Hi Nvidia,
We’re testing the Orin NX 16GB Super Mode full loading.
We chang the DC input to 19V on Devkit and nvpmodel.conf set to Super Mode.
Then we follow the website to run the CPU and GPU Stress test.
https://elinux.org/Jetson/L4T/TRT_Customized_Example#GPU_Stress_Test
But the power loading can’t up to 40W. It’s just 33~35W
Module : Orin NX 16GB
Carrier Board: Orin Nano Devkit
JetPack 6.2 (L4T36.4.3)
How is the tegrastats result when you run this?
$ tegrastats
02-26-2025 16:34:17 RAM 3073/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@70.093C cpu@76.031C soc2@67.625C soc0@69.031C cv1@71.187C gpu@79.468C tj@79.468C soc1@65.031C cv2@63.843C VDD_IN 32157mW/32157mW VDD_CPU_GPU_CV 19734mW/19734mW VDD_SOC 5234mW/5234mW
02-26-2025 16:34:18 RAM 3073/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@70.093C cpu@76.343C soc2@67.718C soc0@69.281C cv1@71.343C gpu@79.875C tj@79.875C soc1@65.5C cv2@64.093C VDD_IN 32007mW/32082mW VDD_CPU_GPU_CV 19734mW/19734mW VDD_SOC 5237mW/5236mW
02-26-2025 16:34:19 RAM 3073/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@70.437C cpu@76.343C soc2@68.125C soc0@69.437C cv1@71.468C gpu@80.125C tj@80.125C soc1@65.531C cv2@64.25C VDD_IN 32021mW/32062mW VDD_CPU_GPU_CV 19585mW/19684mW VDD_SOC 5237mW/5236mW
02-26-2025 16:34:20 RAM 3076/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@70.812C cpu@76.75C soc2@68.25C soc0@69.812C cv1@71.843C gpu@80.281C tj@80.281C soc1@65.812C cv2@64.5C VDD_IN 32007mW/32048mW VDD_CPU_GPU_CV 19734mW/19697mW VDD_SOC 5237mW/5236mW
02-26-2025 16:34:21 RAM 3073/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@71.062C cpu@77.187C soc2@68.5C soc0@70.218C cv1@72.187C gpu@80.593C tj@80.593C soc1@66.218C cv2@64.781C VDD_IN 31871mW/32013mW VDD_CPU_GPU_CV 19585mW/19674mW VDD_SOC 5237mW/5236mW
02-26-2025 16:34:22 RAM 3074/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@71.281C cpu@77.781C soc2@68.75C soc0@70.25C cv1@72.562C gpu@80.718C tj@80.718C soc1@66.375C cv2@64.937C VDD_IN 32007mW/32012mW VDD_CPU_GPU_CV 19734mW/19684mW VDD_SOC 5237mW/5237mW
02-26-2025 16:34:23 RAM 3074/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@71.5C cpu@77.531C soc2@69.062C soc0@70.625C cv1@72.562C gpu@81.312C tj@81.312C soc1@66.531C cv2@65.25C VDD_IN 32157mW/32032mW VDD_CPU_GPU_CV 19734mW/19691mW VDD_SOC 5237mW/5237mW
02-26-2025 16:34:24 RAM 3074/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@71.687C cpu@77.875C soc2@69.312C soc0@71C cv1@72.968C gpu@81.562C tj@81.562C soc1@66.812C cv2@65.5C VDD_IN 32157mW/32048mW VDD_CPU_GPU_CV 19875mW/19714mW VDD_SOC 5085mW/5218mW
02-26-2025 16:34:25 RAM 3074/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@71.843C cpu@78.281C soc2@69.468C soc0@71.218C cv1@73.281C gpu@81.843C tj@81.843C soc1@67.218C cv2@65.906C VDD_IN 31980mW/32040mW VDD_CPU_GPU_CV 19717mW/19715mW VDD_SOC 5080mW/5202mW
02-26-2025 16:34:26 RAM 3074/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@72.156C cpu@78.218C soc2@69.781C soc0@71.343C cv1@73.531C gpu@82.187C tj@82.187C soc1@67.25C cv2@65.906C VDD_IN 32129mW/32049mW VDD_CPU_GPU_CV 19867mW/19730mW VDD_SOC 5080mW/5190mW
02-26-2025 16:34:27 RAM 3074/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@72.562C cpu@78.593C soc2@70.156C soc0@71.531C cv1@73.781C gpu@82.718C tj@82.718C soc1@67.5C cv2@66.062C VDD_IN 32157mW/32059mW VDD_CPU_GPU_CV 19734mW/19730mW VDD_SOC 5085mW/5181mW
02-26-2025 16:34:28 RAM 3074/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@72.718C cpu@78.843C soc2@70.375C soc0@71.875C cv1@74.25C gpu@82.562C tj@82.562C soc1@67.718C cv2@66.406C VDD_IN 32157mW/32067mW VDD_CPU_GPU_CV 19734mW/19731mW VDD_SOC 5087mW/5173mW
02-26-2025 16:34:29 RAM 3074/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@72.718C cpu@79.25C soc2@70.437C soc0@72.156C cv1@74.406C gpu@82.812C tj@82.812C soc1@68.125C cv2@66.718C VDD_IN 31871mW/32052mW VDD_CPU_GPU_CV 19742mW/19731mW VDD_SOC 5087mW/5166mW
02-26-2025 16:34:30 RAM 3074/15656MB (lfb 9x4MB) SWAP 0/7828MB (cached 0MB) CPU [100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984,100%@1984] GR3D_FREQ 99% cv0@73.062C cpu@79.468C soc2@70.687C soc0@72.218C cv1@74.812C gpu@83.437C tj@83.437C soc1@68.25C cv2@66.812C VDD_IN 32021mW/32050mW VDD_CPU_GPU_CV 19734mW/19732mW VDD_SOC 5087mW/5161mW
In another topic it states that you cannot enable super mode on the orin nx with the nvidia jetson orin nano dev kit. Or it at least seems to imply that if you can actually enable it on the software side, it may be unstable or possibly damage the board.
since you questioned what the tegrastats results were instead of saying that the super mode for orin NX of the orin nano super developer kit is not allowed, i’m confused as to if we can actually enable the super mode on the Orin nx 16GB on the developer kit or not.
You can file your own topic for this. No need to ask in other user’s case.
But for your question, a quick answer here is both Orin NX and Nano are able to run in super mode. It is just previous jetpack6.1 does not support all of them and now Jetpack6.2 can.
Hi Wayne,
For this topic,
Do you have any ideas?
Hi Wilson,
You could push the GPU loading first and then add the CPU loading one by one per core but not every core all at once.
The key here is you cannot push over 40W. That will trigger throttling and even give you lower power consumption.
flashing Orin NX 8G to L4T R36.4.3, set power mode to MAXN, however GPU frequency is till Maximized at 765MHz instead of 1.2GHz.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.
