Download throttling and reseeding?

[HeadPhoneGuy]

Hi there..im still new in this...i have similar problem with updated bit torrent.......yes i has read the FAQ but it still doesnt work...please...i want to reseed though......and become a good member..

[Freedom Kira]

How to re-seed.

Throttling = setting a download speed limit.

[HeadPhoneGuy]

But im using bit torrent 7.7.....and i even repeat the procedure, still it dosnt work...

[megido-rev.M]

Where are you stuck at?

[HeadPhoneGuy]

by referring to the procedure based on the FAQ for bit torrent.........act ually there is no problem about understanding the procedure, it just dont want to scan and reseed it....

[megido-rev.M]

You gotta do the recheck: it's the only way!

[Freedom Kira]

Make sure you have pointed at the folder that the torrent was downloaded to. Most torrents have all of the files stored inside a folder. You must direct BitTorrent to the folder that contains the torrent's folder, not that folder itself.

[Al_Sleeper]

That's not true for the earlier versions of uTorrent.

[megido-rev.M]

AFAIK uTorrent demands the download root: either the folder displayed upon opening the torrent, or the area of the single file.
No doubt it's confusing.

[Bozobub]

1) Most clients I know of, including uTorrent, require the top level folder that the torrent resides in (for example, your "Torrents" folder) when reseeding and/or specifying a new location for a torrent.  If you give it a path with "Torrentname", you'll end up with a new copy in "Torrents>Torrentname>Torrentname".

2) uTorrent tends to suck badly past 2.2.1, especially for seeding performance.  The newer versions of the official BitTorrent client are their own giant ball of fail.  "Sponsored" torrents, my ass.

3) Remember that having torrents available for upload, in no way guarantees there are peers who want to download from you.  Most torrents on BakaBT are quite heavily seeded, and you're competing with all of the other seeds for a limited pool of leechers, if there are any at all.

That lousy seeding performance you're seeing under uTorrent 3.2.1 is a combination of uTP (automatic bandwidth management) and bad coding in later versions of uTorrent.

[Bob2004]

As someone who ise very happily using utorrent 3 with uTP enabled, I can assure you that that is not the reason for his poor seeding speeds. The modern versions of utorrent work just fine. They may or may not be more or less efficient than older versions - I have yet to see any kind of reliable study demonstrating any kind of correlation - but at worst, they are not poor enough to cause any problems.

[Freedom Kira]

You probably have a good router then. Not all routers can handle it. When used in the wrong (right) places, uTP will cause slowdown.

[Bob2004]

You probably have a good router then. Not all routers can handle it. When used in the wrong (right) places, uTP will cause slowdown.

Hm, my current router (can't remember what model it is because the admin page won't load, but it's the default router Virgin Media give customers of their 60Mb internet package) is infamous for hating P2P. There are loads of stories of it choking, throttling speeds for no good reason, triggering the IP flood detection on its firewall and blocking it completely for a while, etc. Then again, it is designed to be used with internet connections of 100Mb/s, so it does have plenty of throughput.

My previous router was an old Netgear 832G (or something like that). Had it for a few years (and got it second hand to start with), and it tended to crash every day or so for no good reason. Before that, I had an older Virgin router which seemed to be incapable of doing any kind of useful load balancing (there were 8 of us using it).

Anyway, my point is that I've never had any particularly good routers. Maybe I've just been lucky, but I believe that for the majority of people, uTP will work exactly as intended. If it didn't, Bittorrent wouldn't have released it.

[Freedom Kira]

Hmm, most routers are 10/100 or 10/100/1000 these days, so having a 10/100 router (assuming yours is not gigabit) and getting good uTP performance suggests that there might be something else at play here.

It really is hard to say. You might say you want to see some hard evidence about the correlation between uTP and seeding performance, but such research is not likely to be done by professionals, and anyone else would not likely do it simply because of the sheer costs such an experiment would incur. So, the best we can do for now, and probably for a while, is what we get from people's reports on their own experiences. I intend to suggest disabling uTP to people and see what their reactions are. So far, the results are 1-1 even.

[MyonMyon]

To be honest, I don't really think it is important if uTP is enabled or disabled. The only thing that really matters is what gives you the best performance. Just test it and see which one works better for you.

For me switching to old uTorrent (without uTP) happened to boost the upload speed in a major way. For someone else it might not. But having uTP enabled or disabled has no absolute value alone if it does not affect your upload/download positively.

[Bob2004]

By reliable study, I don't mean a full blown lab experiment or something, I'd be happy with anything where anyone has even attempted to apply some kind of proper scientific method to testing which is more efficient. But the only thing I've ever seen to suggest that uTP has any negative effect at all are a bunch of random people on the internet's opinions, most of whom belong to that group of people who rabidly oppose any change that's made to the client at all, regardless of whether it's actually good or bad - and thus tend to just reinforce my opinion that there's really no difference.

I agree with MyonMyon that it doesn't really have a huge effect either way though. It's clear that there isn't a massive bonus from having it turned on, but there certainly don't seem to be any real negatives to it either.

[Freedom Kira]

By reliable study, I don't mean a full blown lab experiment or something, I'd be happy with anything where anyone has even attempted to apply some kind of proper scientific method to testing which is more efficient. But the only thing I've ever seen to suggest that uTP has any negative effect at all are a bunch of random people on the internet's opinions, most of whom belong to that group of people who rabidly oppose any change that's made to the client at all, regardless of whether it's actually good or bad - and thus tend to just reinforce my opinion that there's really no difference.

In that case, I hope you'll find my efforts acceptable. I'm not looking for any kind of bias - I just suggest turning off uTP and ask if they noticed anything about their seeding performance.

I agree with MyonMyon that it doesn't really have a huge effect either way though. It's clear that there isn't a massive bonus from having it turned on, but there certainly don't seem to be any real negatives to it either.

Perhaps you misread... The statement clearly says that switching to the older client, which is similar (similar enough, IMO) to turning off uTP, resulted in a major boost in performance. The score is 2-1 now, for uTP disablers.

[Bob2004]

By reliable study, I don't mean a full blown lab experiment or something, I'd be happy with anything where anyone has even attempted to apply some kind of proper scientific method to testing which is more efficient. But the only thing I've ever seen to suggest that uTP has any negative effect at all are a bunch of random people on the internet's opinions, most of whom belong to that group of people who rabidly oppose any change that's made to the client at all, regardless of whether it's actually good or bad - and thus tend to just reinforce my opinion that there's really no difference.

In that case, I hope you'll find my efforts acceptable. I'm not looking for any kind of bias - I just suggest turning off uTP and ask if they noticed anything about their seeding performance.

I agree with MyonMyon that it doesn't really have a huge effect either way though. It's clear that there isn't a massive bonus from having it turned on, but there certainly don't seem to be any real negatives to it either.

Perhaps you misread... The statement clearly says that switching to the older client, which is similar (similar enough, IMO) to turning off uTP, resulted in a major boost in performance. The score is 2-1 now, for uTP disablers.

He said that switching to an older client increased speeds. Yes, that means losing uTP. It also means going back to a very different version of the bittorrent protocol - they made loads of protocol changes and the like in between, say, version 2.1 and the current release. uTP was just one of them. And, occasionally, it's the case that the older versions work better on certain connections - I'm not sure why, and generally that shouldn't be the case, but it does happen.

It might be due to losing uTP, but since there are so many other things that could cause it by changing client versions, it's likely that it could have been caused by one of the other protocol changes, too.

EDIT: Not sure that protocol was the right word there actually. I'm talking more about the way the client operates than the Bittorrent standard itself.

[megido-rev.M]

I think it's the right word, just the one internal to the client as opposed to the BT one.

[Freedom Kira]

No, he's right. Protocol refers to a standard, usually a communications one when used in the context of electronics and computers. The BitTorrent protocol hasn't changed.

He said that switching to an older client increased speeds. Yes, that means losing uTP. It also means going back to a very different version of the bittorrent protocol - they made loads of protocol changes and the like in between, say, version 2.1 and the current release. uTP was just one of them. And, occasionally, it's the case that the older versions work better on certain connections - I'm not sure why, and generally that shouldn't be the case, but it does happen.

It might be due to losing uTP, but since there are so many other things that could cause it by changing client versions, it's likely that it could have been caused by one of the other protocol changes, too.

Fair enough. The score is still 1-1 even, then. To clarify, though, you do not see any difference in performance between having uTP enabled or disabled?