Fullscreen HTML5Med Res (??MB)HTML4

Holy Spheres, Batman! Techno Talk Max the Magnificent My mind is boggled with respect to emerging tools and technologies that are allowing us to create awesome artifacts and to explore the universe in ways that would have appeared impossible a few scant years ago. I ’m sure that, like me, you’ve been entranced by what you’ve seen and heard about the Sphere at The Venetian Resort in Las Vegas. Costing $2.3bn (which means I won’t be buying one of my own in the foreseeable future), this bodacious beauty is the largest spherical building in the world at 81,300m2 of exterior area. Inside the Sphere we find seating for 18,600 people. The multi-layered audio system boasts 160,000+ speakers and uses 3D audio beamforming technology that promises to present a targeted, crystal-clear, and uniform sound experience to every seat in the house. Some say this will be an audio experience to die for, which might be both figuratively and literally true if hackers ever manage to take control of the system. In addition to the awesome audio, 10,000 of the seats incorporate an infrasound haptic system that will allow their incumbents to experience 4D features, such as changing temperatures, breezes, scents, and so forth. The interior flaunts a 16K-resolution (19,000 × 13,500 pixels!) wraparound LED screen, measuring 15,000m2, thereby making it the largest and highest-resolution LED screen in the world. The venue is scheduled to open on 29 September, 2023, with U2 as its first performer (I wish I could see that). I have no doubt that the inside of the Sphere will be stunning (I’m already straining at the seams for superlatives), but it’s the outside of the beast that makes me want to squeal in excitement. The exterior surface will features 54,000m2 of programmable lighting in the form of 1.2 million hocky-puck-sized LED arrays, each containing 48 individual tricolor diodes, all of which can be used to present the most awesome displays. Just perform a Google search for ‘YouTube Sphere Las Vegas’ and prepare to have your mind well and truly boggled: https://youtu.be/wKCY1Ph7T0k How much power are we talking about here? Well, these are superbright LEDs, so let’s assume 10W per 8 puck, which gives us (1.2 × 106 pucks) × 10W = 12MWh over the course of an hour. According to the Office of Gas and Electricity Markets (Ofgem), a typical household in Britain uses approximately 3MWh of electricity annually. This equates to an average of (3 × 106) / (365 days/year) / (24 hours/day), which equals 342Wh per hour, give or take. Thus, whenever the outside of the Sphere is operating at its full potential, it’s consuming the same amount of power as around 35,000 houses in the UK! (I’m glad I don’t have to pay that electricity bill.) I can’t take the strain. All this talk of LEDs and spheres has made me want to build my own, like the one created by Jiri Praus: https://bit.ly/45oTWGA What time is it? For reasons we don’t need to discuss here, I’ve recently been learning a lot more about packet-based networks than I ever really wanted to know. To work their magic, all the elements forming the network must be synchronised in time. Establishing and measuring the level of simultaneity – the temporal ordering of events – requires those events to be ‘timestamped.’ In turn, it’s necessary for all the devices forming the network to be synchronised to a common ‘grandmaster clock.’ Of course, we have access to tremendously accurate clocks these days. For example, according to the boffins at the National Institute of Standards and Technology (NIST), there’s a strontium atomic clock so accurate it would not have gained or lost a second had it started running at the dawn of the universe. The problem is that clocks of this ilk are the opposite of cheap, and you certainly can’t afford to stick one in every switch and router forming a network. The solution for packet-based networks is to use something called the Precision Time Protocol (PTP), a.k.a. IEEE 1588, which can achieve accuracy in the sub-microsecond range. In this case, whichever subsystem is connected to the grandmaster clock – let’s call it the upstream unit – sends a packet to a downstream unit stamped with the time of departure derived from the grandmaster clock. The downstream unit stamps the packet with its time of arrival based on that unit’s internal clock, which may be out of sync with the upstream unit. The downstream unit then returns the packet to its originator, adding the new time of departure. In turn, the upstream unit appends the packet’s new time of arrival. Using these four timestamp values, the upstream unit can determine the error in the downstream unit’s clock and instruct it to adjust itself accordingly. Like many things, this seems simple if you say it quickly and then point out of the window and shout, ‘SQUIRREL!’ In reality, it’s horrendously complicated because each outgoing and incoming packet may take a different path (and hence have a different delay) through the network. Also, the downstream unit’s clock may drift over time. This means the two units are obliged to keep on passing packets back and forth, constantly refining their level of synchronisation. The problem is only exacerbated by the fact that the primary unit may be connected to multiple downstream units. In turn, each of these downstream units will assume the role of a source clock to any of its subordinates, and so it goes down the chain. My head hurts just thinking about all this. Great balls of fire There’s so much more I wanted to talk about, like the fact that the telescopes and technologies used by astronomers are now so incredibly accurate and precise that they’ve discovered a star 1,300 light years from us that rotates once every 15 minutes. The truly amazing thing about this ‘two-faced’ star is that one side is hydrogen while the other side is helium. How can this be? I have no idea. One last thing I really have to tell you is… OK, next month! Practical Electronics | October | 2023