PDA

View Full Version : tRandomNumbers(); tRandom();



HKbotz
02-21-2015, 01:15 PM
Went ahead and made these because I was curious how much of a difference a truly random number would make, so far mostly in antiban, instead of a pseudo random number.
These get random numbers from https://www.random.org/. I recommend getting a big list of them all if you're using them for antiban or something where you'd want a lot, since it takes a second or two to get the page.

// intArr := tRandomNumbers(1, 100, 10);
function tRandomNumbers(min, max, amount: integer): TIntegerArray;
var
s, ts: string;
c, i, n: integer;
begin
if (min >= max) then
begin
writeln('Error: Min must be less than max.');
exit();
end;
if (amount <= 0) then
begin
writeln('Error: Amount must be greater than 1.');
exit();
end;
//had to use a proxy since getPage() doesn't do https
s := getPage('aproxy.info/browse.php?u=https%3A%2F%2Fwww.random.org%2Fintege rs%2F%3Fnum%3D' + intToStr(amount) + '%26min%3D' + intToStr(min) + '%26max%3D' + intToStr(max) + '%26col%3D1%26base%3D10%26format%3Dplain%26rnd%3Dn ew&b=4&f=norefer');
setLength(result, 1);
c := 1;
n := 0;
for (i := 1) to (length(s)) do
if (execRegExpr('^[0-9]', s[i])) then //if it's a number
begin
setLength(ts, c);
ts[c] := s[i];
inc(c);
end else //if it's the space between numbers
begin
setLength(result, n + 1);
result[n] := strToint(ts);
inc(n);
c := 1;
end;
end;

// int := tRandom(1, 100);
function tRandom(min, max: integer): integer;
begin
result := tRandomNumbers(min, max, 1)[0];
end;

// int := tRandom(100);
function tRandom(n: integer): integer; overload;
begin
result := tRandom(0, n - 1); //works like traditional random()
end;

the bank
02-23-2015, 12:57 AM
This scares me simply because of the fact it isn't using a secured network.

Its not difficult at all for Jagex to monitor traffic especially with the native libraries they already have running on their system. Not to mention I'm paranoid as is with everything going on in terms of domestic spying as of late.

Check out my native plugin to accomplish the same thing (and more, plus secured network traffic) here (https://villavu.com/forum/showthread.php?t=112007).

Olly
02-23-2015, 03:23 AM
This scares me simply because of the fact it isn't using a secured network.

Its not difficult at all for Jagex to monitor traffic especially with the native libraries they already have running on their system. Not to mention I'm paranoid as is with everything going on in terms of domestic spying as of late.

?? You think they could legally track everyone's internet traffic? That doesn't seem right...

But they wouldn't look for someone accessing random.org through Simba. They would look at the 99.8% of people who send a nice 'hello, I am using Simba' as SMART spawns. :p

HKbotz
02-23-2015, 09:21 AM
This scares me simply because of the fact it isn't using a secured network.

Its not difficult at all for Jagex to monitor traffic especially with the native libraries they already have running on their system. Not to mention I'm paranoid as is with everything going on in terms of domestic spying as of late.

Check out my native plugin to accomplish the same thing (and more, plus secured network traffic) here (https://villavu.com/forum/showthread.php?t=112007).

I wouldn't worry about it too much since Simba is independent of RS and SMART, and like Olly said, I doubt they care if people access random.org anyway. Nice plugin though, I'll look at it more closely a bit later when I have some more free time :)

cosmasjdz
02-23-2015, 09:41 AM
Basicly i am confused or i think the wrong way. So far my opinion is and with few tests done even 100 times generated random is like 90+% certainty of being generated by random(compared to real gameplay), theorically true random deviation should be ideally the same, unless jagex catches repeating sequence. But while some people write like random(100,150+(random(15,30))) the pattern (after it starts repeating) will be very wavy(technicly a waving wave:D) and unnatural(again to start repeating it may take like from weeks to months depending on how often it generates it). So why jagex should bother by quessing which number is going next by using pseudo random generator if they for like 90+ % certainty can tell you using either random or pseudo random in 100 times used (random in most cases arent human like)? How long would jagex take quessing next random number you gona get and how intense it could be to their standart data comparison their resources wide. I quess they need to know which random generator function you use and it could take like 20-25 times in a row to 90+ % certaity tell you use this, not really worth it. And prolly they doesnt look for unnatural random they indeed look for any random which again in most cases is unnatural.

Still gauss method of times deviation is few lightyears ahead of random numbers. Very very close to actual gameplay. I mean by times distribution its nearly perfect. But during long time if they count pattern longetivity over long periods it could give up you botting. Adding some little biometric stuff to change could be nearly unrecognizable as botting without using very complex analysis and still then jagex dont want much false bans and wont ban you even for 99% certainty they will to like 99.95% or so.

Also what gets you banned is doing something or anything for 100% certainty or one thing after another 100% certainty or antiban only in one place of mainloop. Doing large amount of cicking,actions per hour ist very fast for jagex to realize you may be botting. Had lost before legitly played account, on brothers pc, new ip, i mena no connections with botting or anything, runned browser and i think the hook was papaya bot Always went clockwise checking spots as counterclockwise case was out of range by my mistake. Got banned as fast as 3rd day botting 4-6h a day. Had also few banwaves i believe cause of that. Now making some very ambitious bots for higher proffit and dont want lose accounts so my idea is not make anything 100 percent after anything. I mean real examples how could you spot bots, they tend to prefer for example one tree against the other All the time and all those little stuff adds up. Also wait times in gauss proved to be god and changing those every few days is advisable to prevent same patterns over prolonged periods of time. Test accounts surviving months now so far so good :)

HKbotz
02-23-2015, 04:22 PM
Basicly i am confused or i think the wrong way. So far my opinion is and with few tests done even 100 times generated random is like 90+% certainty of being generated by random(compared to real gameplay), theorically true random deviation should be ideally the same, unless jagex catches repeating sequence. But while some people write like random(100,150+(random(15,30))) the pattern (after it starts repeating) will be very wavy(technicly a waving wave:D) and unnatural(again to start repeating it may take like from weeks to months depending on how often it generates it). So why jagex should bother by quessing which number is going next by using pseudo random generator if they for like 90+ % certainty can tell you using either random or pseudo random in 100 times used (random in most cases arent human like)? How long would jagex take quessing next random number you gona get and how intense it could be to their standart data comparison their resources wide. I quess they need to know which random generator function you use and it could take like 20-25 times in a row to 90+ % certaity tell you use this, not really worth it. And prolly they doesnt look for unnatural random they indeed look for any random which again in most cases is unnatural.

Still gauss method of times deviation is few lightyears ahead of random numbers. Very very close to actual gameplay. I mean by times distribution its nearly perfect. But during long time if they count pattern longetivity over long periods it could give up you botting. Adding some little biometric stuff to change could be nearly unrecognizable as botting without using very complex analysis and still then jagex dont want much false bans and wont ban you even for 99% certainty they will to like 99.95% or so.

Also what gets you banned is doing something or anything for 100% certainty or one thing after another 100% certainty or antiban only in one place of mainloop. Doing large amount of cicking,actions per hour ist very fast for jagex to realize you may be botting. Had lost before legitly played account, on brothers pc, new ip, i mena no connections with botting or anything, runned browser and i think the hook was papaya bot Always went clockwise checking spots as counterclockwise case was out of range by my mistake. Got banned as fast as 3rd day botting 4-6h a day. Had also few banwaves i believe cause of that. Now making some very ambitious bots for higher proffit and dont want lose accounts so my idea is not make anything 100 percent after anything. I mean real examples how could you spot bots, they tend to prefer for example one tree against the other All the time and all those little stuff adds up. Also wait times in gauss proved to be god and changing those every few days is advisable to prevent same patterns over prolonged periods of time. Test accounts surviving months now so far so good :)

I made this mostly for my own curiosity to see if I noticed a significant difference in true vs pseudo random(I haven't really :p). I also use gauss in most cases for randomness, like mouse moving or wait times or whatever, but real randomness can be useful in certain instances, so that's mostly what it was for. I agree that it's highly unlikely jagex tries to predict randomness, rather they probably just watch for patterns that have a mostly even distribution to them.

Like I said it was more out of curiosity and I thought I'd share it in case anyone else was intrigued by it as well :)

So far I haven't received any bans on any accounts(5 of them since last September). Then again I do my best to use antiban and I never bot for more than a few hours at a time on each account, and I think time is probably the most significant factor in bans. When I first started writing scripts I didn't really pay attention to randomness or antiban, so they were very bot-like, I just didn't run them very long, and I'm assuming that's why I've been able to dodge the ban hammer, at least for now lol.