View previous topic :: View next topic |
Author |
Message |
awsentinel How do I cheat?
Reputation: 0
Joined: 17 Jul 2010 Posts: 8
|
Posted: Fri Jul 23, 2010 7:30 am Post subject: Predict player position |
|
|
Hi all!
What about predicting of some player position by having list of last positions.
Let suppose I have list of 3D points and its timestamps. Every point described player position at some moment of time. What do you think about some logic to predict players position after some period of time. It can be used for advanced aiming with slow speed weapon. E.g. aiming with a rocket at jumping (dodging) player.
PS. There is no problem if player moves on a stright line. More intersting is to predict trajectory of jumping player.
_________________
push bp |
|
Back to top |
|
 |
Slugsnack Grandmaster Cheater Supreme
Reputation: 71
Joined: 24 Jan 2007 Posts: 1857
|
Posted: Fri Jul 23, 2010 7:59 am Post subject: |
|
|
Definitely possible but I have a feeling the rocket time will often be much greater than the time taken for a jump. Other than that it's simple 3D projectile motion. Also unfortunately, many games unrealistically allow players to change direction in midair - accelerate towards another x/y whilst z still changes with a constant gravity acceleration.
|
|
Back to top |
|
 |
sangeli Master Cheater
Reputation: 0
Joined: 07 Dec 2006 Posts: 406
|
Posted: Fri Jul 23, 2010 10:24 am Post subject: |
|
|
don't predict, use a sniper and shoot the guy in the head
_________________
Dark Byte wrote: | ce can certainly damage hardware let's say you have a robotarm attached to your computer, and the software limits usually block it from ripping out it's own cpu. If you remove that limit and then issue the command to rip out the cpu, sure, say goodbye to your hardware |
|
|
Back to top |
|
 |
atom0s Moderator
Reputation: 205
Joined: 25 Jan 2006 Posts: 8587 Location: 127.0.0.1
|
Posted: Fri Jul 23, 2010 12:53 pm Post subject: |
|
|
Someone was doing this exact thing on Game Deceptions forums recently with Team Fortress 2. You can see the results and such here:
http://forum.gamedeception.net/threads/20213-TF2-Projectile-Bot
Further into the post is the math behind the logic of whats being done.
_________________
- Retired. |
|
Back to top |
|
 |
awsentinel How do I cheat?
Reputation: 0
Joined: 17 Jul 2010 Posts: 8
|
Posted: Sat Jul 24, 2010 1:19 am Post subject: |
|
|
sangeli wrote: | don't predict, use a sniper and shoot the guy in the head |
Even for sniper rifle you have to predict position regarding lags (ping) to server.
_________________
push bp |
|
Back to top |
|
 |
Jani Grandmaster Cheater
Reputation: 2
Joined: 29 Dec 2006 Posts: 804
|
Posted: Sat Jul 24, 2010 4:24 am Post subject: |
|
|
awsentinel wrote: | sangeli wrote: | don't predict, use a sniper and shoot the guy in the head |
Even for sniper rifle you have to predict position regarding lags (ping) to server. | Games' netcode already does this (well, should). You need to predict the time the bullet takes to travel. And height the bullet drops.
|
|
Back to top |
|
 |
Pingo Grandmaster Cheater
Reputation: 8
Joined: 12 Jul 2007 Posts: 571
|
Posted: Sat Jul 24, 2010 5:16 am Post subject: |
|
|
What game?
_________________
|
|
Back to top |
|
 |
awsentinel How do I cheat?
Reputation: 0
Joined: 17 Jul 2010 Posts: 8
|
Posted: Sat Jul 24, 2010 1:10 pm Post subject: |
|
|
The game is UT3. There is no ping correction wihout UTComp. I have a source code of UT3 engine. And there is some logic to synchronize players with a lags. May be there is some logic to predict position. I know about high level bots who able to predict any movements:) But have no idea how to find and use that code.
I'm just need to predict position of player position within 100-400ms:) Now I'm working on Lagrange polynom, it can be right solution. Not sure, but looks very promissed.
Really, I'm just need a help in prediction algorythm, but not in assistance about how to avoid this.
_________________
push bp |
|
Back to top |
|
 |
awsentinel How do I cheat?
Reputation: 0
Joined: 17 Jul 2010 Posts: 8
|
Posted: Mon Jul 26, 2010 11:43 am Post subject: |
|
|
Solved. Just some spline and then easy logic:
- calculate vectors for every last step
- add vectors to last position with some KF:
first - 1/2
second - 1/4
etc.
Now aiming is smooth and efficient.
_________________
push bp |
|
Back to top |
|
 |
|