![]() ![]() by caching the wait time and speed for each IP address). But I appreciate that this might need enhancements to the BitTorrent protocol and that this might therefore need to be a long-term aspiration - in which case I would like LibTorrent to do the best it can to optimise connections which actually download over those which wait a long time (e.g. Ideally I would like the BitTorrent protocol to allow seed to estimate the time before it starts downloading and to estimate the maximum upload speed, so that LibTorrent can decide whether to hold on to that connection or try to find another connection which will start faster and / or download faster (using the above algorithm for the latter - just without waiting for the download to actually start and stabilise). ![]() 100Kbs the bottom one is < 5% of this (i.e. So if I have 3 connections at > 100Kbps and two which are below 1Kbps then the average speed of the top 2 would be c. For a torrent with many more seeds than allowed connections (setting to define the multiple), drop any connection which downloads at a speed much slower than the average of the fastest 50% of all other connections (setting to define the %). ![]() I also get a lot of connections to seeds which wait some time before downloading. When downloading a torrent with a large number of seeds, I get some connections which download at several 100Kbps, and some that download at a few bytes per second - and I would like LibTorrent to drop these in the hope that it can find a significantly faster seed. Libtorrent version (or branch): 1.2.6.0 (in latest version of qBitTorrent) ![]()
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |