I might consider programming one as well. I'll send it to velo if I ever get around to it. I'd most likely use c++ for mine. Since you can easily set it to create windows keypresses I don't really see how it would be detectable if I make some kind of +1 -1 frame offset between notes.
I might consider programming one as well. I'll send it to velo if I ever get around to it. I'd most likely use c++ for mine. Since you can easily set it to create windows keypresses I don't really see how it would be detectable if I make some kind of +1 -1 frame offset between notes.
Technically, the one I made that emulates the very statistics of my playstyle is not detectable (95% confidence interval)
If everything's hitting the same frame, it'll be easily detectable. If it's simply randomized, that can be detected too.
The only way to be undetectable is to randomize it according to your playstyle, which means doing more than just reading/hitting notes: You have to know what KIND of patterns you're hitting as well as the local arrow densities. You also have to know your own playstyle intimately (and quantitatively -- via testing).
In other words, most people here aren't smart enough to figure out how to do that -- so Velocity is more or less correct, lol.
Technically, the one I made that emulates the very statistics of my playstyle is not detectable (95% confidence interval)
If everything's hitting the same frame, it'll be easily detectable. If it's simply randomized, that can be detected too.
The only way to be undetectable is to randomize it according to your playstyle, which means doing more than just reading/hitting notes: You have to know what KIND of patterns you're hitting as well as the local arrow densities. You also have to know your own playstyle intimately (and quantitatively -- via testing).
In other words, most people here aren't smart enough to figure out how to do that -- so Velocity is more or less correct, lol.
Ya I always wondered if someone took a different approach in making a bot like this. Most bots just hit everything in the same frame so it straight up easy to tell its a bot. Randomizing it would lessen that automatic assumption a bit but still is detectable unless you sit there and can do as you said above.
Pretty much if you take this much effort to make a bot this well just to cheat a score on a flash game thats pretty sad. lol. < for serious players
A bot like this would be something neat to mess around with though.
Technically, the one I made that emulates the very statistics of my playstyle is not detectable (95% confidence interval)
If everything's hitting the same frame, it'll be easily detectable. If it's simply randomized, that can be detected too.
The only way to be undetectable is to randomize it according to your playstyle, which means doing more than just reading/hitting notes: You have to know what KIND of patterns you're hitting as well as the local arrow densities. You also have to know your own playstyle intimately (and quantitatively -- via testing).
In other words, most people here aren't smart enough to figure out how to do that -- so Velocity is more or less correct, lol.
honestly, this is what I was going to try to do, I was going to randomize the frames it hits on, but that's too obvious, to make a successful human emulator, I should, as you said match my playstyle, so right handed trill will hit horribly off frame, and jumpstreams have plenty of split jumps, etc. Gotta see how this works though, haven't started yet cause' I have shit to do but I need to have this done by April 3rd or some shit
It's getting better all the time
I used to get mad at my school (No, I can't complain)
The teachers who taught me weren't cool (No, I can't complain)
You're holding me down (Oh Oh)
Turning me round (Oh Oh)
Filling me up with your rules (Oooh)
Comment