Common Bittorrent Mistakes, Part 1

There’s a number of common mistakes made by people, when they decide to torrent things Some of these are long-standing mistakes, and others are more common. The one thing they have in common, is that they’re all easily avoidable.

We’ve covered some typical mistakes and some handy tips in the past, but it’s always good to have a refresher. To many of the more advanced users, much of this will seem old hat, but as the saying goes, “we were all n00bs once”. In fact, some of these mistakes are not only made by those who have recently come to bittorrent, but by those who consider themselves ‘experts’ as well.

Part 1 here concentrates on mistakes by people who just don’t understand bittorrent.

Running more torrents than is advisable / trying to micromanage torrent activity

People are, by nature, impatient. When we’re trying to get something, we want it now, or better yet, 20 minutes ago. That torrent not going fast enough? Start another one! Got lots of torrents to seed? Start them all, so you can seed a little bit to all of them.

The problem with this is, your connection is only a finite size. The more you split it, the less each gets. Worse, each torrent uses data to maintain itself, called ‘overhead’. With a lot of torrents going at once (over 150 in some cases) the overhead consumes most of the bandwidth, leaving very little for actual transfer.

µTorrent, for instance, will start a queued torrent if there’s no activity on one already running, and will do that until it reaches it’s limit for active torrents, at which point it’ll queue the others. In other words, your client had manage eleventy billion queued torrents to try and keep the optimum number running at once. And unlike you, it’s keeping an eye on how much bandwidth is being used… and it knows the bittorrent protocol better too.

High numbers of connections

Again, it’s impatience. People see ‘there’s 20 seeds, and 80 leechers. I want to connect to them all! Because that’s going to be faster than connecting to half of them, right? Wrong. As above, more peers, means more overhead. Significantly more, as much of the overhead is telling peers what pieces you have, and being told what pieces they have.

In both cases, these efforts to speed things up, actually slow things down. Worse, it can set up a negative feedback loop, where a slow speed causes people to do one – or both – of these things. As that can slow things down more, they start more torrents or try for more connections, and make the problem worse, thinking they’re making it better.

Settings based on download speed, not upload speed.

This has the same effect as those above. It’s it’s more a mistake of misunderstanding, rather than intentional though, as most internet connections are advertised and sold based on their download speed, not their upload speed. It’s fixed by a simple speed-test and then changing settings accordingly.

Making downloads “faster”

This is always amusing to see. Someone wanting a way to ‘make’ their download speed up, because it’s too slow for them. It’s not like a file-locker, where there’s a set server, with a fast connection, and it’s deciding to limit you. It’s a collection of other peers, just like you, running the same kind of software as you’re using. Your download is their upload, and vice-versa, your upload is their download.

The only way to MAKE your download faster, would be to force these peers to upload to you. There’s only one problem with that. If you can do it to them, they can do it back to you. That’s why they don’t exist.

There are only two ways to speed up your torrents.

  1. Use the optimum settings for your connection
  2. Buy a faster connection.
Even funnier is when someone wants their UPLOAD to be faster. Doesn’t happen much (mainly on overseeded ‘private’ torrents – but that’s for part 2) but it does happen.

Next time

In part 2 we’re going to cover mistakes made by ‘experts’, which require a deeper understanding of the bittorrent protocol