| Author |
Message |
hav3k
Joined: 18 Mar 2007
Posts: 8
|
 Posted: Mon Mar 19, 2007 5:11 pm Post subject: |
 |
|
| admin wrote: | | As for now, GreedyTorrent is impossible to detect if used properly. |
Just to put this in context, its not impossible to detect if used correctly. The one thing this ratio 'fixer' along with all the others can't (and will never be able to do) is balance the total up+total down for a specific torrent.
Take for instance the case where you have a 0:0 ratio, you download a 1GB file and set GT to upload say 0.9MB for every 1MB downloaded (hence producing a 1:0.9 ratio which is recommended). When you have finished downloading the file you will have 1GB down and 900MB up, this will mean the torrent is now out by 900MB (i.e. the other members of the swarm will report uploading the 1 GB to you but not downloading the 900MB) so it will be attributed to you as the difference will match what you "claim" to have uploaded.
This is exactly the same issue if you have it increasing your upload.
I think the statement should more be a "if the software is used rarely for very small amounts or on a swarm full of ratio 'fixers'" then its hard to detect. Your relying more on the fact that the site doesn't have the facilities or knowledge to detect it as opposed to it being undetectable.
Just thought I would clear that up as its wrong to claim something is impossible to detect when used correctly as there is no situation in which it is "impossible".
Edit: Correction - there is a situation in which it would be impossible to detect, on public sites where your ratio isn't logged... but that would be pointless now wouldn't it. |
|
| Back to top |
|
 |
admin
Site Admin
Joined: 25 Feb 2007
Posts: 42
|
| Posted: Mon Mar 19, 2007 6:03 pm Post subject: |
|
|
| hav3k wrote: |
Take for instance the case where you have a 0:0 ratio, you download a 1GB file and set GT to upload say 0.9MB for every 1MB downloaded (hence producing a 1:0.9 ratio which is recommended). When you have finished downloading the file you will have 1GB down and 900MB up, this will mean the torrent is now out by 900MB (i.e. the other members of the swarm will report uploading the 1 GB to
you but not downloading the 900MB) so it will be attributed to you as the difference will match what you "claim" to have uploaded.
|
Yes. You're right, but only up to an extent.
You can detect a cheating user using this method only if total(all other peers download) < upload(gt user). Larger the swarm, the impossible it gets to detect. Smaller multiples of upload the user use, the difficult it becomes again. However, you can definitely understand from the values that some one in the swarm has fooled the tracker, but not quite 'who'.
You cannot however take this as a proper condition to detect a GreedyTorrent user, due to the following reasons.
1. Some peers may not have reached the tracker update timing to report the actual downloaded quantity.
2. Some might have even failed to report or may not report at all, due to a crash of the client, lost internet connectivity, offline tracker, selfish clients etc.
3. There are some tools available that report less than the actual download quantity. In such a case, the user you suspect as a GreedyTorrent user could be a legitimate one, and a hidden user the swarm may be reporting a low download to maintain his ratio. GreedyTorrent's next version which is under development might feature this.
Btw, I'm glad that finally I meet someone who talks sense and not foul mouth. Good luck. |
|
| Back to top |
|
|
natedogg
Joined: 19 Mar 2007
Posts: 11
|
| Posted: Mon Mar 19, 2007 7:40 pm Post subject: |
|
|
haha! hot topics and strong arguments.. how can I keep silent?
by now you guys know every bit of the protocol spec, rt? 
@admin
i think you forgot these two. 
--- torrents with DHT enabled. reported ratio may not be accurate, as peers from outside swarm could be leeching.
--- torrents whose tracker update is disabled... i do this often, (but I will endup in trouble if torrent is stopped halfway)... i remove tracker info from utorrent immediately after 1st update. now how 'bout that... how bout u add that too in next version????
now @hav3k,
wht u say abt these two???? you'll end up banning the wrong, legitimate, poor, seeding user. wanna do that???? |
|
| Back to top |
|
|
hav3k
Joined: 18 Mar 2007
Posts: 8
|
| Posted: Mon Mar 19, 2007 8:26 pm Post subject: |
|
|
| natedogg wrote: | haha! hot topics and strong arguments.. how can I keep silent?
by now you guys know every bit of the protocol spec, rt?
@admin
i think you forgot these two.
--- torrents with DHT enabled. reported ratio may not be accurate, as peers from outside swarm could be leeching.
--- torrents whose tracker update is disabled... i do this often, (but I will endup in trouble if torrent is stopped halfway)... i remove tracker info from utorrent immediately after 1st update. now how 'bout that... how bout u add that too in next version????
now @hav3k,
wht u say abt these two???? you'll end up banning the wrong, legitimate, poor, seeding user. wanna do that???? |
To start with DHT is disabled on private trackers as its defeats the object of them being private. Most of the mainstream clients will adhear to the private flag being set on the torrent (i.e. DHT off) and will only accept peer information from the tracker and not other members of the swarm (as DHT relies on).
Sure you can disable update (aka ghosting) but you will be pulling from more than 1 member of the swarm so there will be a difference in upload (which matches your download) across the members of the swarm who report, not 1 spike in 1 user as I was detailing before. When somebody comes across this they will notice its not 1 user its attributed to but there is 1 user who has snatched a few MB if not less and also you have more than likely done it on a number of torrents. Now it could be chance that you only download a few KB of a release or it could be that you ghost - i'll let you guess what they believe. |
|
| Back to top |
|
|
Melvinmeow
Joined: 21 Mar 2007
Posts: 31
|
| Posted: Wed Mar 21, 2007 2:47 pm Post subject: |
|
|
| hav3k wrote: | | natedogg wrote: | haha! hot topics and strong arguments.. how can I keep silent?
by now you guys know every bit of the protocol spec, rt?
@admin
i think you forgot these two.
--- torrents with DHT enabled. reported ratio may not be accurate, as peers from outside swarm could be leeching.
--- torrents whose tracker update is disabled... i do this often, (but I will endup in trouble if torrent is stopped halfway)... i remove tracker info from utorrent immediately after 1st update. now how 'bout that... how bout u add that too in next version????
now @hav3k,
wht u say abt these two???? you'll end up banning the wrong, legitimate, poor, seeding user. wanna do that???? |
To start with DHT is disabled on private trackers as its defeats the object of them being private. Most of the mainstream clients will adhear to the private flag being set on the torrent (i.e. DHT off) and will only accept peer information from the tracker and not other members of the swarm (as DHT relies on).
Sure you can disable update (aka ghosting) but you will be pulling from more than 1 member of the swarm so there will be a difference in upload (which matches your download) across the members of the swarm who report, not 1 spike in 1 user as I was detailing before. When somebody comes across this they will notice its not 1 user its attributed to but there is 1 user who has snatched a few MB if not less and also you have more than likely done it on a number of torrents. Now it could be chance that you only download a few KB of a release or it could be that you ghost - i'll let you guess what they believe. |
You also didnt take notice to the fact that many trackers have systems to catch GHOST leechers. If your going to try and cheat a tracker your probably better off using the same method each time. As using a couple differant methods makes it easier for the staff to notice. They might not flag you using the single method but multiple methods adding up makes it a little more obvious.
Example: I steal an apple from the local fruit stand... nobodys notices. 2 minutes later I steal an orange.. Few minutes after that I steal a banana... after a while the fruit cart idiot is bound to notice you. Even if he doesnt notice you that 1 time... if he starts to notice more and more fruit missing it will alert him to possible theft and he will start to watch his potential customers more. (Possibly even install a video camera hehehe)
Just thought I would present my theory and opinion on the matter. |
|
| Back to top |
|
|
ace696
Joined: 15 Mar 2007
Posts: 109
|
| Posted: Wed Mar 21, 2007 4:08 pm Post subject: |
|
|
| Nice theory. Hope it works! |
|
| Back to top |
|
|
ADSL (l)user
Joined: 16 Mar 2007
Posts: 19
|
| Posted: Tue Mar 27, 2007 8:06 pm Post subject: |
|
|
| admin wrote: |
Then, about using it in USB. Right now you can copy the executable GTor.exe if you want, to a USB or removable drive and use it from there. However, the settings are stored in windows registry, which is machine dependent. You will need to configure both GreedyTorrent and the BitTorrent client the first time on each machine. |
Maybe you should store GT's settings in .\Documents and Settings\UserName\Applications Settings\Greedy Torrent\? I think that that is the proper way to store your setting nowdays.
If you would have that, you could make a "portable" GT client version, which would read its settings from current dir, and that would be it.  |
|
| Back to top |
|
|
Whatever
Joined: 08 Apr 2007
Posts: 1
|
| Posted: Mon Apr 09, 2007 12:23 am Post subject: |
|
|
I'll second request on proxy chaining.
I guess if someone really needs it it can be done with current version using external app like sockschain, but would be much better if it was inherently supported.
| Quote: | 1. Even if you use a free internet proxy, it cannot be a transparent one, but an anonymous one (one that does not use the 'via' header, and forwards traffic as it were originated from it). If you are using a transparent proxy the tracker can get your real ip from
the Via field.
2. Only the tracker communication can be made anonymous by using such a proxy. The peer-wire protocol is still open, and your ip does not stay anonymous.
3. If multiple users start using the same anonymous IP, tracker admin can list and ban the users connecting via it, and ban the anonymous proxy from connecting to it in the future.
4. If used a transparent proxy, tracker admin can always ban users connecting from publicly available proxies. |
There're way too many proxies out there to make banning them of any use.
| Quote: | | Only the tracker communication can be made anonymous by using such a proxy. The peer-wire protocol is still open, and your ip does not stay anonymous. |
Obviously when talking about proxies in context of greedytorrent it's not about keeping your ip anonymous but about bypassing tracker bans or trying to avoid getting your real ip banned in case you're going to do something stupid like uploading at abnormally high speed.
I guess some people _need_ to use proxies in original setup so no inherent proxy chaining support in greedytorrent means extra issues to hem.
| Quote: | | There are some tools available that report less than the actual download quantity. In such a case, the user you suspect as a GreedyTorrent user could be a legitimate one, and a hidden user the swarm may be reporting a low download to maintain his ratio. GreedyTorrent's next version which is under development might feature this. |
Yes, download reduction sounds like an important ratio-cheating tool.
Maybe simultaneously reducing amount of downloaded data + multiplying amount of upload could be less detectable than doing either of that on it's own.
On other hand obvious problem arises is that you can get to point of reporting yourself as seeder to tracker having downloaded less than full torrent worth of data. Which is plain stupid giveaway.(not reporting yourself as seeder having downloaded full torrent worth of data is also bad in terms of cheating detection i guess)
While modified original clients probably can have facilities to prevent such giveaway that'll probably become a big issue with proxy-like app.
I guess udp and ssl private trackers won't be affected by proxy-like ratio-cheating tools though? |
|
| Back to top |
|
|
gtuser001
Joined: 26 May 2007
Posts: 4
|
| Posted: Sat May 26, 2007 6:58 am Post subject: Feature request |
|
|
Getting back to the original post in this thread, I had the same idea, which is how I found this forum. But I have since had a better idea.
There are two problems with using the download multiplier option. First, all your torrents have exactly the same ratio - very suspicious. Second, it seems you no longer get any credit for any files you seed! This is because if you already have the file complete you aren't downloading anything so GT is always setting your upload amount to 0. Correct me if I'm wrong, but this is what it looks like to me from viewing the log at least.
I think I have the solution to both problems. Use both a download and upload multiplier then add them to get the reported upload. The way it solves the first problem is the upload multiplier 'randomizes' your ratio. And it solves the second because upload amounts are credited even when the download is 0. This method also solves another problem which isn't solved by simply using an upload multiplier. If you are the only leecher and everyone else is a seeder you get no credit because there is no one to upload to. In this case the download component would give you your ratio.
Finally, a comment about ethics. I for one am uploading MORE in real terms since I discovered GreedyTorrent. The reason is before I had no chance of maintaining my ratios so basically I wouldn't even try. If I got banned I would just change my IP and email and create a new account and start over. I mean, there isn't much more I can do since my upload bandwidth is 100% utilized 24 hours a day, but now that I at least have a fighting chance of maintaining a proper ratio I am much more concerned with uploading every way I can.
________
Rolling a joint
Last edited by gtuser001 on Wed Feb 16, 2011 1:46 pm; edited 1 time in total |
|
| Back to top |
|
|
mmmmm
Joined: 15 Jun 2007
Posts: 8
|
| Posted: Sun Jun 17, 2007 3:25 am Post subject: |
|
|
| good information |
|
| Back to top |
|
|
|
|
