Experimental broadband is 1000's of times faster!

Check this out over at the New Scientist web site.

http://www.newscientist.com/article.ns?id=dn8177

Woo - very impressive.

Maybe those of us that live 5 miles outside the supposedly "most connected city in europe" (yeah right!) might finally get a decent connection. I won't hold my breath though :-)



Bitter? Me? Damn right!

I think you may've misunderstood the article, it isn't suggesting that it will offer broadband that is thousands of times faster, but that it offers a single link of that speed. Once you divide that link by 1000 or more subscribers, you will get something that isn't very fast.

That is unless we all have a balloon each. :D

You should know prof! Come on! :P

The fastest broadband will be when they set up a photon network :P

You mean fiber optics?

Ok. Its not bi-directional, but its a step in the right direction especially if they get a bi-directionsl system working.

No, I dont mean fiber optics.... Theyre faster, but theyre not quite Photon.

ok, so by Photon, you don't mean light?

hehe you've torn it now, Cyg! Here we go, eyes down for the full house... :P

Not in the same way as Fibre-optic cable :)

Just a quick link:

http://www.theinquirer.net/?article=3977

[edit]

Gonna find a better link.

Basically they can transfer info usnig two photons that are not connected to each other. That will be a fast way to transfer ;)

Awww!

Lol, not sure I'd want Scotty to Beam me up, thats for sure.

My question is, if the data was sent using a radio frequency- and the laser was destroyed- well, Radio frequencies can go through stuff- solid stuff- so Im not sure what theyve actually discovered anymore :P

There was a better article, but i cant find it- that described it all in much better detail!

Still all very experimental, and a bit misleading as well. With most residential broadband now in the 1.5-3.0 mbit range, it is at best 'hundreds' of times faster, not thousands... Plus a major catch is that the bandwidth of such a floating platform is shared among -all- of the people uplinking to it, making the realistic throughput rates significantly lower, especially if such a system becomes popular. While there are bottlenecks with wired services as well, they appear much later in the chain and can much easier be worked around than with a wireless service where you are limited by the size of the wireless spectrum itself.

Also, something else to put this 'breakthrough' a little bit more in perspective:

http://www.convergedigest.com/Bandwidth/newnetworksarticle.asp?ID=14545
"Hong Kong Broadband Network (HKBN) officially launched its 1 Gbps symmetric service for the residential market. Approximately 800,000 households, out of a total of 2.2 million households in Hong Kong, are wired to receive the service."

(Typical broadband speeds and availability in South-East Asia and western Europe are currently greater than in North America -- the US now ranks #16 worldwide when it compes to the percentage of its population that has broadband internet, and typical available speeds are lower)

Cooooooooorrrrrr, imagine that. You could download an entire DVD in 38 seconds. Now all we need is a solid state hard drive to save that fast :)

Unfortunately current solid state technology has its own range of problems -- in addition to being very expensive and not available in sizes more than a couple of gigabytes, most of the solid state drives are based around Flash RAM technology.
The major downside of that is that they can only be written a limited number of times. From what The older generations could do ~10,000 writes, the newer apparently ~100,000 using a bunch of additional tricks like purposefully spreading the usage across the entire bank of memory instead of writing the same places over and over... But they still wouldn't last very long if you put very dynamic files like your windows swap file on such a drive...

We need another breakthrough in the solid state drive area...

I read something about a new harddisk buffer thats being developed by Intel and some other company.
Basicly just a buffer memory of 128mb or so which stores the latest data, or the most often used stuff, decreasing harddisk access by a major amount.
This made Flash drives more viable as they wouldn't be written to often at all.

I'll see if I can dig up the article.

If the buffer is designed to hold on to stuff for longer times, this brings up another potential problem: A crash can undo a lot of work, if the actually data never made it to the harddrive before your computer takes a dive...

Even more so if you don't have a properly journaling filesystem like EXT3 (which is a lot more resilient than FAT or NTFS which windows uses)