Question on Forum Rules

15791011

Comments

  • killahKlown
    killahKlown Posts: 583 Critical Contributor
    I finally saw the video. However, that doesn't prove the accused is cheating directly. All these are circumstantial evidences. The logic seems to go like this:

    1) Player A whose roster isn't as good as mine is clearing pve faster than me.
    2) It's impossible that he is clearing faster than me because my roster is stronger than his; therefore, he's using speed hack.
    3) I have video evidences of myself speed hacking the game; therefore, it's true that the player is using speed hack.

    Also, I find the idea that low level champed 4* can't beat 5* team in clearing pves frightening. I was getting top 10, and sometimes top 5 in scl 7 and 8 pves when I was in 4* land. If the dev believes the above idea, I and other 4* players who got top 10 in pves regularly, without sniping brackets, would be banned eons ago. Afterall, top 10 placements in scl 7 and above are dominated by 5* players.
    This is all well and good, but frankly, we’re not talking about CL7 and 8 PvE. Honestly, not even CL9 PvE either. You can clear those quickly with Polaris and some 3stars. We’re talking top-end competitive CL10 play here. I have a decent sized A team in CL10 PvE with a 500+ Okoye and I can sleepwalk to a T25 placement. Even with that, I’m nowhere close to clearing as fast as some of the top-end rosters I’ve been in brackets with, such as some of the commenters on this thread and some of my ally-mates. Hard to say this without sounding pompous or whatever, but I’m fairly good at PvE. They’re just bigger, and bigger is faster. Yet, somehow myself (and other rosters of similar size bigger and smaller than me) are consistently being out-paced by much smaller rosters (450s-mid 460s) who are finishing ~right behind~ 550 rosters? I just can’t believe that. I’ve been in CL10 since the beginning and I know the clear times for 550s and 460s, as I was a 460 player when CL10 opened and I’m constantly discussing PvE strategies with 550s. There’s just simply not enough top-end punch to be clearing these mega-health nodes that quickly and consistently. Do I have absolute proof that these players are speed-hacking? No. But we know overclocking exists, and a lot of these ~questionable~ rosters just so happen to be Steam players. I’ve submitted tickets in the past. Numerous times. However, I don’t flood them with inquiries, I allow reasonable time for due process of investigation. And what has been the outcome of such inquiries? Bupkus. Crickets. So while I’m not as extreme about swinging the banhammer personally, I do think it’s absolutely a necessity that we at least have some communication of the topic from the Dev team. Nuking threads and pretending this doesn’t exist is not only not the correct procedure, it’s a slap in the face to anybody who’s taken the time to file reports under the correct procedure and has allowed them to proceed at their own pace to investigate these claims. When you respond with absolute silence, the aggrieved parties will react with louder and louder responses, as we’ve been seeing on the forums and Reddit lately. 
    Maybe those of us who finish right behind the top 550 rosters are just better and smarter at the game than you?  

    It's dangerous to start pointing your accusing finger at everyone who in your expert opinion should not be placing better than you.
  • Southside84
    Southside84 Posts: 93 Match Maker
    I finally saw the video. However, that doesn't prove the accused is cheating directly. All these are circumstantial evidences. The logic seems to go like this:

    1) Player A whose roster isn't as good as mine is clearing pve faster than me.
    2) It's impossible that he is clearing faster than me because my roster is stronger than his; therefore, he's using speed hack.
    3) I have video evidences of myself speed hacking the game; therefore, it's true that the player is using speed hack.

    Also, I find the idea that low level champed 4* can't beat 5* team in clearing pves frightening. I was getting top 10, and sometimes top 5 in scl 7 and 8 pves when I was in 4* land. If the dev believes the above idea, I and other 4* players who got top 10 in pves regularly, without sniping brackets, would be banned eons ago. Afterall, top 10 placements in scl 7 and above are dominated by 5* players.
    This is all well and good, but frankly, we’re not talking about CL7 and 8 PvE. Honestly, not even CL9 PvE either. You can clear those quickly with Polaris and some 3stars. We’re talking top-end competitive CL10 play here. I have a decent sized A team in CL10 PvE with a 500+ Okoye and I can sleepwalk to a T25 placement. Even with that, I’m nowhere close to clearing as fast as some of the top-end rosters I’ve been in brackets with, such as some of the commenters on this thread and some of my ally-mates. Hard to say this without sounding pompous or whatever, but I’m fairly good at PvE. They’re just bigger, and bigger is faster. Yet, somehow myself (and other rosters of similar size bigger and smaller than me) are consistently being out-paced by much smaller rosters (450s-mid 460s) who are finishing ~right behind~ 550 rosters? I just can’t believe that. I’ve been in CL10 since the beginning and I know the clear times for 550s and 460s, as I was a 460 player when CL10 opened and I’m constantly discussing PvE strategies with 550s. There’s just simply not enough top-end punch to be clearing these mega-health nodes that quickly and consistently. Do I have absolute proof that these players are speed-hacking? No. But we know overclocking exists, and a lot of these ~questionable~ rosters just so happen to be Steam players. I’ve submitted tickets in the past. Numerous times. However, I don’t flood them with inquiries, I allow reasonable time for due process of investigation. And what has been the outcome of such inquiries? Bupkus. Crickets. So while I’m not as extreme about swinging the banhammer personally, I do think it’s absolutely a necessity that we at least have some communication of the topic from the Dev team. Nuking threads and pretending this doesn’t exist is not only not the correct procedure, it’s a slap in the face to anybody who’s taken the time to file reports under the correct procedure and has allowed them to proceed at their own pace to investigate these claims. When you respond with absolute silence, the aggrieved parties will react with louder and louder responses, as we’ve been seeing on the forums and Reddit lately. 
    Maybe those of us who finish right behind the top 550 rosters are just better and smarter at the game than you?  

    It's dangerous to start pointing your accusing finger at everyone who in your expert opinion should not be placing better than you.
    Skill is defintely a factor.   I usually open about 25-30th in speed, but can usually grind my way into top 20.  I'm easily one of the smallest rosters finishing T20.  I could see people looking at my roster and saying "he's definitely cheating."  The top players all seem to finish their grind with under 1 minute left in the sub.  Every additional point helps at that point.  While hacks need to be fixed, I agree that speculation shouldn't be the deciding factor on whether or not someone gets sandboxes or banned.
  • Chipster22
    Chipster22 Posts: 299 Mover and Shaker
    It has been suggested that this hack only affects the very top players.  I'm not sure I agree with that.  If someone plays more casually but still cheats with the hack they might get into the top 100 or top 50 when they wouldn't have otherwise.

    That would mean people who finished 51st or 101st would be robbed of rewards.  

    Actually, regardless of where the cheaters place, they are pushing everyone else below them down in the rankings from where they should have finished.
  • DAZ0273
    DAZ0273 Posts: 10,275 Chairperson of the Boards
    You might finish last in a 100m sprint but if the guy who finished first cheated then the whole thing is tainted.
  • killahKlown
    killahKlown Posts: 583 Critical Contributor
    It has been suggested that this hack only affects the very top players.  I'm not sure I agree with that.  If someone plays more casually but still cheats with the hack they might get into the top 100 or top 50 when they wouldn't have otherwise.

    That would mean people who finished 51st or 101st would be robbed of rewards.  

    Actually, regardless of where the cheaters place, they are pushing everyone else below them down in the rankings from where they should have finished.
    Good point, but it will be a lot harder to identify those that cheat their way to the middle of the pack.  
  • HoundofShadow
    HoundofShadow Posts: 8,004 Chairperson of the Boards
    At least The rockett is honest or detailed about what happened. It also show that something was done about cheater, with at least one getting boxed, and it's not true that nothing was done by the developer. It's unfortunate that he was unboxed, which is probably due to the flawed logic of the codes. And it's probably due to this flawed programming code that caused a longer than expected time to solve the problem.

    While it's true that the CS said that the video is a hack, but you still can't tie it down to the cheater conclusively.

    I want to address one of the points about the hack being possibly used in scl 8 and 9. I'm not sure what placement we are talking about here: top 10, top 5 or top 1. If it's the first two, by the logic put forth about how non champed 5* or 4* players shouldn't be able to beat 5* players in clearing nodes quickly, my personal experience is that it's not true. I was getting top 5 and top 10 consistently in SCL 8 as a 4* player, without bracket sniping, and without cheating. Once in a while, I get top 1 because 5* players screwed up their wave nodes. Potentially, I could be labeled as one of the cheaters because I was beating 5* players as a 4* player. 


  • The rockett
    The rockett Posts: 2,016 Chairperson of the Boards
    It has been suggested that this hack only affects the very top players.  I'm not sure I agree with that.  If someone plays more casually but still cheats with the hack they might get into the top 100 or top 50 when they wouldn't have otherwise.

    That would mean people who finished 51st or 101st would be robbed of rewards.  

    Actually, regardless of where the cheaters place, they are pushing everyone else below them down in the rankings from where they should have finished.
    Good point, but it will be a lot harder to identify those that cheat their way to the middle of the pack.  
    What we actually found out is some people that we suspected started to wait for flips instead of in 1st brackets. Flips are so much harder to hit perfectly, as you know.  2-4 mins late on a flip will kill sub 1 and could destroy the whole event.  
  • helix72
    helix72 Posts: 996 Critical Contributor
    edited March 2021
    There are two questions at play here:

    A ) Specifically, what is or is not considered a violation of the Terms of Service (ToS)?
    B ) What is being done to discover, prevent, and punish people who are violating the ToS?

    A lot of the conversation has jumped to question B, but (and maybe I'm the only one) I still don't have a clear answer to question A. So to the ToS:
    2. You shall not use the Services to develop, host, or distribute cheats, automation software (bots), modded lobbies, hacks, mods or any other unauthorized third-party software in connection with the Services, or engage in any form of cheating, boosting, or booting.
    For example, is there a list somewhere of what is considered unauthorized software? I assume playing in iOS on my iPad is authorized, but I don't see that spelled out anywhere. Is there a definition of boosting?  Again, don't see it spelled out anywhere.

    All I'm asking for at the moment is an answer to question A. I would like the devs to be very specific with:
    1. a list of what hardware/software/settings is/are permitted (so anything not on the list is a violation), AND/OR
    2. a list of what hardware/software/settings is/are not permitted (so anything on the list is a violation)
    And the ability to freely discuss for clarification purposes on the forums items that are not clearly spelled out as either 1 or 2. I have sent a note to Customer Service asking a similar question, we'll see what kind of response I get.
  • Bad
    Bad Posts: 3,146 Chairperson of the Boards
    I noticed from few months ago it is a bit harder to rank t100 on pve scl10 but I thought it was for more people adquiring okoye and entering scl10. 
    Possibly it could be them. 
    Obviously on pvp that could apply too, being able to hit several times faster with a hulkoye team. 
    If that is right then great damage could be done to pvp added to the superiority of hulkoye teams. 
  • The_Boogie
    The_Boogie Posts: 25 Just Dropped In
    At least The rockett is honest or detailed about what happened. It also show that something was done about cheater, with at least one getting boxed, and it's not true that nothing was done by the developer. It's unfortunate that he was unboxed, which is probably due to the flawed logic of the codes. And it's probably due to this flawed programming code that caused a longer than expected time to solve the problem.

    While it's true that the CS said that the video is a hack, but you still can't tie it down to the cheater conclusively.

    I want to address one of the points about the hack being possibly used in scl 8 and 9. I'm not sure what placement we are talking about here: top 10, top 5 or top 1. If it's the first two, by the logic put forth about how non champed 5* or 4* players shouldn't be able to beat 5* players in clearing nodes quickly, my personal experience is that it's not true. I was getting top 5 and top 10 consistently in SCL 8 as a 4* player, without bracket sniping, and without cheating. Once in a while, I get top 1 because 5* players screwed up their wave nodes. Potentially, I could be labeled as one of the cheaters because I was beating 5* players as a 4* player. 


    Oh I think you make a completely valid point. There is certainly the potential and even likelihood that some are using these same speed hacks to move themselves into t1/5/10/20/50/100 in CL 8 and 9 in addition to what is happening at the top of CL 10. It just stands out more at the top of CL 10, and that's common sense. If two people were cheating at golf - one is cheating to get a score of even par and another is getting a hole in one on every hole, most people would notice the second person before the first.
  • DeNappa
    DeNappa Posts: 1,396 Chairperson of the Boards
    Bad said:
    I noticed from few months ago it is a bit harder to rank t100 on pve scl10 but I thought it was for more people adquiring okoye and entering scl10. 
    Possibly it could be them. 
    Obviously on pvp that could apply too, being able to hit several times faster with a hulkoye team. 
    If that is right then great damage could be done to pvp added to the superiority of hulkoye teams. 
    Still, would the use of this hack/exploit really be that widespread? I made the same observation in the past months (I never play for placement, but without very significant changes in gameplay style I noticed I'd gone from t200 to t300 placement), but always thought this was due to shardpocalypse generating an influx of SCL10 players.
  • Beefhammer69
    Beefhammer69 Posts: 58 Match Maker
    Maybe those of us who finish right behind the top 550 rosters are just better and smarter at the game than you?  

    It's dangerous to start pointing your accusing finger at everyone who in your expert opinion should not be placing better than you.
    In no way was I implying that my clear/grind times were the be all/end all and that I should be T1 every event. That’s ludicrous and a deliberate misrepresentation of what I was saying. Of course there’s people that are better/smarter at the game than I am. However, I can use my experiences as a benchmark and it’s very unlikely that there’s a deluge of players who are so much more skilled/optimal at PVE than I am that it would outweigh the much lower roster strength that they have comparatively. I can’t confirm anything because I haven’t taken the time to extensively research the issue such as previous commenters have, but I can say that it had certainly piqued my interest as suspicious behavior before I had even known the speed hack was a thing. That’s all. 
  • Phumade
    Phumade Posts: 2,496 Chairperson of the Boards
    DeNappa said:
    Bad said:
    I noticed from few months ago it is a bit harder to rank t100 on pve scl10 but I thought it was for more people adquiring okoye and entering scl10. 
    Possibly it could be them. 
    Obviously on pvp that could apply too, being able to hit several times faster with a hulkoye team. 
    If that is right then great damage could be done to pvp added to the superiority of hulkoye teams. 
    Still, would the use of this hack/exploit really be that widespread? I made the same observation in the past months (I never play for placement, but without very significant changes in gameplay style I noticed I'd gone from t200 to t300 placement), but always thought this was due to shardpocalypse generating an influx of SCL10 players.
    It really does explain a lot.  I have seen my rankings falter.  I just asssumed it’s because people got bigger rosters, but seems clear they installed something
  • ThaRoadWarrior
    ThaRoadWarrior Posts: 9,454 Chairperson of the Boards
    anecdotally I'd been feeling the same thing about getting edged out of where I usually fall playing SCL10 to full progression. And that's not me crying about placement, I've been openly a progression-only player since forever, just me saying I noticed the way I'd been playing had dropped out of t200 with no changes on my part. If enough people jump over the "regular" players, the "true" tWhatever starts below however many of them are doing it. It's not a 51/101 problem if you have 10,20, 100 players playing like this, right?
  • jp1
    jp1 Posts: 1,081 Chairperson of the Boards
    anecdotally I'd been feeling the same thing about getting edged out of where I usually fall playing SCL10 to full progression. And that's not me crying about placement, I've been openly a progression-only player since forever, just me saying I noticed the way I'd been playing had dropped out of t200 with no changes on my part. If enough people jump over the "regular" players, the "true" tWhatever starts below however many of them are doing it. It's not a 51/101 problem if you have 10,20, 100 players playing like this, right?
    I’ve noticed this as well. I kind of shrugged it off, placement rewards aren’t really my bread and butter anyway...but it does feel a little annoying to think it’s due to others cheating.
  • Southside84
    Southside84 Posts: 93 Match Maker
    I think all the okoye stores plus the prevalence of the BRB/Polaris winfinte options are pushing more people into SCL10.  When you get half of a 5 star cover for progression alone, it's a no brainer to jump up if you were not T10 in SCL9
  • ThaRoadWarrior
    ThaRoadWarrior Posts: 9,454 Chairperson of the Boards
    I noticed it when the grind went down to 4x, figured it was just that.
  • HoundofShadow
    HoundofShadow Posts: 8,004 Chairperson of the Boards
    edited March 2021
    2x, you meant?

    The theory about more players having bigger Okoyes sounds plausible, which in turn leads to players jumping up to SCL 10. And who are the one asking players to hoard for Okoye?   :D

    The last few times I jumped into SCL 10, my rankings were either high 2x or mid 3x and my 5* are all 45X.
  • ThaRoadWarrior
    ThaRoadWarrior Posts: 9,454 Chairperson of the Boards
    edited March 2021
    To me it’s 4x instead of 6x lol; PVE is just a grind for resources to me.

    also this idea that we are being swamped by a wave of new Okoye havers is comical to me.
  • HoundofShadow
    HoundofShadow Posts: 8,004 Chairperson of the Boards
    edited March 2021
    We had massive shards injection in December last year and we had Okoye store in October 2020 and last month. So, it's plausible we had a wave of new Okoye players, and higher levelled Okoye players.