Pace same on ORIN nano AND nano

Hi i use openGL / CUDA 11 Jetpack 5.1.2. on Jetson Orin Nano for a fluid simulation, why is the speed of the animation the same as on a Jetson Nano 4Gb ?
It should be 80 X faster , wouldn’ t it ?

How do i speed up the pace ?

Hi,

Have you maximized the device performance first?

$ sudo nvpmodel -m 0
$ sudo jetson_clocks

Thanks.

thanks for the quick reply, now where can i find the manual for nvpmodel, i found a document

(https://docs.nvidia.com/jetson/archives/r35.3.1/DeveloperGuide/text/SD/PlatformPowerAndPerformance/JetsonOrinNanoSeriesJetsonOrinNxSeriesAndJetsonAgxOrinSeries.html)

what i want to know is in which state is my platform now, and what happens if i nvpmodel -m 0 or do something with Jetson_clocks. It is not clear what the consequences are, nor how to return to the previous situation.
I see there is something of a RTOS kernel on my board ? maybe general information on how to tinker with speed issues would be very helpfull, especially complemented with a way to set it all back.

Hi,

nvpmodel and jetson_clock is a tool for Jetson to adjust the clock.
You can find more information in the below doc:
https://docs.nvidia.com/jetson/archives/r35.4.1/DeveloperGuide/text/SD/PlatformPowerAndPerformance/JetsonOrinNanoSeriesJetsonOrinNxSeriesAndJetsonAgxOrinSeries.html#supported-modes-and-power-efficiency

Mode ID 0 indicates the performance mode (10W for Orin Nano 4GB, 15W for Orin Nano 8GB).
The suggestion here is to compare two devices that are all in the maximal performance mode.

Thanks.

well off course i have done that, all boards are performing at max capacity, what is missing is a guide to tune the Jetsons.
thanks for your effort

Hi,

You can set a custom nvpmodel mode based on your requirements.
Usually, it is a tradeoff between power and performance.

Please see the below for the detailed steps:

Thanks.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.