With you: Problem selfish download torrent
Anderson paak anderson paak malibu album download | 750 |
How to use thaumcraft mod after downloading it | 65 |
Dccm album download | 161 |
Fortnite aimbot mod download | 722 |
|
Subaru 1-3-2-4 Premium Member join:2001-05-31 Greenwich, CT |
Subaru
Premium Member
2007-Jan-3 10:06 am
Re: hrm
said by tp0d i`ll stick with azureus, thanks.
[/BQUOTE :
same
[/BQUOTE :
same
Doctor Four My other vehicle is a TARDIS Premium Member join:2000-09-05 Dallas, TX |
This will do for Bit Torrent what QuickMX has done for
the WinMX network: ruin it for other users.
It deserves to be blacklisted.
the WinMX network: ruin it for other users.
It deserves to be blacklisted.
yabos join:2003-02-16 London, ON |
I don't think you understand what it does. It seems that it prefers to communicate with peers that are uploading faster to you which in turn makes the downloads faster. It will not send as much data to the slower peers so they are encouraged to upload faster.
I'm not sure how azereus chooses peers right now so I can't compare it to that. I tried the software for one torrent and I didn't notice that it was extremely fast or anything. I got 200+ KB/sec though which was pretty good for a torrent with a fairly small amount of seeders.
I'm not sure how azereus chooses peers right now so I can't compare it to that. I tried the software for one torrent and I didn't notice that it was extremely fast or anything. I got 200+ KB/sec though which was pretty good for a torrent with a fairly small amount of seeders.
peter_m
Premium Member
2007-Jan-3 4:28 pm
Re: hrm
It's interesting. But why are many people assuming that it's bad for the swarm?
Does it just give priority to faster peers or does it actually ignore slow peers completely and waste upload bandwidth waiting for faster peers?
Does it just give priority to faster peers or does it actually ignore slow peers completely and waste upload bandwidth waiting for faster peers?
Combat Chuck Too Many Cannibals Premium Member join:2001-11-29 Verona, PA |
Re: hrm
said by peter_m:
It's interesting. But why are many people assuming that it's bad for the swarm?
Because it's being reported that way and they don't know enough, or are too lazy to actually look into it.It's interesting. But why are many people assuming that it's bad for the swarm?
said by peter_m:
Does it just give priority to faster peers or does it actually ignore slow peers completely and waste upload bandwidth waiting for faster peers?
It gives priority to faster peers, as does every other Bittorrent client, and it doesn't in it's current implementation snub slow clients if there is extra capacity.Does it just give priority to faster peers or does it actually ignore slow peers completely and waste upload bandwidth waiting for faster peers?
MxxCon join:1999-11-19 Brooklyn, NY |
it is NOT a leecher client!
developers used bad wording for their name. that client DOES NOT HARM NETWORK!
go read their whitepaper!
developers used bad wording for their name. that client DOES NOT HARM NETWORK!
go read their whitepaper!
FFH5 Premium Member join:2002-03-03 Tavistock NJ |
FFH5
Premium Member
2007-Jan-3 9:45 am
Looks like it is designed to punish those who slow uploads
Looks like this version is designed to punish those users who configure their clients to do only downloads and either block or severely slow down uploads. I guess many people were worried about the legal danger of uploading copyrighted material and set themselves for downloading only - minimizing their legal risks. But, of course, this makes bittorrent less workable, so this new version is designed to shut out those cautious or nervous about legal risk.
Re: Looks like it is designed to punish those who slow uploads
One way to solve this is by designing a client that hides the users IP addresses. I never quite understood why my IP needs to be seen or is shown to other users.
vpoko Premium Member join:2003-07-03 Boston, MA |
vpoko
Premium Member
2007-Jan-3 10:18 am
Re: Looks like it is designed to punish those who slow uploads
said by compton:
One way to solve this is by designing a client that hides the users IP addresses. I never quite understood why my IP needs to be seen or is shown to other users.
Hiding it in the client doesn't really help - you can still see the destination IP address of any packet sent from your computer.One way to solve this is by designing a client that hides the users IP addresses. I never quite understood why my IP needs to be seen or is shown to other users.
yabos join:2003-02-16 London, ON |
Because then the internet wouldn't work at all. They don't have to have the IP address visible in the program but you can't hide an originating IP address with TCP data transfers. Just typing netstat in a command prompt you can find out the IP address you are connected to.
ElJay join:2004-03-17 Portland, ME |
How would that work in a peer-to-peer system like BitTorrent? Beyond the tracker system, all of the data is shared directly between the client computers and to make that happen you obviously need to know the IP address(es) to connect to.
It's also designed to punish those with smaller upload pipes. Not everyone has 1+ MBPS up.
Re: Looks like it is designed to punish those who slow uploads
said by Morac:
It's also designed to punish those with smaller upload pipes. Not everyone has 1+ MBPS up.
It's also designed to punish those with smaller upload pipes. Not everyone has 1+ MBPS up.
I agree, if had more upload bandwidth, I would share more of it. I am not about to give up 75% of my upload speed. More than me using internet in my household, plus I dont like high ping times for gaming.
yabos join:2003-02-16 London, ON |
My upload is about 85KB/s max and I usually set my upload to be 60KB. Browsing the web is unaffected. The remaining 25KB/s is enough for that.
MxxCon join:1999-11-19 Brooklyn, NY |
that's incorrect.
a torrent that does not have excessive upload capacity should not spend equal amount of time on low capacity clients.
high bandwidth clients will get data faster and will spread data faster.
a torrent that does not have excessive upload capacity should not spend equal amount of time on low capacity clients.
high bandwidth clients will get data faster and will spread data faster.
nekote
Member
2007-Jan-3 10:24 am
Maximizing bandwidth utilization?
Doesn't this have the effect of making maximum use of available bandwidth - of course driving utilization towards an undesirable 100% ?
OTOH, the sooner the data transfer is completed, the sooner other data transfer request can be serviced? Unfortunately (?), potentially additional requests from those who have already (just) benefited from the quicker download?
The rich get richer, so to speak?
OTOH, the sooner the data transfer is completed, the sooner other data transfer request can be serviced? Unfortunately (?), potentially additional requests from those who have already (just) benefited from the quicker download?
The rich get richer, so to speak?
Morac Cat god join:2001-08-30 Riverside, NJ |
Morac
Member
2007-Jan-3 10:27 am
How does this help the downloader?
The client rewards other clients that upload to you faster by unchoking them first. That doesn't necessarily mean that those clients will unchoke you first so you might not see any faster download speeds unless the other clients are also running BitTyrant and you are uploading faster than anyone else in the swarm.
If everyone is running BitTyrant than the people who upload the fastest will download the fastest. The rest will have to wait for the faster ones to finish. If those people seed this may speed things up overall, but if they don't seed then people with slow upload speeds will be out of luck. You'll see cases where instead of people uploading to 4 people at 10 kbps, they'll set up to upload to 2 people at 20 kbps or 1 person at 40 kbps to get better download priority.
An interesting thing about this client is that if there are multiple people in a swarm running this client who happen to have smaller upload pipes than the other non-BitTyrant client users, than the BitTyrant clients will not upload to each other since their ratios will be too low.
If everyone is running BitTyrant than the people who upload the fastest will download the fastest. The rest will have to wait for the faster ones to finish. If those people seed this may speed things up overall, but if they don't seed then people with slow upload speeds will be out of luck. You'll see cases where instead of people uploading to 4 people at 10 kbps, they'll set up to upload to 2 people at 20 kbps or 1 person at 40 kbps to get better download priority.
An interesting thing about this client is that if there are multiple people in a swarm running this client who happen to have smaller upload pipes than the other non-BitTyrant client users, than the BitTyrant clients will not upload to each other since their ratios will be too low.
hmm
doesn't work well with tor
MxxCon join:1999-11-19 Brooklyn, NY |
MxxCon
Member
2007-Jan-3 3:33 pm
Re: hmm
only brain dead retards use TOR with P2P networks!
click_310 Eat my shorts join:2002-12-06 Savannah, GA |
So thats why the ToR network was slow earlier, you were choking it with P2P traffic.
raisch Premium Member join:2006-11-30 West Newton, MA |
raisch
Premium Member
2007-Jan-3 11:19 am
A Commons(sic) Problem
"That which is common to the greatest number has the least care bestowed upon it. Every one thinks chiefly of his own, hardly at all of the common interest; and only when he is himself concerned as an individual." - Aristotle
JimF Premium Member join:2003-06-15 Allentown, PA |
JimF
Premium Member
2007-Jan-4 3:33 pm
Re: A Commons(sic) Problem
I am not sure it is a problem of the commons. That refers to private overuse of a common resource. Here, we want the common resource (bandwidth) to be used to the maximum extent. It is just a question of "fairness" (whatever that is), and efficiency. You don't want people who never upload to be able to download, but we all are in situations where we have not uploaded to a particular area, and so might never be able to download. We depend on the generosity of others.
snipper_cr Premium Member join:2002-01-22 Wheaton, IL |
GAG
I guess i have to find out what "Choke" means in the overall scheme of torrenting. I am pretty sure i know how it works in general but this is a new term for me.
jdong Eat A Beaver, Save A Tree. Premium Member join:2002-07-09 Rochester, MI |
jdong
Premium Member
2007-Jan-3 3:56 pm
Re: GAG
Depending on your client's wording, choke means that one client is not giving the other pieces for whatever reason (i.e. max upload slots exceeded, your client has yet to give the remote client a chunk back)
Leech Clients not welcome
I will be finding a way to automate the banning of all BitTyrant clients in my swarms, I already banned 2 of them that were leeching without uploading anything to me.
yabos join:2003-02-16 London, ON |
yabos
Member
2007-Jan-3 4:08 pm
Re: Leech Clients not welcome
It's not a leech client, read it. If you manage a tracker you should understand the BT protocol and from what I can tell this doesn't leech any more than any other client does.
Re: Leech Clients not welcome
said by yabos:
It's not a leech client, read it. If you manage a tracker you should understand the BT protocol and from what I can tell this doesn't leech any more than any other client does.
No but it rewards those with faster uploads, therefore affecting the neutrality of the system, in return those that download it faster will most likely not remain to share their download with those with slower uploads thus impacting the speed of the swarm. Now if the ratio was tracked by trackers over a period of time rather then the most common individual torrent basis that would be acceptable, but that bring other cons to BT to it. It's not a leech client, read it. If you manage a tracker you should understand the BT protocol and from what I can tell this doesn't leech any more than any other client does.
Ignite Premium Member join:2004-03-18 UK |
said by pkarlos_76:
I will be finding a way to automate the banning of all BitTyrant clients in my swarms, I already banned 2 of them that were leeching without uploading anything to me.
You ban clients that don't upload to you? It's the way BT works, you can't always expect a peer regardless of their client to upload to you even if you are uploading to them. This client appears to more strongly reward uploading in a swarm. It'll be interesting to see how this pans out.I will be finding a way to automate the banning of all BitTyrant clients in my swarms, I already banned 2 of them that were leeching without uploading anything to me.
This doesn't look like a leech client, not even close. Read »www.cs.washington.edu/ho ··· rant.pdf for more information on how the client works and why it is as it is.
Doctor Four My other vehicle is a TARDIS Premium Member join:2000-09-05 Dallas, TX |
said by pkarlos_76:
I will be finding a way to automate the banning of all BitTyrant clients in my swarms, I already banned 2 of them that were leeching without uploading anything to me.
You're not alone. I just read on Slyck.com that BitmetvI will be finding a way to automate the banning of all BitTyrant clients in my swarms, I already banned 2 of them that were leeching without uploading anything to me.
and Oink have banned it.
sounds like Emule/Edonkey
Emule has a system like this, and it basically slows everything down in the name of fair sharing. It takes absolutely forever to start downloads, since you're stuck in a vicious "chicken and egg" circle of "I can't upload anything to download" and "I can't download anything to upload".
As it stands, Bittorrent is how the Edonkey protocol used to be before ratio systems were added to the clients; Fast. After Edonkey started adding anti-leech systems to the clients, the speed went into the toilet, and the queues started skyrocketing.
I suspect that if this catches on, you can kiss 300kb's downloads goodbye.
As it stands, Bittorrent is how the Edonkey protocol used to be before ratio systems were added to the clients; Fast. After Edonkey started adding anti-leech systems to the clients, the speed went into the toilet, and the queues started skyrocketing.
I suspect that if this catches on, you can kiss 300kb's downloads goodbye.
••••
MxxCon join:1999-11-19 Brooklyn, NY |
MxxCon
Member
2007-Jan-3 3:36 pm
Not too bad.
i'm talking to some developers in #bittorrent on freenode and they think this client's ideas are solid and can improve the network.
bittyrant's whitepapers are detailed and thought out..
bittyrant's whitepapers are detailed and thought out..
just call it...
bitleech, then that way you know exactly what you are getting...
jig join:2001-01-05 Hacienda Heights, CA |
jig
Member
2007-Jan-3 5:43 pm
it's a hobble
anything that treats a section of peers deferentially negatively disrupts the efficiency of the swarm as a whole. if you make the assumption that the people running bityrant will stick around and upload just as much as they would had they been using the normal client tweaks, then you are probably making the wrong assumption. any white paper that doesn't take into account the psychology of the user properly (generally as selfish as possible) isn't an accurate accounting/prognosis.
it's the tragedy of the commons, if it's allowed to enter the swarms. my guess is that it'll get banned at the tracker, though to do it properly is going to take resources since the tracker can't rely on the name of the client (there will have to be some kind of analysis).
to me, this sounds more like a subtle poisoning of the swarms by the **aa.
it's the tragedy of the commons, if it's allowed to enter the swarms. my guess is that it'll get banned at the tracker, though to do it properly is going to take resources since the tracker can't rely on the name of the client (there will have to be some kind of analysis).
to me, this sounds more like a subtle poisoning of the swarms by the **aa.
Re: it's a hobble
what you are saying is, is that if something is for free then in essence has little to no value and therefore the mechanism of sharing gets tied to something with a value and in this case it's a ratio... however, something like bittyrant only gives preference to those with the highest ratios within a given swarm size... the only side to that is if it takes advantage of DHT and then it could become a trackerless BT client and by then no one would care about leeching or seeding...
jig join:2001-01-05 Hacienda Heights, CA |
jig
Member
2007-Jan-3 11:25 pm
Re: it's a hobble
i don't think that's what i'm saying. i'm saying that once you look to make the swarm as efficient as possible, and consider seriously that each client wants in and out of the torrent as quickly as possible (you can't rely on some sense of sharing on the part of the client), you realize any shortcuts to get out quicker than a client's upload dictates is a detriment the the global efficiency.
basically, you are trunking off a segment of the bandwidth, and not necessarily giving back to the portions of the swarm that want it. of course it's less efficient.
basically, you are trunking off a segment of the bandwidth, and not necessarily giving back to the portions of the swarm that want it. of course it's less efficient.
Since it hasn't been pointed out already,,,
This is just snake oil for the script kiddies. This client appears to do nothing substantially different than any other BT client.
Before you tracker admins break out the bansticks I feel the need to point out that ALL BT clients choose who to upload to based on how fast they're downloading from those peers. If you're going to ban this client for being selfish you need to ban every other client for doing the same thing.
And in fact the better a client is at making sure it's uploading to the fastest up loaders the better it is for the swarm as it means fast up loaders will have more pieces of the file to upload which means more distributed copies will appear faster.
Before you tracker admins break out the bansticks I feel the need to point out that ALL BT clients choose who to upload to based on how fast they're downloading from those peers. If you're going to ban this client for being selfish you need to ban every other client for doing the same thing.
And in fact the better a client is at making sure it's uploading to the fastest up loaders the better it is for the swarm as it means fast up loaders will have more pieces of the file to upload which means more distributed copies will appear faster.
-
-
-