Aimbot Detection & Auto-Ban System

[Wazuki]

Your name: W4zuki
Suggestion headline: Add a reliable anti-cheat system that detects aimbots and automatically punishes offending players (e.g., auto-kick followed by auto-ban if repeat offenses occur).
Importance of suggestion: 10/10
Why should it be accepted: After implementing the system mentioned below, admins will no longer have to spend months or even a year monitoring a suspected player and collecting excessive evidence just to ban someone for aimbotting.
Once this anti-cheat is active, the responsibility shifts from individual admins to the system itself.

• This is actually healthier for both sides:
  It reduces admin workload and hesitation.
  It removes personal bias or human error.
  It provides consistent, automated detection instead of subjective judgment.

Most importantly, it offers a benefit of doubt, false positives from admins are possible, especially in fast combat situations, whereas a properly configured system relies on repeated, statistical detection rather than one person’s perspective.

• In short, this change protects:
  Legit players.
  Admins from wrongful accusations.
  The server’s credibility and fairness.

• Implement a complete automated anti-cheat system covering:
  A proven anti-cheat framework.
  Advanced detection logic.
  Automated punishment rules.
  Admin alerts & logging.
  Full server integration.
  Built-in safeguards against false positives.

[Karu_Tepes]

While I love this idea and would support it... what you're asking for has been an SA:MP golden ticket for devs, modders, and so on, for the last 20 years. Many attempts have been made at an automatic bot that can detect and ban aimbots, and in those 20 years, the best success has come from one able to determine player accuracy as a percentage. You're either going to have an inaccuracy toward the aimbots, where if they know what they're doing they can get around it, or toward regular users in which someone who is really good will be auto-banned and accused of something they didn't do.

You can make a bot capable of determining the likelihood of someone aimbotting, but you'll never be able to fully make one that has the human element needed to fully determine the end result of if someone is using an aimbot or not.

[apapaoemama]

I support this suggestion. On some servers, you need to have an anti-cheat system before you can enter the server. The anti-cheat that was developed detects files that are not legal. If a developer is interested in this, send me a message on Discord and I can show it.

[Karu_Tepes]

I support this suggestion. On some servers, you need to have an anti-cheat system before you can enter the server. The anti-cheat that was developed detects files that are not legal. If a developer is interested in this, send me a message on Discord and I can show it.

Interesting if true, but my question there lies in is it compatible with 0.3.7R (some of those require 0.3.7DL, which we cannot force onto players for security reasons) and it may pick up false positives, such as my ELM mod which some servers detect as "command flood" because of how it cycles the headlights.

[Wazuki]

Interesting if true, but my question there lies in is it compatible with 0.3.7R (some of those require 0.3.7DL, which we cannot force onto players for security reasons) and it may pick up false positives, such as my ELM mod which some servers detect as "command flood" because of how it cycles the headlights.

Regarding compatibility, the suggestion should be limited to solutions that are fully compatible with 0.3.7R, without forcing 0.3.7DL on players. There are anti-cheat frameworks and server side detection methods that work purely on 0.3.7R and don’t rely on client enforcement, so this wouldn’t necessarily require pushing players to DL.

As for false positives, especially cases like ELM or similar mods, that’s exactly why the proposal focuses on a system based approach rather than instant bans. Detection --> logging --> escalating actions over repeated flags significantly reduces the risk of punishing legitimate behavior. In that sense, a properly tuned system can actually be less prone to false positives than manual admin judgment in fast combat situations.

The goal isn’t a "hard lock" or adjusting over aggressive settings as it might ban legit players inaccurately if tuned badly. but a consistent, automated detection system with safeguards, where admins supervise logs and appeals instead of having to personally spectate players for months to get some evidences.

If done carefully, this shows fairness, reduces admin workload and avoids the exact issues you mentioned.

[saeeddi]

I gotta be real with you doing it for aimbots is tricky. If they set the system to be too strict, we might accidentally ban our legit players because they have good flick shots. If we set it too loose, the cheaters just tweak their settings to bypass it.

[Karu_Tepes]

I gotta be real with you doing it for aimbots is tricky. If they set the system to be too strict, we might accidentally ban our legit players because they have good flick shots. If we set it too loose, the cheaters just tweak their settings to bypass it.

Exactly this. I have been playing SA:MP for literal decades. Aimbots are not new, they've been around forever. What is new is how they function. As time goes on, and aimbots get discovered, admins across all servers learn to spot them. Before it was easy, it was always on and you could see them just nailing every shot. Then they upgraded: Now they could turn it on and off at the press of a button, making it harder to spot. But once you knew the signs, a ban was incoming. And so they upgraded again, now it's a toggle AND misses every 4th shot to make sure they aren't "100% accurate".

When I say server admins, across all servers, have been in a literal arms race against this shit, I am being literal. While a bot that can find and determine illegal mods might sound good in theory... they will immediately obfuscate the code to where a bot cannot read it. If you're wondering, Ls-rcr has already been nailed with obfuscated code to get around that sort of thing and it worked incredibly well, leading to a large scale security breach. The idea of a bot that can do that is a genius idea on paper, and a nightmare in execution. Like I said, how am I to know that it won't pick up my ELM mod? What if it scans my whole computer for files I might have linked into my SA:MP client, but aren't located in my actual files? Do you suggest we scan a users ENTIRE MACHINE for aimbots? That's highly invasive and violates about 120 countries internet privacy laws. You'd be making Ls-rcr into literal malware doing that.

Now, I do like W4zuki's idea, because it's a good idea... on paper. In execution, even if it could be done, it's a very tricky thing to do and opens up many questions for security, privacy, and so on. Now, are there things that could be done server side to help admins find and punish aimbots? Absolutely, they fully exist. Now, if you want a fun fact... those systems already exist. They are already implemented. I am forbidden by Energ, JohnCarer and Astrl from discussing that any further, because I put the server at risk knowing what I know. But I can tell you that yes, in general, there are already systems in place, but I cannot tell you what those are, how they work, or what they look like. But I can assure you our admins are very well equipped to handle the problem, and the only reason they are not filing more bans is because it takes time to gather proof, and our staff are limited on time.

[Crimson]

It's actually a good idea and I support it but the thing is that it is not really possible. As saeedi said we might ban innocent players for having such a crazy aim but also it is unnecessary. Our playerbase is low and we are moving on, Developers working hard everyday for a move on and requesting this system in an old game is exhausting for them and kinda impossible. Waste of time and no more updates are needed till we move on.

[Commander]

It seems like a good idea if the developers work on it and add the necessary features. Let's see how it actually works to avoid any unfair actions, especially if no admin is online


Also, that doesn't mean the staff team is going to be useless. They are available to spectate players and check their computers to find any forbidden

[Wazuki]

If you want a fun fact... those systems already exist. But I can tell you that yes, in general, there are already systems in place, but I cannot tell you what those are, how they work, or what they look like. But I can assure you our admins are very well equipped to handle the problem, and the only reason they are not filing more bans is because it takes time to gather proof, and our staff are limited on time.

I should not claim I know their hidden systems.
Instead, I can outline the well-known, generic server side systems that exist across SA-MP.
When you say "systems already exist", that makes sense and to be clear, I’m not asking for server specific details or anything confidential.

• In general, when we talk about server side anti-aimbot systems in SA-MP, they usually fall into a few known categories that are widely existed across the community:
  
  • Shot & hit analysis:
  (tracking hit ratios, headshot frequency, weapon spread consistency) and consistency over time to flag unnatural aiming.
  • Aim behavior pattern checks:
  (snap speed, angle changes between frames, impossible tracking)
  • Weapon specific behavior profiling:
  Different thresholds for M4, AK, sniper, etc., since each weapon has known recoil, spread, and fire-rate characteristics.
  • Statistical flagging over time:
  not single incidents
  • Automated logging & evidence collection:
  Time-stamped logs, damage data, target tracking, and contextual information stored for later review.
  • Internal logging systems:
  that store combat data for later review

What these systems look like in practice is usually logs, flags and alerts, not instant bans. They silently collect data, mark anomalies, and require repeated abnormal behavior before escalating.

That’s exactly why my suggestion focuses on automation and escalation, not disclosure.

• If systems already exist, the idea is simply to:
  
  • Let them handle first-level detection consistently.
  • Reduce the reliance on admins manually spectating for months.
  • Use system-generated data to support or dismiss suspicions objectively.