So, not really sure where to post this, but after implementing an update to my games randomgen system that adds more areas (some with more parts/models), every time I’ve attempted to join the game I get met with:
Error Code: 288, Channel: zmimallocgpuoom.
In all honestly I’m not sure what to make of this, and I can’t find anything online that mentions a Channel following a 288 error code. Is my randomgen system so large it’s causing an OOM GPU crash on the Roblox server?
I’ve made sure that the issue is isolated to the new randomgen areas, rolling back the game (but keeping the areas) did nothing, as soon as the areas are removed it works again, everything in them is anchored and there are no scripts added with them. This problem isn’t only on my client, as it happens for my QA testers as well.
There isn’t any new code being added with the new areas, it’s all running on the same central randomgen script, it’s just the areas have more content in them than the older ones they’re replacing (more parts, models).
As far I know, error 288 happens when the game shuts down a server and players are in It, and they get this error.
However, this specific error has another context.
The GPU issues you had are not random, the gpuoom part of the Channel indicates the GPU running out of memory.
Roblox used your Memory also because It Is simulating rendering and physics.
However, this error Is not common and if It happens more often, you should consider posting It in Bug Reports
This is typical when you’re kicked on a debug channel that isn’t LIVE. Any debug channel will show this, it has nothing to do with the GPU not having enough memory.
I’m an IT professional so I’m used to Mimalloc = memory allocation, and GPU OOM = GPU out of memory. Do you have any clue why this may be happening if it’s not a memory error?
