[J_Tanzanite]

Originally Posted by lemeshovich i changed map few times and also made full server restart im not sure about ammo but no, i do not see any animation everything looks like i'm not pressing lmb but i am)

In case I wasn't clear in my reply, I asked you this:

So that leaves this function as the only other thing I can think of that could possibly ever block you from shooting. ... Other than a broken mouse, other plugins altering your weapons, sourcemod being bugged (doubt) or the game itself being bugged (extra doubt)... It makes me wonder tho, does this "blocked from shooting" bug only happen when your server has been on for a long time? Or can it also happen right after you've completely restarted your server? As in, within 30 minutes of a complete restart?

You also say this, which is a clue:

when this happened (im clicking lmb but my weapon isnt shooting) if i continue to press lmb nothing will change and to fix this i have to stop pressing lmb like for 3-4 secs after that everything becomes normal additional observations, that can happen with any weapon and with any enemy (even with bot)

I doubt it's a mouse problem if other players get this too and bots also aren't able to shoot sometimes...
Hence why I ask about how long it takes before this bug happens.
Because you also say:

played ~1.5 hour to reproduce bug, just happened

Kinda starts to sound like it's an issue with the server running for too long?
Which I kinda doubt, as I haven't heard others get this issue before... And floating point precision issues shouldn't happen this early... At least as far as I know.

But that's what it sounds like.

Wondering if others are getting the same issue, with servers running for a long time, and if they also have this bug with the Beta version of Lilac.

---

But yeah, I'd like to get an answer on when this bug happens.
Does it only happen if your server has been running for more than 30 minutes?
And does it get worse the longer the server runs without a restart?

[J_Tanzanite]

Originally Posted by azalty As I discussed about it on github, I can try to add that but I need more precise informations and things: - What should be logged? All detections/bans only/both - What do you want exactly? Detection #number (aimbot)/Delta/TotalDelta/Timestamp.. - Which format do you want? Store a string of the detection/store each info in a row (no sentence) - What are the different detections and things to store || design decisions: one table for each detection method? - What will be the use of that? Knowing what you want to do with the DB can help me design it better

Since I don't use SQL and never have (or ever will), I don't know what people need.
So... I don't know what would be the most useful structure.

Perhaps it should be a discussion on my github rep, so people can discuss how this should be set up.
But as far as I can tell, it's only a few people who have asked for this, and I somewhat doubt the discussion will be used much.

Since this is entirely out of my domain, I can't really answer ;~;
I guess we'll have to wait until someone else gives concrete answers :/

[Ryan2]

Originally Posted by azalty As I discussed about it on github, I can try to add that but I need more precise informations and things: - What should be logged? All detections/bans only/both - What do you want exactly? Detection #number (aimbot)/Delta/TotalDelta/Timestamp.. - Which format do you want? Store a string of the detection/store each info in a row (no sentence) - What are the different detections and things to store || design decisions: one table for each detection method? - What will be the use of that? Knowing what you want to do with the DB can help me design it better

Not that my opinion really matters but I think it would be best to just log each detection in it's own row.

name, steamID, IP, cheat, position, angles, map, team, weapon, connection time, gametime than we can work with the data however we need to.

[OneMore]

Originally Posted by J_Tanzanite So I'm not sure what you're talking about.

No problem, never mind. It is not the usual behaviour of players. But what I can say, that the beta version is more tend to false-positive aimbot ban than the previous one.
Anyway, if I meet these false-positive bans in the real-life, I'll send you demos or we can discuss it in more details.

[azalty]

Originally Posted by Ryan2 Not that my opinion really matters but I think it would be best to just log each detection in it's own row. name, steamID, IP, cheat, position, angles, map, team, weapon, connection time, gametime than we can work with the data however we need to.

Perfect! I'll try to work on it soon

And your opinion matters

[Le1T]

https://www.youtube.com/watch?v=aqM03TiQQmU

https://pastebin.com/9cP8aZcy

Why the ANTI-CHEAT has detected a cheater only 3 times?
about 10 mins he was playing with no ban :/

[J_Tanzanite]

Originally Posted by Le1T https://www.youtube.com/watch?v=aqM03TiQQmU https://pastebin.com/9cP8aZcy Why the ANTI-CHEAT has detected a cheater only 3 times? about 10 mins he was playing with no ban :/

Because no two cases are the same, and the world isn't black and white.
The Aimbot detection tests for 4-5 things, and some of those things can't be tested if a player is too close to an enemy, in Lilac, that range is 350 units (Which is larger than you think). So that can have an effect and is dependent on how far the cheater is from the kill victim.
Which means that different scenarios will play out differently.

Also, if you're referring to the wall-banging (Lilac doesn't check for that). Nor does it detect or deal with Wallhacks.

Another thing you have to consider is that not all cheats are the same, some cheats behave differently to others, even if they look the same (Networking has an effect too).
But even between the same cheats, different configs can yield different results. Plus, some cheaters play differently, which also can change things.

Lastly, the first detection isn't logged, and all aimbot detections expire after 10 minutes, so there may be even more detections that weren't logged.
The reason for this is simple, a single player may get a false positive once in a while (either just unlucky or lag can cause it), and if a player plays on a server for several hours, those detections can add up until an automatic ban is issued; even tho they didn't cheat.
To counter this, detections expire after 10 minutes.
As to why the first detection isn't logged, some admins may freak out over a player having lots of "detections: 1", especially if they play a lot. To prevent admins from misinterpreting cheat logs, the first detection isn't logged.

Looking at your cheat log tho:

... (Detection: 2 ...

Appeared three times, which means he was at minimum detected 6 times.

If I made the 10 minute expire time longer, than the possibility of false bans increase.

[Le1T]

Originally Posted by J_Tanzanite Also, if you're referring to the wall-banging (Lilac doesn't check for that). Nor does it detect or deal with Wallhacks.

No, im not about the wallhack. Dont pay attantion on it.
I'm talking about the moment 0:16-0:20 on video.

He killed 5 mans with aimbot and Anti-Cheat did nothing (i banned him by hand)
Playing 10 mins with a cheater is too long.

I like this AC. And it banned well and quickly on previous maps. But on this map from video the AC being a slowpoke for some reason.

https://pastebin.com/4zR6VswM

[J_Tanzanite]

Originally Posted by Le1T No, im not about the wallhack. Dont pay attantion on it. I'm talking about the moment 0:16-0:20 on video. He killed 5 mans with aimbot and Anti-Cheat did nothing (i banned him by hand) Playing 10 mins with a cheater is too long. I like this AC. And it banned well and quickly on previous maps. But on this map from video the AC being a slowpoke for some reason. https://pastebin.com/4zR6VswM

What I'm saying above is that: Some cases will take longer than others, and there isn't much I can do about that.
Not without causing issues.

Besides, this cheater may have a different config to that of other cheaters on your server which were banned faster, and there definitely isn't much I can do there.

As for the map, it could be that since the map is smaller than other maps, the 350 unit check interferes more often, leading to fewer detections.
SMAC has a 200 unit check, mine is longer because Lilac tests for more things and is drastically more sensitive.


Also, I saw your post on hlmod (assuming that is you). Setting Bhop ban length to "forever" is not recommended.
The default ban length is 1 month for a good reason.
Still, thanks for letting them know

[asdfxD]

Originally Posted by Le1T No, im not about the wallhack. Dont pay attantion on it. I'm talking about the moment 0:16-0:20 on video. He killed 5 mans with aimbot and Anti-Cheat did nothing (i banned him by hand) Playing 10 mins with a cheater is too long. I like this AC. And it banned well and quickly on previous maps. But on this map from video the AC being a slowpoke for some reason. https://pastebin.com/4zR6VswM

Maybe it's because his hack is proof against this anti cheat? I mean you guys use a "public anti cheat" where the source code is public for everyone, hack coders easy bypass this. thats why smac is dead too.

Rly... it makes no sense to use a public anti cheat lol.