Percussive learning

Percussive learning in a world of automation is what it says in the subtitle to this blog. What that means is me making mistakes. My old boss used to say “if you think you’re good, you’re not comparing yourself with the right people”. In my case, I think the right people have me well outnumbered. It’s been very humbling to find that stuff I have just learned is well-known to everyone else. So this blog is not instructional – I have no secrets to impart. Where I witter about something technical, try it yourself before assuming I’m right.

Slightly uneven development, perhaps?

But one thing I have learned is that it’s ok to make mistakes. And if you accept that you will make them and that being wrong doesn’t mean you’re stupid, then you get to learn. Learning means thinking about your mistake and working out why you didn’t get what you wanted. Or maybe looking at a picture and asking yourself why you are less than delighted. What would have made it better? Then think about how to do more of that.

This means reflection. It means saying honestly to yourself “I wish this was more…” and then working out what more means. If the definition of madness is repeating something and expecting a different outcome, then reflection is the road to sanity. The aim of reflection is to not repeat the same mistakes. Many of them will be similar, but what you are hoping for is to break things differently each time. Or if not break things, then to hone in on that thing in the picture you want to do more of.

I’d like to introduce you here to another idea – that of reducing variability. Anyone involved in manufacturing will know this inside-out (see above for me being outnumbered). Every process has a natural level of variation. What the manufacturing people strive to do is understand what this natual level is and then reduce it as far as possible. Then, if the output moves outside the expected range, you know that something is wrong.

Wave your hand under incandescent light and you will see the same effect

What I have learned in photography is that my methods need to be precise, so that I can understand why something went wrong or be able to repeat something that went right. Mostly this means consistent exposure and development. I always use a light meter for example. Then, by looking at my results, I learn to understand and use the light meter better. I have been a bit of a tart for different developers in the past but I’m over it now. When I started out I developed just about everything in Aculux. These days it’s Rodinal for its keeping qualities. Like the metering, I always use a thermometer to get my chemicals to the right temperature and I use the same method of development and agitation every time (unless I am deliberately using stand development). So the results are that the negatives should be correctly exposed and consistently developed. Any variations that show up were either a deliberate choice or a mistake.

Anyway – that’s the theory: reduce variability so that it’s easier to understand what happened when the unexpected occurs. And then keep trying new things to see what happens. Break things with deliberate care.

 

Advertisement

Ars longer

I’ve been scanning a load of my parents’ and grandparents’ old negatives. Amongst all the fuzzy shots of relatives on holiday was one film’s worth of long-term shots taken in my parents’ garden. They were taken by me of a family occasion. I had obviously given both the prints and the negs to my parents so they could make reprints. I’d forgotten all about the pictures and the occasion. But aside from that, I’d forgotten how good a long lens is for pictures of people.

These days I tend to use standard to wide angle lenses. I get environment and context in the shot. But the simple joy of seeing a single figure separated from background, not particularly aware of your presence – it’s great.

It was a sunny Summer day and I was using 200ISO colour negative. I only had two long lenses at the time and it was obvious from the shots that I didn’t use the zoom. So these were shot on a 135mm lens. It was bright sunlight and I know I was shooting with flash, so the fastest shutter speed I could have used is 1/125 which means this lens must have been stopped-down and not wide open. It still blurred the backgound nicely though.

I think this focal length has gone out of fashion: it’s thought to be too long for portraits. One of the photo podcasts described it as a focal length that was invented to let German hill-walkers pick out a detail on the other side of the valley. I guess it would be too long for indoor portraits, but it worked perfectly in my parents’ garden. I’ve even got full-length portraits.

So I’m back in love with the 135. The one I was using at the time was a Pentax – the SMC Takumar f2.5. In the years since the aperture blades have become oily and slow, so I need to send if away for some TLC. But a Vivitar f2.8 came up on eBay at under ten squids and now it is mine.

Back in the old days we used to know that with a 50mm lens on 35mm (or full frame) held portrait, a person would nicely fill the frame at about ten feet. I just did the sums again and the field of view for a 50mm lens on a vertical 35mm frame is two meters at nine feet (ooh, nice mixed metrics – it’s 2m at 2.8m distance).

Slap a 135mm lens on and the distance increases to 24′ (7.4m). So yes, you’re unlikely to be shooting full length portraits indoors. I do like the effect though and it will come in handy with our current distancing and separation.

Down sides? I had obviously not used the Pentax lens much in recent years. Perhaps it really was a bit too long for people and not long enough for sport. I do remember that it was never out of my camera bag at the time. Perhaps then it was just me changing the sort of thing I shot? Maybe I got old and slimmed-down the camera bag?

So it looks like I’ve rediscovered the modest tele lens. Let’s see what I do with it during my government-allowed exercise period.

Changing DX codes

The DX code is a black and silver block-pattern code on 35mm film cassettes, introduced in 1983. It is used by some cameras to set the ISO and by film processing machines to tell them about the film. All well and good, but sometimes we want to push or pull the film and this means setting a different ISO in the camera. If the camera can’t be controlled or over-ridden, you can change the code on the film cassette itself.

This is a typical code:

Three silver, black, silver, black = 200 ISO

The code is read as shown – with the barcode and film at the top. The ISO coding is the top half of the code and forms six panels read left to right. The first panel is always silver/ metal.

And an aside – the lower half of the code panel shows the length of the film and its tolerance for over and under exposure. This one is 24 exposures and +3 to -1 stops.

You change the ISO by scraping the black paint off one or more squares or by covering them with tape. The camera uses electrical contacts to read whether a square is conductive (silver) or not (black).

There is a table below of the codes that correspond to each ISO.

So one of the common hacks would be to rate a 400 ISO film as 800. To do this you need to scrape the black paint off panel 2. To push it to 1600, leave panel 2 and scrape off panel 3.

For more information, plus how to decode the lower section, see here.

It’s also possible to make a completely new code by scraping all of the panels and covering some of them with tape. There has been a revival in using some of the other films in Kodak’s former catalogue. Many of these are known by code number instead of a common name so it may not be obvious what ISO to use. There is a handy decoding list here for Kodak and here for Fuji. So should you find yourself trying to shoot Kodak 2430 in an automated camera or trying to reload old film cassettes with a different film, help is at hand.

… And another aside – follow that link for an appreciation of just how many types of film Kodak made.

Simple lenses and choppy bokeh

Remember the film title Things to do in Denver when you’re dead? Well this is things to do in the house when you are trying to not be. In the house I mean, although not being dead is important too.

Despite all the therapy and support, in times of idleness one’s fancy can turn to things of bokeh. Or the unimportant bit, as we used to call it. That’s when you realise you have several lenses of the same focal length but different construction. And you wonder why, and what the difference is between them. And if the really clever or expensive lens is actually better than the cheap one. And what better means. And if this means you need to buy more lenses. Or find better backgrounds.

So you find the adapter that mounts your cheapo nasty analogue lenses on your digital camera. This takes two hours and uncovers more interesting old lenses. But you only have one version of each of them. So when it gets dark and eBay closes for the night you come back to the plan. One focal length, one aperture, one scene: the ultimate shoot-off. But is it fair to compare a lens that can do f1.2 with a lens that has a maximum aperture of f3.5? Do you compare them all at the same aperture or at their widest? More time passes.

After a few more beers you decide that the reason you started this (and it is totally your fault) is to see what the backgrounds looked like. Kind of an aide memoire of which lens to use to get which effect. So it’s fully open aperture on all lenses to get the best of their aberrations. But to use a scene where the point of focus is close, so that we get lots of the fuzzy. And we reassure ourselves that we haven’t gone all Photography With Classic Lenses and that we still have a sliver of pride and self respect. And then we laugh and drink more beer.

So – big question: what sort of fuzzy does it for you?

Me? I’m a smooth guy. I like the out of focus areas to just look less sharp. I want my background to be background. Smooth tones. No magic circles or swirly. Nothing that looks like the subject is stood in a whirlwind or bubblebath. The only reason to use the weirder lenses might be that the pattern in the background somehow adds to the shot. Otherwise it’s the background. But that justifies the other lenses, right?

Now, some of the theory to this is that the rendering of the background is affected by how well-corrected the lens is for spherical aberations. And a simple magnifying-glass lens gets better corrected for other aberations by adding more elements to the design along with different types of glass. It’s fairly easy to make a reasonable lens at small aperture sizes, but the wider you want the lens to be the more clever the design has to be, which often involves yet more bits of glass. So the argument is that a simple lens ought to have more textured out of focus areas than a more complex and better-corrected lens.

So where does that leave me on a Sunday afternoon? I have, in the same focal length, lenses with from 3 to 7 elements. I have lenses that span the all famous design formulae. To be honest, I have a sufficiency of lenses. And I have time.

So, the simplest lens is no lens. But I have no pinhole cameras at the moment so this is out. Same with a simple meniscus (but been there, done that on medium format) or a doublet. Got a triplet though.

So I’ve got:

• Triplet – a Meyer-Optik Goerlitz Domiplan 50mm F2.8.
• Four elements in three groups (Tessar) – an Industar 50-2 50mm f3.5.
• 6 elements in 4 groups and probably a Planar design – Rikenon 50mm f1.7.
• 6 elements in 4 groups, but a Biotar design – Helios 44 58mm F2.
• 7 elements in 6 groups, probably a version of the Planar – an Auto Yashinon 55mm F1.2.

This is what I had to hand – I am not spending time on eBay looking for a five element lens.

Now, everyone buys the Domiplan lens because it does bubble bokeh: out of focus highlights turn into circles. With only three lens elements, it doesn’t really correct for much at all.

At the other extreme, the Yashinon has seven elements, not so much to make the lens sharper but to manage the aberations in a design that is more than two stops wider than the Domiplan. Somewhere in the middle might be the sweet spot of fast enough and smooth enough – the Goldilocks lens.

My first idea was to shoot them with a nearby lamp as my subject and a window behind to give me some highlights. That didn’t really work, as they all looked quite similar. The only one that stood out was the Yashinon. This was because the design that was pushed to work with a front element almost two inches in diameter caused some crescent-shaped highlights away from the centreline of the lens. This is the downside of pushing a lens design this far.

So instead of dithering indoors, I persuaded Wilson‘s cousin Gilbert to stand in as my model in the garden. He’s a rugged fellow with plenty of skin texture, so ideal.

The background is quite busy and showing bubble highlights

Smoother background. This is a lens I like.

Nice – I’ve had this lens a long time and I can see why.

The background is getting a little swirly.

What background? Was it foggy?

So, what do I think?

Well, one thing was that some of the lenses had noticeable field curvature. In each case I focused on the B in the centre of the lens than reframed the shot. In some, like the Domiplan, the sharpest point has moved nearer the camera. Something to be aware of – I can’t just reframe some of these lenses if I use them wide-open.

The Domiplan is a cheap lens that does the bubble trick on background highlights. It’s OK for that but it can make the background look quite busy.

The Industar 50-2 is great – I really like this lens. It’s tiny and it renders backgrounds smoothly. A bit of a pain to use though, as the aperture is manual.

The Ricoh/ Rikenon is the standard lens that came on my first SLR camera. It does a good job and doesn’t intrude. It works on my film kit and on the digital and gets the highest praise: it just works.

The Helios is a real cult lens because it can make backgrounds look swirly. It’s slow to use, as mine is one of the older ones with a preset aperture. Given the choice I would usually take the Industar, because it is so much smaller and lighter than the Helios.

The Yashinon is bonkers. It does crescent-shaped highlights, it does completely fogged backgrounds, it can see in the dark. On digital it also gives coloured fringes and a bit of a glow to things – see the shots above. Unless you want to shoot at F1.2 all the time it needs to be used on a proper M42 mount, so I use it for film work.

My conclusion? The Domiplan and the Helios are special effects lenses. The Industar is a superb pancake lens with good rendering. The Ricoh is a rock-solid standard. The Yashinon is there for when you really do need to separate the background or to take pictures in the dark. And your mileage may vary.