All Networks Temporary Implementation of a Prediction Engine Until Sentinel is Fully Improved

[yurisatori0312]

I suggest a realistic, temporary solution to our severe cheating crisis: implementing a Prediction-based Anti-Cheat as a provisional measure until Sentinel is upgraded.

The Current Crisis:
With other networks adopting strict prediction engines, cheaters have migrated to CubeCraft. Sentinel is struggling with this influx, resulting in Fly, Speed, and Reach ruining games daily.

Why a Temporary Prediction Engine Makes Sense:

1. Provisional Defense: We know updating a custom anti-cheat like Sentinel takes time and resources, but the player base cannot wait months.
2. Immediate Relief for Players: A prediction engine will immediately filter out aggressive cheaters.
3. Time for Sentinel’s Development: This system will buy the development team time to properly code and test Sentinel without rushing unstable patches.

Conclusion:
This is not a request to abandon Sentinel, but an immediate shield until it is improved.

 

[Wqsknijper]

I'd love to see Sentinel improve, but I don't think a regular prediction AntiCheat does the job. I do believe they could (or should) check out GrimAC's reach check. I have used it on previous servers in the past and noticed it doesn't false flag generally, some exceptions (which can easily be filtered out). I have seen the speed/flight (simulation) checks generate false positives regularly. Seeing how (regular) false bans have a worse impact than some cheaters that can be reported both in game and via the website, I think they should just focus on Sentinel itself and we as players should keep reporting cheaters via /sr, /report and the website/forums.

 

[yurisatori0312]

I'd love to see Sentinel improve, but I don't think a regular prediction AntiCheat does the job. I do believe they could (or should) check out GrimAC's reach check. I have used it on previous servers in the past and noticed it doesn't false flag generally, some exceptions (which can easily be filtered out). I have seen the speed/flight (simulation) checks generate false positives regularly. Seeing how (regular) false bans have a worse impact than some cheaters that can be reported both in game and via the website, I think they should just focus on Sentinel itself and we as players should keep reporting cheaters via /sr, /report and the website/forums.

With all due respect, telling players to "just keep reporting via /sr and forums" ignores the reality of how overwhelmed the reporting system currently is.
I regularly record and submit clear evidence of blatant cheaters to the Bedrock Player Reports section. As you can see from my recent reports, many of them remain completely unreviewed and sitting open for more than 18 to 24+ hours.

While these manual reports wait days in a queue to be checked by a human staff member, those exact same cheaters are free to ruin dozens of games and drive away hundreds of legitimate players.
Manual reporting cannot scale with the massive migration of cheaters we are experiencing right now. We cannot rely on a slow report system to do the job that an automated, modern simulation layer should be doing instantly. That is precisely why an immediate, temporary anti-cheat upgrade is necessary.

 

[Mohammed Edries]

I suggest a realistic, temporary solution to our severe cheating crisis: implementing a Prediction-based Anti-Cheat as a provisional measure until Sentinel is upgraded.

The Current Crisis:
With other networks adopting strict prediction engines, cheaters have migrated to CubeCraft. Sentinel is struggling with this influx, resulting in Fly, Speed, and Reach ruining games daily.

Why a Temporary Prediction Engine Makes Sense:

1. Provisional Defense: We know updating a custom anti-cheat like Sentinel takes time and resources, but the player base cannot wait months.
2. Immediate Relief for Players: A prediction engine will immediately filter out aggressive cheaters.
3. Time for Sentinel’s Development: This system will buy the development team time to properly code and test Sentinel without rushing unstable patches.

Conclusion:
This is not a request to abandon Sentinel, but an immediate shield until it is improved.

Heyy @yurisatori0312 hope you good

It's great idea! Please allow me to support your proposal with an additional idea from my side that might make it even better, This concept is actually inspired by PUBG

My idea is to introduce new in-game report system, like new command for Ex /hax I don't find any command better than that now

However, this specific command can ONLY be used by players who frequently report cheaters here on the forums and have very high accuracy rate of correct reports.

Only these trusted members would have the privilege to use this command in-game.
Once they use it during match the report gets instantly linked to the forums and triggers live alarm/alert inside the game and on the forums for the moderators at the exact same time with the highest priority.

I believe on this way, Active cheaters can be caught and banned much faster and reduced the rate of false ban.

Best regards

 

[Legendary_op_30]

I suggest a realistic, temporary solution to our severe cheating crisis: implementing a Prediction-based Anti-Cheat as a provisional measure until Sentinel is upgraded.

The Current Crisis:
With other networks adopting strict prediction engines, cheaters have migrated to CubeCraft. Sentinel is struggling with this influx, resulting in Fly, Speed, and Reach ruining games daily.

Why a Temporary Prediction Engine Makes Sense:

1. Provisional Defense: We know updating a custom anti-cheat like Sentinel takes time and resources, but the player base cannot wait months.
2. Immediate Relief for Players: A prediction engine will immediately filter out aggressive cheaters.
3. Time for Sentinel’s Development: This system will buy the development team time to properly code and test Sentinel without rushing unstable patches.

Conclusion:
This is not a request to abandon Sentinel, but an immediate shield until it is improved.

This is actually genius!!!

 

[Wqsknijper]

I agree there are a lot of hackers, there should be less and yes indeed sometimes reports stay open for longer than they should.

With all due respect, telling players to "just keep reporting via /sr and forums" ignores the reality of how overwhelmed the reporting system currently...........

Here you're not at all responding to my worry that it will cause more false bans, which will directly cause players to lose interest in cubecraft and will then also overwhelm the report system. This should never be the direct cause of the anti-cheat. I find enough games where there are no cheaters and enough times there's staff online to actually ban a hacker.

Literally just while I am typing this, xThunder_ banned a hacker I came across a few minutes ago. Again, rather focus on updating Sentinel, but like Lyriie has mentioned in the thread below, you can't simply 'plug and play' a random AntiCheat, false bans are just too risky and it's a constant cat and mouse game anyway:

Anticheat
I can’t comment on the state of our anticheat specifically as I’m not a developer, but I can confirm that we do have an anticheat (Sentinel), and we also have an anticheat developer whose job it is to update it. However, there are a few general things that I know based on what developers have said in the past.

First of all, developing an anticheat is essentially a cat and mouse game. This means that, whenever Sentinel gets updated, cheat client developers are quick to create bypasses for it, meaning it needs to get updated again, and the process repeats. So Sentinel will never ever reach a ‘finished’ state; it’s going to need constant development for as long as CubeCraft exists. Additionally, every update that Sentinel goes through has to be thoroughly tested first, and even then there’s still a risk of it causing false bans when the update goes out, which obviously isn’t ideal.

Furthermore, Sentinel is custom built for CubeCraft, and Hive’s anticheat is custom built for them. It’s not as easy as saying ‘just implement [server]’s anticheat,’ because what works for them would most likely not be compatible with most of our gamemodes.

I will fully admit that Sentinel certainly needs some work, and I don’t think any other team member would deny this at this point. But it’s not as simple as the team ‘refusing’ to develop the anticheat. I myself have no coding abilities and very little technical understanding of anticheats or how they are developed, so I prefer not to make judgments like this.

I am not willing to appeal a false ban every now and then, and I've had a few of them actually.. Of which one wasn't even accepted.. So please just work on improving Sentinel as a whole.

 

[xWalin]

Hi! @yurisatori0312.

I think you idea is very good but I don't know how cheap it is to develop a prediction engine from scratch.

Maybe using an existing one could be a solution but some of these plugins prohibit the usage for comercial purposes (as we know Cubecraft uses Minecraft as his main bussines model).

I don't know how far is Sentinel from a prediction anticheat. I think they are very similar.

 

[arni9136]

I'd love to see Sentinel improve, but I don't think a regular prediction AntiCheat does the job. I do believe they could (or should) check out GrimAC's reach check. I have used it on previous servers in the past and noticed it doesn't false flag generally, some exceptions (which can easily be filtered out). I have seen the speed/flight (simulation) checks generate false positives regularly. Seeing how (regular) false bans have a worse impact than some cheaters that can be reported both in game and via the website, I think they should just focus on Sentinel itself and we as players should keep reporting cheaters via /sr, /report and the website/forums.

even if it doesnt ban 100% of the hackers, it would remove atleast like 25-65% of hackers. and thats better than nothing

 

[Wqsknijper]

even if it doesnt ban 100% of the hackers, it would remove atleast like 25-65% of hackers. and thats better than nothing

I think you're missing the fact that there is an active anticheat and most hackers do actually get banned by the Sentinel eventually. Also, 'even if it doesn't ban 100% of the hackers' is a very rough statement, since no anticheat ever will ban 100% off the hackers without banning atleast 20% of legit players. And like I stated, and which actually was my main point, Sentinel is built to decrease false bans as much as possible. Cubecraft doens't have regular games, which means an anticheat that wasn't custom build will 100% generate (a lot of) false bans. That will never be the right solution.