|
I don't know if this can be implemented or is an old idea, but here i go!
Cant anyone create an anti-hack that takes screenshot of the players screen at a random time every 30sec, then combines it with the replay?
Or uploads the screenshots from the game to iccup/other ladder?
Updated Idea:==========================
5.38kb big and 50% of starcrafts picture size. Sufficient to see if maphack is being used.
lets say that a game is 30 minutes long, 5kb x 60screenshots = 300kb over 30 minutes. Easily uploadable to iccup for even those with the slowest connection.
also, the replay could have a higher picture capture of lets say every 5-10 seconds. Which would allow anyone to view the entire game practicly in first person.
(if the player releases it of course)
Updated idea 2==========================
If 5kb is too big, then you could reduce it and make a constant video of the map area.
this screen is 746 BYTE big!
or a black and white one with slightly deleted edges of the map area
568BYTE
You could easily make a constant super low quality video of the map area that way, then save in the replay.
Updated idea 3==========================
On September 04 2007 15:47 Black Pearl wrote:
"I think it would be a great idea to take information about mini map, but not in screen - in other ways programming like code or sth. We have old bwchart or RepASM4, where we can see APM or listing of actions so ... why no implement info about mini-map ?"
This would be perfect if you somehow could save both players complete minimap actions thruout the game, then afterwards save it and make it viewable.
It would make hackers extremely obvious.
|
Random time every 30 seconds, huh?
Regardless, this isn't a bad idea. It's not foolproof, but it would make hacking harder.
|
Anti-Hack idea dont hack .
|
51548 Posts
Sound's like Octinium
It's used alot in CS Online Competitions (CEVO-P, CAL-I etc) to stop hackers.
|
|
|
I'm not expert on BW programming but I don't see a single reason why it should be hard to implement.
|
|
|
Hong Kong20321 Posts
whos gonna keep track of all the screenshots being uploaded all the time to iccup?!
|
If uploaded w/a replay, a crafty program could probably see if something's fishy.
It could definitely work, but it'd only be a matter of time before the hack writers found a way around it.
|
On September 03 2007 20:11 SonuvBob wrote:
If uploaded w/a replay, a crafty program could probably see if something's fishy.
It could definitely work, but it'd only be a matter of time before the hack writers found a way around it.
There will obviously be a way to stop the screenshots from being taken.
But turning the maphack off while screenshots are being taken is probably harder, after all.. if there are no screenshots its rather suspicious...
And I guess you could make the program take screenshots every second and then randomly pick two of the 60 screenshots per minute to be added to the replay/uploaded.
|
making it randomly during the game would make it as safer and would take less kb. good idea
|
"
There will obviously be a way to stop the screenshots from being taken.
But turning the maphack off while screenshots are being taken is probably harder, after all.. if there are no screenshots its rather suspicious..."
yes
|
Harder, but in all likelihood possible. Hack/anti-hack is always an ongoing battle.
|
good idea, but
1) It might cause lag in game.
2) creating a server that accepts jpg uploads from unidentified IP addresses is a security risk. It's like saying pls DDoS me tnx.
3) it would be hard to gauge how Much bandwidth it would take
4) 5.38kb might not sound like a lot, but every 30 seconds? That's about 150kb per game. Times how many games a day? Couple thousand give or take. The database would only be able to be cleared a few times a week to be reliable for catching hackers, so it would get pretty bulky.
|
Ummm easy to hack. Here is how:
1) Make a hack that turns off every 30 seconds
--- Then you counter by one that uses a randomizer to take them on average every 30 seconds.
2) Then they make one that turns off in synch with your randomizer.
See?
|
I know of tourneys who demanded screenshots to be taken from time to time in game. Like beginning, mid & end or when an obsing turney admin said so.
Apparently you can (or atleast could) se if a player hacked even if maphack is turned off. I know of atleast two cases were people got caught with this method. One for sending in screens with complaints iirc.
However I dont know if it is a way to bypass this. Does anyone who have tried a hack knows if this method still works?
|
why dont blizzard just make their own antihacks... thyey are so stupid to make their own antihacks? are there any games out there that is unhackable? blizzard nver learn...frist starcraft then warcraft 3..they r all hackable.. i hope they dont repeat the same mistake with SC2
|
On September 03 2007 21:39 pooper-scooper wrote:
Ummm easy to hack. Here is how:
1) Make a hack that turns off every 30 seconds
--- Then you counter by one that uses a randomizer to take them on average every 30 seconds.
2) Then they make one that turns off in synch with your randomizer.
See?
But cant you see on the map if they have used maphack?
But if you do manage to make fog of war reset with your cheat, then i'm still not entirely sure you could just beat a randomizer and in time turn off the maphack in synch with the screenshots.
maybe you could, but i still think there is potential in the idea
|
|
|
8751 Posts
i like idea 2. just the mini-map. take one on average 3-5 seconds, but randomize the lengths in between. it'll still be well under .5MB for a 30 minute game.
|
On September 03 2007 22:25 NonY[rC] wrote:
i like idea 2. just the mini-map. take one on average 3-5 seconds, but randomize the lengths in between. it'll still be well under .5MB for a 30 minute game.
yes, we really need a programmer to comment on the idea however.
Anyone know someone?
|
8751 Posts
there are programmers that visit this forum. just be patient and they will comment eventually
|
I like the idea 2 a lot.
It is not that hard at all to implement. The only problem that I see about that is not related directly to starcraft. Since the screenshots come from an user's computer, that means that a user will potentially, be able to send any jpg files to BNET. As someone stated, it can be a pain to deal for the bnet server (since they can put viruses into a jpg file and so on). Or even sending virus to the opponent who dl the replay. It is always a bad concept to let some files coming from from any players to the servers that handle the game. The less the player have to send, the better it is networkig and security speaking.
So for sure it is a good idea but it would need some fixes on BNET to handle that correctly.
On the other hand you can imagine a third party anti hack program developped by blizzard, but this kind of tool need a lot of patches and fixes all the time. So I don't know if blizzard wants to consume time and money every day on this tool.
|
GrandInquisitor
New York City13113 Posts
look, let's not bitch and complain
this won't catch every single maphacker out there each and every time
but it's probably good enough to catch a lot of them or at least deter them if they know that random screenshots are being taken of them
|
The good thing about the map-area screenshot is that even if they somehow manage to remove so the maphack does not appear in the map area.. Youll still see if they are moving on areas where their opponents units are at.
I think this makes it perfect
|
Somebody mentioned that it would be too much data on the server eventually.
I suggest the screenshots are kept for max 2 day and then deleted if nobody makes a complaint related to the game the pictures are from.
|
On September 03 2007 23:57 niteReloaded wrote:
Somebody mentioned that it would be too much data on the server eventually.
I suggest the screenshots are kept for max 2 day and then deleted if nobody makes a complaint related to the game the pictures are from.
Well of course files should be deleted in maybe 6 hours if no complaint is made.
If 20 screens are taken per minute, thats about 10kb per minute.
10 minutes= 100kb
30 minutes = 300kb and so on..
So if 6000 games get played in 6 hours(not likely) and the average game time is 15 minutes then it's 900mb of data. That's not much.
|
Hell I could make it, not very difficult to do. But anything that is done, will be shortly exploited by one of your friends at valhala or bwhacks.com.
|
On September 04 2007 09:44 alphablend wrote:
Hell I could make it, not very difficult to do. But anything that is done, will be shortly exploited by one of your friends at valhala or bwhacks.com.
exploited in what manner tho?
I mean if a screen was taken at random intervals every 2-3 seconds, then you could still easily spot strange behaviour.. if they somehow managed to swap the maphack off all the time.
|
All someone would have to do is figure out how to stop the program from taking a screen shot and replacing it with other random photos from a different game. Then it comes down to people taking the time to look at every picture to make sure that what was uploaded is right.
Anyone willing to volunteer for that job?
|
On September 04 2007 12:20 bErAtEd- wrote:
All someone would have to do is figure out how to stop the program from taking a screen shot and replacing it with other random photos from a different game. Then it comes down to people taking the time to look at every picture to make sure that what was uploaded is right.
Anyone willing to volunteer for that job?
You would obviously need some kind of player of the screenshots!
And you would probably be capable of easily see if they maphack or not by just jumping around very quickly.
|
On September 04 2007 12:20 bErAtEd- wrote:
All someone would have to do is figure out how to stop the program from taking a screen shot and replacing it with other random photos from a different game. Then it comes down to people taking the time to look at every picture to make sure that what was uploaded is right.
Anyone willing to volunteer for that job? Which is why you don't check every game. You would only check games in which you suspect you were hacked. Then you can see if the screen shots match up with the game.
|
It's a decent idea, in theory, but its very easy to prevent people from knowing you're hacking. All you'd have to do is hook the screenshot function, and any time its called, disable maphack, then at the end, re-enable it. Voila, all the screenshots look legit.
Edit: And if you use an external screenshot function, it would still have to use DirectX functionality in order to account for people that play in windowed mode and such, so thats also hookable.
|
On September 04 2007 12:29 Blind wrote:
Which is why you don't check every game. You would only check games in which you suspect you were hacked. Then you can see if the screen shots match up with the game.
Then IMO this isn't a very strong anti hack. If the majority of the basis of the hack still relies on peoples intuition, then I think you've failed. I personally am terrible at spotting hack // knowing when I've been hacked.
|
|
|
On September 03 2007 22:49 MarklarMarklar wrote:Show nested quote +On September 03 2007 22:25 NonY[rC] wrote:
i like idea 2. just the mini-map. take one on average 3-5 seconds, but randomize the lengths in between. it'll still be well under .5MB for a 30 minute game. yes, we really need a programmer to comment on the idea however. Anyone know someone?
Taiche should post soon.
|
Well what if you took 10 screenshots a second but only saved 0.5 of them (randomized)
surely you cant turn it off and on 10 times a second?
And even if you can wouldnt you be capable of seeing where the person is looking on the map a ton of times during the game? Where they are using their maphack to spot lets say builds/ army size?
|
On September 04 2007 12:41 Wizard[pl] wrote:Show nested quote +On September 03 2007 22:49 MarklarMarklar wrote:On September 03 2007 22:25 NonY[rC] wrote:
i like idea 2. just the mini-map. take one on average 3-5 seconds, but randomize the lengths in between. it'll still be well under .5MB for a 30 minute game. yes, we really need a programmer to comment on the idea however. Anyone know someone? Taiche should post soon.
but where is he
Anyway, I searched on him and realised he just made a thread about how nobody came up with anti-hack ideas, about 2 weeks ago.
He seems like the guy
|
I don't think this idea is going to fly... at all.
In order for this to work, you have to trigger the print screen function of the game (which is by default bound to the "PrtScn" key on your keyboard), the easiest way to do this is probably simulate a keyPress/keyDown or w/e in whatever language. Another method is to find another way to trigger the Brood War printScreen method, which is much harder.
Now, if you can trigger this Print Screen method (say, by the keyPress idea since it's the easiest to do), then the hackers can definitely do that as well. All they would have to do is write another program in addition to the hack and just listen to the key press events. Whenever the PrtScn key is pressed (it makes no difference whether it's physically pressed or simulated):
onKeyDown: turn off hack
onKeyUp: turn on hack
It's that easy... and your program would not work at all.
As well, it's hard to determine whether the JPG files actually match the replay unless you manually check it...
edit:
Tec27 already said what I said...
|
I'm pretty sure you can create your own screencapturing function.
|
How does map hack work anyway?
Does it show you everything (so nothing is under the fog of war) including the entire mini-map?
|
On September 04 2007 14:51 MarklarMarklar wrote:
I'm pretty sure you can create your own screencapturing function.
Yes, you can.
What I wanted to emphasize was that if you could do it, the hack could also do it. All the hack has to do is to listen to the event that triggers the screen-capturing method, and turn off the hack when that happens...
|
i'm not a programmer but.. if you always use the trigger (multiple times per second) then can they turn it off/on all the time? Maybe 10 times a second and only keep 0.5 of the screens(randomized) = 1 every 2 seconds.
AND EVEN IF THEY DO manage to hide it, it will capture where they are looking on the map, which means even if you turn it off you will still see him looking at a black area serveral times during the game.
|
On September 04 2007 15:07 MarklarMarklar wrote:
i'm not a programmer but.. if you always use the trigger (multiple times per second) then can they turn it off/on all the time? Maybe 10 times a second and only keep 0.5 of the screens(randomized), 1 every 2 seconds.
AND EVEN IF THEY DO manage to hide it, it will capture where they are looking on the map, which means even if you turn it off you will still see him looking at a black area serveral times during the game.
If you keep your finger on your PrtScn button the entire game, your game will be slowed down. By the same logic, if you keep on taking screenshots and discarding most of them, you are effectively slowing down your game. And it definitely takes more resources to take the screenshots than toggling the map hack.
Even if you see a black screen, it does not necessarily mean that the person is a hacker. For instance, I click the dark areas before I use my scan... There are countless reasons why someone may have nothing but black on there screen. As well, since you are only taking a very small set of screenshots, it's more likely than not that the hacker is looking at something that he should be able to see.
Now, to comment on something that many have already commented: the validation process. I don't see a easy way around this validation process except humans.
You can probably verify the screenshots taken during early game more easily (still very difficult and requires a lot of advanced image processing) with a program because, if the minimap shows the enemy's base without a grey area leading to it, then the player hacks. Or if the area shown is much bigger than a scouting unit's sight area, then it's likely that the player hacks (what if it's a fast pool rush, worker rush...).
Once it gets into mid- or late-game, I'd say it's almost impossible to tell whether the player is hacking from the picture alone, the replay then needs to be analyzed to see whether hack is actually used.
|
The problem with ur solution is that it is too easy to implement. Meaning conversly easy for a hacker to work around it.
Lets assume that a program is made based on ur idea of randomising screen shot capture. The problem here is that NOTHING in a computer program is truely randomised. All the hacker needs to do is figure out the seeding value and then time the hack accordingly (which isn't very hard).
There is no point in making a seperate screencapturing function unless you want to save the file like a mini-vod and that is unviable due to the size and people are already saving first person vods.
|
Cambium considering that normal print screen takes a high quality screenshot of 150kb-250kb, there is a difference i'm sure in taking a picture smaller than 5kb in very low quality. It shouldnt be demanding at all...
And haduken, yes maybe. But how will they hide the fact that they are looking at a certain part of the map at a given time? it WILL capture that.
edit
and cambium stop posting HUGE posts on technical aspects you seem to be at least as clueless as me on.
|
One more thing.
What ever solution you comes up with, you are required to at certain timeframe of program execution to save pic on to a harddrive which requires HD I/O which is the slowest internal operation of any computer.
|
I think it would be a great idea to take information about mini map, but not in screen - in other ways programming like code or sth. We have old bwchart or RepASM4, where we can see APM or listing of actions so ... why no implement info about mini-map ?
|
Quick question how exactly will this stop your opponent from hacking when it only takes pictures of yourself?
In all honesty good idea HARD to implement.
EDIT: The only thing this would be good for is proving YOU don't hack.
|
I'm pretty sure you wont have to write it to your harddrive til the game is over.
|
On September 04 2007 15:47 Black Pearl wrote:
I think it would be a great idea to take information about mini map, but not in screen - in other ways programming like code or sth. We have old bwchart or RepASM4, where we can see APM or listing of actions so ... why no implement info about mini-map ?
well it would be amazing if you could capture a players minimap actions, and completely follow how its used thruout a game. But is that possible?
|
On September 04 2007 15:44 MarklarMarklar wrote:
Cambium considering that normal print screen takes a high quality screenshot of 150kb-250kb, there is a difference i'm sure in taking a picture smaller than 5kb in very low quality. It shouldnt be demanding at all...
And haduken, yes maybe. But how will they hide the fact that they are looking at a certain part of the map at a given time? it WILL capture that.
edit
and cambium stop posting HUGE posts on technical aspects you seem to be at least as clueless as me on.
Doesn't bwtv or whateva it was called do that already? every replay file it saved is something like 2mb.
but i see ur point. the problem is keeping the screenshots truely randomised which isn't possible in present day computers. if it isn't randomised then its given that a pattern or a function is followed then a counter-measure is also a given.
maybe we can distribute the program and update the randomization variable every once in a while. ~_- who knows...
from a programmer's view, the implementation is not hard at all, maybe it would work.
|
On September 04 2007 15:47 MarklarMarklar wrote:
I'm pretty sure you wont have to write it to your harddrive til the game is over.
then how do you purpose to save it then? i was understand the impression that some kinda of centralised server is retrieving the pics during a game. that is just a bad idea.
if not then i don't really see the point of the program when so many previous programs does the same thing.
|
dude i just heard bnet has a built-in "antihack" of a sort:
apparently when bnet detects your hack it will give you a loss on your record even if the game is ums or if you won in the other game modes.
it's pretty primitive but if bnet is detecting the hacks and adminstering losses to those who cheat, couldnt we develop something to use what their doing?
|
On September 04 2007 15:54 haduken wrote:Show nested quote +On September 04 2007 15:47 MarklarMarklar wrote:
I'm pretty sure you wont have to write it to your harddrive til the game is over. then how do you purpose to save it then? i was understand the impression that some kinda of centralised server is retrieving the pics during a game. that is just a bad idea. if not then i don't really see the point of the program when so many previous programs does the same thing.
nono, that concept was just if lets say a ladder like iccup would implement it. I'm just brainstorming.
|
On September 04 2007 15:49 MarklarMarklar wrote:Show nested quote +On September 04 2007 15:47 Black Pearl wrote:
I think it would be a great idea to take information about mini map, but not in screen - in other ways programming like code or sth. We have old bwchart or RepASM4, where we can see APM or listing of actions so ... why no implement info about mini-map ? well it would be amazing if you could capture a players minimap actions, and completely follow how its used thruout a game. But is that possible?
i think you would need the brood war API and substantial support/help from blizzard to do this without making the task astronomical. this is why people who at www.bwprogramming.com give up because it is simply too much work for a few guys with sparetime to do.
|
On September 04 2007 15:44 MarklarMarklar wrote:
Cambium considering that normal print screen takes a high quality screenshot of 150kb-250kb, there is a difference i'm sure in taking a picture smaller than 5kb in very low quality. It shouldnt be demanding at all...
And haduken, yes maybe. But how will they hide the fact that they are looking at a certain part of the map at a given time? it WILL capture that.
edit
and cambium stop posting HUGE posts on technical aspects you seem to be at least as clueless as me on.
You said you would take ten in a second (I don't know what you mean by "take", probably means writing them to disk somewhere and deleting all but one), it is a lot of extra processing regardless how big they are. And also, you can't just "take" a lower quality version of the screenshot (unless you write your own algorithm, of course) with the BW print screen function. In order to shrink it, you would have to process it to make it smaller. Ten times a second is a lot of unnecessary processing time.
And nothing in my previous posts had technical details (except the keyPress), anyone could understand them.
And just fyi, I am a software developer...
edit:
And you still haven't addressed my concerns about the validation process...
|
On September 04 2007 15:57 haduken wrote:Show nested quote +On September 04 2007 15:49 MarklarMarklar wrote:On September 04 2007 15:47 Black Pearl wrote:
I think it would be a great idea to take information about mini map, but not in screen - in other ways programming like code or sth. We have old bwchart or RepASM4, where we can see APM or listing of actions so ... why no implement info about mini-map ? well it would be amazing if you could capture a players minimap actions, and completely follow how its used thruout a game. But is that possible? i think you would need the brood war API and substantial support/help from blizzard to do this without making the task astronomical. this is why people who at www.bwprogramming.com give up because it is simply too much work for a few guys with sparetime to do.
Well if that's the case then i guess minisceenshots of the minimap is a better solution for BROODWAR. However, maybe blizzard could implement it for starcraft 2?
to cambium: i didnt actually use real numbers. I just wanted anyone who has the programming ability to come along and say whats possible and whats not.
so lets say you take 2 every seconds then how possible is it? I'm sure you dont have to take a screenshot of the ENTIRE screen! Only the required area which is about 130x95pixels
811byte in lowest quality from photoshop
and at 100% quality its 3.2kb big. That shouldnt be too big a problem.
|
On September 04 2007 16:02 MarklarMarklar wrote:Show nested quote +On September 04 2007 15:57 haduken wrote:On September 04 2007 15:49 MarklarMarklar wrote:On September 04 2007 15:47 Black Pearl wrote:
I think it would be a great idea to take information about mini map, but not in screen - in other ways programming like code or sth. We have old bwchart or RepASM4, where we can see APM or listing of actions so ... why no implement info about mini-map ? well it would be amazing if you could capture a players minimap actions, and completely follow how its used thruout a game. But is that possible? i think you would need the brood war API and substantial support/help from blizzard to do this without making the task astronomical. this is why people who at www.bwprogramming.com give up because it is simply too much work for a few guys with sparetime to do. Well if that's the case then i guess minisceenshots of the minimap is a better solution for BROODWAR. However, maybe blizzard could implement it for starcraft 2? to cambium: i didnt actually use real numbers. I just wanted anyone who has the programming ability to come along and say whats possible and whats not. so lets say you take 2 every seconds then how possible is it? I'm sure you dont have to take a screenshot of the ENTIRE screen! Only the required area which is about 130x95pixels 811byte in lowest quality from photoshop and at 100% quality its 3.2kb big. That shouldnt be too big a problem.
If you don't use the print screen function that BW provides, you would have to write everything from scratch, which would require a ton of work, and I have nothing more to argue. Only then can you actually take a section of the screen.
edit:
If you write your own little thing to take the screenshot from scratch (through w/e means), the hack won't know when the screenshot is taken, and everyone is happy.
|
i don't see the point. no system is unfallable.
Even a perfect anti-hack is only going to last 6months at most. i think the most blizzard or any company can do without spending too much is have strong procedure for discouragement, prevention and even prosecutation of hacking.
|
On September 04 2007 16:16 Cambium wrote:Show nested quote +On September 04 2007 16:02 MarklarMarklar wrote:On September 04 2007 15:57 haduken wrote:On September 04 2007 15:49 MarklarMarklar wrote:On September 04 2007 15:47 Black Pearl wrote:
I think it would be a great idea to take information about mini map, but not in screen - in other ways programming like code or sth. We have old bwchart or RepASM4, where we can see APM or listing of actions so ... why no implement info about mini-map ? well it would be amazing if you could capture a players minimap actions, and completely follow how its used thruout a game. But is that possible? i think you would need the brood war API and substantial support/help from blizzard to do this without making the task astronomical. this is why people who at www.bwprogramming.com give up because it is simply too much work for a few guys with sparetime to do. Well if that's the case then i guess minisceenshots of the minimap is a better solution for BROODWAR. However, maybe blizzard could implement it for starcraft 2? to cambium: i didnt actually use real numbers. I just wanted anyone who has the programming ability to come along and say whats possible and whats not. so lets say you take 2 every seconds then how possible is it? I'm sure you dont have to take a screenshot of the ENTIRE screen! Only the required area which is about 130x95pixels 811byte in lowest quality from photoshop and at 100% quality its 3.2kb big. That shouldnt be too big a problem. If you don't use the print screen function that BW provides, you would have to write everything from scratch, which would require a ton of work, and I have nothing more to argue. Only then can you actually take a section of the screen.
Maybe using a third-party program that video-capture the mini screen based on the horizontal + vertical parameters. Again, i don't see the point when you can record the entire screen in a VOD.
|
On September 04 2007 16:18 haduken wrote:
i don't see the point. no system is unfallable.
Even a perfect anti-hack is only going to last 6months at most. i think the most blizzard or any company can do without spending too much is have strong procedure for discouragement, prevention and even prosecutation of hacking.
Well, going along with his idea...
Say he writes his own screen-snatching program and builds into the anti-hack, and write the screens to files that are encrypted, it will be very difficult to hack (provided that his program and the screens do not get cracked).
However, this would take forever to finish... How long you spend in developing your software is directly proportional to the time it takes for hackers to crack it.
|
On September 04 2007 16:20 haduken wrote:Show nested quote +On September 04 2007 16:16 Cambium wrote:On September 04 2007 16:02 MarklarMarklar wrote:On September 04 2007 15:57 haduken wrote:On September 04 2007 15:49 MarklarMarklar wrote:On September 04 2007 15:47 Black Pearl wrote:
I think it would be a great idea to take information about mini map, but not in screen - in other ways programming like code or sth. We have old bwchart or RepASM4, where we can see APM or listing of actions so ... why no implement info about mini-map ? well it would be amazing if you could capture a players minimap actions, and completely follow how its used thruout a game. But is that possible? i think you would need the brood war API and substantial support/help from blizzard to do this without making the task astronomical. this is why people who at www.bwprogramming.com give up because it is simply too much work for a few guys with sparetime to do. Well if that's the case then i guess minisceenshots of the minimap is a better solution for BROODWAR. However, maybe blizzard could implement it for starcraft 2? to cambium: i didnt actually use real numbers. I just wanted anyone who has the programming ability to come along and say whats possible and whats not. so lets say you take 2 every seconds then how possible is it? I'm sure you dont have to take a screenshot of the ENTIRE screen! Only the required area which is about 130x95pixels 811byte in lowest quality from photoshop and at 100% quality its 3.2kb big. That shouldnt be too big a problem. If you don't use the print screen function that BW provides, you would have to write everything from scratch, which would require a ton of work, and I have nothing more to argue. Only then can you actually take a section of the screen. Maybe using a third-party program that video-capture the mini screen based on the horizontal + vertical parameters. Again, i don't see the point when you can record the entire screen in a VOD.
That would ideally be the best solution...
edit:
The VOD solution, that is.
And also, the validation would be really complicated...
|
how could they crack the 30 screenshots a minute capturing everywhere they are looking on the map?
Even if their maphack DOESNT show up in the screenshots, how can they hide that they are viewing a certain location at a given time? You CAN see that...
THATS what i want to hear from you........................... ???
|
What is the point of capturing screen shots then? when you can just record the entire game in a video?
i'm not sure what you mean by cracking the 30 screenshots a minuate thing and i don't even think that is a good idea.
You idea requires a significant pool of samples to make an accurate analysis and you still need a human to make the analysis. Wouldn't the better idea is just video record the game and view it?
|
okay program a video capturing program that captures 1frame per second and saves it as a video file in the replay afterwards, it's the same concept in a different format.
It has to be small tho so it can be easily used. tops 1mb per game
|
Okay, that might work but its still very akward and is selective sampling. No point in reinventing the wheel my man.
|
why are you so negative towards this concept? the same with cambium?
People want a good anti-cheat, and this seems TO ME to be something that you cant beat if you work it out well.
|
Russian Federation25 Posts
Cambium, why do you always address the BW print screen function? Noone is going to use it, of course. And yes, it can be written "from scratch", and no, it won't "require a ton of work". In fact, that's pretty easy.
"All the hack has to do is to listen to the event that triggers the screen-capturing method, and turn off the hack when that happens..." - looks like you are in the world of Borland VCL events and Windows messages here, no offence  Your own function won't trigger any events or be triggered by any. It's not that trivial, but not that hard either. Yes it can be hooked, but it's waaaay harder than you think of key events. Especially, if the antihack will be kept updated.
Second, 1 Kb large minimap screen doesn't needed to be saved to the disk all the time. All screenshots throughout the game can be easily maintained in the memory. 30 mins / 30 secs * 1Kb = 60 Kb total, that's not even a piece of memory nowadays. After the game finishes, this "ton" of info can be saved or sent to the server immedientely. Or, those 1 Kb can be sent to the server each time they are taken. This may require a bit of bandwidth for, say, modem users, but well, that's the price 
About the defence of randomizer. Nice solution is, as already mentioned, the constant update of the antihack. In fact, this is where we shall anyway come to.
About the "guy, who's gonna view all that shit". As already stated (i'll just emphasize that) it's YOU, who is going to view that. You play a game, smth like "fuck, that guy is a haxor" pops up in your head and you just go and check the screens on the server. That's it. Of course, there's no 100% guarantee that if the screens are clear, the guy didn't hack. But if they aren't, you've got him.
About the viruses in the JPG. Funny statement, correct me if I'm wrong, but JPG isn't an executable in any way possible. Please, don't start the unaware virus panic.
About the DDoS. You are going to accept data only from the antihack program. It can be of course bypassed, but what can't?
About the storage size. Again, as already stated, it will require maybe a 1 GB HD space per day, and one day is enough to make up your mind and check the screens if you want. If you have found a hax on some screens you can save them on your computer or report the admin not to delete them.
Lastly, Id' like to say, that there are already some implementations of this idea in other games. I know at least Screen Shot Slient (SSC) for Counter-Strike 1.6. This project is several years old (maybe dead now), but it worked very well. Yes, it was bypassed, but we must understand, that this is normal. The only way of defence is constant updates. SSC was kept updated, while it was still alive. Maybe some consulting on technical details with SSC developer will free us from already made mistakes during coding. If he'd be so kind, of course.
|
On September 04 2007 16:54 MarklarMarklar wrote:
why are you so negative towards this concept? the same with cambium?
People want a good anti-cheat, and this seems TO ME to be something that you cant beat if you work it out well.
We are not negative. We just don't think the reward equates the effort. What you are suggesting doesn't actually prevent hacking, it only lets you somehow make an educated guess (although questionable) at who is hacking. If thats your goal then why develope something when simple alternatives exist?
|
On September 04 2007 17:00 6AP6APblCKA wrote:Cambium, why do you always address the BW print screen function? Noone is going to use it, of course. And yes, it can be written "from scratch", and no, it won't "require a ton of work". In fact, that's pretty easy. "All the hack has to do is to listen to the event that triggers the screen-capturing method, and turn off the hack when that happens..." - looks like you are in the world of Borland VCL events and Windows messages here, no offence Your own function won't trigger any events or be triggered by any. It's not that trivial, but not that hard either. Yes it can be hooked, but it's waaaay harder than you think of key events. Especially, if the antihack will be kept updated. Second, 1 Kb large minimap screen doesn't needed to be saved to the disk all the time. All screenshots throughout the game can be easily maintained in the memory. 30 mins / 30 secs * 1Kb = 60 Kb total, that's not even a piece of memory nowadays. After the game finishes, this "ton" of info can be saved or sent to the server immedientely. Or, those 1 Kb can be sent to the server each time they are taken. This may require a bit of bandwidth for, say, modem users, but well, that's the price About the defence of randomizer. Nice solution is, as already mentioned, the constant update of the antihack. In fact, this is where we shall anyway come to. About the "guy, who's gonna view all that shit". As already stated (i'll just emphasize that) it's YOU, who is going to view that. You play a game, smth like "fuck, that guy is a haxor" pops up in your head and you just go and check the screens on the server. That's it. Of course, there's no 100% guarantee that if the screens are clear, the guy didn't hack. But if they aren't, you've got him. About the viruses in the JPG. Funny statement, correct me if I'm wrong, but JPG isn't an executable in any way possible. Please, don't start the unaware virus panic. About the DDoS. You are going to accept data only from the antihack program. It can be of course bypassed, but what can't? About the storage size. Again, as already stated, it will require maybe a 1 GB HD space per day, and one day is enough to make up your mind and check the screens if you want. If you have found a hax on some screens you can save them on your computer or report the admin not to delete them. Lastly, Id' like to say, that there are already some implementations of this idea in other games. I know at least Screen Shot Slient (SSC) for Counter-Strike 1.6. This project is several years old (maybe dead now), but it worked very well. Yes, it was bypassed, but we must understand, that this is normal. The only way of defence is constant updates. SSC was kept updated, while it was still alive. Maybe some consulting on technical details with SSC developer will free us from already made mistakes during coding. If he'd be so kind, of course.
Would you code it then
|
Russian Federation25 Posts
On September 04 2007 17:32 haduken wrote:Show nested quote +On September 04 2007 16:54 MarklarMarklar wrote:
why are you so negative towards this concept? the same with cambium?
People want a good anti-cheat, and this seems TO ME to be something that you cant beat if you work it out well. We are not negative. We just don't think the reward equates the effort. What you are suggesting doesn't actually prevent hacking, it only lets you somehow make an educated guess (although questionable) at who is hacking. If thats your goal then why develope something when simple alternatives exist?
An educated guess? You say, that if I check the screens and see that the guy has, say, the whole map opened up that's a guess? He's busted, that's what it is.
And IF there's nothing on screens, then yes, it's not 100% that he's not hacking, but IF there is - that's why it's worth of it.
|
On September 04 2007 17:32 haduken wrote:Show nested quote +On September 04 2007 16:54 MarklarMarklar wrote:
why are you so negative towards this concept? the same with cambium?
People want a good anti-cheat, and this seems TO ME to be something that you cant beat if you work it out well. We are not negative. We just don't think the reward equates the effort. What you are suggesting doesn't actually prevent hacking, it only lets you somehow make an educated guess (although questionable) at who is hacking. If thats your goal then why develope something when simple alternatives exist?
So if a good anti-hack requires effort then its not worth it?
And of course it can prevent maphacking..
|
On September 04 2007 16:39 MarklarMarklar wrote:
okay program a video capturing program that captures 1frame per second and saves it as a video file in the replay afterwards, it's the same concept in a different format.
It has to be small tho so it can be easily used. tops 1mb per game
Your ideas suck for one very simple reason: you exclude anyone that doesn't have a modern PC. If you're going to make an anti-hack, you can't exclude a large part of the BW community by requiring them to play with stuff that will seriously slow down their client. You keep talking about hard drive space, but really, thats the least of the concerns. CPU activity would be too great for many people.
On September 04 2007 17:00 6AP6APblCKA wrote:1.Cambium, why do you always address the BW print screen function? Noone is going to use it, of course. And yes, it can be written "from scratch", and no, it won't "require a ton of work". In fact, that's pretty easy. "All the hack has to do is to listen to the event that triggers the screen-capturing method, and turn off the hack when that happens..." - looks like you are in the world of Borland VCL events and Windows messages here, no offence Your own function won't trigger any events or be triggered by any. It's not that trivial, but not that hard either. Yes it can be hooked, but it's waaaay harder than you think of key events. Especially, if the antihack will be kept updated. 2.Second, 1 Kb large minimap screen doesn't needed to be saved to the disk all the time. All screenshots throughout the game can be easily maintained in the memory. 30 mins / 30 secs * 1Kb = 60 Kb total, that's not even a piece of memory nowadays. After the game finishes, this "ton" of info can be saved or sent to the server immedientely. Or, those 1 Kb can be sent to the server each time they are taken. This may require a bit of bandwidth for, say, modem users, but well, that's the price 3.About the defence of randomizer. Nice solution is, as already mentioned, the constant update of the antihack. In fact, this is where we shall anyway come to. 4.About the "guy, who's gonna view all that shit". As already stated (i'll just emphasize that) it's YOU, who is going to view that. You play a game, smth like "fuck, that guy is a haxor" pops up in your head and you just go and check the screens on the server. That's it. Of course, there's no 100% guarantee that if the screens are clear, the guy didn't hack. But if they aren't, you've got him. 5.About the viruses in the JPG. Funny statement, correct me if I'm wrong, but JPG isn't an executable in any way possible. Please, don't start the unaware virus panic. 6.About the DDoS. You are going to accept data only from the antihack program. It can be of course bypassed, but what can't? 7.About the storage size. Again, as already stated, it will require maybe a 1 GB HD space per day, and one day is enough to make up your mind and check the screens if you want. If you have found a hax on some screens you can save them on your computer or report the admin not to delete them.
1.When dealing with DirectX, you can't merely copy pixels of the screen like you might when taking a screenshot of a web page. Therefore, some function must be called, and this function will involve using a DirectX library. A hack could in fact hook this called function, allowing them to disable the hack whenever you take a screenshot. Its not exactly rocket science.
2.HD space isn't that big of a deal. Be VERY careful with how much CPU time you're using though, as that IS a big deal.
3.See 1.
4.This would require some sort of site to store all the screenshots, and a database to keep track of which one goes with what game. If it doesn't then it requires some sort of client-to-client transfer, which is altogether a bad idea with todays routers and firewalls and such. So, going back to the site, it would require keeping track of loads of data, due to the sheer number of SC games that go on day-to-day (even if you were to prune them weekly or monthly or something)
5.There actually have been viruses in JPG's using their thumbnail capability to execute themselves.
6.Basically true, I suppose.
7.This assumes you're using client-to-client transfers, which, as I already stated, is a bad idea.
|
On September 04 2007 23:01 tec27 wrote:
1.When dealing with DirectX, you can't merely copy pixels of the screen like you might when taking a screenshot of a web page. Therefore, some function must be called, and this function will involve using a DirectX library. A hack could in fact hook this called function, allowing them to disable the hack whenever you take a screenshot. Its not exactly rocket science.
You can still copy them from the video memory via custom kernel mode driver. Try to hook that.
|
Russian Federation25 Posts
On September 04 2007 23:01 tec27 wrote:Show nested quote +On September 04 2007 16:39 MarklarMarklar wrote:
okay program a video capturing program that captures 1frame per second and saves it as a video file in the replay afterwards, it's the same concept in a different format.
It has to be small tho so it can be easily used. tops 1mb per game Your ideas suck for one very simple reason: you exclude anyone that doesn't have a modern PC. If you're going to make an anti-hack, you can't exclude a large part of the BW community by requiring them to play with stuff that will seriously slow down their client. You keep talking about hard drive space, but really, thats the least of the concerns. CPU activity would be too great for many people. Show nested quote +On September 04 2007 17:00 6AP6APblCKA wrote:1.Cambium, why do you always address the BW print screen function? Noone is going to use it, of course. And yes, it can be written "from scratch", and no, it won't "require a ton of work". In fact, that's pretty easy. "All the hack has to do is to listen to the event that triggers the screen-capturing method, and turn off the hack when that happens..." - looks like you are in the world of Borland VCL events and Windows messages here, no offence Your own function won't trigger any events or be triggered by any. It's not that trivial, but not that hard either. Yes it can be hooked, but it's waaaay harder than you think of key events. Especially, if the antihack will be kept updated. 2.Second, 1 Kb large minimap screen doesn't needed to be saved to the disk all the time. All screenshots throughout the game can be easily maintained in the memory. 30 mins / 30 secs * 1Kb = 60 Kb total, that's not even a piece of memory nowadays. After the game finishes, this "ton" of info can be saved or sent to the server immedientely. Or, those 1 Kb can be sent to the server each time they are taken. This may require a bit of bandwidth for, say, modem users, but well, that's the price 3.About the defence of randomizer. Nice solution is, as already mentioned, the constant update of the antihack. In fact, this is where we shall anyway come to. 4.About the "guy, who's gonna view all that shit". As already stated (i'll just emphasize that) it's YOU, who is going to view that. You play a game, smth like "fuck, that guy is a haxor" pops up in your head and you just go and check the screens on the server. That's it. Of course, there's no 100% guarantee that if the screens are clear, the guy didn't hack. But if they aren't, you've got him. 5.About the viruses in the JPG. Funny statement, correct me if I'm wrong, but JPG isn't an executable in any way possible. Please, don't start the unaware virus panic. 6.About the DDoS. You are going to accept data only from the antihack program. It can be of course bypassed, but what can't? 7.About the storage size. Again, as already stated, it will require maybe a 1 GB HD space per day, and one day is enough to make up your mind and check the screens if you want. If you have found a hax on some screens you can save them on your computer or report the admin not to delete them. 1.When dealing with DirectX, you can't merely copy pixels of the screen like you might when taking a screenshot of a web page. Therefore, some function must be called, and this function will involve using a DirectX library. A hack could in fact hook this called function, allowing them to disable the hack whenever you take a screenshot. Its not exactly rocket science. 2.HD space isn't that big of a deal. Be VERY careful with how much CPU time you're using though, as that IS a big deal. 3.See 1. 4.This would require some sort of site to store all the screenshots, and a database to keep track of which one goes with what game. If it doesn't then it requires some sort of client-to-client transfer, which is altogether a bad idea with todays routers and firewalls and such. So, going back to the site, it would require keeping track of loads of data, due to the sheer number of SC games that go on day-to-day (even if you were to prune them weekly or monthly or something) 5.There actually have been viruses in JPG's using their thumbnail capability to execute themselves. 6.Basically true, I suppose. 7.This assumes you're using client-to-client transfers, which, as I already stated, is a bad idea.
1. I understand, that it's also hookable. Almost everithing is. But as I mentioned, it's harder. I just wanted to say, that key events are ridiculous. Constant updates are needed to deal with hacks. Here, I can't argue too deep due to the lack of knowlage of actual implementation, but I suppose, that SSC author had some sort of defence, strong enough to last until next update.
2. I don't thing that a copy of small area of screen once in about 30 seconds is a big deal for CPU. [Edit] By "memory" I meant RAM, not HD.
3. Didn't actually get, what to see in 1. If you mean to hook randomizer, then see 1.
4. Yes, it will require a site, for example ICCup will maintain it. I don't think, that 1 Kb per 30 seconds from players is a "load" of data. Front page of their web site is much heavier. Yes, firewall is needed to be set up to allow antihack to send data, but setting up your firewall is becoming normal today, isn't it? All in all, the "sheer" number has to be concretely estimated. You think it would be a lot, and I think that 1 Kb per 30 secs even multiplied by couple of thousands of games simultaneously isn't a lot for a modern internet site.
5. Wow, to be honest, I didn't know that, but I suppose, that it's not the problem with JPG format itself, but with the viewer you're using. If your viewer is trying to execute JPG somehow then, well, even txt is not safe if you'll rename it to exe and launch. If JPG is dangerous then we'd better close our browsers right now, turn off computers and hide somewhere, because JPG is everywhere today.
7. I do not assume any client-to-client transfers. I, of course, agree that it's a bad idea. Actually, screenshots are sent to the site, saved there, and then viewed by users through their browsers. And from the site's point of view, storage space required is not big (esp. for the modern web site).
|
GrandInquisitor
New York City13113 Posts
jpg's can have viruses in them
but LOL at some of you people. "what's the point it's going to be defeated anyway" holy shit
i don't see the point. no system is unfallable.
Even a perfect anti-hack is only going to last 6months at most. i think the most blizzard or any company can do without spending too much is have strong procedure for discouragement, prevention and even prosecutation of hacking.
We are not negative. We just don't think the reward equates the effort. What you are suggesting doesn't actually prevent hacking, it only lets you somehow make an educated guess (although questionable) at who is hacking. If thats your goal then why develope something when simple alternatives exist?
come ON haduken. i can understand the complaining from a technical perspective (though other games already do this, so you need to demonstrate why it can work for say CS but not possibly for SC), but face it - we're not going to code the mother of all unstoppable anti-hack. we don't need to. this program doesn't need to cure AIDS, it just needs to maybe catch a couple hackers, or, i mean, even if this *deters* just a couple hackers it would be totally worth it
like, we're not forcing YOU to do it. if SOMEONE feels like they want to invest the effort into coding this, who are you to say, no, you shouldn't go do it?
|
lol. you just don't get my point did ya? what he is suggesting can already be accomplished by simple alternatives such as fps vod. I'm assuming the main benefit of his solution is a viable way to transfer a reasonably small sized "screens" files across the internet which i personally find very hard because of all the world constraints + committments. Hey, if anyone got the time and motivation to do this then by all means, don't let a pessimistic fuck like me get in ur way.
|
GrandInquisitor
New York City13113 Posts
FPS VOD is impossible for me. I can never run such a plugin because it'd be an unbelievably huge resource drain to Camtasia every single game.
|
It doesn't have to be every 30 secs. It can be programmed to screencap on random time within a certain range. Anyway, it sounds like there is a lot of info to send if the game goes on long.
|
On September 04 2007 23:32 sundance wrote:Show nested quote +On September 04 2007 23:01 tec27 wrote:
1.When dealing with DirectX, you can't merely copy pixels of the screen like you might when taking a screenshot of a web page. Therefore, some function must be called, and this function will involve using a DirectX library. A hack could in fact hook this called function, allowing them to disable the hack whenever you take a screenshot. Its not exactly rocket science.
You can still copy them from the video memory via custom kernel mode driver. Try to hook that.
I have no knowledge in this area what so ever, but it sounds hard as fuck...
Look, your idea WILL work (almost flawlessly in theory, so it's a good idea) if you find people who are willing to do the following things:
1) Find a non-trivial way to get the screens - sundance's idea sounds the best (although I don't even know if it's possible  ) without slowing down the computer considerably.
2) Find a secure and fast way to encrypt and decrypt your image files.
3) Get a large enough database and fast enough server that can store and handle the increased traffic of replays and image files, as well, automatically process them with efficiency and accuracy, while still not require a fee (optional I guess).
4) Code, implement, debug and maintain all of the above.
|
that sounds like.. lots of hard work man... anyone up for it ?
|
The validation part sounds somewhat exciting, I can probably do that during x-mas (when I have copious free time), not really interested in the rest =\\
|
this is how punkbuster works (and punkbuster doesn't work well)
|
Punkbuster worked well for CS until the hacks were modified to turn off every time a screen shot was taken (which won't be as easy with the method sundance suggested).
FPVODs are clearly not feasible on anything but the smallest scale.
Really, just having an anti-hack that actually works should be enough to scare off a lot of people who think they can get away with maphacking, even if some hack makers start finding ways around it.
|
There are maphacks that just remove fog of war in play screen and avoid the minimap all together.
Just implement a FP for every replay. That way even if they hack and find a way around it showing that, you would still see them looking in dark spots on the map all the time, what their mouse does , etc.
|
It's a step forward if you can stop maphackers from seeing things on the map.
And you would probably need to make an antihack that takes a screenshot every 3-5 seconds or so
Which is possible if they are about 500byte
445BYTE big and sufficient to spot maphack behaviour
this one is 548byte
749byte
every 3 seconds of 500byte screenshots for one minute equals 10kb of data
100kb for 10m
200kb for 20m... And so on
which would result in a 300-500kb replay, small enough?
Just the function of saving this into a replay would be useful i think
|
|
|
|
|
|
EPT![[image loading]](http://img179.imageshack.us/img179/3483/scscrnshot090307123058fc8.jpg)
![[image loading]](http://img264.imageshack.us/img264/8614/mapverisionyw7.jpg)
![[image loading]](http://img131.imageshack.us/img131/3392/mapverision2pu4.jpg)
![[image loading]](http://img104.imageshack.us/img104/6682/untitled1vw2.jpg)
![[image loading]](http://img232.imageshack.us/img232/9748/445kbmi2.jpg)
![[image loading]](http://img411.imageshack.us/img411/9307/548bytese6.jpg)
![[image loading]](http://img210.imageshack.us/img210/4653/749bytetn2.jpg)
