Scion Bot - Runescape botting revolutionized

[Luuke]

Does it handle randoms?

 

[sikander33]

Read the thread please, the alpha(beta) will be released soon to scripters, and anti-randoms are what scripters will make..

 

[sex°]

I'll keep checking back with it, the only way I see this being better than rsbot is if it uses less cpu and has better antiban, less cpu is unlikely and antiban has to do with the people who are scripting it not the bot itself.

 

[Xooley]

Well its already 1.2. Today so is it released yet?

 

[rivgen]

the alpha(beta) will be released soon to scripters, and anti-randoms are what scripters will make..

I'll be happy to make anti-randoms :]. I believe scionbot scripts will look like rsbot scripts, because Java is similar to C#/C++/C.

 

[Castform]

I tried to get on and it says that I have been I.P. banned :O Help....

 

[xinFam0usx]

I'll keep checking back with it, the only way I see this being better than rsbot is if it uses less cpu and has better antiban, less cpu is unlikely and antiban has to do with the people who are scripting it not the bot itself.

Hello, I thought that I would get my say in about the bot since I just found the thread. I am the owner and admin of ScionBot.com along side of Jaco. Many of you, if you have been around the coding/rs scene for a long time know Jaco (skeletron). He helped out on JMacro, as well was well known on sites like RSUnlocked, and Cheatscene, and Cruels.

The bot is still pending an official release. Jaco wants to make sure the bot is perfect. No point to have a flawless color library and a not so stable core like other bots. The bot does support multiple bots, and has much added to it to make it faster and use less cpu and memory than a java bot.

It disables graphics when the window isn't focused which saves alot of cpu, especially for Vista users (if you play windowed games in vista you may know this already). The bot also uses the AForge Framework which gives it some exceptional image processing, neural networks, genetic algorithms, and other image detection methods. It also has an extremely fast detection of vertexes and other images.

At this point the framework is comprised of 5 main and some additional libraries:

* AForge.Imaging – a library for image processing routines and filers;
* AForge.Neuro – neural networks computation library;
* AForge.Genetic – evolution programming library;
* AForge.Vision – computer vision library;
* AForge.Machine Learning – machine learning library.

The bot will in time have alot of anti ban and be able to solve randoms like the school house random or even the Maze random because of its ability to find edges of walls.

I hope to see many of you future scripters, present coders, and anyone else looking to further the development of this bot.

This bot will always remain absolutely free, as will using our community. Thanks again .Kamikaze for posting this!

I tried to get on and it says that I have been I.P. banned :O Help....

This is probably because we use RBL, a product that checks the RBL database for IP's linked to spam. If you would contact me with the account you registered with, I will gladly unban it.

 

[odt]

Looks good, let me know when it's up and i'll try it

 

[Terrankiller]

Hello there! I would like to say I have personally tested the current compiled beta for Scion and I am impressed. Scion will definitely be the fastest bot for Rune Scape cheating. I am also glad the developer has decided to make the bot free.

 

[Four Aces]

Nice, hopefully I'll use it one day. Good work, nice to see your spending your time programming and such, .

- Four Aces

 

[darksilver]

Hello there! I would like to say I have personally tested the current compiled beta for Scion and I am impressed. Scion will definitely be the fastest bot for Rune Scape cheating. I am also glad the developer has decided to make the bot free.

thats good to hear... i'm a scripter, but i havent wrote a script in years, literally, so maybe ill start again if this bot lives up to the expectations that ppl are giving it. but definitely looking forward to this bot

 

[sikander33]

Scion - Preview
http://screencast.com/t/4blNcfj8

Recorded using Jing.


Completed:
-> Basic Mouse/Keyboard Functions
-> Bot Loader
-> Offscreen Screen Capture
-> GUI
-> X/Y Locations
-> Interval Antiban
-> Mouse Splines (Benland's Splines) - Antiban is the scripter's option to implement by setting the ReactionTime and Speed.

TODO:
-> Reflection Scripting Engine
-> Colour Methods (OCR, etc.)
-> Motion Detection Using AForge
-> Vertex Finder Using AForge (Wall Detection, etc.)
-> RS Related Functions (Bank, etc.)

More to see on the Scion website: www.scionbot.com

 

[arsh]

omg i cant wait much longer

 

[XAQ]

I registered and i reffered u too

 

[sikander33]

I registered and i reffered u too

Thanks alot mate, appreciate it

Btw guys, register there, we will have more info coming out soon, register to the best Runescape Botting site now :] and ............ Jaco is ftw.

 

[best991]

Bump*

 

[Parox]

Going to try the bot later, it seems like it's going to be better than RsBot. =O

 

[robotzindisguise]

I've worked with neural networks before. The only way I can see them (and machine learning) being useful is if you taught the bot to recognise objects/shapes. It would be cool to have a network that could recognise shapes/objects from various angles but that could take a very long time to teach the network, and could be unreliable, especially if you want to teach it to recognise a large number of shapes.
The US military about 10 years ago, tried to use this technique to develop software to identify different types of tank from images. They took 100 images of two different types of tank from various angles, and used a re-inforced learning algorithm to propogate the weights to the point the network was identifying the tank successfully in the photographs with a 100% success rate. When they took it into the field, it kept messing up and eventually the programmers realised the only distinction the network was making between the two sets of pictures was the position of the sun in the sky. The two different tanks had been photographed at different times of the day and the position of the sun is what the network was identifying, it hadn't actually learned the two tank's shapes.

The point, networks are complicated (you have to decide on the structure, connections, learning algorithms) they can give your bot a symbolic understanding of shapes but even if scripters can incorporate this somehow in my opinion it's overkill for image processing in RS, where the images are simple, and there are also easier and more reliable ways. Using a network would also be slower than a streamlined color based scan, and slower than what RSbot does to identify objects, even though I don't rate RSbot.

Compiled C# code will run about 5-10% faster than the equivalent Java, will use less memory, and is nicer to program in so it does have this advantage over RSbot. And you can certainly use some parts of the AForge framework such as edge recognition to your advantage, but it's not something other bots can't do. RSbot and NeXus use tiles right? So they can navigate around easily enough, and I use my own libraries to recognise walls, read maps, etc, and give the bot a 3D understanding of it's surroundings.

Jagex have encountered Benland's mouse splines plenty of times before in other (detectable) bots. The chance they don't have an automated system capable of detecting them is low.

Will the bot use the Genetics library?!

I don't think you should hype any bot based on the libraries you're inheriting from, because it's how well you (and scripters) understand them, and how you manipulate them (in the end, your code) that will ultimately make the bot good... or not

 

[sikander33]

Scion - Preview 02.09.2009 (FTP)

The source code has been uploaded to the website to give developers a reference of what has been accomplished on this bot. To note, all of the source has been licensed under the GPL License 3. The license applies to what you can do with the bot.

The source code must be compiled with Visual Studio 2008 or Visual C# Express Edition 2008. For early enthusiasts, the bot is also compiled in binary format. I will open a Scion Help section where questions regarding the bot can be posted.

The bot is basically done, just the colour functions remain to be finished. With help from DrD and other, this will be achieved and another an Alpha build will be posted by the end of this week. This build will have similar capabilities to that of JMacro. Work will then progress over the course of this month to produce a more feature-rich library.

Thanks for the wait! Remember to follow the first two message dialogs carefully or the bot will not be setup properly.

Downloading Info (FTP):
CAN BE FOUND ON SCIONBOT OFFICIAL WEBSITE (Look at my sig <3 )

FileZilla works just fine.

Direct Link:
Can be found at scionbot.com

 

[Cho]

I guess we'll have to test the ban rate when i comes out and things. Can't wait to see what its all about.