Coldplay Wristbands Turn Audience Into Giant LED Display | Hacker News Reader

Coldplay Wristbands Turn Audience Into Giant LED Display

mashable.com

45 pointskator19 comments

Myself and my brother built a prototype of a very similar wristband last year, completely independent of these and without knowing about them (we built ours in the summer, we first saw Coldplay's in autumn). At first we were pretty dismayed and annoyed to see that somebody had patented the concept (we never looked into exactly what the patent covers), though one of the guys we were working with has connections in the Coldplay management team and assured us its fine for us to continue.

But then we thought about it some more and realised that we didn't need to, that the coldplay wristbands have some flaws (for what we wanted to do with them) that we can better tackle in other ways. Our goals are very different from Coldplays anyway - they seem to be using their wristbands more to make the DVD look awesome than to please the crowd present (though theres obviously some of that too). Our goal was always interactivity and engaging the crowd in more direct ways.

So we cooked up a similar concept which is not only more suited to interacting with the crowd, but also less expensive to produce. So win-win.

The wristbands are cool and maybe we'll still do something with that (depending on the patents of course - the Xylobands don't really do exactly what we want, so we'd probably want to build our own anyway) but we did the startup thing and pivoted and came up with something better (for our purposes).

The laptop would be connected by cable to a transmitter box and antenna. These are supplied by us on a free rental basis with a £500 refundable deposit.

Our own transmitter cost us about US$20 to build. Sure, it could only be controlled by USB-MIDI (while afaik theirs supports stuff like DMX), but meh, it wouldn't require a £500 deposit.

In any case, I'm now repositioning myself as more of a backseat tech consultant for this project rather than hands on engineer as I'm working on an unrelated startup and don't have time to stay involved in the day to day building of audio/visual gadgets, so over the next few weeks I hope to finish off a few more prototype devices and then leave the production up to someone who has a better idea what their doing.

The most obvious improvement they could have made would be to vary the transmitter's power so that they could create a (very very approximate) circle of light around the transmitter point. Raise and lower the transmitter power, and you increase and decrease the size of this light circle. Great for pulsing, explosion style effects.

Different transmitters in different locations could trigger different colours.

Thinking further, they could have introduced multiple low power transmitters across the audience, or highly directional transmitters pointing at the audience from a central spot. You could then create colour clusters, rainbows, spinners, rather than just a uniform effect.

Fun project:

- add an IR LED to each band and make it transmit a unique 2 byte MAC address on the IR channel(s)

- get a high speed, high res panoramic camera and some good computing gear

- figure out from the sequence of frames where each MAC address is

- control individual pixels

- add an IR LED to each band and make it transmit a unique 2 byte MAC address on the IR channel(s)

I would think it would be easier to just put a serial number bar code on the back lining up to a MAC of sorts and scan it when the person arrives correlating the the serial number to their seat. That should give you all the location information you need, at least for the non-standing room only areas of the stadium.

They already have visible LEDs. You could just use that. Flash different subsets of bracelets on and off while an algorithm correlates bracelet IDs to pixels. It would only take a few seconds. Hell, the audience would think it was part of the show.

I'd be more impressed if they were able to individually (or in small geographic groups) address the wristbands. Then you might be able to get some sort of interesting visualizations as opposed to pulsing.

But, it is a neat trick, and certainly looks cool from above.

They need to incorporate location of the wristband with some sort of triangulation between transmitters. Then they can make it an actual display. I'm sure this will happen in the not too distant future and I'm excited to see it.

Using highly directional antennas on the transmitter should be able to accomplish similar results without increasing the cost of the wristbands.

What if these bands are paired to phone via bluetooth so that the base station can send commands to the phone (through the mobile network) which in turn relays it to the band - would it allow a more fine grained control with acceptable latency?

Latency would be huge with that setup. 900mhz radio has more than enough range and bandwidth to handle 50,000 - 60,000 devices updating several times a second. I've used LSR [0] modules in the past for prototyping and they have a gateway that would allow you to control the entire thing through ethernet so any computer could run the program and just spit out ethernet packets to send control information.

[0] http://www.lsr.com/

Modulating the antenna's attenuation would add depth-control, which may also be interesting.

You wouldn't even need direction antennas to do this, just some number of reasonably spaced omni's that could feed signal strength information. The problem with this approach is that the wrist band needs to transmit to make this at all useful which adds cost to its design vs being a simple receiver only.

I mentioned this in another comment but it should be easy to correlate a seating assignment to a wrist band MAC giving you reasonably good location information. It wouldn't work "in the pit" obviously but you would be very hard pressed to get any accurate enough location data in an area like that.

Another interesting option would be to have the wristbands communicate by radio with nearby wristbands. Probably not useful to draw precise stuff, but could be fun to have possibly beautiful abstract patterns like cellular automata.

As ZTO and bluesnowmonkey pointed out, if the wristbands became individually addressable, you could determine their locations in the crowd using an overhead camera and a quick calibration routine (which the audience would just think was part of the intended effect; It would just look like more random flashing).

More info: http://xylobands.com/glowbands-product_info.php

> How do they work?

> Xylobands™ are operated using our proprietary software that can be downloaded onto your laptop.

> The laptop would be connected by cable to a transmitter box and antenna. These are supplied by us on a free rental basis with a £500 refundable deposit.

Anyone reminded of the scene in Snow Crash when YT calls in a Code, and everyone at Vitaliy's concert thinks it's a visual effect?

I went to their website to try to find pricing but they require contact info. Anyone have any idea on how much these thing cost when you're purchasing in bulk?

They used to have pricing information, unfortunately I don't remember what it was now. I do remember that at the time, their minimum order was around 1000 wristbands.

Chris Martin is quoted here[1] as saying the bands are costing GBP 400,000 (USD $640,000) a night. $20 a band!

Looking at the teardown[2] there's a radio receiver chip, microprocessor, couple of discretes/inductors/popcorn, then the plastic case and band with hi-bright LEDs and a battery. I'm thinking most of the cost is in the LEDs and battery need to drive those LEDs for two hours.

[1] http://blog.chron.com/celebritybuzz/2012/06/coldplay-spends-...

[2] http://hackaday.com/2012/02/19/ask-hackaday-did-you-catch-th...

[3] Top that, Bono! =)

Reply on news.ycombinator.com