Hard crashing PC locks up

"
KissFC#2626 написал:
After getting tired of my personnal computer getting freeze with a 100% CPU usage, i got the great idea of putting and trying to see what happens on my ES debug station, which have an out of band CPU thread and memroy analyzer (namelly a debug and software profiling station).

The way they're handling multi threading is threatened by modern standard, upgraded UEFI and well secured OSes as a core threat, thus stopping the suspicious threads marked as threat. Similar to how speculative/transient execution CPU vulnerabilities works.
But as the main program isn't flagged as a threat and is considered secure, the OS do not stop it from generating new threads, and the whole process works like a fork bomb.

At this point, i strongly suggest GGG to make a full review of their networking code and to enable fork bombing detection on their compiler because it is something that should never happens.



GGG, this is the post. ^
bump
😑
"
After getting tired of my personnal computer getting freeze with a 100% CPU usage, i got the great idea of putting and trying to see what happens on my ES debug station, which have an out of band CPU thread and memroy analyzer (namelly a debug and software profiling station).

The way they're handling multi threading is threatened by modern standard, upgraded UEFI and well secured OSes as a core threat, thus stopping the suspicious threads marked as threat. Similar to how speculative/transient execution CPU vulnerabilities works.
But as the main program isn't flagged as a threat and is considered secure, the OS do not stop it from generating new threads, and the whole process works like a fork bomb.

At this point, i strongly suggest GGG to make a full review of their networking code and to enable fork bombing detection on their compiler because it is something that should never happens.


Imagine having your community debug your game for you.

I'm just saying...
Woke up, watched the stream.
They didn't address the issue, only some NVIDIA driver-related crash to desktop.
And they replied to the other thread, but not this one.
Whoop-dee-doo!
This is a freaking joke. I would unnderstand if there was a sever issue within their networking code and they didnt know how to fix in until release as long as they would let us know about it. The silent treatment make me f*cking angry!

and what a f*cking joke of a intterview stream, makes me wonder if Ghazzy and DM were even allowed to bring up the multithreading issue or if thet were under some kind of NDA.

Lost my respect for Ghazzy last night cause the chat brought in up with him after the interview but he kept saying they asked and got answered by GGG and refused to acknowledge that they didnt say squat about it.
I played since day 1 without issues on an older processor.
i7 6700k with 4070 graka on win10.

I made an upgrade previous week to a 7700x amd processor and new mobo + windows 11.

And since then i also have these total locks that where i have to use the power button.

Only with POE2.

I played a lot of other games without these issues after the upgrade.
(stalker 2, ashes of creation, FF remake, Once human, Marvel rivals etc...)
"
KissFC#2626 написал:
After getting tired of my personnal computer getting freeze with a 100% CPU usage, i got the great idea of putting and trying to see what happens on my ES debug station, which have an out of band CPU thread and memroy analyzer (namelly a debug and software profiling station).

The way they're handling multi threading is threatened by modern standard, upgraded UEFI and well secured OSes as a core threat, thus stopping the suspicious threads marked as threat. Similar to how speculative/transient execution CPU vulnerabilities works.
But as the main program isn't flagged as a threat and is considered secure, the OS do not stop it from generating new threads, and the whole process works like a fork bomb.

At this point, i strongly suggest GGG to make a full review of their networking code and to enable fork bombing detection on their compiler because it is something that should never happens.

Bumping this.
Bump to wake them up.
"
KissFC#2626 написал:
After getting tired of my personnal computer getting freeze with a 100% CPU usage, i got the great idea of putting and trying to see what happens on my ES debug station, which have an out of band CPU thread and memroy analyzer (namelly a debug and software profiling station).

The way they're handling multi threading is threatened by modern standard, upgraded UEFI and well secured OSes as a core threat, thus stopping the suspicious threads marked as threat. Similar to how speculative/transient execution CPU vulnerabilities works.
But as the main program isn't flagged as a threat and is considered secure, the OS do not stop it from generating new threads, and the whole process works like a fork bomb.

At this point, i strongly suggest GGG to make a full review of their networking code and to enable fork bombing detection on their compiler because it is something that should never happens.


Based on the way they addressed the issue of crashes and freezes when asked about them by Darth Microtransaction in their interview yesterday, it doesn't seem like they are even looking at their own code as the culprit. They immediately kicked the ball to Nvidia and seem to be waiting to see if Nvidia's next driver update fixes all thing things. Nevermind the fact that there have been issues with POE1 as well over the years that have never been addressed.

They built POE2 on POE1's engine, netcode and all, and they don't seem too motivated to make any changes.

Windows 11 24H2 (Steam), AMD 5800X3D, RTX 4070 Ti Super 16GB, 32BG DDR4 3200, Samsung 980 Pro SSD

Пожаловаться на запись форума

Пожаловаться на учетную запись:

Тип жалобы

Дополнительная информация