All Networks Does Cubecraft even have an anti-Hack?!? Well, I think I may have a way to make it insanely better

[JCBlueHornet]

Guys, so people keep mentioning an anti-hack, but is there really one? shouldn't an anti-cheat be able to detect a player flying across the map and ban them, or the cant computer can just check how far away players are from their opponents when getting hit, or even just sense that when people are being hit, their character doesn't move back, so its anti-kb. Isn't that what other servers do? Why doesn't cubecraft use this? I could see that it could cause hella lag if it runs all the time, but maybe if players a high enough level to prove they aren't brand new (like let's say level 10+) then when they report someone an anti-cheat bot could monitor them for the next 1-2 minutes and take in all the data to see if they are cheating. This could cause lag with spam reports, but we could limit each player (above level 10+) to 1 report for game, players 50+ to 4, and players 100+ to ten. then if the anti-cheat bot figures they are hacking they kick them, which would at lease remove them from the game, without the problem of them being banned. Then if someone is being multiply kicked from games and reported (5+ times) then it would be put on top priority for a mod to check. This would work because it would allow all reports to be checked, the bot can then kick them from the game removing the problem for the reporters and doubling as a warning to the cheater, and then the moderator can come and ban them if the report gets moved up to that high, which would allow mods to have a lot less reports, and discourage hackers who would have to keep rejoining. Lmk what yall think!

 

[TijmenDeEngeMan]

Yeah I know what you mean honestly. Its especially annoying when you can tell someone is clearly hacking. But then again, its a robot. It's never as perfect and accurate as a moderator. That's what the report system is for. But I do get the frustration it gives when Sentinel won't detect it as frequent as it should.

Also there shouldn't be prioritized reports. What if, for example, someone gets reported like 50 times for no reason because of some "joke", and they get prioritized over someone who is actually breaking the rules. People would abuse that system just for laughs probably.

And for the level cap to report shouldn't be necessary. Anyone who spots a hacker or someone who shows inappropriate behaviour should have the option to report that person. It wouldn't be fair to low level players if there was a minimum level required.

 

[Sauud]

Guys, so people keep mentioning an anti-hack, but is there really one? shouldn't an anti-cheat be able to detect a player flying across the map and ban them, or the cant computer can just check how far away players are from their opponents when getting hit, or even just sense that when people are being hit, their character doesn't move back, so its anti-kb. Isn't that what other servers do? Why doesn't cubecraft use this? I could see that it could cause hella lag if it runs all the time, but maybe if players a high enough level to prove they aren't brand new (like let's say level 10+) then when they report someone an anti-cheat bot could monitor them for the next 1-2 minutes and take in all the data to see if they are cheating. This could cause lag with spam reports, but we could limit each player (above level 10+) to 1 report for game, players 50+ to 4, and players 100+ to ten. then if the anti-cheat bot figures they are hacking they kick them, which would at lease remove them from the game, without the problem of them being banned. Then if someone is being multiply kicked from games and reported (5+ times) then it would be put on top priority for a mod to check. This would work because it would allow all reports to be checked, the bot can then kick them from the game removing the problem for the reporters and doubling as a warning to the cheater, and then the moderator can come and ban them if the report gets moved up to that high, which would allow mods to have a lot less reports, and discourage hackers who would have to keep rejoining. Lmk what yall think!

This is what hackers are looking for because they get very strong protection for their hacks but I havent seen a hack in CubeCraft that wasnt banned before the end of the game

 

[Siza]

The anitcheat (Sentinel) is always being further developed. If you'd like to know a bit more about Sentinel, you can read about it here:

Behind the Cube #2 - Sentinel

Hello CubeCrafters! Welcome to the second edition of "Behind the Cube"! We hope these development blogs can give you guys some insight into what we are doing behind the scenes. In this edition, we will be discussing the development of Sentinel...

www.cubecraft.net

This might give some insight on how Sentinel works and why hackers don't always get instantly banned. It's a cat and mouse game ;)

To add to that, kicking suspected hackers would feel like a votekick system to me, which is when multiple players can vote to kick a certain player from the game. There is a risk that this will get abused if implemented.

 

[Sauud]

The anitcheat (Sentinel) is always being further developed. If you'd like to know a bit more about Sentinel, you can read about it here:

Behind the Cube #2 - Sentinel

Hello CubeCrafters! Welcome to the second edition of "Behind the Cube"! We hope these development blogs can give you guys some insight into what we are doing behind the scenes. In this edition, we will be discussing the development of Sentinel...

www.cubecraft.net

This might give some insight on how Sentinel works and why hackers don't always get instantly banned. It's a cat and mouse game ;)

To add to that, kicking suspected hackers would feel like a votekick system to me, which is when multiple players can vote to kick a certain player from the game. There is a risk that this will get abused if implemented.

Well said I agree!

 

[JCBlueHornet]

The anitcheat (Sentinel) is always being further developed. If you'd like to know a bit more about Sentinel, you can read about it here:

Behind the Cube #2 - Sentinel

Hello CubeCrafters! Welcome to the second edition of "Behind the Cube"! We hope these development blogs can give you guys some insight into what we are doing behind the scenes. In this edition, we will be discussing the development of Sentinel...

www.cubecraft.net

This might give some insight on how Sentinel works and why hackers don't always get instantly banned. It's a cat and mouse game ;)

To add to that, kicking suspected hackers would feel like a votekick system to me, which is when multiple players can vote to kick a certain player from the game. There is a risk that this will get abused if implemented.

Thanks so much for the link, it does help me understand better. I don't think this would add a vote kick system. I was saying that when someone reported it would be more like the sentinel then monitored the player who was reported, not that the player who was reported would be kicked. then if the sentinel catches the player the sentinel kicks them, and if the player gets caught multiple times, the sentinel sends a high priority ping to the moderators to come look at the player.

Yeah I know what you mean honestly. Its especially annoying when you can tell someone is clearly hacking. But then again, its a robot. It's never as perfect and accurate as a moderator. That's what the report system is for. But I do get the frustration it gives when Sentinel won't detect it as frequent as it should.

Also there shouldn't be prioritized reports. What if, for example, someone gets reported like 50 times for no reason because of some "joke", and they get prioritized over someone who is actually breaking the rules. People would abuse that system just for laughs probably.

And for the level cap to report shouldn't be necessary. Anyone who spots a hacker or someone who shows inappropriate behaviour should have the option to report that person. It wouldn't be fair to low level players if there was a minimum level required.

If a player got reported a ton, each time the player is reported the sentinel would keep an eye on them for a short period of time, collecting data to determine if they are cheating. then if the sentinel determines they are cheating, the sentinel kicks them. so, unless the player is caught cheating multiple times (by the sentinel) then the reports don't get moved up to a high priority for a mod, otherwise the reports just cause the sentinel to be watching someone a ton, just to determine they aren't cheating. the only way for a high priority to get to the mod is if someone is caught cheating by the bot multiple times, I said this because I don't want false bans by the bot.

As for level caps, I don't want tons of false reports, cause I worry it may cause lag (I have no clue about in game stuff so this part is free to be corrected by someone smart) I worry that when the sentinel collects so much data from a player that it could cause lag, so I don't want the sentinel to cause tons of lag with everyone spam reporting(and if a player is reported and the bot is investigating it would send a message like, "Ongoing investigation, try again later" this way the players who play more often (and are more trusted, probably not trolls) can actually report multiple hackers, without trolls being able to spam the system. I would have the system use a different type of report so that way everyone can use the old system, for example people wanting to report to the sentinel could use "/s report"

This is what hackers are looking for because they get very strong protection for their hacks but I havent seen a hack in CubeCraft that wasnt banned before the end of the game

Well I have, me and my friend Og true pain were in a match against 4 hackers, and they got our egg and killed me. and I had to wait 45 MINUTES for Og true pain to slowly and painstakingly kill them all.

 

[Siza]

Thanks so much for the link, it does help me understand better. I don't think this would add a vote kick system. I was saying that when someone reported it would be more like the sentinel then monitored the player who was reported, not that the player who was reported would be kicked. then if the sentinel catches the player the sentinel kicks them, and if the player gets caught multiple times, the sentinel sends a high priority ping to the moderators to come look at the player.

On Java, there’s already a command for this, /sr. When used, Sentinel will 'pay closer attention' to the player. If the player is actually hacking, Sentinel will issue a ban rather than kicking them from the game.
There’s also a /report command, which alerts Moderators that a player is suspected of hacking. A Moderator can then spectate the player to verify the report. I’m not aware of any kind of 'priority system' for these reports, but I'd assume that all reports are handled with equal priority. :)

 

[TijmenDeEngeMan]

If a player got reported a ton, each time the player is reported the sentinel would keep an eye on them for a short period of time, collecting data to determine if they are cheating. then if the sentinel determines they are cheating, the sentinel kicks them. so, unless the player is caught cheating multiple times (by the sentinel) then the reports don't get moved up to a high priority for a mod, otherwise the reports just cause the sentinel to be watching someone a ton, just to determine they aren't cheating. the only way for a high priority to get to the mod is if someone is caught cheating by the bot multiple times, I said this because I don't want false bans by the bot.

As for level caps, I don't want tons of false reports, cause I worry it may cause lag (I have no clue about in game stuff so this part is free to be corrected by someone smart) I worry that when the sentinel collects so much data from a player that it could cause lag, so I don't want the sentinel to cause tons of lag with everyone spam reporting(and if a player is reported and the bot is investigating it would send a message like, "Ongoing investigation, try again later" this way the players who play more often (and are more trusted, probably not trolls) can actually report multiple hackers, without trolls being able to spam the system. I would have the system use a different type of report so that way everyone can use the old system, for example people wanting to report to the sentinel could use "/s report"

Usually when a false ban happens, its either a game bug which makes them seem like a hacker, or some bug within Sentinel. As I said before, Sentinel is a robot and will never be as accurate as a human moderator. It's bound to make mistakes and that's why it's always being improved as time goes on.
And even IF you get false banned by Sentinel, you can always contact the mods to appeal it.

Also, I understand that you worry that a lot of reports might cause lag, then what if the reports get disposed of when it's been checked by Sentinel or a mod. Once they've taken action, the report data will be deleted and there will be more space for other reports. That way a level cap wouldn't be necessary. But I don't know how frequent people get reported on Bedrock though, so I don't know how fast that would pile up on a day.
And I don't know if Bedrock has it, but on Java we have /sr (/sentinel report). Which basically is the monitoring function you suggested. It would look at the reported player's behaviour and check if they're cheating or not.

 

[Thunder]

I want to clear up some misunderstandings here.

First of all, we have an Anticheat (Sentinel) for a very long time and it is constantly being updated. However, it is a cat and mouse game between the Anticheat and Cheat client developers. That's why a perfect Anticheat will never exist. You can read this Behind the Cube if you are interested in learning more information about Sentinel: https://www.cubecraft.net/threads/behind-the-cube-2-sentinel.270731/

Thanks so much for the link, it does help me understand better. I don't think this would add a vote kick system. I was saying that when someone reported it would be more like the sentinel then monitored the player who was reported, not that the player who was reported would be kicked. then if the sentinel catches the player the sentinel kicks them, and if the player gets caught multiple times, the sentinel sends a high priority ping to the moderators to come look at the player.

Sentinel does not work like that. It always monitors every player online in the network. If a player triggers sentinel, a ban may be applied. And, moderation team is already being alerted about suspicious players who are detected by Sentinel as you can see in the Behind the Cube #2. I don't think kicking cheaters would benefit us. If a cheater is caught, they should be banned. Also, Sentinel shouldn't be making decisions based on player reports. Otherwise, it would be a vote-kick system, which is unnecessary. If you are suggesting to make a system to alert the moderation team about the most reported players, we already have this system. We have a Sentinel report command (/sr) and we receive your reports made using this command if the same player is reported several times.

On Java, there’s already a command for this, /sr. When used, Sentinel will 'pay closer attention' to the player. If the player is actually hacking, Sentinel will issue a ban rather than kicking them from the game.
There’s also a /report command, which alerts Moderators that a player is suspected of hacking. A Moderator can then spectate the player to verify the report. I’m not aware of any kind of 'priority system' for these reports, but I'd assume that all reports are handled with equal priority. :)

And I don't know if Bedrock has it, but on Java we have /sr (/sentinel report). Which basically is the monitoring function you suggested. It would look at the reported player's behaviour and check if they're cheating or not.

/sr command has nothing to do with the Anticheat system. It is an in-game report command which is used to send a report to the moderation team. Unlike /report, it requires several players to report the same player. /sr command exists on both networks and /report command exists only on Java.

 

[TijmenDeEngeMan]

/sr command has nothing to do with Sentinel. It is an in-game report command which is used to send a report to the moderation team. Unlike /report, it requires several players to report the same player. /sr command exists on both networks and /report command exists only on Java.

Oh I thought it did. I thought it was an abbreviation. My bad D:

 

[JCBlueHornet]

I want to clear up some misunderstandings here.

First of all, we have an Anticheat (Sentinel) for a very long time and it is constantly being updated. However, it is a cat and mouse game between the Anticheat and Cheat client developers. That's why a perfect Anticheat will never exist. You can read this Behind the Cube if you are interested in learning more information about Sentinel: https://www.cubecraft.net/threads/behind-the-cube-2-sentinel.270731/


Sentinel does not work like that. It always monitors every player online in the network. If a player triggers sentinel, a ban may be applied. And, moderation team is already being alerted about suspicious players who are detected by Sentinel as you can see in the Behind the Cube #2. I don't think kicking cheaters would benefit us. If a cheater is caught, they should be banned. Also, Sentinel shouldn't be making decisions based on player reports. Otherwise, it would be a vote-kick system, which is unnecessary. If you are suggesting to make a system to alert the moderation team about the most reported players, we already have this system. We have a Sentinel report command (/sr) and we receive your reports made using this command if the same player is reported several times.



/sr command has nothing to do with the Anticheat system. It is an in-game report command which is used to send a report to the moderation team. Unlike /report, it requires several players to report the same player. /sr command exists on both networks and /report command exists only on Java.

...Well, I obviously don't understand the anti-cheat.

Well let's take a situation where someone is reaching, and call him Dave

Dave fights George

in their fight Dave hits George from 5 blocks away

ok so the sentinel can't just measure the distance by recording both players location at the time it was hit and then ban the hacker?!

How does the sentinel miss when people do that?! or when someone is playing with 30cps with no breaks in between (obviously an autoclicker) or even just ban people with over 15 cps? or how does it miss when someone gets hit and doesnt take KB? could somone please explains this ToT

 

[Siza]

/sr command has nothing to do with the Anticheat system. It is an in-game report command which is used to send a report to the moderation team. Unlike /report, it requires several players to report the same player. /sr command exists on both networks and /report command exists only on Java.

I'm not sure as to why I thought /sr would alert Sentinel instead of Moderators. Thank you for correcting me. :)

 

[Thunder]

...Well, I obviously don't understand the anti-cheat.

Well let's take a situation where someone is reaching, and call him Dave

Dave fights George

in their fight Dave hits George from 5 blocks away

ok so the sentinel can't just measure the distance by recording both players location at the time it was hit and then ban the hacker?!

How does the sentinel miss when people do that?! or when someone is playing with 30cps with no breaks in between (obviously an autoclicker) or even just ban people with over 15 cps? or how does it miss when someone gets hit and doesnt take KB? could somone please explains this ToT

Sentinel is already being developed to detect unusual behaviours/use of cheat modifications. However, cheat clients are also being developed to bypass these detections. This continues in a cycle.

 

[JCBlueHornet]

Sentinel is already being developed to detect unusual behaviours/use of cheat modifications. However, cheat clients are also being developed to bypass these detections. This continues in a cycle.

oh ok, sorry about that Thuinder

 

[neptuneenchantment]

the anticheat is bad, you can fly, scaffold really fast (100 bps) etc

 

[thetigerpython]

the anticheat is bad, you can fly, scaffold really fast (100 bps) etc

Me when the anti cheat devs are lazy and there are full disablers:

Sentinel is already being developed to detect unusual behaviours/use of cheat modifications. However, cheat clients are also being developed to bypass these detections. This continues in a cycle.

On both java and bedrock there are much better anti cheats available. Even free anti cheats on java detect more than Sentinel with fewer false positives.

 

[Rabidsyllab]

Me when the anti cheat devs are lazy and there are full disablers:


On both java and bedrock there are much better anti cheats available. Even free anti cheats on java detect more than Sentinel with fewer false positives.

I ran into some hackers on bedrock, one got banned 30s after I died from them, one banned within the day. Its a robot, its not perfect and as Thunder said, mods are already notified if Sentinel thinks they are hacking, even if he doesn't ban them.

Also as Thunder says:
Sentinel is already being developed to detect unusual behaviours/use of cheat modifications. However, cheat clients are also being developed to bypass these detections. This continues in a cycle.

Plus just get a recording software(PC: Xbox Game Bar, Xbox: Built in recording software, PS: Must have something because Bico streams of it and Nintendo: up to 30s). Then just upload it to somewhere link Imgur or Youtube or other services like that and make a player report with the clip of that player cheating.