Ahahah! Oh, wow.

Raw doggin the people who cry hax.

If i understood correctly,

You mean you have an application that scans a demo and can detect use of aimbot?

Yes. Kind of can)

"Kind of" sums that up for me atm really.

You've claimed you've been working on something like this for some time, yet have shown no actual sign of competence in the field.

I'm well aware that that can just be your in game manner coupled with English not being your first language, but either way, If you can actually demonstrate an actual program that works at all, let alone with any actual level of accuracy, I will be impressed.

But until that time, I'm not going to be positive about what I can only conclude is a mechanical turk.

I think you are last person whose opinion I want to know.
I already tasted your cooperation and I think this is enough for me.

If you can please not comment on my posts and threads, then it will be just great.

I've never been asked to cooperate with you that I can currently recall, had any reason to cooperate with you, or any wish to do so, all you've ever done is spam offensive cheat accusations at me even back when I tried to help you understand that playing vs 20ms players on 80ms was giving you a distorted view.

I can't fathom your stupidity in pushing things so far, despite warnings, that you got yourself permanently banned. Nor why you would continue to circle around a community based around the thing you were permanently banned from instead of taking it as a sign and moving on in your life.

If you've actually produced anything worth praise, then I will praise it on its own merits, similarly I will not hold back in offering my opinion if I believe it to be either flawed or some kind of con.
I think your abusive in game attitude stinks, but that doesn't mean you might not be able to produce software, however until that is actually seen it is pure vapourware.

It does actually concern me greatly that even if it is capable of processing demos, it will be based on the subjective opinion of what Pall thinks is humanly possible, which I can't help but feel from your tirades in game is going to be heavily biased towards if Pall can't do it then no-one can. Instead of actual inhuman behaviour, and as humanized aimbot code is almost certainly going to improve then that distinction is going to be incredibly hard to recognise as it is.

If it's real, and it doesn't suffer from these problems then I'll be very happy that someone has managed to help improve things for everyone in this way.

If you don't want the public commenting on your threads, then don't post them on public forums.

Next he'll cry anti-semite.

I wonder if the clients viewangle granularity could be used to determine if aim is being assisted. It's a fairly safe assumption that most players are not using a low enough sensitivity to make 1 mouse count translate to the smallest angle change possible (I believe the sensitivity value is ~0.25 with 0.022 yaw). If a player always makes angle changes larger than the minimum of 0.005 degrees in smooth strafe jumping motion, but starts going a lot lower during LG tracking, it could be suspicious.

Old QW players may remember the Qizmo proxy 'f_sens' command, which would display the sensitivity value of all players in the server. I believe this calculation was based on the smallest player angle change that was received, and the assumptions that everyone makes a 1 count mouse motion at some point and keeps a constant sensitivity value.

However, low sensitivity combined with +zoom, higher m_filter values and scripts that reduce sens for certain weapons or only during firing would throw this method of detection off. Also, bots that inject themselves as mouse movements and don't directly influence the viewangle would also avoid detection.

Just wanted to throw some thoughts out there!

Thanks!

Well, currently I have features vector of length 4. So, every demo takes single place in 4 dimensions search space.
One of these features is built up on some modification of your idea of small angles.
I have in plan additional 20+ features, but check of most of them has really heavy power consumption (I see it even without implementation). Some algorithms work good, but inventors of these algorithms don't care about money, lol. They are in the sky!

And sure, aimbot user can pass some features check, but it will fail on others.

Currently I can not read player's config from demo file, but I don't think its a big problem to change game code and add couple of values to demo file. This information will allow to run much cheaper and precise analysis.
For now I just work with what I get from demo file - angles, times, some special events. That's all.

Bots that inject themselves as mouse movements can not avoid detection if these movements are done not in human manner.

For example human can correct mouse aim only after some time passed after sudden change (not predictable) of target movement.
So, as example of feature, if you analyze whole demo for this, then you should get percentage of guesses under 70 (on, lets say 50 cases). And if you got this percentage above? Good sign for cheat - one of your features says its aimbot.
But reality for such feature - not much frags are done in each demo, so you can not calculate such percentage precisely. So, finally this feature can not be used as I defined it.

That sounds a lot like you're running the risk of instantly penalizing people that use different fovs on LG, or the slightly less common different fov when pressing fire while holding LG though (or simply manually +zoom, which could also be used backwards to give high fov for rarer very close range battles)

I doubt there is a blanket solution.

I have tried to persuade syncerror to sacrifice some packet space and put the players current FOV in so spectators/demo watchers can by default get an absolute view of what the player is doing (unless they set a cvar to only ever use their own settings), but to no avail yet.

If it's just fov on/off it's a single bit, but sending fov values uses space they could want for other things.

By "kind of" I mean that if person puts 1 slightly aimbotted rail in 10 minutes game, then its hard to detect such a thing and it won't be detected (kind of failure for my solution).
Having 10 such demos will give more information and maybe it will work better. But currently I work with single demo of FFA and duel aimbot users only. They do a lot of aimbotted frags that give enough information for decision maker.
Besides that, currently whole analysis if pretty much power consuming and check of even single demo can cost some real money. So, for now its additional reason to say "kind of".

10/10 meh

Can you please comment in English?
I have some difficulties with understanding what is written )

meh

Ha, I don't spend enough time in internet to learn such things!
Thanks.

Smells like buying cat in a bag, but its not

I can't deny that this is what it smells like.

OK)

Costs money to scan through a demo? lol what kind of bloatware are you writing?

Read about bitcoin.
There everything is open to everyone.
They also explain why they need some cpu.
I am not bitcoin, sorry.

build it as an ethereum app. :)

Think of such service.

Site, where you can creat threads about the game (q1, q2, q3, ql, cs go or others) and get oppinions if there is any exists.
No check for grammar and other stuff. Writing text only.

Site will have public stats created for all threads with all oppinions. Search over whole data base (name of user, dates, words, etc).

If everything will go well enough, then at launch site will have all threads from sites esr (or some other thread base) analyzed, so user will not start with some empty content.

No technology explanation (how decision is made) will be provided in public access. Smells like buying cat in a bag, but its not.

Are you ready to pay for such service?
How much can you pay?

For example 0.1$ per 10 words in thread.
Maybe 10 tokens for 10 threads for 1$.

What do you think.

Nice edit, bro. You have great talent!
If you can rewrite my post using good English, then it will be even better.

you cant bait me with your jewish tricks

I just did, comrad.

Are you ready to pay for such service?

You're not going to like the answer

I don't think that I really liked any answer in this thread.
So, shoot, don't be shy.

Nothing \:D/

payPALL

If it's any good - just sell it to id software.
At the moment I'm not even sure there is a cat in that bag of yours.

They will not buy it, since this is not something that will bring them money.

PALLPOWER

If you expect to make money out of something like this, you'll need the cooperation of the game developers. And to get their attention, you have to show some results.

As a player, I'll never bother using the service. Don't need a program to tell me I got owned by a hacker, my ego isn't that fragile. Also not too interested to play internet police. I rather play the game itself.

Such kind of cooperation is not that will bring me any good.
And if I already spent months playing this game, then I want some more from it besides only fun.
But I see your point.

I really doubt that you will not let yourself to visit such service site, too look on ads, to run some searches around nicknames from game, looking at demos and decisions. It is the same as to say that you don't know what qlranks is. But anything is possible, sure. Other question about paying for your personal demos to be checked.

No police here)
I have cool ideas on how to build permanent ban system that will work in 100% cases (per individual player - any config, any computer, just by game style). Such ideas already implemented (but in other field then computer games). But I don't want to ban cheaters, since its too cruel. This is also wanted by game developers - keep ALL players in game and not lose anyone. I want them return, but with better cheat, that will pass all exams. And if I will build correct exams system, then to pass these exams this new cheat must look really good in game, less annoying, more human, more fun for all.

Target of such service is to earn some money and set new horizons for cheat (aimbot) developers, force cheats to be more sophisticated, make them more human. With my help or not this is what will happen in the future of fps games.
And such player like you (plays for fun, no care of ranks or other stuff) will have more fun without even being aware of whole process behind the scene.

btw cs go have builded in overwatch system for free

Its not automated, comrad.
There legit players watch demos with eyes and try to detect cheat.
And its very hard to do properly, since many stuff that can be
extracted from demo is not visible for eye.

Think of sound signal when carried on two separate frequencies:
one frequency for noise part of signal (very strong and high) and other frequency for desirable part of signal (very weak and low).
All together (for human ear) it sound like just noise.
But if two these frequencies are at good distance one from other, then you can filter out all the noise and stay with only desirable part of signal.
Sounds like cat in the bag, but its not)

csgo overwatch its just filter for real investigation and its automated enough

I know that automated part there is is just submitting demos and retrieving them for watching by small group of players (judges?) who put their verdicts on each demo.
I am not aware of any automated processing of demo for cheat detection.

Correct me if I am wrong.

its works the same as your suggest. one guy reporting (uploading demo) another guy making verdicts.
you do not come up with any new things but asking for money. such impudence

I know that what I offer is something new, that is not implemented in any game yet. And this is also pretty original, required and still requires a lot of hours of work. Why do you expect to get this for free? Let me know if this is still about jews and stuff, I will stop replying to you.

What is real "impudence" is to run game without anti cheat at all.
You better look there.

At first quake is dead and its too late for any addings. other games wont your shit.
At second i am so much sure you cant show better performace than vac/be/punkbuster or any other anticheat system.
and at last you are just jew which looking for some earnings

he is suggesting a program that automatically scans the demo and makes a verdict, not a player

do you belive that its possible ?
he talk something about sounds signals... this is bullshit. programm cant judge. there alot of options how ppl make decision. And programm cant be so smart. His shit will be so usefull as pankbuster which will ban you if you hit some random spam.
So, he just kid, leave him alone.

I think it is possible to detect the less sophisticated cheats with a program. As a very simple example, imagine that a new player spawns 90° from a suspect. Then the suspect proceeds to move its cursor towards the enemy at exactly 2 degrees per frame consistently until the target is aimed at then suddenly stop at the right time. Too robotic, isn't it? You can pick something else instead of a linear function if you want (like perhaps a cursor speed proportional to the square of the distance), the pattern can still be detected and if it's respected too well for a human being, this is a sign.
All of that could indeed be automatic.

Yes, cheats will adapt to this (those who haven't already anyway) by randomizing their input, using additional delays, more complex functions, etc. At this point I guess you could still fight by imagining the base aimbot function (without randomization) most fit for the general cursor pattern you see, calculate what the differences are between the expected and real results and try to determine if this serie was obtained from a pseudo-random algorithm. Well anyway as soon as you get randomization in this, it becomes a cryptoanalysis challenge. It also doesn't help that the player himself is gonna add his own entropy in the system.

Anyway, yes, it could be possible with enough work to get a decent program to detect dumb cheats. Against randomized cheats it just seems like too much work to see the light of the day, at least for QL. In any case, that doesn't tell if OP can do any of that and he has sown no result whatsoever so far so I wouldn't get my hopes up.

yeah i know less sophisticated cheats detector is possible to make. but its non sense to have it for fee.
and yes cheats will adapt and anticheat will always stay outdated as vac.

anyway this all doest matter because he offer offline anticheat which just incredible useless and he asking money for this which incredible jewish

i mean if i will have choice between fee for qlranks or fee for this simple-anticheat i will choose qlranks.

because anti cheat should not let you play if you using cheats. and not like caught you ... maybe ... somewhen.

I would say without hesitation that such a program is possible; flashsoul gave some very good examples.

Whether this guy has created such a program I cannot say...but I doubt it somewhat.

Smells like buying cat in a bag, but its not.

That's something a "cat-in-a-bag" seller would say.

Thread title: Automated Overwatch service?
last post: "Its not automated, comrad."

Yep, I'm totally giving you money for this.

I was talking about Overwatch in CS GO, its not automated there.
People watch demos and put their decisions, not software.

Oh ok.
I am convinced, where do I put the money for a cryptic biased banning system?

)
I promise to tell you as soon as I finish with it!

)
read again

What? I say that as soon as my biased system will be ready, I will let you know where to pay!

Look up the word biased :p

He's saying that you're not trustworthy enough to have the power to choose who is banned and who isn't.

Set it up people will donate if its good and useful

Already exists in CSGO and other FPS games are too irrelevant for anyone to care.

It doesn't exist in Cheater Strike.

im not paying to get cheaters banned.

but yes, i do want overwatch for QL with trusted members of the community. there are probably as many cheaters running round QL as they do in CSGO.

:DDDDDDDDD it all began here folks

haha, a bit longer ago actually...

))))))))))

ouch gg

was going to make a serious comment on this thread, but this says it all.

thanks!

Nice comments guys.
Now I understand on my own skin why ID developers don't ask "community" before doing anything.
But I gave you a chance.
1200+ views tells something.

Any way, thanks for all fair comments.

Welcome to the internet!

You just crossed the Rull Fetard Line.

Good jab.

Such Service
Wow

Anyone who has played with pall will already know where this is headed. This service will 'detect' that everyone is cheating, followed by a statement about how stupid cheaters suck.

Sounds like a winner.

they already have this type of service in CS:GO

high level players get access and can review demo's and make their decisions..

I've recently gotten access to be an overwatch viewer and if I don't like you I vote hax