By: Reader's Write

Reader's Write — Fri, 16 Jul 2004 16:12:11 +0000

Well, the trend goes to fat cable/dsl/E1/T1 connections.
While even in 2000 still most users had 56k, now the fat connections are cheaper and provide MUCH more bandwidth, so ppl can download a full album, not a single song.
The number of files doesn’t matter – at least every 5th file on FastTrack is a fake and every 5th download is corrupted in some way by BayTSP for example. The bandwidth is important – it is the amount of data transferred and it can be split into just mp3s or just apps or just isos etc., but the main point is: more data has been transferred on BitTorrent than on dying KaZaA (should be already at least 1 year before dead – it would be better for all of us.)
Even the number of files on KaZaA is technically irrelevant because you see a very small chunk of the network. If you start klite and land on the klite cluster (which is a second FastTrack network not bigger than 90,000 users), you will get similar results as with the whole network connected.
BT is built in another way and is also not built for tens of thousands of users on a single torrent, but technically (and theoretically), it would also work with 100,000 of users on a single torrent – the only question: will the tracker hold it out?

By: Reader's Write

Reader's Write — Fri, 16 Jul 2004 04:46:49 +0000

It’s worth mentioning that CacheLogic measures bandwidth consumed by the various P2P networks, not the number of files downloaded. The Bittorrent “network”, being the best for transferring large files like movies to multiple users, naturally sucks up the most bandwidth since it has the largest average file size by far. But when it comes to total number of file (mainly MP3) downloads, networks like Fasttrack still get the most traffic an a per-file basis. Edonkey occupies middle ground, having both large movie files and much smaller software titles/cracks.

Comments on: BitTorrent usurps Kazaa

By: Reader's Write

By: Reader's Write