Today theyve formally revealed that the new Jetson Nano 2GB will go on sale later on this month for just $59. Lets take a close look at this new model of the Nano to see whats altered (and what hasnt) from last years model.
The TK1 was an impressive piece of hardware, but not something the hacker and maker neighborhood was necessarily interested in. For something, it was fairly pricey. However maybe more significantly, it was plainly tailored more towards industry types than consumers. We did see the periodic task using the TK1 and the subsequent TX1 and TX2 boards, however they were rare.
Came the Jetson Nano. According to the companys own figures, the number of active Jetson designers has more than tripled because the Nanos introduction in March of 2019.
NVIDIA started their line of GPU-accelerated single board computers back in 2014 with the Jetson TK1, a $200 USD advancement system for those wanting to get involved with the growing world of so-called “edge computing”. It was developed to put high performance computing in a little and energy efficient enough plan that it might be integrated directly into products, rather than connecting to an information center half-way throughout the world.
Cutting the Fat
To be clear the new Jetson Nano 2GB is not a brand-new device, its essentially just an expense optimized version of the hardware that was launched back in 2019. Specifically because the most significant modification isnt noticeable: as the name suggests, the brand-new design just has 2 gigabytes of RAM compared to four in the initial Nano.
The board has lost a few ports as part of the effort to get it down to half the initial cost. The Nano 2GB drops the DisplayPort for HDMI (the previous variation had both), deletes the second CSI electronic camera connector, eliminates the M. 2 slot, and lowers the variety of USB ports from four to 3. Losing a USB port most likely isnt a deal breaker for most applications, however if you require high-speed data, its worth noting that only one of them is 3.0. Overall, it seems clear that NVIDIA took a close take a look at the sort of devices that folks were connecting to their Nano and changed the type and number of ports appropriately.
The 2GB removes the old school DC barrel jack and replaces it with a USB-C port. On the initial Nano you might run it off of the micro USB port for most tasks, but it was recommended to use a laptop design power supply if you were going to be pressing the hardware.
Of course, the 40 pin header on the side stays the same so the new board needs to remain pin-compatible with anything youve currently constructed. The Gigabit Ethernet port is still there, but sadly cordless still didnt make the cut this time around. So if you need WiFi for your task, rely on one of those USB ports being completely taken up with a dongle.
A Tight Squeeze
Considering that the hardware is nearly similar between the 2 versions of the Nano, theres truly no point running any new standards on it. If your software application worked on the $99 Nano, it will run just as well on the $59 one.
To help combat this, NVIDIA recommends disabling the GUI on the Nano 2GB and running headless if youre preparing on doing any computationally extensive jobs. Itll be intriguing to see if some lightweight operating system options are offered down the line to assist resolve this issue.
NVIDIA sent me a review system so as an easy test, I ran the detectnet.py script that makes up part of NVIDIAs AI training course on the live video from a Logitech C270 electronic camera. While the Nano kept a decent 22 to 24 frames per 2nd, the system ran out of RAM almost instantly and needed to dip into swap to keep up. Naturally this is pretty problematic on an SD card, and definitely not something you d wish to do for any prolonged amount of time unless you happen to own SanDisk stock.
Increase of the Machines
Its probably not fair to call the Jetson Nano 2GB a direct competitor to the Raspberry Pi, but plainly NVIDIA desires to close the gap. While the lack of integrated WiFi and Bluetooth will likely offer lots of makers pause, theres no concern that the Nano will run circle the Pi 4 if youre looking to experiment with things like computer system vision. At $99 that might not have mattered for budget-conscious hardware hackers, today that the Nano is basically the very same cost as the mid-range Pi 4, its going to be a harder choice to make.
Especially because NVIDIA is using the release of the new board to assist kick off the Jetson AI Certification Program. This totally free self-paced source is made up of tutorials and video walkthroughs that cover everything from the principles of training up to practical applications like collision avoidance and object following. To finish the Jetson AI Specialist course and be approved the certification, candidates will need to submit an open source project to NVIDIAs Community Projects online forum for review and approval.
To help fight this, NVIDIA advises disabling the GUI on the Nano 2GB and running headless if youre planning on doing any computationally intensive tasks. Its most likely not fair to call the Jetson Nano 2GB a direct competitor to the Raspberry Pi, however clearly NVIDIA wants to close the gap.
Picking up a Jetson Nano at $99 was a fantastic option if you desired to get your feet damp with AI and maker knowing. Now that theres a $59 variation that includes access to a training and accreditation program, theres hardly even an option left to make. Which in the end, is exactly what NVIDIA wants.
Today theyve officially announced that the brand-new Jetson Nano 2GB will go on sale later this month for simply $59. To be clear the brand-new Jetson Nano 2GB is not a new device, its basically just an expense enhanced variation of the hardware that was released back in 2019. In general, it appears clear that NVIDIA took a close look at the sort of devices that folks were connecting to their Nano and adjusted the type and number of ports appropriately.
