Hi! I’m trying to find out information on power consumption for the Tx2/Tx2i when in Linux idle, for example. And when using just one of the Denver or A57 cores. Our application is mostly dormant, or doing simple things like monitoring inputs, needing high performance only eventually. I have a Tx1 here, but it’s sadly a -100 revision, and the INAs are, apparently, not connected (I do get readings, but bogus, and then I read it has to be a -300 revision to properly work).
While waiting for my Tx2i, I’d like to know the power consumption when not using GPUs, and also, how much of the SoM can/must be turned off (I figured out there’s a big difference between the CPU power readings from INA and total system power consumption, even when not using GPUs, right?)
The best report I found is https://www.phoronix.com/scan.php?page=article&item=jetson-tegra-x2&num=2
I’m also trying to understand the same for the Xavier AGX, and I found even less information on that. While I could buy both an measure, power requirements when doing simple linuxy stuff is a selection criteria for me, and I’d love to be able to choose based on that before buying.
Also, I don’t quite understand the differences in power consumption between Tx2 and Tx2i. I guess most of the difference (15W vs 20W) is because the Tx2i can operate at higher temperatures where it needs to dissipate (and consume?) more. But is there other differences in power consumption, maybe due to differences in the components to support industrial temperature ranges?
Any help? pointers?