|
07:06 KST - method linked here has been disproved here10:54 KST - Find a full timeline of pro comments (including Spades) in the topic here.08:47 KST - Summary:Accusations of maphacking have the potential to destroy a player's career if left unaddressed. Because of the potential consequences, we should be careful about accepting unproven accusations. The principle of 'innocent until proven guilty' should be applied here. That does not mean that there has been a conclusion about this case, however, which is why this thread remains tentatively open. Please discuss with caution and use evidence to back up your claims. (also a summary post by an unnamed pro on reddit here) |
On June 07 2012 09:53 EtherealDeath wrote:Show nested quote +On June 07 2012 09:51 JustTray wrote: Just would like to say that the "the current public hack doesn't allow actions while in camera lock" was debunked by a hacker about 75 pages ago, around page 190-205, I don't remember exactly. In fact, in this thread, a hacker came out saying he was one of a few select Grandmasters who developed a private hack and have been circulating it, and that he was going to contact someone in TL with names.
I would really like to know what came of this.
This person specifically stated that the camera lock feature could easily be adjusted, and in some verison of the hack I believe was an option to turn it on or off.
But ultimately, if you are going to say that this MOUNTAIN of evidence proving he's a hacker isn't enough, you simply cannot then make the argument that because one hack does not allow this feature that is proof he isn't a hacker. That kind of double think is simply retarded.
There should be zero more posts about the hack in question. It's irrelevant. It's just white knights grasping at straws who know it is incredibly likely Spades is a hacker. If it's a private hack I can't say. If you're talking about a public hack, then as far as I can tell he's full of shit. But, he's referring to a private hack so I have no competence to comment on it. He edited his post to "rainbow unicorns" or something like that btw == However I think some of the more technical-related evidence of Spades hacking may be bogus, and we should stick with the analytical evidence from the games themselves. Keep to that which is most evident and least improbable.
Private hack. I'm shocked so many people are unaware of the prevelance of private hacks. Sorcery used one back in Seasons 1-4, was called out by Protech, and admitted it. Why does this still shock people that there are undetected private hacks circulating? It's rampant. It has been rampant since the first season.
|
On June 07 2012 09:47 Wolvmatt. wrote: How much better does a map hack actually make a player? It seems to me that a gm-level player would never lose a game if given a map hack.
In particular, I'm referring to someone that was a no name toss player. The guy had exactly 100 apm and didn't use hot keys at all. This guy was pure bad, but he was like number 1 or 2 on the ladder for quite a bit. White-ra, who was one of the better toss players, obv, seemed to lose the majority of the games versus this player. Not to brag... but I did beat this guy, every time, because I was terran and actually used comsat... Really, you have no idea how bad you have to be to choose to hack as a terran player, when you can already comsat anything, at any point in the game. It's beyond crazy to me. You have to be so bad and greedy. Like, really, you need more than that...
|
On June 07 2012 09:54 Shiori wrote:Show nested quote +On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks.
The very fact that HRGzack is willing to use an obvious blink hack means he simply has no caution about getting caught. What is your standard for judging careful hackers? How can we surmise that someone is hacking except through examining subtle inconsistencies in the games in question vs the ones on ladder as we have done with Spades?
|
On June 07 2012 09:55 EtherealDeath wrote:Show nested quote +On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. Actually his play is really fucking shady lol, all technical-related "evidence" aside. Some parts are shady, some parts are really not. Initially when I read the OP's post I was convinced that it was all shady, but pretty much all of it has plausible explanations. Even Illusion's analysis rested rather heavily on some particular siege moments which have been shown to have plausible motives.
|
On June 07 2012 09:57 StarStrider wrote:Show nested quote +On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. The very fact that NRGzack is willing to use an obvious blink hack means he simply has no caution about getting caught. What is your standard for judging careful hackers? How can we surmise that someone is hacking except through examining subtle inconsistencies in the games in question vs the ones on ladder as we have done with Spades?
HRG not NRG, don't drag a legit NA team into this by typoing
|
On June 07 2012 09:57 StarStrider wrote:Show nested quote +On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. The very fact that NRGzack is willing to use an obvious blink hack means he simply has no caution about getting caught. What is your standard for judging careful hackers? How can we surmise that someone is hacking except through examining subtle inconsistencies in the games in question vs the ones on ladder as we have done with Spades? It could be the case that there really is no way to tell when someone is hacking, and that we'll just have to deal with that until Blizzard figures out how to make it less of a liability. IMO, the accusations are just way too weighty and devastating to be made on such little evidence. Yes, I'm aware of the fact that we probably won't get a super conclusive proof no matter what, but we might just have to bit the bullet and say "yeah, it's possible to hack sufficiently well that we can't discern it from plausibly innocent play." There's nothing shameful about that.
|
On June 07 2012 09:55 caradoc wrote:Show nested quote +On June 07 2012 09:50 toiletCAT wrote:That does it; In the name of hardworking men (and possibly women) of the Teamliquid community and in the name of justice, I, toiletCAT, + Show Spoiler + hereby concludes, that the accusing portion of the community of Teamliquid.net, have yet to acquire definitive evidence of Spades hacking. By the power of the burden of proof, Spades is not guilty. + Show Spoiler +I'm dead serious guys, enough is enough, let's all end it now that we're cool with each other, once and for all. You joined TL yesterday, and all your posts are in this and related threads. It's fair that you have an opinion, but I wouldn't expect it to be given much weight given this. It's like if I said 'I think he hacks cuz my friends' friends' friend thought so and I think that guy is smart, and I think I'm smart too'. I can say it, it can be my opinion, but it's probably not worth much to the majority of people reading it given the context.
Yeah, it's a shame, if only I had a larger post count, then I would make a great judge. In all seriousness though, judge aside, enough is enough.
|
On June 07 2012 09:57 StarStrider wrote:Show nested quote +On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. The very fact that NRGzack is willing to use an obvious blink hack means he simply has no caution about getting caught. What is your standard for judging careful hackers? How can we surmise that someone is hacking except through examining subtle inconsistencies in the games in question vs the ones on ladder as we have done with Spades?
Apparently, in SC1, the hackers did not click on the enemy unit once for 15 games.
We don't have anything like that here
|
On June 07 2012 09:57 Shiori wrote:Show nested quote +On June 07 2012 09:55 EtherealDeath wrote:On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. Actually his play is really fucking shady lol, all technical-related "evidence" aside. Some parts are shady, some parts are really not. Initially when I read the OP's post I was convinced that it was all shady, but pretty much all of it has plausible explanations. Even Illusion's analysis rested rather heavily on some particular siege moments which have been shown to have plausible motives.
Each incident has plausible explanations....by itself. They would be plausible if they were isolated and sparse. They are too frequent in all the same series to be just dumb luck and 'trying a different way of playing' compared to the ladder control group.
|
On June 07 2012 09:58 EtherealDeath wrote:Show nested quote +On June 07 2012 09:57 StarStrider wrote:On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. The very fact that NRGzack is willing to use an obvious blink hack means he simply has no caution about getting caught. What is your standard for judging careful hackers? How can we surmise that someone is hacking except through examining subtle inconsistencies in the games in question vs the ones on ladder as we have done with Spades? HRG not NRG, don't drag a legit NA team into this by typoing 
So sorry. Thanks for the correction.
|
On June 07 2012 09:59 Shiori wrote:Show nested quote +On June 07 2012 09:57 StarStrider wrote:On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. The very fact that NRGzack is willing to use an obvious blink hack means he simply has no caution about getting caught. What is your standard for judging careful hackers? How can we surmise that someone is hacking except through examining subtle inconsistencies in the games in question vs the ones on ladder as we have done with Spades? It could be the case that there really is no way to tell when someone is hacking, and that we'll just have to deal with that until Blizzard figures out how to make it less of a liability. IMO, the accusations are just way too weighty and devastating to be made on such little evidence. Yes, I'm aware of the fact that we probably won't get a super conclusive proof no matter what, but we might just have to bit the bullet and say "yeah, it's possible to hack sufficiently well that we can't discern it from plausibly innocent play." There's nothing shameful about that.
If we had a large enough pool of games where he was suspected hacking vs. a large enough pool of games where he was confirmed not hacking, there should theoretically be a way of statistically analyzing behaviour patterns that is definitively reliable. Whether this is at all practical or not is another question completely.
|
On June 07 2012 09:57 Shiori wrote:Show nested quote +On June 07 2012 09:55 EtherealDeath wrote:On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. Actually his play is really fucking shady lol, all technical-related "evidence" aside. Some parts are shady, some parts are really not. Initially when I read the OP's post I was convinced that it was all shady, but pretty much all of it has plausible explanations. Even Illusion's analysis rested rather heavily on some particular siege moments which have been shown to have plausible motives.
I don't want you to take offense to this, but what league are you? I mean, what is the highest league in starcraft 2 that you have EVER been?
You make a lot of posts declaring this evidence isn't damning, when myself, someone who has played maybe 5k games of master level of every race, including numerous games versus hackers, conclusive.
I would just like to know what basis of experience you are basing your opinion. My guess is very little to none other than your very opinionated beliefs.
|
On June 07 2012 09:59 Shiori wrote:Show nested quote +On June 07 2012 09:57 StarStrider wrote:On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. The very fact that NRGzack is willing to use an obvious blink hack means he simply has no caution about getting caught. What is your standard for judging careful hackers? How can we surmise that someone is hacking except through examining subtle inconsistencies in the games in question vs the ones on ladder as we have done with Spades? It could be the case that there really is no way to tell when someone is hacking, and that we'll just have to deal with that until Blizzard figures out how to make it less of a liability. IMO, the accusations are just way too weighty and devastating to be made on such little evidence. Yes, I'm aware of the fact that we probably won't get a super conclusive proof no matter what, but we might just have to bit the bullet and say "yeah, it's possible to hack sufficiently well that we can't discern it from plausibly innocent play." There's nothing shameful about that.
Are you kidding me? Lol if all this is "little evidence" Then I would love to see a case that got alot of evidence. Jesus crist man how much more do you need?
|
On June 07 2012 09:59 jacksonlee wrote:Show nested quote +On June 07 2012 09:57 StarStrider wrote:On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote:On June 07 2012 09:22 RezJ wrote: Guys. GUYS. Spades might not have used the hack you found with google. Incredible, right?
Drop that argument already. Hacks are not limited to what you read on some forum or readme. You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. The very fact that NRGzack is willing to use an obvious blink hack means he simply has no caution about getting caught. What is your standard for judging careful hackers? How can we surmise that someone is hacking except through examining subtle inconsistencies in the games in question vs the ones on ladder as we have done with Spades? Apparently, in SC1, the hackers did not click on the enemy unit once for 15 games. We don't have anything like that here
So what do we have?
All I am hearing is "too bad, so sad, nothing can be done to indite a hacker. Hackers, if you're not using an obvious tell like auto production or auto blink, then you can't be caught, so hack away with no worries!"
|
On June 07 2012 10:01 caradoc wrote:Show nested quote +On June 07 2012 09:59 Shiori wrote:On June 07 2012 09:57 StarStrider wrote:On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote: [quote]
You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote: [quote]
You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. The very fact that NRGzack is willing to use an obvious blink hack means he simply has no caution about getting caught. What is your standard for judging careful hackers? How can we surmise that someone is hacking except through examining subtle inconsistencies in the games in question vs the ones on ladder as we have done with Spades? It could be the case that there really is no way to tell when someone is hacking, and that we'll just have to deal with that until Blizzard figures out how to make it less of a liability. IMO, the accusations are just way too weighty and devastating to be made on such little evidence. Yes, I'm aware of the fact that we probably won't get a super conclusive proof no matter what, but we might just have to bit the bullet and say "yeah, it's possible to hack sufficiently well that we can't discern it from plausibly innocent play." There's nothing shameful about that. If we had a large enough pool of games where he was suspected hacking vs. a large enough pool of games where he was confirmed not hacking, there should theoretically be a way of statistically analyzing behaviour patterns that is reliable. Whether this is at all practical or not is another question completely.
There'd probably need to be a crapload of variables controlled here to do a proper statistical analysis with a low sigma.
|
On June 07 2012 09:59 StarStrider wrote:Show nested quote +On June 07 2012 09:57 Shiori wrote:On June 07 2012 09:55 EtherealDeath wrote:On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote: [quote]
You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote: [quote]
You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. Actually his play is really fucking shady lol, all technical-related "evidence" aside. Some parts are shady, some parts are really not. Initially when I read the OP's post I was convinced that it was all shady, but pretty much all of it has plausible explanations. Even Illusion's analysis rested rather heavily on some particular siege moments which have been shown to have plausible motives. Each incident has plausible explanations....by itself. They would be plausible if they were isolated and sparse. They are too frequent in all the same series to be just dumb luck and 'trying a different way of playing' compared to the ladder control group. But let's be honest: probably like 80% of the initially marked "shady moments" were actually not shady at all. You could have a billion of them in succession and it still wouldn't be shady. Look at the sheer number of so-called questionable actions called to attention in this thread. How many of them were not summarily disregarded by the first pro to walk in? Maybe 5? Even Illusion's analysis is pretty dismissive of a lot of the accusations, chalking them up to bad play or mistakes.
So yeah, we have some shady shady moments like the weird siege on CK and that moment on TDA, but I wouldn't say we have enough jarringly shady moments to make a conclusion like the one you're saying, especially since now a decent rationalization of the CK siege is being developed. I think we need to come to terms with the fact that we basically have not got enough evidence to convict Spades. We can't say for certain that he's innocent, but that is the default position.
|
On June 07 2012 10:01 JustTray wrote:Show nested quote +On June 07 2012 09:57 Shiori wrote:On June 07 2012 09:55 EtherealDeath wrote:On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote: [quote]
You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote: [quote]
You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. Actually his play is really fucking shady lol, all technical-related "evidence" aside. Some parts are shady, some parts are really not. Initially when I read the OP's post I was convinced that it was all shady, but pretty much all of it has plausible explanations. Even Illusion's analysis rested rather heavily on some particular siege moments which have been shown to have plausible motives. I don't want you to take offense to this, but what league are you? I mean, what is the highest league in starcraft 2 that you have EVER been? You make a lot of posts declaring this evidence isn't damning, when myself, someone who has played maybe 5k games of master level of every race, including numerous games versus hackers, conclusive. I would just like to know what basis of experience you are basing your opinion. My guess is very little to none other than your very opinionated beliefs. I'm mid Masters as Protoss.
|
On June 07 2012 09:45 Shiori wrote:Show nested quote +On June 07 2012 09:44 playa wrote: I haven't read a big percentage of the thread, but I do know someone was accusing spades of stream hacking. If that's true, who does that? People that are comfortable with hacking. Cheating is cheating. Good players rely on their edge of being better than other players... Anyone that cheats should simply be irrelevant. If you played iccup during its first season, you know even the worst players can achieve number 1 and stomp some of the best players in the non Korean scene, when cheating.
If someone still cheats, they're not good and shouldn't be talked about. Find out the truth and move on. I personally don't even know why they get a second chance. The ones that reform and actually become good is probably so small, considering anyone can be "good" if they cheat. Plus, good players don't need to, even if they didn't have morals. I just hate the idea of considering one form of cheating lesser than another. Even so, both should lead to the same result.
It's worth noting that the same posts which condemns Spades for streamcheating also indicts the entire Korean scene, all of whom have shown that they are not necessarily cheaters when it counts.
The Korean thing was corrected as a communication mistake due to language barrier. A lot of people were questioning specifically that part of his statement from the beginning because all he said was "I heard from someone..." On Spades, those two guys were Spade's manager and teammate who lived with Spades for a while. They even said that they knew specifically which games Spade streamhacked in.
|
On June 07 2012 10:01 JustTray wrote:Show nested quote +On June 07 2012 09:57 Shiori wrote:On June 07 2012 09:55 EtherealDeath wrote:On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote: [quote]
You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote:On June 07 2012 09:25 jacksonlee wrote: [quote]
You're doing exactly what I said in my post. It is unreasonable to accuse someone of one thing that he did, and when that's debunked, you conveniently say, "well that's because tha hxors are so 1337 they can do anything and hide every hacking activity." At what point do we draw the line and say, "well, our theory was wrong, let's look somewhere else." I"m talking strictly about the whole FOW argument Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. Actually his play is really fucking shady lol, all technical-related "evidence" aside. Some parts are shady, some parts are really not. Initially when I read the OP's post I was convinced that it was all shady, but pretty much all of it has plausible explanations. Even Illusion's analysis rested rather heavily on some particular siege moments which have been shown to have plausible motives. I don't want you to take offense to this, but what league are you? I mean, what is the highest league in starcraft 2 that you have EVER been? You make a lot of posts declaring this evidence isn't damning, when myself, someone who has played maybe 5k games of master level of every race, including numerous games versus hackers, conclusive. I would just like to know what basis of experience you are basing your opinion. My guess is very little to none other than your very opinionated beliefs.
I would also be interested in this. Would everyone simply state their race and league and games played on their home realm real quick? Just so we can get an idea of the level of understanding we have of the game?
Terran Rank44 Masters NA 3200+ games played.
|
On June 07 2012 10:03 Shiori wrote:Show nested quote +On June 07 2012 10:01 JustTray wrote:On June 07 2012 09:57 Shiori wrote:On June 07 2012 09:55 EtherealDeath wrote:On June 07 2012 09:54 Shiori wrote:On June 07 2012 09:52 StarStrider wrote:On June 07 2012 09:42 jacksonlee wrote:On June 07 2012 09:40 RezJ wrote:On June 07 2012 09:30 toiletCAT wrote:On June 07 2012 09:28 RezJ wrote: [quote] Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Maybe this, maybe that, maybe the Flux Capacitor. Nice try, what you're saying is "if he's hacking he must be using [known-maphack], and [known-maphack] doesn't work that way, therefore it doesn't make sense!", and I'm saying "X occurs, therefore a hack which causes X is likely involved". On June 07 2012 09:29 jacksonlee wrote:On June 07 2012 09:28 RezJ wrote: [quote] Let me explain then: I'm saying he might have been able to lock his camera without being as limited as the public hacks are. Then how is FOW an evidence that he's hacking, if, as you propose, the so-called hack can cover all grounds in regards to FOW. Not sure I understand what you're asking, but the fact that the fog never comes into his vision could very well be used as evidence for cheating (correct me if I'm wrong). Is there an inconsistency here? o.o But my whole argument was that he DOES in fact look at FOW, whether his units are selected or not He looks at the FOW verrrry infrequently and not in normal ways as defined by the control group of his ladder games (and 95% of other pro's ways of looking at FOW). For instance, in his ladder games, and in any other foreign or korean pro game, the screen is scrolled into FOW to look at enemy base frequently. The screen is scrolled around just outside your army location, followed by scan drops. The lack of scouting in these games, the lack of normal screen movements, the fact that a scan drops THEN he looks at it, the fact that he blindly moves his army time and again into watchtowers and risky positions that just happened to not contain a siege line seven games in a row.....it just doesn't add up. For each fishy incident you can come back and say 'well since we can't definitively prove this isn't just a coincidence in normal play we can chock that up to luck or gosu senses'. HOW MANY TIMES CAN YOU SAY THIS before you must start to question whether these things can happen back to back over and over in several different series? How many inconsistencies can you see in these items as compared to the normalities of his ladder pack mechanics before you start to say it can't be mere coincidences that many times in a row? Where is the line people? How many more fishy incidents do you need? We have 7 games with fishy incidents here, and 3 games of incidents there, then a few random 1v1's where similar weird things happened. What is the number? 20? 30? How many games with unexplainable coincidences and uncharacteristic mechanics does it take before you stop giving him the benefit of the doubt? Or is there just never ever enough suspicious activity that will make up your mind, until you see it with your own eyes or are giving the positive results of a nonexistent hack scanner finding it to be true? Short of him confessing which it's obvious he has no intention of doing, you must rely on the circumstantial evidence that has been provided And then again my question becomes: where is that line? How much more shady shit do you need to see before you decide? How many more games? Say a number. I'd say if I see another series in the future with the same inexplicable behaviour which simultaneously contrasts with his performance at MLG/playstyle there, I'd be more inclined to call him a hacker. As it is, though...it's not really conclusive at all. It's not even close to it, honestly. He doesn't look NEARLY as shady as that Zack guy, even forgoing the obvious Blink hacks. Actually his play is really fucking shady lol, all technical-related "evidence" aside. Some parts are shady, some parts are really not. Initially when I read the OP's post I was convinced that it was all shady, but pretty much all of it has plausible explanations. Even Illusion's analysis rested rather heavily on some particular siege moments which have been shown to have plausible motives. I don't want you to take offense to this, but what league are you? I mean, what is the highest league in starcraft 2 that you have EVER been? You make a lot of posts declaring this evidence isn't damning, when myself, someone who has played maybe 5k games of master level of every race, including numerous games versus hackers, conclusive. I would just like to know what basis of experience you are basing your opinion. My guess is very little to none other than your very opinionated beliefs. I'm mid Masters as Protoss.
Fair enough, thank you. Continue on.
|
|
|
|