-as described on b3ta

Statistically speaking, if you’re reading this site, it’s because of this video:

It’s actually a few months old, and (as the name suggests) the second attempt at a project, Glide, that I described here back in July. I also posted it to Reddit then, garnering a few hundred hits, which is about as much as I can usually hope for. However, Glide 2 was languishing at forty-something views, so last week I thought I’d try and get it some more attention by posting it to b3ta. It made the newsletter, spread in a few directions via twitter, and the hit counter seemed to stabilise just above 5,000- so I smugly posted that on facebook and set off to Edinburgh to graduate.

Except it seems that the hit counter hadn’t so much stabilised as jammed: by now it’s gone thoroughly viral, with just over 100K views- more than everything else I’ve put online, combined, by an order of magnitude. However, along the way plenty of people have been a bit confused as to exactly what they’re seeing, so I thought a proper follow up post was in order.

The Method

Both glides were filmed by sticking a – relatively cheap – digital camera out of the window of a train as it arrived at a station. The ‘trick’ is the camera collects images at a rate of 210 per second – but the film is played back at 30 frames per second. So, every seven seconds of footage that you watch corresponds to 1 real second. At least at the start, one real second is plenty of time for someone to move into, then out of, the camera’s field of view, but isn’t enough time for them to really do much: hence, the frozen effect. It breaks down towards the end not because I’m doing something clever with the frame rates (captured or replayed), but simply because the train was stopping! Thus, as it decelerated, any given person would be in view for longer, and have more time to point an arm, take a few steps along the platform, or maybe even notice me at the window. Any such action captured is still slowed down seven-fold during playback, just as with my usual static captures.

At least one other person has tried this before: Trey Ratcliff captured a station in Japan this way over a year ago, describing the effect as “Stuck in motion“. He also mixes in other slow motion footage and its inverse, time lapse photography, in this gorgeous video, Heartbeats of Time.

The Technology

Was cheap! I used a Casio Exilim FH20, which Amazon offers for just £230 these days. There’s a slighty newer iteration, the FH25, at around £300; the slimmer FH100 with a shorter zoom. Their big brother, the F1, can be had for £600 or so: still reasonable compared to many SLR lenses, which can be just as narrow in purpose. Still, Casio, feel free to send me one for future projects 😉

As for the train (less cheap), I’d seen the service described as everything from high-speed to a bullet train, much to the amusement of anyone who’s spent time on British rail… On the other hand, my original post described how you could get a subjective speed of 5mph by 210fps filming at 35mph, so a lot of posters assumed that’s exactly what I’d done and that it was crawling along. Anyway: I was travelling on a Bristol-London Intercity, which I believe is capable of 125mph, but I’ve no idea how fast it was entering the station. Presumably there’s a limit even for non-stopping services, but obviously this one had to be slow enough not to overshoot the platform. People have tried estimating the speed based on everything from the likely length of benches and the time taken to pass them, to the flickering of lights on mains frequency, but it might be easier just to ask someone at the station next time I’m there. Update: Various people have claimed (without sources!) that the limit at Bath Spa is 35mph due to the curve of the platform.

The Sound

There isn’t any! Lots of people have commented that it needed a soundtrack: but the suggestions varied over several genres, so perhaps it’s for the best that I didn’t impose my own tastes on others! Plus I was hesitant to use copyrighted content lest it get the whole video blocked. Still, if you think you’ve found the perfect score for either Glide, feel free to mash them together with Youtube Doubler or the like and let the world know in the comments.

The Action

There isn’t much of this either! I feel that adds to the experience: it’s repeatedly been described as mesmerising or hypnotic. I wanted to capture a slice of ordinary life that wouldn’t otherwise be seen, and adding a gimick would detract from that simple pleasure. I think it worked, particularly for international viewers who have picked up on everything from British signage (“Way Out” rather than “Exit”), to the relative slimness/attractiveness of average Bathonians compared to Americans. Many comments were to the effect that “this should be boring, but somehow it wasn’t”.

Still, there were plenty of suggestions for seeding the platform with characters to add excitement: gymnasts mid-jump, a timelord rushing through at ‘normal’ speed (any sprinters with a taste for cosplay?), a waiter spilling soup, a streaker… some celebrity cameos wouldn’t hurt either, although the youtube crowd think they’ve spotted a few already!

The Legality

No, I don’t have model releases for anyone, let alone everyone, on the platform… in the UK there’s no default right to privacy in a public place, and I wasn’t trying to film anyone in particular, paparazzi-style. So I believe this is acceptable use as an artistic endeavour, but that probably changes in a commercial context. So, no google adsense revenue for me, and for those who have got in touch asking to use it for a music video or the like, I’m afraid it’s not going to be possible. I suppose you could stage such a thing with consenting extras, but then you might as well ask them to stand still and save yourself the need for slow motion kit…

The Future

So I find myself with dozens of youtube subscribers that are presumably hoping for a follow up, and the fear that I’ve already filmed the most popular thing I’m ever going to… I’m always trying to think of new slow motion projects in general, but there’s still plenty of potential for glide-like captures: I’m thinking of switching from rail to road next time, although not being able to drive could complicate that!