Possible Auto-aim

[ReCycled]

Well a previous suspect showed up Sunday nite so I thought I would demo him. He was winning every level against some good players (eg Strychnine) and was heavily reliant on Rail. Several players complained about him during the match. He goes by the name of Lanky and I've posted him before with the consensus being he's legit. I recorded it on my laptop which only has Quake 3.20 so the demo is 2 levels combined into one. It didn't auto-stop at the end of the first match but kept recording. I've zipped it to shrink the file size.

As a secondary question - when I was watching the match "live" I had 6 or 7 places where I lost 1 or 2 seconds of action due to packet loss. But when I playback the demo, it's perfect. No gliches anywhere. Very odd. I have this same problem with my desktop where my "70 ping" goes to hell but demos are good. I don't get it.

     

[[BTF]Defiant!]

Any particular part of the demo to review?  First couple minutes did not appear cheatworthy.

[|iR|Focalor]

"Heavily reliant" is an understatement. Might as well play instagib.

Looks pretty kosher to me. Nothing consistent really pops out that I see. If he is using something, he's got the FOV on it really really low. Doesn't look like it though.

[Sgt. Dick]

I asked him the other day if he was the same person I'd remembered from Enter.net days and he said yes.  I remember Lanky as being a very tough rival back then.  So he has been cheating for years (seriously doubt) or he is just that good.

[PANTONE 7717C]

I haven't seen him in a while but have played him on occasion here and quite a few times in Lithium Mods. I've never suspected him of botting. Not really an allround player but fragcount-wise he can be hard to beat if there aren't enough decent players on to slow him down. He loiters in one central part of the map(s) and just waits for others to come in. Even on big maps when there is just a handful of players he won't really go look for the action. That's how those guys play, smart and effective without working up a sweat.
You know he'll still be somewhere in the same area with his Rail the soon as you come back, so don't come flying in unprepared with a Chain and 50 bullets... get some armor (shards, stimpacks etc) and at least one powerful weapon. Or just leave him there and avoid the area and focus on easier frags around you to keep up in score, if you care for that sort of thing:D

[playboy]

 

[ReCycled]

It's funny how what you see live and what the demo shows - it's like 2 different versions of the same match. From the POV of the shooter, the demo is all smooth, plenty of time to assess the shot and aim and all very logical. This is nothing like what I saw "live". What I saw was snap, instant shots within a millisecond of seeing the opponent, or in some cases not even seeing the target - just the aftermath of the kill. There's one instance about 3/4 thru the demo where Strych, obviously getting pissed off at getting railed so consistently, does a sneak jump attack from the side of a door opening (left to right) and still gets railed in the air. When I saw this "live" there was no way he could see him and rail/aim that fast. It almost looked like a wallhack shot. In the demo it just looks like a lucky guess.

But does anyone have some insight into my secondary question in my post? Quote:

"As a secondary question - when I was watching the match "live" I had 6 or 7 places where I lost 1 or 2 seconds of action due to packet loss. But when I playback the demo, it's perfect. No gliches anywhere. Very odd. I have this same problem with my desktop where my "70 ping" goes to hell but demos are good. I don't get it."


 

[Whirlingdervish]

any luck on getting in-eyes spectator mode on the TS FFA servers?

from 3rd person it looks sorta normal but that doesn't prove anything compared with a good in-eyes demo.

[Whirlingdervish]

I've been speccin him lately when I see him playin FFA and I havent gotten anything amazing on the demo end, but I still don't think hes 100% legit. First off, he's played this long and still uses 3.20? I've never seen a nocheat say from him, and he doesnt come back on !versions with an r1q2 or apr client.

For such an all around player, he can only seem to use hitscan weapons and the RL from less than 10 feet away to get kills, yet he keeps up 350-500 fph on dm?

I've tooled his ass on railz quite a few times, so his hitscan weps aren't all that great, and from distance he's nothing special.
His movement is quite slow and does not seem to be up to par with his short range hitscan "aim".


I dont have a solid demo to point to, but I recommend watching him use the ssg if you'd like to see some suspect shots.
If you hit any part of the bbox with the rg they're dead, but the ssg you better hit them with a shitload of pellets to make it count.

The euros may be 1hit kill ssg pros, but this guy is far from that skill level.

[console]

First off, he's played this long and still uses 3.20? I've never seen a nocheat say from him, and he doesnt come back on !versions with an r1q2 or apr client.

His version string reports "3.20 x86 Oct 16 1998 Win32 RELEASE".

I have indeed encountered numerous FFA players who have played for years and years who haven't the slightest clue that other versions besides 3.20 exist.  (And occasionally have encountered some who were aware of other client versions but who refuse to use anything but the "stock" id software client.)

When I've spec'd him I haven't noticed anything obvious.  But, there seem to be enough complaints I guess we could force AC...


Regards,

[|iR|Focalor]

That's pretty retarded. A person is has become accustomed to playing with their setup, but it's fair to make them have to re-work their config and use another client alien to them... all because a few people don't like getting their asses kicked.

Innocent until proven guilty?

[Whirlingdervish]

I wonder how he'd play with AC, but don't think it should be forced without at least some sort of oddity in a demo that we can point a finger at and be sure its not legit.

 

[splisskin]

If this is the same guy as aux_lanky (something like that) then I dont think he cheats. I remember him playing alot of railz some time ago so he does like the railgun. I seem to remember him being streaky, good games bad games  nothing unusual.
r1q2 can be a pain in the ass for the occasional player used to 3.20. I recently was forced to use    anticheat again and still get a rubberband effect (with cl_async off and anticheat, doesnt do it without anticheat) when i play. so its not for everyone when you have to mess with it to get it to work similarly to 3.20!

[console]

r1q2 can be a pain in the ass for the occasional player used to 3.20. I recently was forced to use    anticheat again and still get a rubberband effect (with cl_async off and anticheat, doesnt do it without anticheat) when i play. so its not for everyone when you have to mess with it to get it to work similarly to 3.20!

Hmm.  I'm sorry to hear about this rubberbanding thing.  For me, r1q2 with AC and cl_async 0 is essentially interchangeable with 3.20, and because of that, I have seldom looked at forcing anticheat on someone (whose OS can run it) as being much of an imposition.

Since for me, r1q2 with AC and cl_async 0 is essentially identical to the feel of 3.20, my general outlook has been that players who can run AC probably should be running AC.  If AC ran on all operating systems, most of our servers would probably force it globally.

But if many players are experiencing this 'rubberbanding' issue, or whatever, that makes r1q2+AC feel like a downgrade from 3.20 instead of a neutral transition, then, yes, I would be inclined to be more conservative about forcing AC.

But as for 3.20, I must say I have never felt too bad about requiring someone to upgrade away from that client.  I don't suppose most players know about this, but it's public knowledge that 3.20 contains buffer overrun exploits.  Any malicious Q2 server operator could exploit 3.20 to execute arbitrary code (install malware, keyloggers, make your machine part of a botnet, format your hard drive, whatever.)

So from that point of view one could argue nobody should be running 3.20 at all.  (But I suppose it is their business if they're aware of the risks and don't care.  But again, I suspect a lot of players have no idea.)


Regards,

[PANTONE 7717C]

If you finally try an upgraded Q2 client and the mouse just feels awkward then that's probably reason enough not to use that client, no matter how good it is. When I used r1q2 for the first time I also had problems with the mouse. Not really a rubberband effect but more like a floating-on-a-thin-layer-of-air effect when turning the mouse.
After a bit of playing with the settings the mouse is now identical to the feel of 3.20 so as far as I can tell, if for some reason r1q2 doesn't feel right it has to do with the mouse acceleration and input.

R1Q2 has some additional options; in the game menu go to Options > r1q2 options, and just test some different settings and combinations. To keep the changes you'll need to store them in a custom cfg file.

m_fixaccel 0/1 (mouse acceleration fix)
When enabled (1) it slows down the Windows mouse accel but you can try to compensate it with a higher 'sensitivity'. However I've noticed my mouse is more accurate with fixaccel disabled (0) and using a low sensitivity)

m_directinput 0/1/2 (0=disabled /1=buffered /2=immediate)
When enabled (1 or 2) gives a much smoother mouse response and completely fixes the issue of mouse acceleration on Windows 2000 or higher, but it can also make the mouse feel "floaty" and less accurate. Seems to default to 0 though...

I've disabled both fixaccel and directinput and lowered my mouse's sensitivity to get the desired effect, but I know that if I install any mouse software again I need different settings... so I don't know what it is exactly.