Poseidon
Poseidon copied to clipboard
sondernextdoor
Slow
Is it supposed to be very slow to read memory? Takes me about 3 seconds to read 64 addresses.
^^^ If anybody finds the issue do let me know Are you using a KProcess object or direct read?
^^^ If anybody finds the issue do let me know Are you using a KProcess object or direct read?
Sorry for the late reply. I’ve completely deleted the Kprocess class I only use direct read. Have you found one to be quicker?
there’s a couple sleeps in the switch routine for the driver loop but those are mostly for routines where the driver is inactive and even removing them just to be sure did no help. It may just be the hook itself.
Note: mine is a 32bit process. It’s just slow enough to shoot right after I cross over somebody’s body on the Triggerbot I’ve made.
I was just wondering as I thought the KProcess might be slower, however they appear to be the same speed. Looping through entity list in CS:Go gets me 1 frame every 3 seconds of ESP.
^^^ If anybody finds the issue do let me know Are you using a KProcess object or direct read?
Sorry for the late reply. I’ve completely deleted the Kprocess class I only use direct read. Have you found one to be quicker?
there’s a couple sleeps in the switch routine for the driver loop but those are mostly for routines where the driver is inactive and even removing them just to be sure did no help. It may just be the hook itself.
Note: mine is a 32bit process. It’s just slow enough to shoot right after I cross over somebody’s body on the Triggerbot I’ve made.
Hey, did you manage what is the performance bottleneck? I really consider if it is not the MmCopyVirtualMemory itself? Every call internally attaches to the process so maybe this is slow?
Hey, did you manage what is the performance bottleneck? I really consider if it is not the MmCopyVirtualMemory itself? Every call internally attaches to the process so maybe this is slow?
I believe it has more to do with the communication method than the actual copying code. I don't believe there's anyway to resolve this issue other than implementing an entirely different communication method. Perhaps something from here should help guide: https://github.com/adspro15/km-um-communication
there is a way to make it fast enough, just create more memory pools and remove 99% of the used sleeps, because if u sum them up u will notice a big delay between the requesting and response, also use 2 different memory pools, one for the request and the other one for the response.
single call takes 40 ms even after removing all sleeps. Anyone who has a reference implementation that is faster so I can get a better understanding why exactly it is so slow?
