Aimbotter

liqvid

DARKLY Regular
Hello all.
While playing about an hour ago I came across a player who seemed to be cheating. He was using a Scout primarily and would nail headshots on players whom he was not aiming at. When I noticed the behavior, I switched to spectator and watched closely. Eventually, I was kicked from the server for being a spectator for too long (annoyingly), but I believe I gathered enough evidence to convince a TD admin to do the honors.

After I got kicked, I watched the demo again. This time, slowing the playback as much as possible whenever I knew the player was about to take a shot. With 1 exception, every time he took a shot with the scout while his target's head was visible, the bullet would fly off center (in some cases, extremely) and hit the target in the head. The one exception I mention is when the player actually got a headshot that he earned (crosshairs were actually on the target's head).

I then proceeded to do what I usually do in this situation, and paid a visit to the stats page to search for and find the player's page. However, after spending 25 minutes searching, I was unable to find this player. I'm now kicking myself for not doing a "status" command in the console while I recorded the demo.
I did find several players that went by the name 0 (thats the number by the way), but none of them fit the bill. All but one had an activity rating of less then 100%, and the only person who was active enough to be the culprit hasn't used that name in over 3 days.

Regardless, I'd like someone to take a look at the demo for verification purposes. Maybe one of you will know how to find this player's information. Possibly, by going though the server's logs? I don't know...

Instead of PMing the demo to an admin like I was doing previously, I'll just post the link right here for all interested to see. Let me know if you agree that this player is hacking, and if you have any idea how to figure out who the player is. Thanks.

Here's the link:
http://cid-3c305802ad40c64b.skydrive.live.com/self.aspx/.Public/Demos/0 Aimbot.7z

Note: Use Shift+F2 to bring up the playback window for demos. You can slowdown or speed up the demo's playback with this window. Unfortunately, you can't rewind, so instead, use the GoTo command to jump to a certain Tick number. Press the GoTo button once to rewind the demo to the beginning, and then again to jump to the intended tick number.
 

r3volution

TD Admin
I might be wrong, but I don't think aimbot changes the flight of the bullet (I'm not even sure if CSS even models a bullet). Unless you see his/her view snap to the enemy head when it was originally out of view, then I doubt you can pin him/her down for hacking.
 

Remy

TD Member / Gay Gyoza
[quote1272142910=Liqvid]
He was using a Scout primarily and would nail headshots on players whom he was not aiming at.
[/quote1272142910]

Its called fucking retarded hitbox. Youtube.com solves all problems.

edit: theres word this was actually fixed...hmm time to test...
 

OG buckshot jr

TD Admin
Liquiv, I appreciate your concern, however I don't feel it necessary for you to keep doing us a "favour".

We have a lot of admins that can take care of the job.
 

Remy

TD Member / Gay Gyoza
Not to say your opinion isnt wanted Liqvid but theres usually an admin on. if not and you notice something feel free to bring it up. Thank you.
 

liqvid

DARKLY Regular
[quote1272231155=Pyro]
looked fine to me... just good aim.
[/quote1272231155]

I'm not sure if you're seeing something that I'm not while watching the demo, I would appreciate your insight. However if you only watched the demo at 100% speed then I'd urge you to watch it again, but much slower. In fact, watch it tick by tick using the > button. What I hope you'll see is that the player actually does snap toward the head of the victims. Note that when you view ticks individually, all the models seem to move ahead (into the future). This is due to the compensation of lag and whatnot that leads to the hitbox problem pointed out by Remy.

When viewing the demo as a video, the models move to where the effective hitbox is. This also occurs while playing online. When viewing individual ticks in the demo, the models do not move since you are only looking at one frame and no interpolation can occur.

Anyways, the point is, when viewing individual ticks, expect shots to register as hits if they go through a point that the model passed through a few ticks ago. I hope that makes sense.

With this in mind, view the following ticks
(41552 - 41561)
(47107 - 47111)
(48130 - 48136)
(48401 - 48411)
(50943 - till death of victim)

It is fairly easy to see the quick snapping that is occurring when viewing tick by tick. All the snaps take place within the span of 10 ticks which, on a 66 tick server is around 0.15 seconds, which is twice as fast as you can blink your eye...
Also note that during the deagle sections the snaps do not put the crosshair directly onto the victim. Since the player is moving around at the time, automatic accuracy compensation occurs and the crosshairs may move away from the target in order to land a hit. This is standard procedure for aimbots in CS:S

After doing some research, once can find out about the existence of "silent aim" aimbot software, who's design goal is to give very little visable evidence of their existance. The snapping effect here is being hidden by the smoothing interpolation of player's movements that is done by the source engine. By having the snap move to the target location, and then back to where it was originally in a very quick time peroid, the interpolation glosses over the movement, making it very hard to see.

Now, I realize that the demo only gives a short glimpse of this players activity. Be it ligit or not, it is too short of a glimpse to make any real judgement.
For example, if the player were cheating then one would expect the shot taken at 45226 to have landed. Since it did not, maybe the player is legit.
All I'm asking is that your admins give it a serious look over and come up with your own, defendable opinions, instead of simply saying that "it looked fine" or "thanks, but no thanks". I'm putting the effort in, and I'd appreciate if you would too. That is all.
 

liqvid

DARKLY Regular
[quote1272259789=cranswick]
dot on the monitor and good aim
[/quote1272259789]

Why would putting a dot on his/her monitor have helped? None of the kills were no-scoped scout kills.
 

Leroy

2012 Troll of the Year
LiqVid, appreciate your breakdown, we'll give this one a fair amount of attention
 

Pyro

TD Admin
[quote1272265070=Liqvid]

All I'm asking is that your admins give it a serious look over and come up with your own, defendable opinions, instead of simply saying that "it looked fine" or "thanks, but no thanks". I'm putting the effort in, and I'd appreciate if you would too. That is all.
[/quote1272265070]
the reason that i simply said that it 'looked fine' is because:
-he aims, hits, and misses just like any good player
-i've personally gotten similar scouting kills and many of our more dedicated scouting players get several kills like this in single rounds. you zoom in quickly, swing the rifle by their head and click. remarkably, you get a headshot quite often, especially when you practice alot.

Even if I found anything in your demo suspicious:
-your demo is short, and player-side. this means that the sample is not long enough and it is inaccurate to begin with
-due to natural lag, your computer's demo doesn't see what 'actually' happens on the server; also
-the guy you are accusing, for whatever reason, doesn't appear on the stats page and you didn't get his steamID in the first place.

The gist:
-i understand that you are trying to help the server; but
-the best aimbots are extremely hard to catch because of their smooth pathing. it takes shit loads more to prove than one short demo; so
-the lack of concrete evidence prevents me from banning; BECAUSE he

-looked fine to me... just good aim.

Although my opinion is not the be all and the end all, of course.
 

r3volution

TD Admin
Snapping to a person's head while they are in view is just good wrist flicking. Even if the guy turned a full 180 and still managed a headshot it still doesn't mean they are aimbotting. Now, if this happens every single time, and his crosshairs follow a player that isn't visible, then we might have a problem. Counter-strike source has a lot of problems with faulty and lagging hitboxes - issues that are fairly well known, especially to players from 1.6.

It's really cool what you're doing and how concerned you are for our community. We have a lot of good admins, and I'm sure they have watched your replay and taken it into consideration.
 

liqvid

DARKLY Regular
Well, regardless of what you think of the previous demo, I have just finished recording another one today. This guys is definitely aimbotting. There is no question. While in game I messaged the admins, but apparently there were none. I informed the opposing team about this player as well.

The players stats page:
http://www.torontodarkly.ca/hlstatsx/hlstats.php?mode=playerinfo&player=20048

Note his 1.0 Headshot per kill rate. This guy wasn't trying to be subtle.

Demo:
http://cid-3c305802ad40c64b.skydrive.live.com/self.aspx/.Public/Demos/fallentroll.7z

Edit:
I see he was taken care of. Thanks admins.
 
Top