Review of the new NVIDIA Jetson Xavier NX Single Board Computer

You guys can help me out over at Patreon, and that will help me keep my gear updated, and help me keep this quality content coming:
/ paulmcwhorter
In this video we review a major new single board super computer and take it through its paces.
Pick up this Jetson Xavier NX on Amazon HERE:
amzn.to/3dKGNMY
SSD For the Jetson Xavier NX
amzn.to/3btqbYN
NVIDIA Product Page
nvda.ws/3bqcNEx
LInks below are for my work on the Jetson Nano
Jetson Nano
amzn.to/2Z4SGG4
You need a GOOD power supply:
amzn.to/2yWKrkF
You need a GOOD SD card:
amzn.to/2Ktf8nT
I love this cool Acrylic case with fan:
amzn.to/2H4ar1R
Wireless Keyboard and Mouse that works with Jetson Nano:
amzn.to/2YWVkxu
I like to run with minimum wires, so I am running this wifi card, which works very well. If your budget is tight just use an Ethernet cable, as all my lessons will work with just an ethernet cable.
amzn.to/2H26b2R
Finally, you can use either a Logitech C920 WEB cam or the Pi Cam. I am using both, but you need at least one camera working with the nano:
amzn.to/2H4dPd0
amzn.to/2KI5UTW
[Disclosure of Material Connection: I am a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to amazon.com. This means if you visit the link and purchase the item, I will receive an affiliate commission. Regardless, I only recommend products or services I use personally and believe will add value to my readers.]
#JetsonXavierNX
#AI
#Review

[@andrewbeaton3302]

I'M BLOWN AWAY!!! THANK YOU SO MUCH FOR TEACHING ALL OF US!

[@wallymurray620]

I had no idea stuff like this was out for the home hobbyist. I’m about half way through the arduino series of videos and this presentation just makes me want to get the arduino video series done, fully understood and absorbed then get working on the jetson nano video series. Thanks again Paul for all your work.

[@KamalSehairi]

Best teacher ever. I started running tensor RT on my nano also.

[@MonsieurSquidward]

You helped me so much with the GPIO-Button Tutorial.- Thank you for that. You are the best teacher in the World

[@richleyden6839]

Well thought out and informative review. I really like the overview format where you show the elements of your program without doing the programing on screen. Seeing many possible end points in one episode is motivating.

[@paulmcwhorter]

I am really excited about the possibilities. I will have my hands full improving my skills to more fully utilize the capability of this board.

[@FreddiMuliantono]

I really love your enthusiasm!

[@ysiegel5544]

Boom! Thanks for the really great review, very interesting and well done. New to the channel and love it! This board is amazing!

[@peterwooldridge7285]

Wow, that's impressive. Thanks Paul

[@teamplus9362]

I bought an arduino kit an started watching your tutorials and theyre awesome

[@DrDave327]

Great video, Paul!

[@jamespeterson166]

Thanks again for your excellent work.

[@BUBBLETEABOY]

Thank you for everything you do!

[@gerokatseros]

this boom apsets me everytime! good review

[@frobbit30]

Sir this video and all your work is inspiring, thank you for so much teaching in so enjoyable a format! I wanted to get into Fusion360 and python for some art purposes and thought I was lost until I found your channel, thank you!

[@craigphillips1034]

I would like to thank you so much for your video series. Last Friday I received a new Jetson NX and was lost the first day. I then found this video on the NX and it all came together. Thank you. I also appreciate that you did not come out with the standard box cutter and opening the box. Instead, your brought to all of us a hands-on tutorial on how we can do it ourselves. I did take your examples from the video and was able to replicate the blue marker cap example and it worked great. This was the first time that I was actually able to do something useful with the NX. Thank you. I hope that you will have more on the NX too. We'll be watching.

[@paulmcwhorter]

Have you looked at my Jetson Nano series? All of those lessons will work on the Jetson Xavier NX, and will get you up to speed. Things will be just the same, only everything will go much faster on the Xavier

[@billfield8300]

Great review Paul. I will get all the specs from another review but this is what I need to see. Also knowing I will be able to get the lessons here as well is a major point in deciding to make the purchase. BTW... Nice comic relief at 49:45 and 51:30 I look forward to catching up with the Jetson Nano and following in with the Xavier when you are ready. Greetings from Canada and stay safe.

[@luisarellano09]

Great content!!

[@fury4076]

Here in KZbin, the one who really explain very well, the one who got least of views, i hate the algorithm of yt. Thanks for inspiring every one. By the way Im the guy who chatted you on Fb long time ago, and now I made a different projects in arduino. Thank you so much 💓

[@josgrootaers1881]

Great review Paul. Thank you for al l your fantastic teaching work. Btw.. I love the intro!

[@paulmcwhorter]

Thanks, had lots of fun with it.

[@Eden_Maker]

Thanks so much for this series Paul. I'll try to subscribe to Patreon when more bills have been paid.

[@kelvinkwok5800]

This is great stuff!

[@ghostdog5096]

Paul, I'm following the Jetson Nano, tutorials which are excellent. Yes, I've bought the kit. I'm in my 50's and a techi', followed your Arduino tutorials which are the best on the net'. God Bless🙏

[@Smallio1]

Don't know what you age demographic is, I'm in my mid 20's and got to episode 20 on your arduino Tutorials and man I can't thank you enough! Learning so much and love blazing through your videos, having a tone of fun. Can't wait to get my hands on the Nvidia Jetson and start those tutorials once I'm done with the arduino! Keep up the awesome work

[@lalitjain1969]

I am 11 and am on your jetson nano tutorial (after completing your arduino and raspberry pi tutorials)

[@hqmhqm]

very informative, thanks !

[@AndrewDzula]

Lol the "Reptilian 99% confidence" on Hillary's face xD

[@AndrewDzula]

100% true

[@craiggardner5347]

Thanks for this. You have inspired me.

[@chinnudanturi9730]

Hey, the first to comment here and I am still learning from your new and improved arduino lessons. Thank you for the lessons Paul.

[@jamespeterson166]

best darn review I ever saw. Thanks a bunch put my nickels away.

[@paulmcwhorter]

thanks for the kind words. I get kind of weary of the open the box and run the demo typical reviews. Appreciate your feedback

[@rukidding7588]

I am in awe of the power of the Xavier. The price is a little rich for me, but it's almost scary how powerful it is.

[@tobybavin8994]

Its nice to see such an infux of peiple finding Paul's vcideos very useful during quarantine

[@xacompany]

51:31 I'm writhing with laughter lying on the floor... Agree with you 100% you are genius!

[@paulmcwhorter]

Funny how few people saw that.

[@Mouse_007]

@@paulmcwhorter omg , you're a wildman

[@Sikuq]

Love all of it.

[@garyward9668]

I just started your videos on thee nano they are great. I'm also doing your improved arduino lessons. Thank you Paul

[@paulmcwhorter]

Hope you have as much fun watching them as I had making them

[@garyward9668]

@@paulmcwhorter I had gotten the 2560 ultimate kit from Elegoo and was doing their lessons before I found you on youtube. Your lessons are so much more in depth with the math and how the components work.

[@icequark1568]

Iced coffee and fresh NV tech. Sign me up friend.

[@KK-fh1ds]

Lol ma guy sets air times. I love it. Im looking for something that can run Visual slam smoothly, so Im very intrigued by this!

[@moserpeter2386]

The most hilarious moments starting at min 36, I couldn't stop laughing!! 😂

[@kestergascoyne6924]

Wow! Will we see a series on the full module? I'd buy them both!

[@ahadbi1]

By far the best review on the Jetson products ! Great tutorial, really well explained, and great attitude. I would power a NERF shooting drone. Guaranteed headshots after headshots. Alternatively, I'm pretty sure it's powerful enough to be able to track someone's mouth and feed him pizza or whatever you manage to manipulate with robotic arm.

[@rampalgrihdhwajsingh5312]

explained very well and too easy

[@qck57594]

Amazing new content Paul, I'll be getting one. If you don't mind a comment, with respect the background is very busy and distracting due to the movement. Just my opinion. The picture is good though. Thank you for all you do.

[@ghostdog5096]

Thanks for the intro to the Xavier. What about developing a robot implementing the processors and sensors you have taught.

[@pralaymajumdar7822]

Wow..we are highly grateful to you like teacher...what a knowledge you have..you are going in an extraordinary space of knowledge and make us mad..Thank you sir..long live and God bless.

[@fury4076]

More power and life, Im wished it for you. Im your greatest fans, greeting from Philippines 💓

[@rickt1866]

I'm here to learn and grab a gear. Thank you for taking the time to make these videos. I'm starting at zero. I got a really good deal on a LENOVO ThinkEdge SE70 Jetson Xavier NX 1.5GHz Nvidia Integrated what I felt was a good deal anyways. I'm hoping to pick a few new skills up along the way. again thank you for the videos.

[@RupertBruce]

You could get a higher frame rate by turning off logging. Great video, thank you.

[@davidbooth529]

Hi Paul, I picked up an Xavier AGX! You NEED to do a course :o)

[@chrisingram7277]

Two comments .. Wow and .. it feels like a video from the year 2030, thanks Paul

[@paulmcwhorter]

Thanks, i was really trying to capture the speed and bleeding edge nature of this board.

[@mukageegee]

I'm sorry Paul I had to cut short watching the Premier as my new Grandaughter called with her Mum errr Mom! We social distanced the visit. I rushed back in when they left started the video up again and saw ..."OK this has been a little bit of a long video........ I had missed about 40 seconds!. Well done Paul another great video. Wonder if the Jeston Xavier will land in the UK?

[@specsix]

Ah! I see the gstreamer errors now that you are using cv 4.1.1 @20:29 . Glad I am not the only one.

[@lorenzoleongutierrez7927]

17:42 fan started for first time... : )

[@danielhoven570]

Im building and programming an arduino drone from scratch using a teensy 3.6 (an awesome development board every tinkerer should have). My plan is to eventually use a model predictive controller in combination with a genetic algorithm so i can fly advanced maneuvers autonomously. Now I'm thinking that this SBC could be a perfect drone core, since I'm using a 500mm platform.

[@DeeGeeDeFi]

If it can run the new Flight Simulator, I’m sold!

[@nrnoble]

WOW! this is the best technical YT video I have seen in 2020. Very very impressed with PIP using multiple cameras without it getting confusing!.... The question that comes to mind: Is it possible to use the same NVIDIA AI development software with a NVIDIA GPU (gaming) card on Windows?

[@sylvainmartens7852]

You can use opencv on windows and start doing stuff like this however your computer motherboard doesnt have GPIO to interact with physical things like a button, servo motor, etc... but you can communicate with a cheap device like an arduino or a teensy via serial/com over USB to make it do what you want from what your computer computed.

[@djricky89999]

HI! What are the video editing applications? for nvidia jetson xavier NX?

[@qzorn4440]

this is a wonderful jetson video... now to take the SBC and a 7" touch-display for a compact cheap system for factory floor machine-vision PLC-projects.. Mmm. 😂 thanks

[@milesklingenberg5885]

Hey Paul, these videos are phenomenal. Do you have any experience sending the results data you are getting from these models to a server on the same network (assuming ethernet)?

[@danrohler1329]

Wow!! I've been waiting decades for this. Progress on PCs and Macs has been unimpressive over the last few decades, especially compared to Moore's law of h/w. I've been frustrated with the poor video performance of Raspberry Pi 4 (though remarkable in many ways). This is the first I've seen of you and the Jetson SBC. I can't wait to get involved.

[@paulmcwhorter]

This is the funny thing to me. No matter how many transistors Intel packs on a chip, microsoft figures out how to bring a system to its knees. I have a state of the art desktop PC, and it takes 5 minutes to boot. They just keep adding more trinkets and doodads, and the systems with incredible hardware run like a dog. I love this system. Crispy and fast linux boot, and then mindbobling speed on the board.

[@jstro-hobbytech]

Alot of the new projects can use the arduino or an stm32 sensor board. It controls the lidar and such that I've bought.

[@TheOleHermit]

Wanting to ween myself off Windows, I was undecided between this SBC and the Latte Panda Alpha 864, as a desktop replacement for my Surface Pro. I had already seen the Nvidia demos, but 30 minutes into this video, I had to pause it to place my order with Nvidia. Can't wait for it to arrive. I'm already 1/2 finished building a WiFi controlled HD surveillance camera w/motorized zoom lens on an X/Y mount. But, now I see that the RPi4b simply ain't got the brains to add this functionality... 💡unless it simply streams the video to the Jetson, which analyses it and sends motion instructions back to the RPi4. 🤔 Then, the $400 Jetson Xavier NX could stay on my desk, where it belongs. I can already hear it begging to become a stereoscopic AI robot head.😎

[@asml-fh3sz]

Mr. McWhorter, thank you for your video series, they are really helpful. I would like to ask you a question. Based on your experience, which Nvidia controller could be the best option for a drone, the Nvidia Jetson Nano or the Xavier NX? I know that Jetson Nano runs at 5V, what about Xavier NX? I mean, could be the power supply requirements of the Xavier NX a problem for autonomous systems like drones?

[@kekocan2900]

Is there a connection between jetson utilities/interfence and yolo object detection? Are we going to learn yolo object detection too?

[@cepwin]

I did get to put the premier on in the background but since I was working I didn't get into chat. Anyway, in addition to the facial recognition we've been studying with the Nano, I can see something like this invaluable for CG artists if it can run popular CG software and renderers (the programs that actually take the images/animation created in popular CG tools and creates files in formats that can be used down the development pipe (eg. go into after effects, fusion, nuke etc. to be put together as a final video.) One of the big issues CG artists face (I'm in the middle of an intense Cinema 4d class so I've faced this myself) is that rendering images takes, in some cases, many hours. Now I know, unlike windows and MacOS, linux can be very version sensitive so even if an app has a Linux version (not all of them do.. the adobe tools are Windows/Mac only...but Cinema, Houdini do have linux versions) there's no guarantees it will run or take advantage of the hardware. But it is something that the CG community and software providers should be looking at.

[@mostafamohsen250]

just picked up an Xavier NX, I'm new to this channel and I can't wait to follow along with this whole series!!! Was it a mistake that i started with the XNX rather than the Nano?

[@paulmcwhorter]

Not a mistake but is you are a beginner go through the Jetson Nano lessons. They work the same on the Jetson Xavier. If you are already familiar with machine learning and opencv jump into these lessons

[@dineshachuthan5027]

Excellent and very exciting. Love the way you present naturally. Do you have these code shared in GitHub? Thanks

[@paulmcwhorter]

No, my goal is to teach you how to code, not teach how to copy and paste. Realy important to learn to code, you need to be putting the commands in and then troubleshooting your work

[@peterxu6681]

Paul you are the best teacher ever. I started learn Jestson NX also. How can I get the source code about your demos. Thanks!

[@yubingyan3262]

hi ,do you know how bootstart(开机启动) fan in the Jetson AGX Xavier

[@johnmoore4248]

Can i use that module on the new nano board (b01)?

[@dinoscheidt]

This timelapse background needs an epilepsy warning. Seriously 😵. - But the rest of your content is great. Would love your thought on Xaviar vs Latte Panda - i.e. what the performance difference might be? I am building a hacked RC CAR with 6 streaming cameras for server side AI work for which the jetson nano turned out to be too weak. Going for a “mini Computer” or giving Xavier a shot 🤔

[@smraro]

it would be nice a tutorial on how to train you own object/facial recognition model for jetson nx/nano =)!

[@paulmcwhorter]

I show training to recognize people in the jetson nano series. That will bork on Xavier as well

[@kryptonic010]

Is there a way to cluster JN NXs for greater frame rates?

[@_scoobydoo7456]

A clean black or green background would have been nice !

[@wayneswan3092]

Hey Paul McWhorter, can you do a lesson series on how you set up your studio, greenscreen and background your secondary cameras and computer screen?

[@paulmcwhorter]

I did couple videos look at my playlists they are on my channel

[@wayneswan3092]

Thank you. I'll take a look. In the main time, I'm jumping back and forth between your Jetson nano and arduino robot car series. Thanks for the Linux refresher! It's been a couple years since I've touched a Linux based computer.

[@musiconanotherlevel]

51:30 it says Hillary is Reptilion-99%-Conf. I think I almost died laughing.

[@whothefoxcares]

Does Xavier NX have UFS card reader like Xavier AGX?

[@adityak3578]

waiting..

[@dgb5820]

Is there any way it’s possible to recognise an object and have an arm with claw to pick that item up and place it into a box

[@paulmcwhorter]

That is possible, but that is not an easy problem.

[@aklmini-makers4304]

Is this new SBC an opportunity to build a ground station for amateur drones? [running two screens.] Or onboard drone camera system.

[@paulmcwhorter]

Lots of interesting possibilities

[@varunchandrashekhar1267]

For this new series of lessons do we need to upgrade to a Jetson Xavier NX or can we continue using our Jetson Nano? And thank you for all your lessons Mr.McWhorter, I'm currently 16 and have been through most of your lessons. It has allowed me to venture into fields that everyone told me I was too young for.

[@paulmcwhorter]

When you can, it would be good to upgrade, but most things you will be able to do on either platform

[@sharyarkhan690]

you made me hero in arduino love you sir from pakistan

[@amos660]

Hello, Sir. I love your channel and I just subscribed. Thanks for making and sharing these videos, they are very helpful. I have a question about the nano. I read somewhere that the nano is reaching its EOL. Is this true? if yes , would you recommend using the modules instead? I have a project that requires reading telemetries (mainly temperatures on the board) on a nano that does facial recognition and I need to do that from a remote server, so I need to buy one for testing etc.. I don't know if I should go for a used nano 2GB or 4GB or take the plunge and buy the xavier. I don't mind using old drivers on the board as long they work. Thanks in advance for your help and thanks again for the video(s).

[@hemangpandhi]

Thank you. Your review is very very impressive. Are you sure all your tutorials provided in jetson nano will will work on Jetson Xavier NX? I don't have Jetson nano so I am planning to purchase Jetson Xavier directly. Can you confirm ?

[@paulmcwhorter]

Everything should work. There are a few lines of code that have changed on this, as my lessons are jetpack 4.2 which has opecv 3. This has jetpack 3.4 which has opencv 4. The command for contrours has changed but most everything else should be the same.

[@hemangpandhi]

@@paulmcwhorter Thank you for your reply. I will get NVIDIA Xavier NX soon. Will inform you in case of any issue.

[@peterhartshauser5176]

Does it make sense to move from completed Arduino tutorials to Jetson Nano tutorials but work through them on the Jetson Xavier device directly? (i.e. skipping the purchase of the Jetson Nano)?

[@paulmcwhorter]

I would buy the Jetson Nano with the two camera ports, then work through the Jetson Xavier lessons, using the Nano. Then maybe go do some of the later lessons on the original nano series.

[@g.s.3389]

very interesting, i would like to have a look to a couple of python you run, do you have a github where you keep them?

[@paulmcwhorter]

I develop these programs line by line on my Jetson Nano playlist on this channel. I develop each program in detail. I dont post code, as I am trying to encourage people to write code themselves, and learn how to troubleshoot.

[@mundotudofacil7624]

What's about CUDA experience?

[@anpr]

Thanks for the vid. Anywhere to download the code presented? Thanks.

[@paulmcwhorter]

No in the series of lessons I teach to code. My belief is you learn to code by coding . . . so generally I will show how to write the code, but will not post it to copy. Because copying code you never learn to do it yourself.

[@leefitzjames]

🖤

[@markomatesic5394]

Great review Paul, as always. I copied the code @53:00 and if I add after line 13 "print(width, height)", I get full cam resolution result. Shouldn't it be 800, 600? Code lines 7 to 13 are exact as yours.

[@paulmcwhorter]

It might depend on your camera. Does your camera support the cam.set command, and are you setting it to a size supported by your camera. Might have to tweak a little

[@markomatesic5394]

And I get as a end results : jetson.utils -- cudaFromNumpy() ndarray dim 0 = 2464 jetson.utils -- cudaFromNumpy() ndarray dim 1 = 3264 jetson.utils -- cudaFromNumpy() ndarray dim 2 = 4 ...not 600 and 800 like you do.

[@paulmcwhorter]

Marko, I will make a lesson on this pretty soon. We have not started using the jetson utilities yet in my lessons, and need to show all about that to make these work right

[@markomatesic5394]

Paul, thank you. No prob. I'll wait for your lesson.

[@fun_engineering5710]

Sir I am very excited for this series, but do I need to see the jetson nano series before this or I can just jump in here.

[@paulmcwhorter]

Either way. I do think if you are completely new to this I go over a lot of the basics in the nano series which I will not cover here.

[@fun_engineering5710]

@@paulmcwhorter OK , thanks Sir

[@JordanPriede]

51:30 Epic.

[@paulmcwhorter]

I am going to get myself sent to a reeducation camp

[@conelrad1447]

"Now you see its not recognizing faces in the press, now thats not because they're non human life forms..." LOL

[@jstro-hobbytech]

I'm going to just jump in with the xavier instead of the nano brother. The xavier uses the unreleased Volta architecture with turingl rt cores and Volta tensor 2.0 cores.

[@hemangpandhi]

I was doing face recognization and eye recognization program using har cascad library on Jetson Xavier NX, however I am getting very low framerate. I am not sure why? Currently I am using default image jetpack 4.4 developer preview provided from NVIDIA

[@paulmcwhorter]

I think there are issues with jetpack 4.4. For one, OpenCV is not compiled with CUDA support. Normal openCV functions are not computationally intensive, but perhaps the Harcascades are. I have had good luck using face recognizer library. You can watch my video in my Jetson Nano playlist, and see how to install that. I was getting really good frame rates on the Xavier using that library.

[@hemangpandhi]

Paul McWhorter Thank you for your reply. Today I successfully compiled and install OpenCV in jetpack 4.4 with Cuda support and now it is actually using GPU. I can see from jtop, however still I am getting low framerate for the the face detection and eye detection program. Do we need to install harcascad library separately? I will check your Jetson nano playlist. Thanks

[@hemangpandhi]

@@paulmcwhorter Today I successfully compiled OpenCV in jetpack 4.4 with Cuda support and now it is actually using GPU. I can see from jtop, however still I am getting low framerate for the the face detection and eye detection program. I am not sure why I will try with face recognization library and inform you. Thanks

[@FlyingPhilUK]

How does the performance compare to the TX2 (& Nano)? Thanks!

[@paulmcwhorter]

I dont have experience with the TX2, but can say that there is a very real and perceptible performance advantage over the Nano. By perceived, I mean when I write programs and run on the Xavier NX I get much higher frame rates, or can have higher resolution video at the same frame rate. For inference and detection, I can set it to find much smaller faces/ojects, and still have good frame rates. So, the advantage is large enough that I am going to begin doing work on the Xavier. I will continue my series of lessons on the Nano, but will start a new series on this platfrotm

[@FlyingPhilUK]

@@paulmcwhorter After a quick check on the web - it looks like the Xavier is 6-8x the Nano, and the TX2 is 2-3x the Nano... - I have a TX2, so I will likely stick with that until it gets too painful to continue with! BTW, I do like they way you show how to do actual tasks on the Jetsons rather than just run up a few of the (admittedly good) demos that nVidia provides...

[@jamespeterson166]

when will you have a class on the xavier nx

[@paulmcwhorter]

Hope to start producing videos this week. To get started, just follow my lessons on the Jetson Nano, which will all apply to the Xavier NX. On the Xavier specific lessons I hope to focus more on the NVIDIA specific frameworks, where for the jetson I did more openCV type things.

[@jonesbbq307]

So which one would be faster at machine learning? A $700 PC or $700 Xavier?

[@paulmcwhorter]

The Xavier NX is about $399. A little price fluctuation on amazon right now, but should be about $399. This thing is blazing fast, and for machine learning will outperform a laptop by a long way.

[@jonesbbq307]

Paul McWhorter That’s amazing. I made a mistake and I was looking at the price for Jetson AGX Xavier. Is that even faster than the NX?

[@paulmcwhorter]

Not sure, but really feel like the NVIDIA program is moving to Jetson Nano and Jetson Xavier NX. Those are the future direction in my mind.

[@jstro-hobbytech]

Sorry for nattering. Are you gonna touch on ros? It's the tensor cores and Cuda butting heads because of the ineffiencies of python compared to c++. C++ sings on that hardware

[@paulmcwhorter]

No, Really ROS has not worked well for me as I have played with it.

[@jstro-hobbytech]

@Paul McWhorter I'm kinda committed to getting it working and figuring out a project haha. I was thinking a mobile garbage can. Haha

[@bootyholeman5659]

Those cascades are old tech. You should try CNN's to detect objects, and try facenet to do the recognition.

[@aklmini-makers4304]

Hsve you seen the new Raspberry Pi 12mp camera board with lenses?

[@paulmcwhorter]

I have, but have not been able to verify that it would work with these nvidia boards. Would be eager to try them out.

[@sigmiami]

Hi Paul any projects on asus tinker boards in the near future

[@paulmcwhorter]

Probably not. Likely, I will continue new lessons on the Jetson Nano, then this series on the Jetson Xavier. I will probably start an arduino robot car this summer, and then the Jetbot robot with Jetson Nano

[@sigmiami]

@@paulmcwhorter looking forward to the arduino robot... my poor tinker board is there collecting dust, cant even find decent projects on it.

[@Jure1234567]

How about Goggle Coral USB stick? I've heard it can give impressive results even with Jetson Nano...

[@paulmcwhorter]

I am just not a fan of google . . . very hard company to work with. I have found the NVIDIA guys to be great guys to work with, and really believe their hardware platform is the future of AI.

[@Jure1234567]

@@paulmcwhorter I understand.. Here is the article: blog.raccoons.be/coral-tpu-jetson-nano-performance I was shocked how much fps jetson+coral gives compared to rpi+coral. Must be those nVidia optimisations. BTW maybe you have tried Intel NCS2? Interesting module as well...