4 ms·

The alternative is that the online games they're protecting become infested with cheaters, players depart, and the game closes down because it's no longer a via

The alternative is that the online games they're protecting become infested with cheaters, players depart, and the game closes down because it's no longer a viable business. It's not nice, I don't like running a kernel mode anti-cheat any more than you do, but I can see why it's necessary for preserving the competitive integrity of free-to-play shooter games like Valorant.

imchillyb 1y ago
If one must give up kernel level security of their system to an untrusted 3rd party, was it really free?
- Hikikomori 1y ago
  Stuff you run as your user can already read all your files and memory of processes of that user so you're already very exposed.
__alexs 1y ago
They are already full of cheaters. DMA cards are undetectable even from kernel anti cheat.
- doikor 1y ago
  The DMA card still needs to be installed, have drivers installed, firmware, etc. The anti cheat tests for that (the cheaters do mask the device spoofing the name/vendor/etc). Having to make new driver/firmware every time the anti cheat starts to detect it is way slower then just new software. Though at some point they will probably just automate it so that every customer gets their own driver and firmware matched with only their cheat software making this way harder but we are not there yet (and you would have to get all of these signed so following the cert chain should make it easier to find at least whos stolen cert they are using) Also the amount of people willing to buy another pc and DMA cards is way smaller making the chance of running into cheater in your match smaller.
  - az09mugen 1y ago
    That's a good idea indeed.
  - __alexs 1y ago
    Giving every customer their own firmware is already automated. Computer vision based cheats are also rapidly on the rise. You don't get wall hacks but you do top 1% reaction times and perfect tracking.
    - fc417fc802 1y ago
      Elephant in the room is that any behavior based solution that can reliably detect computer vision based cheating obviates the need for kernel level anti-cheat because it more or less solved the detection problem.
- abigail95 1y ago
  are you writing your own firmware or hotwiring the bus? there are many methods to detect this.

imchillyb 1y ago

If one must give up kernel level security of their system to an untrusted 3rd party, was it really free?

Hikikomori 1y ago
Stuff you run as your user can already read all your files and memory of processes of that user so you're already very exposed.

__alexs 1y ago

They are already full of cheaters. DMA cards are undetectable even from kernel anti cheat.

doikor 1y ago
The DMA card still needs to be installed, have drivers installed, firmware, etc. The anti cheat tests for that (the cheaters do mask the device spoofing the name/vendor/etc). Having to make new driver/firmware every time the anti cheat starts to detect it is way slower then just new software. Though at some point they will probably just automate it so that every customer gets their own driver and firmware matched with only their cheat software making this way harder but we are not there yet (and you would have to get all of these signed so following the cert chain should make it easier to find at least whos stolen cert they are using) Also the amount of people willing to buy another pc and DMA cards is way smaller making the chance of running into cheater in your match smaller.
- az09mugen 1y ago
  That's a good idea indeed.
- __alexs 1y ago
  Giving every customer their own firmware is already automated. Computer vision based cheats are also rapidly on the rise. You don't get wall hacks but you do top 1% reaction times and perfect tracking.
  - fc417fc802 1y ago
    Elephant in the room is that any behavior based solution that can reliably detect computer vision based cheating obviates the need for kernel level anti-cheat because it more or less solved the detection problem.
abigail95 1y ago
are you writing your own firmware or hotwiring the bus? there are many methods to detect this.