14 күн бұрын
2 ай бұрын
3 ай бұрын
4 ай бұрын
6 ай бұрын
7 ай бұрын
11 ай бұрын
Жыл бұрын
Жыл бұрын
Жыл бұрын
Жыл бұрын
Жыл бұрын
Жыл бұрын
Жыл бұрын
Пікірлер
@ProaliusVideos 7 сағат бұрын
Hi Again!, imposible to use csi camera with new Super versión, libraries as board.py fails when detection de pinout... i can't fix it.... any help?
@JetsonHacks Сағат бұрын
I don't know what board.py is.
@Glanmire3 11 сағат бұрын
Hi there, I'm just before this long procedure so I'd like to ask is there any tip or trick to speed up the lengthy (8-12h) build process? Like turning off certain functions or programs that are running in the background. Also what versions of SW elements was at the time of the record? I know not always the latest one is the best to get a working system. (Hope you can get what I mean, I'm not native English)
@JetsonHacks Сағат бұрын
I would suggest that you use the Docker versions of OpenCV with GPU support, it's much easier. It's difficult to advise what to do, because everyone has different support needs. github.com/dusty-nv/jetson-containers/tree/master/packages/opencv
@geojaya 13 сағат бұрын
Where I can buy Orin Nano Super
@JetsonHacks 13 сағат бұрын
Right now they're on back order, here's the link: nvda.ws/3OVTwBj Thanks for watching!
@mohammadadnansiddiqui3374 Күн бұрын
How can i use this cv2 in my conda environment
@JetsonHacks Күн бұрын
You will probably need to build it for the specific version of Python that you are using, and the JetPack version of libraries. To build it for Python: kzbin.info/www/bejne/eHSxn6SKe85gfdU Hope this helps.
@bhaveshdamania750 Күн бұрын
Thank you very much for this video...I was having hard time running thru SDK Manager due to GUI Display error. First try following this video the FAt32 partition was not copied because I had Swap memory as well as used the SSD for Docker following the guide...Formatting the SSD caused the nano to go in Emergency mode. Using Chatgpt it gave me steps to update the Fstab file and commenting out the entries... That worked and then I followd your steps and now I am sucesfully able to login thru the SSD. but using the JTOP commmand I still SWP as OFF in red how do I resolve that ?
@JetsonHacks Күн бұрын
Sounds like quite the process! You can use jtop in the MEM tab to create a new Swap file. Thanks for watching!
@bhaveshdamania750 Күн бұрын
@@JetsonHacks Thank you will google for steps..your videos are very helpful for a beginner like me...
@JetsonHacks Күн бұрын
@@bhaveshdamania750 You are welcome. I am glad you find the videos useful.
@BenjiNotknown Күн бұрын
has anybody been successful running a jetpack .img in a virtual machine? is there anyway to change an .img to an .iso in order to try it?
@JetsonHacks Күн бұрын
That's a good question, hopefully someone here sees it. I think the issue people run into is the way VMs handle USB. During the flash process, the Jetson disconnects and reconnects, and VMs tend to have trouble handling that. Over the years, people have come up with workarounds, but it is very dependent on which VM you are using. Thanks for watching!
@jreicher 2 күн бұрын
Thanks so much for this video! Things flaked out a bit for me, but the second time around everything worked fine (I think I had a bad spot on my SSD - it was one of those losely Kingston ones!)
@JetsonHacks Күн бұрын
You are welcome. If you can't trust your memories, who can you trust? I am glad you were able to get it to work, and thanks for watching!
@julianvanderkraats408 2 күн бұрын
You can also use SDK Manager to flash eveything to NVMe, works well, no scripts needed.
@JetsonHacks Күн бұрын
That's a great (and approved) way to set up your system! Unfortunately some users (like me) have not had the same 'works well' experience. Thanks for watching!
@rreybier 2 күн бұрын
Hello JetsonHacks , i am happy follower from your channel , thank you fro you support 🎉
@rreybier 2 күн бұрын
Thank you for video , you mix cm and mmm by the way
@rreybier 2 күн бұрын
No problem is am un uneducated européen i only understand inches (because of my thumb ) 😅
@JetsonHacks Күн бұрын
You are welcome. You caught me! All this new fangled metric stuff here in America always gets me. Thanks for watching!
@JetsonHacks Күн бұрын
Lol. Americans don't have much less of an education, that's why we used our feet instead of something reasonable like a meter for measuring.
@jpinyon 2 күн бұрын
Thanks for the vidoe. Do you think this will also work for AGX Orin's eMMC to SSD migration?
@JetsonHacks Күн бұрын
I have not tested that particular case, but I think there is a high probability of success. However, the AGX Orin is such a different class of machine that it's worth it to invest in learning how to use the BSP and rootfs tools and/or use the SDK Manager. Thanks for watching!
@jpinyon Күн бұрын
@@JetsonHacks Thank you for your kind reply. I will try it when I have the opportunity and report back.
@BerndKircher 2 күн бұрын
For my whole setup I used a much easier way: I directly copied the downloaded image onto my SSD with balenaEtcher. Then I just needed to modify /boot/extlinux/extlinux.conf on the SSD: replace root=/dev/mmcblk0p1 by root=/dev/nvme0n1p1
@airatooo 2 күн бұрын
I have Orin nano dev kit from a year ago. Would this work on it? Is it the same hardware as the new super? Thanks!
@JetsonHacks 2 күн бұрын
The hardware is the same, this method will work on it. First, you will need to upgrade the firmware and create a bootable SD card with Super mode enabled: kzbin.info/www/bejne/fWmynKWLf92hh8U Thanks for watching, and let us know if you get it to work.
@JetsonHacks 2 күн бұрын
In the video, I state that the 80 in 2280 stands for 80 centimeters. It means 80 millimeters, of course. 80 centimeters would have much more storage! What can I say, I'm an uneducated American 😉
@Kw1161 2 күн бұрын
I'm holding out for a 80 kilometer SSD...! Happy New Year JetsonHacks!
@sansmojo 2 күн бұрын
Maybe the Super Duper will support 80 cm SSDs.
@JetsonHacks 2 күн бұрын
@@Kw1161 I like it! Thinking big! Happy New Years to you and yours, and thanks for watching!
@JetsonHacks Күн бұрын
@sansmojo One can only hope. Thanks for watching!
@jatu4180 2 күн бұрын
I think you mean mm, not cm?
@JetsonHacks 2 күн бұрын
Yes. 80 CM would have much more storage capacity. I'm a silly American, I always seem to get that meter stuff wrong. Thanks for watching!
@jatu4180 2 күн бұрын
Same here. Love your channel, thanks for all your good work. It makes my job so much easier.
@JetsonHacks 2 күн бұрын
@@jatu4180 You are welcome, I'm glad you find it useful. Thanks for watching!
@TT-it9gg 2 күн бұрын
Thanks for the vidoe. Seems mm? not cm?
@JetsonHacks 2 күн бұрын
Yes that's correct. My American is showing. Thanks for watching!
@garykingston1662 2 күн бұрын
I’ve been waiting eagerly for this! Not only did you make a video, but you wrote scripts to do it. Bravo! I can’t to get this running. Excellent video, as well. Update: just completed the steps and it works exactly as described in the video. Perfection. Thanks again!
@JetsonHacks 2 күн бұрын
Thank you for the kind words. I am glad you were able to get it to work. Thanks for watching!
@abdshomad 2 күн бұрын
I've been waiting for this 🎉 So, firstly, we needed to install JetPack in our Jetson Super on SD CARD, then when installation completed, we move it to SSD. This is because we can not "burn" the JetPack directly to SSD?
@jakesullivan3017 2 күн бұрын
I used the SDK manager and flashed on the SSD directly. This is for a host free setup
@JetsonHacks 2 күн бұрын
You could write the image to a SSD if you have the hardware to do so. However, and the part that is tricky, is updating the firmware/QSPI. It should be straightforward if you are running firmware version 36.3.0 or above; there is a program in the image which will schedule the firmware update on the next startup. However, if you have an older firmware (as many current Jetson Orin Nanos do), you may need to do the upgrade multiple times, using older versions of JetPack. I don't know how much testing has been done on SSDs as an upgrade path from older JetPacks. Instead, what is presented here is a firmware upgrade method which people have been using for several months, the SD card approach. Once you get the firmware upgraded, you're most of the way home. Of course, the only NVIDIA approved way to run from SSD is to use SDK Manager to download JetPack 6.1 rev1 and install it. This is in part because this is an interim release, and should be fixed when the next JetPack version is released later in January. Thanks for watching!
@JetsonHacks 2 күн бұрын
That's a good way to do it, but is more painful for more casual users. I noticed that the SDK Manager had some major hiccups over the last week. Hopefully the next JetPack release will be smoother. Thanks for watching!
@energyideas 3 күн бұрын
If I get a new Jetson Nano Orin Super delivered in 2025, is the firmware up to date for Jetpack 6? Thanks.
@JetsonHacks 3 күн бұрын
I do not know. Please ask this in the official NVIDIA Jetson Orin Nano forum where a NVIDIA engineer can give you the correct answer: forums.developer.nvidia.com/c/agx-autonomous-machines/jetson-embedded-systems/jetson-orin-nano/632 Thanks for watching!
@energyideas Күн бұрын
@@JetsonHacks Received new Jetson Nano Orin Super today 1/5/25, had old firmware, not updated. What a hassle. Thanks for your help!
@JetsonHacks Күн бұрын
@@energyideas Yes, it's a hassle. The out of box experience needs some work. I'm glad you were able to get yours. Good luck on your projects!
@kinsung85 3 күн бұрын
Do you think the flashing on AGX Orin will work if Ubuntu on your host computer is in a virtual machine?
@John-lv4zy 3 күн бұрын
I installed a debian desktop live cd on an usb stick, had my windows laptop boot from it, installed sdkmanager and could flash the agx without problems. Use a 64gb+ stick if you can.
@JetsonHacks 3 күн бұрын
Virtual machines tend to be a hit or miss proposition. During the flashing process, the Jetson will disconnect and reconnect via the USB port multiple times. This tends to confuse the Virtual Machine. Some people have workarounds, but they tend to be VM specific. Thanks for watching!
@energyideas 4 күн бұрын
Is this still the recommended way to boot off of the <2 drive for a Supper? Thanks.
@JetsonHacks 4 күн бұрын
No. The recommended way is to use the NVIDIA Jetson SDK Manager. For the Super release, there's an interim set of libraries that can only be installed that way on the Orin Nano Super. Thanks for watching!
@bluesjeon 5 күн бұрын
Can Windows 11 64bit arm version run on the Jetson Orin Nano Super Developer Kit?
@JetsonHacks 5 күн бұрын
Not at the moment, it's strictly Linux. Thanks for watching!
@frankpetiprin876 5 күн бұрын
Cost of Jetson Orin Nano Developer Kit
@JetsonHacks 5 күн бұрын
Now $249: kzbin.info/www/bejne/bp2pfqF6q611pdU Thanks for watching!
@asdfsdfsadfasdfsf 8 күн бұрын
hello. i tried do as you exactly but my nano still need sd card to boot. if i remove the sd card it will display No SD card and bootloop at nvdia logo. but i can confirm if i still let the sd card there, it will boot from ssd. really hope this can reach you
@EstellaWhite-ws7gh 8 күн бұрын
Super super Jetson ..❤
@JetsonHacks 8 күн бұрын
I love that song! Thanks for watching.
@innuendo70 10 күн бұрын
Ordered the cheaper Jetson Orin Nano Super 10 days ago, I have a 2TB NVME that I'd like to put in it, and looking at videos like this one right now while waiting for it to get shipped. Should this tutorial work for the Jetson Orin Nano Super too? And are there any details that should be a bit different that I'll need to be aware of when trying to apply this tutorial.
@JetsonHacks 9 күн бұрын
It depends on how you want to use it. If you boot from the SD card, you can format the SSD and use it as extra storage. If you want to boot and run from the SSD, you will need to use the SDK manager to flash the SSD. The issue with using the method in this video is that it is currently not possible to download the Super disk images directly from the JetPack SDK pages. You must use the NVIDIA SDK Manager to do that. I'll be making a video that goes over the options. Thanks for watching!
@JFlogerzi 8 күн бұрын
@@JetsonHacks Do you have an ETA. I just received my Orion Nano and would like to boot from my 1TB SSD
@JetsonHacks 7 күн бұрын
@@JFlogerzi No ETA right now, been working on getting everything together. You should be able to use the SDK Manager if you're really in a rush.
@JFlogerzi 7 күн бұрын
@@JetsonHacks I will be doing so.
@under_14 10 күн бұрын
Can I install an ssd to this setup for running my machine learning programs and have the Jetson boot from the sd card?
@JetsonHacks 10 күн бұрын
Yes you can! There are several ways to do that, here's one example from Jetson AI Lab: www.jetson-ai-lab.com/tips_ssd-docker.html Thanks for watching!
@SaleheddineBouberhane 11 күн бұрын
So i want to aquire audio signal through i2c if thats possible i would realy appreciate if u show me how to this or if only revising ur code in github is enough
@JetsonHacks 11 күн бұрын
Not sure what you're asking. This is a video about how to do that. Did you try it?
@oceanicdrop 12 күн бұрын
I want to save up for one of these, but I wonder if I can use it similarly like ChatGPTs API. I want to connect to the Jetson over network, chatting to it, and supplying tools (functions) that LLama can call on my computer. I kind of just want to use it as an AI server.
@JetsonHacks 11 күн бұрын
You can do that. However, if you are just calling it from a PC it may be better to invest in a better graphics card instead to run your LLM locally on the PC. Thanks for watching!
@oceanicdrop 11 күн бұрын
@JetsonHacks after some testing for my use case (an AI powered terminal shell) I think the 8b models and below are just too unpredictable. I see that a graphics card could benefit me in several ways. I just don't know which cards to consider. I got a 5700xt at the moment which doesn't accelerate ai
@JetsonHacks 10 күн бұрын
@@oceanicdrop If you want to use models larger than 8B, such as 70B, you will need a lot more graphic card memory. There are many alternatives if you have a PC. You will need a NVIDIA graphics card. The Ampere architecture (as in the Orin Nano) is 30X0, such as 3090. The next generation up is 40X0, such as 4090. There are many articles which discuss the tradeoffs in price and configuration, but mostly it's how much money you want to spend and how much power draw is acceptable from your PC. Good luck on your project!
@blubberflutsch 12 күн бұрын
Would love to see an update, that enables h264/h265 hardware encoding... I still can not believe that they crippled that on the Orin Nano.
@JetsonHacks 12 күн бұрын
Me too. Thanks for watching!
@Kenwaldek 12 күн бұрын
Thanks for sharing knowledge
@JetsonHacks 12 күн бұрын
You are welcome and thanks for watching!
@JohnWindberg 12 күн бұрын
I went through the steps, and everything appears to be working fine, except one small thing. I no longer get the green NVIDIA bios screen any more. Just black until the mouse pointer appears and then Ubuntu starts. Thus I no longer have access to bios settings or boot options. I've tried pressing various keys at various times. F2, F10, Esc. Still, just black until Ubuntu. Any ideas how to get the green NVIDIA bios screen back?
@JetsonHacks 12 күн бұрын
I have not heard of that issue. It's worth brining up on the Jetson Orin Nano forum so the NVIDIA engineers can help: forums.developer.nvidia.com/c/agx-autonomous-machines/jetson-embedded-systems/jetson-orin-nano/632
@michielsmissaert 13 күн бұрын
Why does it write your spoken sentences a bunch of times instead of once? Can this be prevented?
@JetsonHacks 12 күн бұрын
That's just debug output, you can turn it off. Thanks for watching!
@michielsmissaert 13 күн бұрын
Nice video! Can you also talk freely with the AI, so you have first speech recognition, then ai thinks of an answer and then does text to speech. How hard would that be to set that up? I heard that this single board computer can run ollama 3.2, how good is that if you just want to talk freely with the AI?
@JetsonHacks 12 күн бұрын
Thank you for the kind words. Here's an example. kzbin.info/www/bejne/Y5vKn62JodOYd7csi=DUD6mGMraMw1mjIm
@manorotti 13 күн бұрын
The module looks like laptop ram
@JetsonHacks 13 күн бұрын
Same type of connector. Thanks for watching!
@ardeshiraminian9841 14 күн бұрын
thanks for your video. Concise and to the point.
@JetsonHacks 13 күн бұрын
You are welcome, and thank you for the kind words. Thanks for watching!
@YomTech44 14 күн бұрын
Hey, Jim, I am your all-time fan. Nice to see you lately!🤩🤩 I need one thing badly, can I install pytorch/torchvision in the Nvidia Jetson Orin Nano 8GB devkit with Jetpack 6.1 (cuda 12.6 cudnn 9.3)? I want you to give me a clue, the 'torchvsion' doesn't allow me to build it with CUDA for the above specs.
@JetsonHacks 12 күн бұрын
I don't know how you are attempting to build it. I always use the official NVIDIA .whl files from developer.nvidia.com/embedded/downloads You can ask for specific help from the NVIDIA Jetson forum, where NVIDIA engineers and community members share their experience.
@YomTech44 12 күн бұрын
@@JetsonHacks Okay, thank you, Jim. I tried so many times that is why.
@7BlackJack8 14 күн бұрын
"Super" is just marketing to relaunch an existing item with a simple upgrade patch.
@JetsonHacks 13 күн бұрын
And cutting the price in half, which is the real part of the announcement. Thanks for watching.
@ronakvyas6595 13 күн бұрын
The price drop is "super"
@JetsonHacks 12 күн бұрын
@@ronakvyas6595 The price makes it accessible to a much larger range of projects. Thanks for watching!
@whatsinthename8722 14 күн бұрын
can you please do some benchmarks on super kit, like running some LLMs like Llama3.2 and some object detection models like yolov8 or yolo11?
@JetsonHacks 13 күн бұрын
No thanks, NVIDIA has a full suite that they do. Here's an example of the results: www.jetson-ai-lab.com/benchmarks.html Thanks for watching!
@ESGamingCentral 10 күн бұрын
@@JetsonHacks any idea on where you can find these on stock? scalpers are no fun
@JetsonHacks 10 күн бұрын
@@ESGamingCentral I don't have any good leads right now. The inventory they had sold out fast! Fortunately, they have factories to make more, and more are on the way. Thanks for watching!
@ESGamingCentral 9 күн бұрын
@@JetsonHacks thanks for your content, please keep sharing. I want to replicate some of the things you do once I'm able to find one.
@whatsinthename8722 9 күн бұрын
@@JetsonHacks I had seen them already and seen some other people on social media saying that actual benchmarks is not even close, that's why I asked!!
@JetsonHacks 14 күн бұрын
MAXN (25W) mode is currently only available on the official NVIDIA Jetson Orin Nano Developer Kits. That's part number p3767-0005 if you're keeping track. You can get the part number from the Info tab of jtop. The NVIDIA Developers Kit have a micro-SD reader on the Jetson module.
@garykingston1662 14 күн бұрын
Just in time… almost. I got this working yesterday but the previous method of moving and booting to an ssd doesn’t work… so I anxiously await your video on how to set that up. Early Christmas present! Happy holidays.
@JetsonHacks 14 күн бұрын
I guess Santa's a little late in getting the videos out. I hope you and yours have a great holiday!