Is nvpmodel what you use on Xavier to switch modes? Are there any new modes? I think I remember reading somewhere you have 3 modes now and they are 3 different power levels is this correct?
Yes, if you run
nvpmodel -?
it will show usage info.
nvpmodel is preferred, although technically you could make a script like jetson_clocks.sh that uses the sysfs files that nvpmodel does.
There are 7 preset models IIRC. You can see them defined in /etc/nvpmodel.conf
There is a table that lists the frequencies and ID’s for the nvpmodel presets included in the L4T Documentation under ‘Clock Frequency and Power Management’ in the ‘Power Management for Jetson-Xavier Devices’ and ‘Max-Q and Max-P Power Efficiency’ section.
Thanks dusty. Loving this machine.
When using nvpmodel, is the fan controlled automatically, or should it’s speed be set manually like in the jetson_clocks.sh script?
The fan is controlled automatically with nvpmodel. jetson_clocks.sh disables frequency scaling and pegs the clocks at their maximum frequency regardless of workload, so the fan is enabled. nvpmodel sets the clock governor limits, so frequency scaling is still enabled and scales with workload, hence the fan may not always need to run.
Please also refer to latest L4T document (rel-31.0.1)
https://developer.nvidia.com/embedded/dlc/l4t-documentation-31-0-1
When I click this address, it says “Membership required”. Could I have this document? Or it is only for your company’s engineer？ I hava registered in the address. But it does not seems work.
File a new topic please… please do not use any topic which is already 3 years ago (2018) to ask new question.
Also, the link may change because it has been 3 years…