Carry a *really* long lens

This photo may seem mundane, but the fun thing about it is that it's taken with a silly, silly setup: The EF-S 60mm macro lens plus an 8x CrystalVue teleextender. The thing to note isn't the incredible quality, but the fact that it's of quality at all. There's quite a bit of blue/red fringing around the camera, but the hand actually has quite a bit of detail. I borrowed it from my brother, who kept getting bad vignetting with it on the compact camera in the picture. Not sure why I didn't get any, but we're swapping this for the smaller 8x CrystalVue teleextender I had lying around.

Yes, I know it's really small aperture, cheap glass, etc etc, but it's fun to play around with, and I definitely want to try it out on my 55-250, which it would turn into a 440-2000 (!) mm lens. Dunno if it'll work, but it'll be fun to find out. I'm a fan of the idea of front-mounted tele extenders. They allow for having a larger entry pupil, so they don't necessarily cut the aperture as badly as a rear-mounted one. Unfortunately, they are mostly made for compacts and video cameras, so generally not of the best quality and with small rear pupils. I was amazed that there was no vignetting to speak of in this case.

The image is so small because I'm using a limited version of the LR/Blog plugin

Moving time

If you take a careful look at my blog profile info and the posts in this blog, you'll notice some posts that are older than the profile. How's that? Elementary, my dear Watson, this blog has moved from its old home on LiveJournal (where the ads finally got to be too annoying). And, it got a new name in the process: Walk Softly and Carry a Long Lens.

This name came about as I looked over shots from my various outings. I tend to take many shots at many focal lengths, but was finding that the most striking and interesting images were taken at longer lengths. For instance, the photo on the right was taken in a crowd of costumed people at 250mm. While I have other nice photos from the Fasching party at Viktualienmarkt, this is the only one that really "speaks".

So while I used to have my "walkaround" 18-125 lens on my camera by default, now it's the 55-250 that lives there. It has a nice zoom range for getting details and still getting some context when needed. At the same time, I would like to get an ultra-wide like the Sigma 8-16, but that would call for a different and much more deliberate style of photography.

When normal isn't normal

One of the first things one learns when starting to look at lenses is the concept of a "normal" lens; in 35mm terms, this is usually taken to be the 50mm lens. The reasoning behind this is that at that focal length, the perspective looks similar to what we see with our eyes. Use a wider lens on the same subject (i.e. going closer), and the closer parts of the image will look larger than normal (which is sometimes what you want, e.g. in landscapes). Use a longer lens, and the foremost things will look smaller - which turns out to be useful in portrait photography and can also give a striking effect in other cases. But many photographers like the "naturalness" of a normal lens.

But try looking into how the 50mm definition was arrived at, and things get a little blurry. First of, the "true" normal focal length for 35mm cameras is 43.3mm, since the standard definition of normal is to be equivalent to the image diagonal. 50mm was picked by Leica because it's easier to make a slightly longer lens. Choosing a 45 or 40 mm (equivalent) would be more normal, but only a little bit.

But where did the standard definition of normal as equivalent to the image diagonal come from? This is where it gets interesting. It's picked based on the idea that images are typically viewed at about the same distance as the diagonal, and *at that distance*, the perspective appears natural. View it from further away, and your eyes expect a longer "normal focal length". Stick your nose right up next to the picture, and a wide angle suddenly appears normal. But in a "free" viewing situation, people will supposedly tend to look at things at approximately the right distance, stepping back to take in all of a big picture, and leaning in to see a small one. It would be interesting in a gallery or museum to do a study of what people actually do.

However, these days most pictures are not viewed in a gallery or even on your wall at home. They're viewed on a computer screen, which gives a quite different situation. Let's take a closer look at what that means.

According to http://www.w3schools.com/browsers/browsers_display.asp roughly half of their users - which tend having towards newer and larger machines than the internet average - use a resolution of at most 1280x1024. However, panoguide - probably a better sampling place for photographers - say that 25% of their users (in 2009) used 1024x768 monitors, probably translating to roughly 20" or laptops of about 13". 

The suggested viewing distance for monitors is between 30" and 40", though for laptop screens one is more or less forced to go down to about 20". Thus the viewing distance is roughly (very roughly) 150% of the diagonal, giving a "normal lens" length of 75mm.

But how often do you actually fill your entire screen with an image? Not too often. In the typical viewing environment - a browser - there is space taken by the window system and desktop, space taken by the browser, space taken by headers and footers and side information and whatnot. For horizontal pictures, there's usually not that much loss, maybe 30% of the width. Since the picture is scaled down proportionally, that means the diagonal is now down to 15", or 10" for laptops. "Normal" focal length is now 100mm equivalent. Whip out those macro lenses!

It gets even worse if we consider vertical pictures. Since most of the lost screen real estate is vertical, and the screen typically is placed horizontally, we can easily get down to 10" or less. How's a 150mm lens for normal? 200mm?

You'll notice a lot of "typical" and "average" and other assumptions in the above. Which only goes to show that there's so much variability that when it comes to viewing images online, there really is no such thing as a normal lens.

What I want to do when I win a bazillion dollars is open a gallery with plenty of viewing space, put up some sort of face recognition system that can determine where people are actually viewing various size images from, and get some solid data on the gallery situation at least.

In the meanwhile, all I can do (or hope others will do) is get some statistics from the browser and ask the user about viewing distance. That at least could give some clue as to just how normal normal is.

More ISO testing

or, what a photographer does when stuck in a hotel room on a rainy Sunday.

I should really have done this test many years ago, but now at least I have Lightroom 3 with its better noise reduction, and it really is better. The below is with the default noise reduction settings. Except for the first in each line, all pictures are taken with ETTR - since the motif doesn't have any real highlights, I could do one stop over the suggested exposure. The first one is (comparably) pushed. All photos taken on tripod with mirror lockup and remote delayed release. It's nice to have a tripod where you can mount the camera on one end and the thing to shoot at the other, and just move the tripod for better lighting. Speaking of lighting, it's unfortunately still tungsten, but at least it's the same light source all the way through.



I'm actually pretty impressed with this. There's some noise visible in the flat area, but it's pretty film-like and not very distracting. There is hardly any loss of detail in the focused area. The last row does seem to be a little brighter, but that could just be due to the resolution of the exposure display - the difference in lighting was done by moving the tripod further away from the light until the light meter showed a two stop decrease in light, but that can vary by 1/3 of a stop. A proper fixed light or filters would make it more accurate.

An interesting thing is that the "correctly" exposed images are a little deeper in the darks (may be hard to see on this crop). Like in the previous test, it is due to the black clipping being off in the "pulled" ones. So when doing ETTR, one should really adjust the black point to match, in this case move it up a little in the right-most 5 columns. Doubling it gives about the same effect as having normal exposure, but then I frequently move the black point up a bit just for the extra punch.

The bottom line: With LR3, I am not afraid of high ISO anymore. At the next nice evening, I shall go find myself a nice detailed scene with some sky and dark details and stuff to try a more realistic test. Studio tests only tell so much of the story, after all.

P.S. Stupid Chrome and its lack of color management.

An experiment with ISO and pushing

A long time ago, somebody introduced me to the concept of "pushing", from the film days where you couldn't just change ISO at the press of a button. Pushing means to deliberately under-expose the shot and then over-develop to compensate, giving a way to fake higher ISO. In the digital world high ISO is of course the standard by now, but can you also push digital? Of course you can, and it's even easier. Just underexpose and draw the 'exposure' slider up in your favorite photo program. The question that has long plagued me is: under what circumstances, if any is software pushing better than increasing ISO? The only 'circumstances' that really matter are the exposure times. It is well known that high ISO noise is worse in longer exposures than in shorter exposures, but is there a point where the noise from pushing is less than the noise from higher ISO? My initial guess is no, that there are advantages to "pushing early" that get lost if you wait on the pushing until the picture is shot, but we shall see.

To test this, I set of a simple test shot with my trusty if dated Canon 350D and the very trusty and not at all dated EF-S 60mm macro lens. Using various light sources, I took repeated shots of the same scene (camera tripod mounted, using remote release, mirror lockup and delay, just to be sure), starting with a correct (ETTR) exposure at ISO 1600 and the lowering the ISO while keeping shutter speed and aperture the same. I set the aperture to f/5.6, enough to have a reasonable DOF.

Due to the lack of a proper studio setup, I had to make do with an assortment of lights of decreasing intensity:
Flash aimed at the ceiling (1/160 sec), close-by tungsten lamp (1/25 sec), ceiling light (.5 sec), and a bounced tungsten lamp (4 sec, black-frame subtraction kicked in only on ISO 1600). This unfortunately makes it difficult to compare the effects at the different light levels, however the effect at different ISOs is what I'm after, and here are the results:


And this shows plainly why pushing isn't a commonly used method in digital photography. The noise is not the problem, in fact noise reduction (this is with LR3 default noise reduction) is so good now that you can hardly tell the difference between ISO 100 and ISO 1600 even on an old camera like mine (see below). The problem is posterization. In the top 1 stop (approximately), you have half the information in the picture. The next stop down has half of the rest, etc. So for each stop of underexposure, the amount of data is halved. With 12 bits per channel, you normally have 4096 levels total, of which half are in the top stop. So if the top of the usable dynamic range is 2-3 stops above medium exposure (which other experiments have shown to be the case for this camera), the midtones (say one stop to each side of medium exposure) have between 1500 and 750 tonal levels. That's plenty to hold a nice tone curve. Underexpose by a stop, and you have between 750 and 375, still a fair amount. Underexposed by 3 stops you have around 200-100 levels, which is not a lot.

And that's just the midtones. Consider the shadow areas - say from 1 stop below medium exposure and down. Here you have (again depending on where the top of the dynamic range lies) between 250 and 500 levels, again with half in the top 1 stop. This is already not a lot. The third stop below medium exposure has around 100 levels. With three stops of pushing, that drops to 12 levels. But here's the kicker: To remove the worst noise and to give some more punch, it's common to clip the lowest few levels, known as 'black clipping' or 'black level'. In the case of LR, this is done before exposure adjustment, so when the exposure is pulled up several steps, the black at the bottom is pulled far into the shadows and in the most extreme cases into the midtones. The clipping works well at removing noise, there is almost none, but removes even more of the precious few levels at the bottom, and makes the darks much darker. The result is very obvious in the image above: the darks get darker and start having major color flaws.

So what if we skip the black clipping?[1] Then we have something that makes more sense from a sensor point of view. All levels of pushing end up looking roughly the same at a large level, the darks are approximately the same darks, but now you can see the noise being pulled up, and how. Here are some 100% crops of details:





Voilá, the noise appears. Already at two stops pushing, there are noticable color errors. At 4 stops pushing, it is a mess of color noise, more than the otherwise pretty good noise reducer in LR3 can handle. The difference at one stop pushing is really hard to tell, though, even when flipping back and forth between the two. So it *is* possible to push digital, but you have to go against the normal workflow and you don't really gain anything.

What surprises me most is just how good ISO 1600 actually is with LR3. I've always tried to keep at 400 at the very most, but this gives me hope that ISO 1600 is actually useful. See some details for yourself:




Mind you, these were both taken with flash, I will have to do another test to see how various ISOs handle various levels of light when pulled through LR3. I have to say, though, I am hard pressed to see the difference between the two.

It's definitely much better than I expected. Gives me hope for my li'l photocapture box. And for my next test, I will have a more stable setup:)

[1] Which you can do on a Canon RAW file. I am given to understand that Nikon actually applies the black clipping to the raw data, leaving you without this option.

Old lenses

Took a trip into a local second-hand electronics store for a new CD player today (got a nice Denon, almost exactly the same as we broke after 20 years of service), and looked at a couple old lenses he had standing around while I was there. Nothing directly Canon-compatible, but some that could be interesting to adapt, or are just interesting:

Cosmicar Television zoom 22-66mm f/1.8 - apparently a C mount, made for 18mm video cameras, not big enough for EOS, but could work with Pentax K.

Dixi 135mm f/2.8 - not found on Google at all!

Mamiya/Sekor SX 55mm f/1.8 - M42 mount, so not useable for my tilt/shift ideas, but apparently quite good. May have to get this.

Danubia 500mm f/8 (72mm thread version) - T2 mount, supposedly good quality for the price. And it would be fun to have just for the sheer size of it (it's loooong!)

Also, since I'm looking for a better flash:

Tumax 320 TFZ flash unit - uncertain voltage, and looks a little cheap. I'd better go for the Strobist special if I want to upgrade.

Unless I can find something interesting about the Dixi somewhere, I may spring for the Mamiya and Danubia. They are both mountable with a simple adapter.

Printing: the beginning

A while back, I had promised my DW prints of my 5 best landscape photos so that she could do some fiber magic with them. For one reason or another, I didn't get around to it until today, where I brought out the only printer I have been able to afford so far: A Canon Selphy CP 760 portable postcard printer (not that I couldn't have gotten a bigger one, but there are priorities). It's small, prints only 10x15 cm, and uses a lot of ink and expensive paper, but it's a start. Later I'll get something that can do beautiful 33 cm wide panoramas and all.

At first print, my pictures came out very dark, much darker than they seemed on-screen. A quick bit of browsing revealed that a few brave souls had made printer profiles for the Selphy line, at least some of them. From this page, I downloaded the generic profile and the 720 profile and tried them out. The 720 was a bit better, but the colors still kinda muted. The generic profile, however, is actually really close to what I see on the monitor, with vivid colors and a pretty good feel to it. Makes me hopeful for home printing. I'll have to print the same pictures at my favorite print shop and compare.

This picture shows the difference, the left-most being with no profile, the middle with the '720 profile, and the right-most one with the generic Selphy profile. And before you say anything about overly saturated greens and Velvia disease, the early spring forest is actually that green in Denmark. I am in fact rather impressed that this little dude can reproduce it so well.