penalty unless you actually raise it up.

Surprisingly, raising it just a bit seemed to have the same impact as raising it all the way.

With good technique, you can install and use a center column if you need to, but there's indeed some small penalty to actually using the column.

Like so many things in photography and life, it becomes a balance of convenience vs. quality.

As before, let's see the results now when using a less-good technique (carefully releasing the shutter by hand, rather than using a remote shutter release)....

Test Analysis

The results parallel those in the previous set, but the poor technique amplifies the differences. It now seems apparent that there's a small penalty for just having the column installed. Still, I must reiterate that with a difference as small as we're seeing here, and the long list of caveats yet to be discussed, it's not reasonable to draw any conclusions.

I can't say the same while looking at the “10 cm Up” and “32 cm Up” frames, which show (relatively speaking) considerable shake.

The conclusions here parallel those from before, and basically show the importance of using a remote shutter release with a tripod, or at least not using a center column unless you really need to.... at least not with long exposures with a long zoom lens.

That last “at least...” phrase is important to highlight: the blurring effects of camera shake (whether due to poor technique or inferior equipment) decrease as you use a faster shutter speed and/or less zoom. These tests of one-second exposure at 200mm zoom may well be common in landscape photography (an area I have little experience in), but are unrealistically harsh for many common situations where a one-second exposure would result in blur from subject movement that far exceeded that from the tripod or technique.

Caveats

So, here are some things that may conspire to make any conclusions I've come up with worthless....

1. I should have taken multiple shots in each situation and then picked the best or average result, thereby reducing the chance of a one-time mistake making its way into the results.
2. I should have been more careful to ensure my camera was level, so that I wouldn't have had to make minor rotations in the results to get them to look comparable. This greatly increases the margin for error in comparing mostly-sharp results.
3. I should have been more careful to ensure that the lens was always at the full 200mm. At one point along the way I noticed that it was a touch below 200mm (190mm?). I hadn't moved so much that the Exif data registered anything but “200mm”, but I should have been more careful nevertheless.
4. Since the target was 50+ feet away, I didn't pay strict attention to the exact point I set up the tripod, but in thinking about it now, a difference of just one foot would result in a 2% change in the size of the target, which perhaps has a real effect on how we interpret the results. I should have kept the camera-to-target distance constant (which includes adjusting for the changes in camera level that come with different tripod setups).
5. I'm just testing with my own personal equipment, using techniques that I happen to know. The results may not apply at all to your equipment or technique.
6. It may well be that the relative effects of changes across different setups is completely different at different shutter speeds. There's no way to know without testing, and I haven't done that (yet?).
7. My ballhead is equipped with a Really Right Stuff PLC-1 panoramic quick-release clamp, which may have completely different properties from the more common solid quick-release clamp that I used with the non-ballhead “Ref” images. I really should have taken PLC-1 out of the mix and used a solid quick-release clamp for all the tests (unless I specifically wanted to test the effect of adding the PLC-1 to the list).
8. None of the tests measured “3rd-party vibration,” such as that from people walking by, trucks driving by, wind, etc. It may well be that the relative merits among the various situations tested would change considerably when outside forces are added to the mix.
9. It could be that my house is not a very good platform upon which to conduct tests. I think it should be fine because it's a modern post-Kobe condo built with an abundance of concrete and steel, and the flooring itself has a lot of vibration-dampening built in (that we added as an option while it was being built, as an added measure against the sound of Anthony's running feet bothering the downstairs neighbors). But for all I know, vibration from the fridge or washing machine or birds flying outside make their way to my floor and tripod, and perhaps that happened while I was conducting the tests? I don't know.
10. I should have taken more steps to try to increase sharpness, such as hanging a weight from the center column when one was being used. Another area for improvement would be to have moved from a 200mm zoom to a prime. (Re-doing these tests will be the perfect reason to go out and get a new lens; if someone would like to contribute this fine lens to the effort, I would be most appreciative ).
11. When it comes down to it, I'm not an expert on any of this stuff, so perhaps there are fundamental flaws in my tests or my interpretation of the results.

(Note: this post will be of interest only to camera geeks)

My previous post contained a superficial listing of considerations for building a tripod, and left off wondering how to test the effect of various ways to use a tripod. For example, how detrimental is it to add a center column? How beneficial is mirror lock up? Since getting my really nice tripod, I find myself wanting to learn how to use it properly, and these questions are a big part of that.

I tend to dive into things looking for answers, as evidenced by the posts in my photography-tech category, such as my auto-focus test chart, my analysis of NEF compression, my seven-page writeup on digital-image color spaces, or my Lightroom metadata-panel presets builder. In this case, I'm not really sure how to proceed, but I've come up with some tests and results that at least form a basis for further study.

I'm not really sure how relevant the test I'm about to present are, both in theory and in how I executed them. One of the reasons for my writing this post is to solicit comments on the validity of my methodology and interpretation.

The Tests

I created a simple test-pattern image, shown here at quarter size (click to see the full-sized version):

I set up my tripod at one end of the longest hallway in my home, and displayed the test pattern on my laptop's screen at the other end 16m (53 feet) away. I mounted the big Nikkor 70-200 f/2.8 onto my D200. My goal was to get something that had bulk to challenge the tripod support system, and also reach (high magnification) to exaggerate the effects of any instability. (Frankly, a 200mm lens is not really very much magnification, but it's the best that I have at hand.)

To give you an idea of what I am testing with, here's a picture from my tripod post showing the camera and lens mounted on my ballhead.

I took dozens and dozens of one-second test exposures under various conditions. (See this post for a discussion of why a one-second shutter speed was chosen, and for other technical details about how the tests were performed.)

For reference, here's what one of them looked like, with my laptop showing the test screen in the center:

Cropping out all but the center part of interest, the result might look like this:

I took long one-second exposures to highlight the ill effects of vibration and instability. The example crop above might not look very sharp, but considering that it's a one-second exposure and that it's a ^1/44^th-crop from the full frame, I think it's pretty good.

It seems especially good when you compare it to using a bad tripod, or handheld without a tripod. Mouse over the buttons below this next image, to compare among a few different results...

Most Solid Good Tripod Bad Tripod Handheld (crazy) Handheld (sane?)

mouseover a button to see that image

I'll talk about these and other results in more detail in another post, but these are some of the most broad results for which the conclusions were already fairly predictable: a good tripod is better than a bad one, and a bad one is better than no tripod.

For reference, here is a bit about the images being compared...

All but the last are one-second exposures at f/8 and ISO 100. It's crazy to try to hand-hold a 200mm shot for one second. (Heck, I was lucky to get a 17mm shot crisp at ^1/8^th of a second in this post).

So, the final “handheld (sane?)” shot uses all the technology my camera has to offer to better my odds at a less-shaky result. I opened the aperture three stops (from f/8 to f/2.8) and bumped up the ISO five stops (from 100 to 3,200), which allowed me to increase the shutter speed by eight stops (from 1 second to ^1/250^th of a second). Even then, the shake is still much worse than the “good tripod” result, not to mention the lower quality due to the high ISO and less-sharp aperture.

The “good tripod” represents bad technique with a reasonably good tripod: I used my Gitzo tripod with its center column down and only three of the four leg sections extended. That's all fine, but the bad technique is that I pressed the shutter button by hand rather than using a remote shutter release.

The “bad tripod” version used the same bad technique, but with a flimsy tripod totally inappropriate for the big camera and lens (a Velbon ULTRA MAXi F, described here).

Finally, the “Most Solid” version used my big Gitzo legs without any leg extension, no center column, and no ballhead (using the Really Right Stuff B2-Pro quick-release clamp from my monopod bolted directly to the legs). This is a fairly impractical setup, but extremely solid and stable, so I use it as a reference. I also used mirror-up mode, waiting at least 10 seconds after putting the mirror up before activating the shutter.

In a future post, I'll use some of the other 66 test images to isolates the effects of using mirror lockup, a remote shutter release, all four leg sections (compared to three leg sections), a center column, and more.