Yes, I know. Scraping is returning the amounts of seeders and leechers for a certain info_hash. Not their IPs. Even if a torrent had 1000000 seeders and 1000000 leechers the bandwidth impact would be negligible for a single scrape request.
Now, Torrents.ru doesn't allow scraping because with over 1.5 million users, they save a lot of bandwidth, even if every request takes just ~1kB. It's not because scrape request themselves return the peers, increasing server load.
Wrong. Fire up a packet sniffer and see by yourself.It means getting whole F**king peers from the database, imagine if there are 10k peers and every 10k member scrape fully the whole database for 10k peers, do you think their server will be stable ? HELL NO !
I'd like to think you're confusing things due to your lack of knowledge rather than favoring TD here. It has already been explained before. You can control how many peers should be returned via the numwant parameter, but in the end it's up to the tracker. uTorrent asks for 200 peers on every announce, but T.ru only appears to return 50, which is also the default value from when no numwant is specified. Now, that isn't a percentage like you're saying - they'll give 50 peers regardless of it being the totality of the swarm or the torrent having a million seeders.Very less trackers optimized their trackers not to allow FULL SCRAPE, i saw this option at torrents.ru and 3-4 more Russian sites.
They just do not allow full scrape, some of them allow 10%, some allows 30% or 50 or 60%.
Bookmarks