A brief hiatus

Another pause in blogging from me, I'm off on vacation for a couple of weeks far, far from civilization and the digital world.

More on my return...

The job of Auto Focus

With all the discussion going on in web land about Rob Galbraith's analysis of Canon 1D III AF performance, I have been missing one important thing: objective analysis of whether this is expected behaviour or not, and how it relates to the tolerance of the system. I've mentioned this before. This is the engineer in me wanting to analyse the thing and see if it's doing what I'd expect and thence, what changes would be needed to make it do what I want (if possible). This is a general discussion, applicable to any camera. It will be fairly technical but I hope even non-scientists will be able to follow it.

This got me to thinking, what exactly is an AF system doing and what results should we expect? This, then, is the result of me breaking down the job of AF so I can get a feel for how much we should expect. I figured readers might find it interesting, hence the post. Please, if you find any factual errors, leave a comment and I'll make corrections.

I will not talking about how focus is measured. I am going to assume that whatever AF is implemented, there is an objective, measurable system for determining focus that is in place. There are a two main ways of achieving this (phase detection and contrast detection) and from the perspective of AF, the actual focus measurement method is irrelevant here (as will become clear).

Defining terms

First of all, there are some important terms I'm going to use in the discussion, and some abbreviations.

Focus Tolerance (AFTol) - the range in which the subject is determined to be in focus by the AF
Focus Accuracy (AFAcc) - the degree to which the focus approaches optimal focus
Sample Frequency (FSamp) - the rate at which the AF system checks for focus

The red dot is the subject, the solid line is plane of focus, dotted lines represent focus tolerance (AFTol). Gap from solid line to red dot is the AFAcc.

Open loop - a system of control where a measurement is taken, error calculated and the controller moved to desired position. There is no feedback whether the point that the controller moved to is correct. This is how most AF works in SLRs: degree of out of focus is determined and the lens moved to where the AF thinks focus is.
Closed loop - a system of control where the degree of error is constantly checked. For AF, this involves moving the lens and then rechecking whether focus has been achieved.
Feed forward - A system of control where the measurement predicts into the future and then moves the controller where is expects the system to be needed. For AF of moving subjects, this would be the system employed for predictive AF, where the subject speed is determined and the focus moved to predicted next location. it can be applied in either open or closed loop mode.

The gamut of AF

Next up, what are the situations we need the system to cover? I reckon we've got three of them: static subjects, slow approaching subjects, fast approaching subjects (a retreating subject is just a negative approach). For the moving cases there are also steady and erratic speeds. We also need to consider AF under single shot and rapid fire. So that's a lot of operating cases to cover.

In the industrial world with a wide variety of operating modes, we would normally expect either to cover each case with a separate control scheme, or accept compromises in performance on the lesser ones, to hit the important ones. With cameras, that's going to be difficult: who's to say what's important? A landscape photographer is going to want different things from a sports shooter.

Allied to this are the AF control modes used to achieve the desired results. In general, there are 3 types: single shot, continuous and predictive. In the first case, focus is acquired once and applied to all shots in a sequence (which may have just 1 frame), the second checks focus for each frame, the third tries to get ahead of things with moving subjects by predicting where the subject will be on the next frame. In any case, open or closed loop may be applied but it is normal for current systems to use open loop control (i.e. measure once, move as desired).

Let's go through the main cases:

Static subject - fairly simple. Subject sits there, nice and steady, focus is achieved, shutter gets pressed. Think mountains, castles, cooperative portrait subjects. Focus needs only be achieved once.

Slow approach - subject moves forward slowly relative to the response of the camera. It is important to consider relative speed, compared to the camera, rather than absolute speed. More on this later.

Fast approach - subject moves relatively quickly relative to the sped of the camera. Of course high absolute speeds will be involved but the point when a subject moves from being slow to being fast will vary from camera to camera.

And now a quick look at the main AF modes

Single shot, single frame - focus is achieved once, there is one press of the shutter. Best used for static subjects. If open loop control is in place it will be almost impossible to catch moving subjects in this mode.

Continuous focus, multiple frames - With open loop control, the focus is acquired between each frame, producing a continuous shift of focus across the frames.

Predictive focus, multiple frames - The AF feeds forward the estimated speed to move the focus to where it expects the subject to be. This will require at least 2 measurements of focus (and therefore subject) position to be able to sense motion.

What are the implications?

OK, so we know what it's all doing, how does this work in practice? And how do things change as camera specs change?

First up is the notion of focus tolerance versus desired accuracy. A camera will think that a subject is in focus if it lies in the the tolerance bands. If it moves within the bands, the AF is likely to be unable to detect the motion (important later on). If, however, we can perceive, in the final picture, out of focus even within the tolerance range, then the tolerance isn't tight enough. In general, as megapixel count goes up, so this discrepancy gets bigger (manufacturers don't move their definitions of tolerance much).

Then we come to the relationship of subject speed and focus speed. If a subject moves slowly relative to the focus tolerance, then it may be hard to detect the motion (see figure below). If the first couple of points are within tolerance, it would take a third for motion to be detected. This would then lead to over-compensation 9front focus) as the subject is deemed to have moved twice as much as it actually has.

Slow movement relative to focus tolerance. Subject moves 1-2-3 but system only detects movement 1-3, shifting focus A-B for point 4 (now front focussed).

If rate of fire (fps) increases, then the speed that is "slow" in relation to the focus actually increases - the subject moves less between shots. Implication is that higher fps cameras have harder job with approaching subjects that may have been OK for slower models. Just slowing shutter speed may not help as the sample frequency won't change, unless the camera is set up to spend more time on focus for slower modes.

For relatively fast motion, things are better because the motion can be detected between frames (see below).

Fast speed - subject moves 1-2-3. AF detects motion 1-2 and moves A-B for point 3. Subject is in focus for subsequent shots.

If fps goes up, however, the fastest speed that can be focussed may go down as it is harder to move the lens between frames.

My conclusion

Faster camera, with higher resolution actually make focus worse, unless the focus tolerance is changed. Doing anything else won't help much. As fps increases, AF has less and less time to react, and will find it harder and harder to detect subject movement, unless tolerance is smaller or focus is able to detect movement within the range of focus tolerance. Resolution will make focus appear less and less precise.
Closed loop control doesn't help if the AF ha the same tolerance as open-loop. all it would do is minimise the chance of overshoot and provide a chance to correct, but that takes longer than just controlling to a more accurate position in the first place.

Image qualities aren't necessarily measurable

Some great points at the Landscapist on the techno-weenie discussions on-going about micro 4/3. Summed up: the mockers state that IQ will be poor due to smaller than other sized sensors.

There's not a thing as "Image Quality", there are image qualities, and because they are qualities (rather than quantities) they are by nature subjective.

The current talk that all shot have to be perfectly noise free at the pixel level at EI 1 gazillion, misses the point about qualities of images. back when getting ISO1600 meant grainy black and white, I don't think there was a hankering after ISO25 like smooth tones. The high speed leant itself to particular kinds of situations and images and conferred certain qualities therefrom.

There is one aspect to the digital techno babble that I can relate to (but rarely gets mentioned) and that is about lost changes. By that I mean desirable things left out, or undesirable things added, which mean we are stuck with undesirable results. Mostly this gets talked about in terms of lost resolution or added noise. The point being that once we have an unrecoverable result, there is nothing more we can do. However, there are too few talking about what can be achieved with given parameters. Many "noisy" cameras produce very nice black and white in low light, a digital return to the days of grainy film. It is a different set of qualities conferred on the pictures.

One group who seem to buck the trend are the Sigma DP-1 fans. They seem to talk very much about what the camera can do, rather than what it can't. Of course, if you actually want it to do something it can't, it's not the device for you.

But of course, all that requires thinking, which seems to be the antithesis of modern cameras.

Kinesis Journeyman: first impressions

I've just received my new Kinesis Journeyman backpack system. I'll run through some first impressions, but first some background.

What I want in a photobackpack

I was in the market for a new backpack, my old alpine style one being unsuited for camera gear and the padding long since compressed to nothing.

I've posted before about photo backpacks. I'm disappointed by all the offerings from big names. Pressing a regular backpack into photo service is also a non-starter.

What I need is a 30-40l daypack that can carry my photo gear and my regular stuff (like lunch, water, waterproofs) comfortably in the hills. I can end up carrying 10-15kg easily on a day out. I also need to to carry my tripod and a pair of walking poles, plus all the camera bits and pieces in an organised manner.

I also want to be able to carry the photo gear near the top of the pack for better weight distribution. I prefer the tripod lashed to the side. While this sets the weight a little one-sided (I've often got a side pouch on the other side anyway), it is also closer to my back, which is more important to me.

Must have a proper harness for comfort over longs days in rough terrain.

Where the market falls down

Photo backpacks come in 2 camps: the huge expedition size things or small set-ups. Almost all are designed with only photo gear in mind - clearly not aimed at those of us who hike stuff into the hills. Split compartments always put the photo gear at the bottom. Harnesses are mediocre at best. Weight is high due to all the padding. Quite frankly, I'm just as well off using my Think Tank carry-on.

Regular backpacks are really designed for hiking or climbing only. No real way to put camera gear in well. In the day-pack size there are almost no dual-compartment designs to allow me to stow camera and other gear separately. Lots are also coming with these new ventilated back systems where there's a tensioned mesh against the back and gap to the pack - this moves the weight away from your back which is bad, especially for heavy loads.

I figure in today's world, I don't need to compromise or buy lots to find something suitable. I just keep plugging away until I find someone who's got it together.

Enter Kinesis.

A bit of a one-man-band enterprise, this is a novel line up of photo gear. All modular in away that fits all kinds of packs, pouches, belts and harnesses together in anyway you chose.

Their backpack offering is the Journeyman, which can be combined with all sorts of options to create the perfect set-up for your application. It's a proper backpack, with the ability to hook-in camera pouches. Pole loops & tripod mounts as standard. A host of accessories to add on.

What I ordered

A lot of gear (see picture). Pack, hydration pouch, accessory pouch, interior pouch, straps & doohickies for lashing stuff on. Clip system to attach camera to the harness.

Kinesis Journeyman (centre) and components

Configurations

The stuff I've ordered allows me to do the following:

• Backpack, with or without camera.
• Removable harness, removable belt - great for carry-on.
• Extendible with the hydration pack as an extra pouch.
• Slap everything on for a big day out.
• Hydration pack only.
• Camera pouch & belt as a waist pack. Add hydration pack for a lightweight all-day set-up.
• Camera pouch as a shoulder bag.

Loads of options, great for travelling or sightseeing or hiking.

What do I think

Without using it in anger, here are the thoughts out of the box.

This thing is pretty large. Bigger than I'd expected. Pack measures 55x30x20cm (22"x12"x8"). It's a square shape, too, so you get full volume for your size. Weight is reasonable, not the lightest or heaviest. The harness is good, though, so the weight is light in the shoulders, which is what counts.

Journeyman harness system opened out
Back padding folded down & belt removed to show aluminium struts

Materials and construction are all top quality. harness sizing is pretty large. I would say larger than the website credits. I'm 6' tall, not too fat, and the medium/large harness is cinched right down when I've got it on - the smaller size would have been fine. Even then, the chest strap is at maximum extension. I went down a size on the belt and am glad I did. Side mesh pockets will actually be useful. Most sacks have very flat designs or they're small so everything falls out. Not these - expanding sides, large and a good stiff top band.

The internal pouch is excellent. The lightest I've come across for the size, and it's large. Fits easily my 4x5" field camera or the Mamiya RZ67. It's SLR plus 3 lenses territory - 70-200 stands up easily inside. Top opening is well designed. Most of the padding can be removed which is also useful.

Journeyman opened with inner pouch in place
The Pouch can it attached anywhere through the height of the pack

All of the accessories and pouches snap together easily. This is truly modular design. There are all kinds of ways to do things and I've been spending hours working out which will be best for me.

There are a few negatives. Harness strap design doesn't quite come down over the back of my shoulders as I'd like. I'd prefer the top attachment a bit further down the pack for this - allows better with distribution over the shoulders.
Attaching the inner pouch to the belt for a waist pack has a problem. The pouch is a "triple wide" which means it spans 3 attachment points. Trouble is, the attachments are evenly spaced from the belt centre so the waist pack is offset from centre.
Inner pouch attachment only has one attachment point so it hangs off the fixings. This means it's loose at the bottom which may cause it to swing around or gear to get trapped underneath. Time will tell, but I think a bottom attachment would have been good.
The front pocket, whilst useful for putting small stuff in, has no extra pockets or attachment points. In effect everything has to be loosely stored. even some Velcro strips would be useful.
Colour - it's not uniformly black but I'd like the grey side panels to have a red option. Red is a much better colour for visibility in the hills or poor weather.

I'll follow this up with an in depth review after my up-coming trip to India. It should all get a thorough working over on that one so I'll really know what it can do.

UPDATE 25/9/08: my field report now online.

The moment in time

Cartier-Bresson had his decisive moment. Now comes the Landscapist talking about the notion of freezing time. Interesting idea, that coincided with earlier thoughts I had on "The Americans". Coincidentally, I was studying Frank's book the night before reading Mark Hobson's post. I then went and looked at his new "Shore Light" gallery. As seems to happen to me frequently, these things all came together in a series of thoughts.

When a photograph freezes a moment in time, is it any moment in time, or a very specific moment from a specific time? How is it we are able to place the picture in space and time? We often, consciously or not, derive whole stories, histories from the contents of a single image. Is the photographer helping us in that effort, or deliberately obfuscating to give only a hint or glimpse?

The stronger thought that came through, however, was the fact that many photographs give a distinct sense of specific place and time both geographically and historically. There is no getting away from the fact that "The Americans" is a slice of 1950s America, and I wrote before that the effort could easily be taken up once a generation without losing impact. It is also clear that "Shore Light" shows a sense of now about the place.

I see two reasons why this is the case. Firstly there are the obvious cultural references - clothes, cars, hairstyles. The second is in photographic style, which is much harder to put a finger on. Some of it comes from materials and technology. Some of it comes from means of presenting the subject. I wondered if this second aspect would work for other subjects? I don't know but it would be interesting to see.

Take a subject well covered over many years but I would think something without an obviously iconic shot. A great landmark that's been photographed since there were photographs. I reckon Mount Fuji in Japan would make a good example. Then take 1000 shots spread across the almost 150 years, and I bet you'd be able to pretty closely mark the timeline. There are obvious differences in processes available from late 19th century to early 21st. But I also think there would be a great deal of difference in the way the subject is presented that would change too. even such a static, simple one as that.

Maybe, then, we are capturing as much about the times of the photographer as the times and moment of the subject.

Manual, what manual?

Picked up via dpreview, this survey about technology bafflement. Doesn't surprise me. Like every bit of technology, the way you're going to get the most out of your camera is to know how to use it properly. trouble is, most people I meet want to be able to pick up the camera, point at the subject, push the button and get a good picture. back to the Instamatic days.

Therein lies the problem. There are 2 types of camera user: the Instamatic crowd who truly want point and shoot (i.e. no functions, adjustments or doodads) and those of us who want full control over everything.

It make me wonder why all those whiz bangs get added to cameras when neither group wants them.

A passing thought - the number one question I get asked by friends about their cameras: how do I turn off the flash? (Yes, I seem to be helpdesk man for cameras because I'm into photography).

Neither here nor there follow up

Following up from yesterday, Wikipedia is already up and running with micro 4/3. Plus, it's got all the lens mount details too.

With this size, looks like an absolute winner for digital RF, given that LTM & M bayonet should be easily adapted.

That's the optimist bit. The pessimist in me thinks that this will never happen and we'll end up with a limited line of camera with crappy interfaces, no optical VF and no hot shoe for accessories either.

Neither here nor there

but somewhere in between.

Quite a bit of buzz on the new micro 4/3 format. RAWsumer thinks it may mark the death of digicams offering RAW, Michael Reichmann thinks it's a dead end. In a "let's weight up the evidence, fair hearing" kind of way, my thoughts (for what they're worth):

It's an interchangeable lens format. Will that really mark the doom of fixed lens cameras? I don't think so. I don't think micro 4/3 will really be small cameras, in the shirt pocket category. Look at the bigger brother: supposedly a smaller DLSR format yet the cameras aren't much smaller than their bigger sensor rivals. No one ever accuse RFs of being pocket cameras, and they're the smallest interchangeable lens cameras available (AFAIK).

I think it might offer an interesting platform for a digital rangefinder. OK, so it's got a 2x crop factor. Plus it uses what seems to be a larger mount than Leica thread mount or M. But the deal with the format is to get around all those nasty edge of sensor problems that The M8 has hit hard. I don't agree with Reichmann that the M8 has proven the way for large sensor RF, quite the opposite. I think it's so beset with inherent problems that different solutions may be needed (many of which may be software, rather than hardware). Still, I think there is potential here.

Will it rival DSLRs? I don't think so, at least not the larger ones. Optical viewfinding is still superior to electronic and this new format removes the possibility of reflex viewing (hence my thought on RF). Might make an interesting alternative to the entry level DLSRs, but I bet it'll never compete on price (which is almost the number one differentiator in brands for real buyers).

That mount is still large, as I've mentioned. Smaller it may be, but only that the big 4/3. Compared to RF mounts, I pretty sure it's jolly large. which starts to preclude really small lenses. For me, a small lens needs to have a small diameter for comfortable holding, not just be pancake flat. I don't think this large mount, aimed at motor driven lenses, will give that small diameter I'm looking for (as a point of comparison with SLRs, I think only the smallest, cheapest, non-IS lenses come close to being small diameter, and optically they're pretty rubbish).

So where do I think it fits: in that medium size, carry all day slot currently held by RF cameras. Give me a small package with a good optical VF and small, fast primes and I think it would be useful. With a decent range zoom as well, it would make a strong travel companion, that might convince me to leave the SLRs at home on trips, especially with the bike. I still want a high quality compact with RAW, for the shirt-pocket option.

Too familiar or too superficial?

Picking up on a running theme of late, and following from Doug Stockdale's comment on my wet weather post, is the idea of becoming over familiar with things.

For residents of Northern Europe, rain is an everyday occurrence (OK, maybe not every day, but pretty frequent), whereas in sunny SoCal it's more like a desert. As Doug points out, rain for him is pretty unusual.

So can I be too familiar with rainy conditions. Have I blocked out the rain effect so that everything looks the same to me wet or dry? Can we start to tune out subtle variations because they are too familiar to us? That may not be a bad thing, instead of observing the superficial differences (in this case the weather), we might be able to get past it and deeper into seeing the subject. On the other hand, it might be that an important part of the character of a subject is lost to us as we've tuned it out.

I think I may have to make an effort to walk familiar streets on a rainy day (and when I've got suitable clothing to hand) to see if I can spot the differences it brings, delving into this conundrum.

Wet weather photography

It's raining here, has been since mid-afternoon yesterday. Not the weather to get me out on the bike (but I enjoyed the sunny morning conditions yesterday), nor is it weather to inspire me to take the camera out.

This is a recurring theme - I don't take photographs in the rain. Even if I happen to be out in the weather, with a camera to hand, I don't press the shutter. Why not? Partly, obviously, is the general desire not to be out getting wet, nor risking screwing up the camera. Main, though, I think it's because I don't really see images in the rain. There aren't subjects that inspire me that wouldn't be done better when it's dry.

What is a wet weather aesthetic anyway? I've seen some stuff done in the wet, and it seems to focus around reflections & puddles. Gets to being the rain equivalent of glorious sunsets.

Anyone out there actually making pictures in the rain? What are the things you're looking for that's particular to the conditions?

The megapixel race is not over

There was much made of the new Panasonic DMC-LX3 camera announcement that stated that sensor had been enlarged (to a whopping 1/1.63") and resolution restricted to 10.1Mp. So is it a slow-down in resolution. Not likely, all the other, more basic, cameras announced at the same time had increased resolution on their predecessors to 14+Mp. Likewise other manufacturers. Until the sales process (from marketing to the stores) can get it into their collective heads that people need better pixels, not more pixels, the more this rubbish will continue.


The good news is we're getting another compact with RAW mode, burst shooting an wide-angle. The lens is even pretty fast. I've seen some (low res) first shots and low light performance seems decent. One wonders if there will be a Leica-branded version, too. For once I'm actually pretty interested from a buying pov in small cameras. Both this and the Ricoh GX200 look like pretty good contenders for my cash.