Peak IPC

In [1], it is said that

The theoretical maximum peak IPC is a device limit and defined by the compute capabilities of the target device.

The web page for the CC as described in [2] has no such information as I saw. Maybe I missed something. Can someone explain?

[1] https://docs.nvidia.com/gameworks/content/developertools/desktop/analysis/report/cudaexperiments/kernellevel/instructionstatistics.htm
[2] https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#compute-capabilities