Life, Photography, and Security

←September→
Mon	Tue	Wed	Thu	Fri	Sat	Sun
		1	2	3	4	5
6	7	8	9	10	11	12
13	14	15	16	17	18	19
20	21	22	23	24	25	26
27	28	29	30

2006-04-03

camillo.särs.net → www.ged.fi

Now that .fi domains are also available to private individuals, I registered “ged.fi”. This means that my site will primarily use “www.ged.fi” in the future. Or at least until I change my mind.

[/computing/internet] permanent link

2005-12-01

I have followed the Finnish discussion on national broadband strategy pretty closely. In short, it’s a mess. The discussion seems to focus a lot on how consumers will buy DSL or cable broadband Internet connections, and how to provide services for those consumers. That’s of course nice, but I don’t think it has a lot to do with strategy. A strategy is a long term plan of action designed to achieve a particular goal, as differentiated from tactics or immediate actions with resources at hand.

But, you may say, the Finnish national broadband strategy has 50 action points. Isn’t that a “long term plan of action”. Well yes, it is. The problem is not with the actions, but with the goal. What is the vision of the Finnish strategy? There is a hint in the document, though. The Government’s broadband strategy is ‘technology neutral’, in that it does not favour one particular technology over another, but instead promotes competition among them and their complementary use.

The strategy fails to look into the future. It is fairly easy to make some estimates on the growth of capacity demand. The strategy would then describe what the biggest challenges to meeting this demand are, and how to tackle those challenges. Let’s try it.

The most bandwidth intensive consumer application currently is video. Current “broadband” connections are already struggling with the demands of video-on-demand. How will this develop during the next few years?

HD-DVD, regardless of technology, will become reality in 2006. This pushes sales of HD television sets. Current DVB broadcasts are optimized at a quality level slightly above VHS, at 2-4 Mb/s. This level of quality is not acceptable on HD sets, which require a bandwidth of up to 8 Mb/s for decent MPEG2 quality. DVB-T networks are resource constrained, so they squeeze 5-6 channels into a 21 Mb/s multiplex. They cannot meet the quality demand, so quality content for HD sets has to come through other channels.

MPEG2 compressed HD can take up to 18 Mb/s. For the sake of this argument, let’s say an average family has four members. The exact number doesn’t really matter, I am aiming for a rough estimate. It is reasonable to assume that at some time during an evening, at least three family members are watching TV. It is also fairly safe to assume that at some point in time, two recordings may be running simultaneously. This gives us a peak estimate of five concurrent video streams. For MPEG2 HDTV, that means 90 Mb/s of video.

The conclusion is evident. In a few years, a typical family can be expected to have a peak broadband demand of close to 100 Mb/s. This number is two orders of magnitude bigger than the 256 kb/s currently considered “broadband”.

The Finnish broadband strategy is technology neutral. That is why it cannot succeed. The real challenge in broadband today is how any country can help the actors on the market embrace new technologies which enable true high-speed connections for the “last mile”.

Fiber optics have a capacity of 10 Gb/s, which easily can be extended using WDM technology into the range of >1 Tb/s. In anything but the short term, the only reasonable technology available is fiber optics. Installing fiber optics is very costly, so there is a very real need for a national strategy for deployment. The national broadband strategy should be to support rapid deployment of a national-wide fiber optics network.

[/computing/internet] permanent link

2005-06-14

BF2 Tactics

The demo has been out for a few days, and already it is very clear that gameplay is radically different. I have managed to get into the top five twice on a 64-player server with only a handful of frags. The key to this is teamwork. BF2 awards cooperation, and the awards are significant.

If teamwork is important, what is the “perfect squad” setup? Any fairly balanced squad will probably do well, but I can already point out a few things that may be important to a killer squad.

Medic: Any squad needs a medic, and the medic should not be the squad leader. If the squad leader is killed, the medic needs to revive him quickly to keep the squad together. OTOH, if the medic is killed, he will respawn with the squad anyway.
Support: To be able to operate for extended periods, any squad needs support. I think that the squad leader is a good candidate—he can set up a supply spot wherever he chooses, and has the option to duck and cover if things get too hot.