So I have been unable to truly understand the semantics behind, would help some clarification.
Below is one of my RTX Ada 6000 Gen GPU
Timestamp : Mon May 11 00:57:11 2026
Driver Version : 580.95.05
CUDA Version : 13.0
Attached GPUs : 4
GPU 00000000:01:00.0
Temperature
GPU Current Temp : 44 C
GPU T.Limit Temp : 47 C
GPU Shutdown T.Limit Temp : -7 C
GPU Slowdown T.Limit Temp : -2 C
GPU Max Operating T.Limit Temp : 0 C
GPU Target Temperature : 85 C
Memory Current Temp : N/A
Memory Max Operating T.Limit Temp : N/A
Now as per the semantics [1] i understood till now, my GPU will undergo software throttle ( min clock shave) at exactly 44 + 47 = 91 aka Max Operating Temp [2], while the Target Temp is 85, it doesnot slows down, the acceptable interval which GPU likes to keep as min,max is {85,91}. After which we consider +2 = 93 as the HW slowdown threshold (more clock cut or safe state) while at a margin of +7 i,e 98. it should shutdown. The issue is however the discrepency in the data sheet by bluechip [2] and my understanding since based on the data sheet it says 100 as slowdown and 105 as shutdown,
what am i missing?
[1] = Understanding Optimal GPU Temperature and Default GPU Fan Curve (NVIDIA RTX 6000 Ada)
[2] = https://www.bluechip.de/media/86/92/28/1726579902/Datenblatt_98216.pdf?ts=1726579902