Memory Reader.

**cloud99** · 03.06.17, 08:01

Hello People/Other Members,

I've noticed many discussion have "memory reader" mentioned in them. Now I know it is a part of a tool which generates a Peer-id, Key and port for ratio cheating tool. Does this capture the current "Peer-id, Key and port" of the Torrent Client used?

i.e., If I were to be using qBittorrent 1.0 and my peer id=5456dga, key=a87aw4rg5 and port=443. I assume memory reader can capture this data live and insert it into the ratio cheating tool.

Can I copy data from the log file of a current emulation (Torrent Client) and insert it into the ratio cheating tool? Does this make sense? Cause every different torrent file loaded in a torrent client causes a change in peer id and key. Will different ids reported to the tracker after completion to seed cause any problems?

Regards,
P.S. Please move it to the right section if wrong!

**anon** · 04.06.17, 04:59

Originally Posted by cloud99

I've noticed many discussion have "memory reader" mentioned in them. Now I know it is a part of a tool which generates a Peer-id, Key and port for ratio cheating tool. Does this capture the current "Peer-id, Key and port" of the Torrent Client used?

i.e., If I were to be using qBittorrent 1.0 and my peer id=5456dga, key=a87aw4rg5 and port=443. I assume memory reader can capture this data live and insert it into the ratio cheating tool.

The memory reader is meant to be used when emulating clients that use a global peer_id and/or key. As the name suggests, it reads the memory space of the client's process until it finds announce data, which is then parsed to get the required strings. After this, if a tool uses that data and does a flawless emulation, it and the client will appear as one and the same to the tracker (using the same port is also a must), which is less suspicious and very desirable.

Many years ago, when generation algorithms were not understood, it had the extra function of obtaining values that were actually valid.

Can I copy data from the log file of a current emulation (Torrent Client) and insert it into the ratio cheating tool? Does this make sense?

Yes, if you prefer a slower path

Cause every different torrent file loaded in a torrent client causes a change in peer id and key.

Not always, and this feature is not meant to be used with clients that generate new data upon starting a torrent (see above).

**cloud99** · 06.06.17, 09:01

Originally Posted by anon

global peer_id and/or key.

The prefix of the global peer id remains constant. (i.e., -TIXATI4.53- not the suffix according to me). Can I use a different peer-id generated by other torrent announce. (i.e., abcmovie.torrent announce "h9c8i3d7h8e9" to private tracker, can I seed defmovie.torrent to a private tracker using "h9c8i3d7h8e9" or do I have to use the same id/key/port as the original client to act less suspicious?

**anon** · 06.06.17, 16:14

Originally Posted by cloud99

Can I use a different peer-id generated by other torrent announce ... or do I have to use the same id/key/port as the original client to act less suspicious?

When the peer_id and/or key are global, the same one(s) is/are used for all running torrents (not just the prefix, all of it). Otherwise, a different one is used for each torrent. That usually also correlates with generating new values every time the torrent is started, but that's not 100% guaranteed. You have to find out how the client you're emulating acts in this regard, then duplicate that behavior.

**anthony-joal** · 18.08.17, 01:01

First things firsts. Thanks for your answer

i finally know what memory reader is.

Originally Posted by anon

It reads the memory space of the client's process until it finds announce data, which is then parsed to get the required strings. After this, if a tool uses that data and does a flawless emulation, it and the client will appear as one and the same to the tracker (using the same port is also a must), which is less suspicious and very desirable.

I don't really understand why is this used. IMO, having two torrent clients seeding at the same time is not an obvious proof of cheat. I can have multiples computers (or even a local seedbox) or familly members that seeds in my house (of course, dont use BitCommet since it sends the local ip along with the announce, but who use that mess anyway...).

Isn't memory reader more "dangerous" as regular key & peerid emulation?
Let's take Mr Scatty as an example. Mr Scatty uses utorrent (which does not reset peerid until restart). And he use a ratio-faker tool that uses memory reader.
First issue
1) Mr Scatty has downloaded a torrent from his favorite tracker.
2) Mr Scatty is inadvertant (what a surprise) and he forgot to remove the torrent that has started to seed in utorrent.
3) Mr Scatty starts faking seed with his ratio tool.
4) Here comes the fuck. utorrent and ratio-faker will both sends announces with inconsistent stats with each others, on the same torrent and with the same peerid.

Second issue
1) Mr Scatty starts to seed a torrent with a ratio-faker that uses memory reader.
2) Mr Scatty closes utorrent, then restart it.
3) Is Mr Scatty going to use the new peerid after that utorrent has restarted or not?

To me, explaining that you have two computers will seems more legit than sending inconsistent stats.

I'd be happy to have some feedback on this.

**slikrapid** · 18.08.17, 17:17

the two issues mentioned should not appear if used properly:
- don't run the same torrent both in your real client and RM at the same time
- don't use RM prior to getting a valid memory reader result

afaik there was a tutorial or few on how to use RM
anyways, afaik this was the way things were several years ago hehe

edit: the RM example may apply to any other tool

**anthony-joal** · 18.08.17, 17:35

Ok. Seems fair, but i'll most likely not implement this feature in joal ^^.
When it comes to users, the less rules there is the better it is :)

CF: murphy's law ^^

**slikrapid** · 18.08.17, 17:45

it would be nice to be able to circumvent these rules as needed (advanced users)

also, regarding that murphy's law: the number of rules alone does not speak about their quality

**anon** · 19.08.17, 04:16

Originally Posted by anthony-joal

I don't really understand why is this used. IMO, having two torrent clients seeding at the same time is not an obvious proof of cheat.

It's not (one of the most proficient What.cd uploaders ran four instances of uTorrent with about 5000 seeding torrents each), but when you're cheating, you want as little attention as possible, and most people don't run multiple clients. Not to mention things won't look good for you if you make a mistake and report suspicious upload that's also coming from a different peer_id.

Isn't memory reader more "dangerous" as regular key & peerid emulation?
Let's take Mr Scatty as an example. Mr Scatty uses utorrent (which does not reset peerid until restart). And he use a ratio-faker tool that uses memory reader.

In both examples, the problem is between the keyboard and the chair, not the memory reader