On Thu, 2013-06-20 at 14:43 -0400, Rob Owens wrote:
> Seems like this could be avoided if the existing repos simply seeded 
> every file that they host.  Then you could always get the files via
> BitTorrent even if no other regular users were seeding.  Unless I'm
> missing something, torrent download speed could not be any slower than
> direct download speed, with the exception of the extra time it takes
> to contact the tracker and locate peers.

You can't compare 20 users downloading from ftp, http/s with 20 users
seeding and leeching a torrent at the same time. Theoretically the
torrent should be faster, but in practise this seldom or never happens.
I suspect that the handling of seeders, leechers and trackers, IOW the
sync, will slow it down. I might be mistaken. However, imagine you
should leech a part of the download from some people seeding, with a
terrible low bandwidth. Some seeders could get on and off line and this
must be handled too etc. pp.. Perhaps somebody with knowledge could
explain why I and at least another one from this list experienced
torrents as slower than a direct download. For sure it's less important
for those who download with a low bandwidth, but many of us for sure
have very fast DSL connections, even if those connections might not be
as fast as the ISP does claim.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1371756505.644.72.camel@archlinux

Reply via email to