I use www.motorworld.net (a private tracker) to get videos I simply can’t buy for love or money in the USA, including Top Gear. If I load a torrent from Mininova and a competing one from Motorworld, the Motorworld one can beat the Mininova one by about a factor of eight, in spite of of having over ten times as many people in the swarm. Is there something inherent to the nature of BT that makes this so, or something else?
Users on private trackers usually have upload/download ratios to maintain, so they hang on after the file is finished downloading and continue uploading. Users of a public tracker are apt to leech and run. So a public torrent might have more users overall, but is likely to have fewer seeders with a complete file.
Indeed. I tend to look at the number of seeders rather than the absolute number of people in a swarm.
It’s a ratio thing. You want to have a high seeder to leecher ratio, something like 2:1 or better, to get good download speeds, something private trackers can supply easily. But Mininova et al normally has a fairly average 1:2 or 1:4 kind of ratio, or worse even than that, because of the high hit and run leechers.
I’ve noticed that if the sheer number of users is large enough (as in, several hundred, at least), then the seed-to-peer ratio doesn’t matter as much and I can max out my connection, or close to it. I’d guess that’s because with so many peers, you still have dozens or hundreds of complete copies among them.