Jump to content

Talk:Micro Transport Protocol

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
This is an old revision of this page, as edited by Meno25 (talk | contribs) at 16:14, 8 April 2015 (Assessment: Internet: class=Start; Software: class=Start; Telecommunications: class=Start; Computing: class=Start (assisted)). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

23 December 2008 AfD

How is this no consensus? It is easily a very clear keep 76.66.198.171 (talk) 07:23, 29 December 2008 (UTC)[reply]

6 January 2009 Snaaio

Page should be renamed and moved, uTP stands for "micro Torrent Protocol" (in analogy to μTorrent being written as uTorrent). Source: [1], under "What is in 1.9:" --Snaaio (talk) 03:08, 6 February 2009 (UTC)[reply]

Inacurracies

A lot of incorrect information in this page.

  • µTP is designed to reduce congestion, not to increase speed.
  • µTP is not "used by BitTorrent implementations", it is used by one BitTorrent implementation (granted, the most popular one).
  • µTP is not hidden, it is used by default by µT 2.0 and 2.1.

Somebody really should fix all of that. Jec (talk) 06:15, 6 November 2009 (UTC)[reply]

  • Okay, I did it. --Jec (talk) 00:33, 15 November 2009 (UTC)[reply]
  • To me it really does increase speed. Sitting behind two NATs (outer one, by ISP, has no port forwarding) i see uTorrent getting an orders more peers (thus speed increased) with uTP on. In theory UDP alone should do it fine, but in reality only together with uTP it works. In Vuze however uTP is not capable of pinning the holes. BTW, it's ironic, but heavy uTP traffic may bring home router (TrendNet BRP-632 gen 1) to its knees, while same speed of TCP traffic does not, that talking of congestion. 79.111.223.5 (talk) 22:08, 16 April 2011 (UTC)[reply]

Neutrality? Sources?

Is this still Wikipedia or what? This article (in its present state) does NOT belong here.

Oh my, care to clarify? Jec (talk) 21:31, 15 April 2010 (UTC)[reply]

1. µTP was developed within BitTorrent, Inc., with no input from the wider networking community.
Where's the actual source of this info? Author provided only a link to a forum thread where nothing like this is mentioned. [1].

The author (me) provided a link to a forum thread where the uTorrent developers acknowledge that there is no uTP specificatin, at a date when uTP was already being fielded. What more information do you need? Jec (talk) 21:31, 15 April 2010 (UTC)[reply]

2. The framing scheme has functionality roughly equivalent to that of TCP (with timestamps and SACK), but realized in a somewhat idiosyncratic manner.
This is biased against µTP (the author needs to explain what "idiosyncratic" is supposed to mean in the context of µTP) AND no source is given.

Fair enough, I'll clarify. I do object to your removing the term altogether, though.Jec (talk) 21:31, 15 April 2010 (UTC)[reply]

3. Ledbat has been described in an Internet-Draft, but the details of the µTP implementation are different from those of the draft.
In the forum thread [2] that is supposed to be the source, one of the µTorrent developers clearly states that there's precisely one difference between LEDBAT and µTP and this difference is a minor one.

Ehm... Reacting to loss less aggressively than TCP is a minor issue? You're kidding, I hope. Jec (talk) 21:31, 15 April 2010 (UTC)[reply]
Given source is no longer publicly accessible: "This group is on a private domain". Removing it from the article. bungalo (talk) 16:51, 28 December 2014 (UTC)[reply]

4.Many ISPs had problems on BRAS and Border routers when customers started to use µTP
The "source" of this information is a link to a forum thread [3] on a Russian community forum where the implied "problems" are discussed by anonymous posters. There's no way to verify their connection to any actual ISP. Furthermore, there are NO offial reports of any major ISP in the world having problems with µTP.

Please, provide some RELIABLE sources or remove the quoted sections from the article. —Preceding unsigned comment added by 95.27.28.247 (talk) 09:01, 26 February 2010 (UTC)[reply]

  • No reply? Alright, I'm removing the unverifiable info now.
That's not the way things are done. Jec (talk) 21:31, 15 April 2010 (UTC)[reply]
There are a couple of research papers[citation needed] that mention this protocol. If someone has time, please search at http://scholar.google.com and write a review. Mange01 (talk) 17:38, 2 May 2010 (UTC)[reply]