Avian’s Blog

I stumbled upon a handful of interesting games in the past few months and I thought I might share the links.

I'm not that much into computer games. The last big-title game I played was probably Quake 2, and even that was long after it got out of fashion. I have the impression that recent games simply require too much effort to play and I have more interesting things to do (I won't touch anything that involves on-line play with the proverbial ten-foot-pole). The only complicated game that I have been playing occasionally for years is Nethack.

It's nice to escape from the white-on-black console sometimes though, put the brain in idle and enjoy something more colorful.

So, first one is Alex the Allegator 4 by Free Lunch Design. It's a cute platform arcade game with an environmental theme. It has grayscale graphics and beeping noises in the best style of old Gameboy systems.

Alex4 is free as in freedom (GPL2) and is in official Debian repositories (apt-get install alex4)

This is Gravity Bone, a first-person adventure with a really twisted James Bond theme and a surprising ending.

There's source available, but it doesn't compile on Linux. Windows binary runs perfectly on Wine though.

In the end there's Morningstar. This is an adventure with the classic theme of being a spaceship pilot stranded on a desert planet. It's got voice acting and a video intro - not something I expected for a game you get to play for free. Needs a bit of thinking for solving some puzzles, but breaks the tradition of Space Quests by not letting you do anything that would be hazardous to the health of your character.

It's a Flash game you play in the browser, so it's only free as in beer. I used the Adobe plugin for Firefox and I doubt that any free Flash interpreter is capable of running it.

I just got back from Institute Jožef Stefan, where I attended TEDx Ljubljana, the first Slovenian independently organized TED event. It was organized under license from the TED conference by three volunteers in the spirit of their motto, ideas worth spreading. The attendance was free, which is quite a contrast to the four-figure annual membership fee that is required at the official talks in California. There was a mandatory registration though, and all tickets were given away in just three days.

The topic of today's event was education and standard of living. Just like the schedule it wasn't published before the start of the event, which is an interesting trick. The idea of TED is that the best way of getting new ideas is to get off your daily routine and listen to people outside your expertise. Indeed, I most likely wouldn't attend if I knew the topic beforehand.

Nevertheless I enjoyed being there. There were four half-hour live talks by Slovenian speakers and two recorded and subtitled English talks from California.

First on the schedule were Boris Pavlih and Andrej Bauer, former a high school and latter an university professor. They both talked about the problems they see with students today, how the public perception seems to be that the parents only need to provide for the material good of their children while everything else about their education is the domain of the school. How school seems to be only the necessary evil that gets you a good pay check and should be gone through with as little effort as possible.

Especially dr. Bauer gave a engaging performance that was truly in the spirit of talks like you see recorded on TED.

Following were Sebastjan Mislej, who gave a tour of the videolectures.net effort and how they're bringing knowledge to the far corners of the planet, and mag. Nada Rotovnik, who talked about the common misconceptions about healthy living (e.g. food and exercise) and how there's no such thing as a single, ideal one-fits-all diet (even pure sugar can be good for your in some situations).

The event was concluded by play-back of two TED talks, that are also available on-line and which I really recommend watching: Everything and How schools kill creativity.

As I said, a wonderful evening and looking forward to the next TEDx.

I had this handheld PMR transceiver (i.e. "walkie-talkie") in my hands for a short while a couple of weeks ago. I took it apart out of curiosity and to see if there were any chances of making a general-purpose receiver for the 400 MHz band out of it.

Except for Mega P on the front side and PMR X3 in the battery compartment it doesn't bear any other identifying marks. I didn't have a manual, but a search turned out an user's guide for Binatone Adventurer 100 which appears to be a very similar device.

Inside there is a single multi-layer circuit board attached to a small piezo speaker and powered by two AA batteries.

The front side has the LCD display, microphone and pads for 5 keys. There's an integrated circuit bonded directly to the PCB under the display, which I guess is the microcontroller that drives the display, keys and provides the user interface to the electronics on the other side. There's a tuning fork style quartz oscillator nearby, which generates the clock signal for it.

The back side has all the RF circuitry soldered under shielding covers. There's a AN6311FA integrated circuit in a 52-pin QFP hidden under the copper foil on the right. I couldn't find any more information on it, short of some other mentions of it in connection to PMR equipment. It's described as "IC ASIC" on a parts list for another transceiver, so I guess it's an application specific circuit, maybe taking care of modulation and squelch.

PMR works on 8 radio channels on the 446 MHz band (hence the 446 MHz SAW filter in the middle), which is legal to use in Europe without a license. The output power in the Binatone manual is specified as 0.5 W, which is the maximum ERP allowed. It uses narrow-band FM modulation (channel spacing is 12.5 kHz).

This particular transceiver also uses CTCSS, which gives you 38 subchannels for each of eight frequencies - you can only hear somebody if both of you have the same subchannel set. Interestingly this system doesn't provide any security: it's only the well-behaved receiver that filters out any transmissions that don't have a sub-tone of the correct frequency added to it.

The 21.250 MHz quartz on the top is most likely the local oscillator from which all other frequencies are derived via frequency dividers and PLLs. Divided by 1700 it's equal to the channel width. There's a trimmer capacitor on the other side for fine-tuning its frequency.

On Saturday I went to see Angels & demons. It was entertaining to watch, although it was apparent at times that whoever wrote the script was more at home in medieval myths than physics. I guess I got used to weird things characters tend to say in Hollywood movies whenever the topic of conversation shifts to science.

One thing that did made me curious is a scene towards the end when they had to get rid of an anti-matter explosive device in the middle of Rome with 5 minutes to spare. So a guy jumps into a helicopter and flies straight up. When the time is up, the helicopter is high enough that the explosion only causes a spectacular fireworks display and shatters some frescoes.

At the first thought, that seemed like a really bad idea to me. They say that the explosion had an equivalent yield of 5 kT of TNT, which is comparable to a small nuclear bomb. This kind of devices do the greatest damage when they're exploded at some height above the target, so is it really feasible to save a city in the way shown in the movie?

First question is how high can you get in an average helicopter in 5 minutes? After sampling some random technical specs on Wikipedia it appears that the rate of climb of a helicopter varies from 6 m/s in civilian to 13 m/s in military craft. In the movie they use a police or a medical helicopter. It's also empty, so take an average, say 8 m/s, to get a nice rounded figure of 500 m/min.

Second question: How dangerous is a 5 kT explosion at 2500 m if you're standing right below it (like the crowds does in the movie)?

Unfortunately the Strangelove Slide Rule doesn't provide this kind of calculation. I did some more searching and found a couple of old DOS programs from the US Defense Nuclear Agency that answer exactly this kind of questions (they're complete with warnings to reset the computer after entering classified data - I guess they haven't yet come to shredding hard disks at that stage).

The result shows that 5 kT explosion will cause a shock-wave with 10 kPa of maximum pressure on the ground zero directly below. That's a little under 1.5 PSI, which is, according to the previously mentioned slide rule, not so bad for any by-standers. Such a shock will shatter windows, but won't knock down houses or even damage most people's hearing.

Surprisingly, that's exactly what you see in the movie. So this may be one case where they actually got it right.

Two engineers at MIT are apparently working on a sophisticated way to make conversations in a large room (like a large office) private. In short, you put a large number of microphones, speakers and PIR sensors around the room and connect them to a central controller. When someone in the room wants some privacy, it sends a command and some algorithm will put just enough white noise through the nearby speakers to make any conversation unintelligible. As I understand this is not some sound canceling technology. It seems that they merely attempt to lower the signal to noise ratio by adding more noise.

I can't help to wonder that if you have resources to implement such a complicated scheme you are probably better off simply putting moderately sound-proof walls around people. I doubt that would be much more expensive than the solution mentioned in the article and it has the added bonus that your coworkers don't need to put up with your noise-enhanced chatter. Plus you also get a shield against unwanted stares.

I never quite understood what's so attractive about large, open-plan offices that every company seems to use these days (well, except for the cost of the setup). At Zemanta's office I find that the amount of work I get done is more or less reversely proportional to the number of people in the office. It's mostly bearable when I'm doing some trivial Python coding that only requires half a brain. But when there's a real problem I need to concentrate on, even the sound of the nearby keyboard giving off agonizing noises under a torrent of keypresses is distracting. Not to mention two simultaneous conversations along the office's diagonals, trying to out-shout each other.

It appears that the sound channel in the office has a similar problem as e-mail: The barrier to sending a message to many people is just way too low (even easier than adding half of employees to the CC list, after all, you just have to open your mouth). So a lot of messages are sent without the sender thinking twice about what he's going to say and merely interrupt work of everyone in the room for no good reason. If the effort needed is higher (say, stand up and go to the next room), I'm sure people think better than to break your line of thought with the latest Apple rumor.