Jeff Geerling
Jeff GeerlingVideo
- Format:
- YouTube channel
- New videos:
- Weekly
- Typical length:
- about 10 minutes, not counting Shorts
- Latest video:
About this show
Father, author, developer, maker. Sometimes called "an inflammatory enigma".
Videos come from the channel's public YouTube feed. The artwork is the newest video thumbnail. Last updated . Names, artwork and text belong to their publishers.
Cosmos
Similar by topic and creator
- The PrimeTime
Shares the topic Technology. - Just Have a Think
Shares the topic Technology. - Neso Academy
Shares the topic Technology.
Videos
A tiny AI supercomputer for your desk
· 11 min
Let's see if Nvidia's GB10 "AI Superchip" is all it's hyped up to be... Thanks to Dell for providing the two Dell Pro Max with GB10 units for testing and evaluation, along with accessories to get them clustered. Resources I mentioned in this video: - Dell Pro Max with GB10 Benchmark Results: https://github.com/geerlingguy/sbc-reviews/issues/92 - Dell Pro Max with GB10 AI Benchmarks: https://github.com/geerlingguy/ai-benchmarks/issues/34 - Dell Pro Max with GB10: https://www.dell.com/en-us/shop/desktop-computers/dell-pro-max-with-gb10/spd/dell-pro-max-fcm1253-micro/xcto_fcm1253_usx - DGX Spark has no power LED: https://community.frame.work/t/dgx-spark-vs-strix-halo-initial-impressions/77055 - Mediatek on Grace CPU: https://www.mediatek.com/press-room/newly-launched-nvidia-dgx-spark-features-gb10-superchip-co-designed-by-mediatek - DGX OS Release Cadence: https://docs.nvidia.com/dgx/dgx-spark/dgx-os.html#release-cadence - Jetson Nano Ubuntu 18.04 only: https://forums.developer.nvidia.com/t/trying-to-install-ubuntu-20-or-22-on-jetson-nano-2gb/327491/2 - CoconutMacaroon's Blender compilation instructions for Arm Linux: https://github.com/CoconutMacaroon/blender-arm64/ - FCLC's post on FLOPs: https://bsky.app/profile/fclc.bsky.social/post/3lc4qpte3ys2o - John Carmack's tweet on the 'petaflop': https://x.com/ID_AA_Carmack/status/1982831774850748825 - Top500 list for June 2025: https://www.top500.org/lists/top500/list/2005/06/?page=4 - Exo blog post on DGX Spark + Mac Studio: https://blog.exolabs.net/nvidia-dgx-spark/ Support me on Patreon: https://www.patreon.com/geerlingguy Sponsor me on GitHub: https://github.com/sponsors/geerlingguy Merch: https://www.redshirtjeff.com 2nd Channel: https://www.youtube.com/@GeerlingEngineering 3rd Channel: https://www.youtube.com/@Level2Jeff Contents: 00:00 - It's not a mini PC 01:07 - It's not a gaming PC 03:06 - It's an Arm Linux PC 03:49 - Improvements over the DGX Spark 04:28 - Grace CPU 05:19 - DGX OS and a concern 07:00 - Benchmarks (and FLOPs) 08:40 - Dual 200 Gbps networking 10:07 - AI on the GB10
Not rated yetNIST's NTP clock was microseconds from disaster
· 5 min
...but most people would never know. The two posts referenced in this video: - Primary time scale failure: https://groups.google.com/a/list.nist.gov/g/internet-time-service/c/o0dDDcr1a8I - Update on Internet Time Services: https://groups.google.com/a/list.nist.gov/g/internet-time-service/c/OHOO_1OYjLY Also check out the status of all NIST Internet Time Servers: https://tf.nist.gov/tf-cgi/servers.cgi Thumbnail photos were adapted from Wikipedia Commons / US Government Public domain images: - https://www.nist.gov/pml/time-and-frequency-division/time-and-frequency-metrology-and-distribution - https://www.nist.gov/topics/physics/optical-frequency-combs Support me on Patreon: https://www.patreon.com/geerlingguy Sponsor me on GitHub: https://github.com/sponsors/geerlingguy Merch: https://www.redshirtjeff.com 2nd Channel: https://www.youtube.com/@GeerlingEngineering 3rd Channel: https://www.youtube.com/@Level2Jeff Contents: 00:00 - 5 microseconds 00:39 - What happened? 02:15 - NTP timing and why it doesn't matter 03:17 - Timely lessons
Not rated yetDo you need a whole PC to run a GPU?
· 18 min
Shop Micro Center Holiday Deals: https://micro.center/e787eb Check out Micro Center's Other Top Deals: https://micro.center/315b05 Shop Micro Center Bundles: https://micro.center/f7b037 Visit Micro Center News: https://micro.center/9b77f2 Can a Raspberry Pi match the performance of a modern desktop PC when it comes to GPU performance? Of course not, in a general sense. But there are specific use cases where you'd be surprised by how close the Pi comes, performance-wise. In some cases, the Pi is actually faster! Resources I mentioned in this video: - AI Benchmark results and methodology: https://github.com/geerlingguy/ai-benchmarks - Multi-GPU benchmark results: https://github.com/geerlingguy/ai-benchmarks/issues/44 - Use AMD GPUs on Pi: https://www.jeffgeerling.com/blog/2025/using-amd-gpus-on-raspberry-pi-without-recompiling-linux - Use Nvidia GPUs on Pi: https://www.jeffgeerling.com/blog/2025/nvidia-graphics-cards-work-on-pi-5-and-rockchip - Use Intel GPUs on Pi: https://www.jeffgeerling.com/blog/2025/all-intel-gpus-run-on-raspberry-pi-and-risc-v eGPU Setup (some links are affiliate links): - Minisforum DEG1 eGPU Dock: https://amzn.to/4s3gUz6 - Micro SATA Cables Oculink to M.2 adapter: https://amzn.to/49dKKcE - Super Flower 850W PSU: https://www.microcenter.com/product/684393/super-flower-leadex-iii-gold-up-850-watt-80-plus-gold-atx-fully-modular-power-supply-atx-31-compatible - AMD Radeon AI Pro R9700: https://www.microcenter.com/product/702853/xfx-amd-radeon-ai-pro-r9700-single-fan-32gb-gddr6-pcie-50-graphics-card Dual GPU Setup (some links are affiliate links): - chenyang PCIe 4.0 M.2 NGFF to SFF-8643: https://amzn.to/4pyqSqt - 10Gtek SFF-8644 to SFF-8643 Cable: https://amzn.to/3MKNbup - Dolphin PCIe HBA MXH932: https://dolphinics.com/products/MXH932.html - Dolphin 3 slot PCIe Backplane: https://dolphinics.com/products/IBP-G4X16-3.html HUGE thanks to Patrick from @ServeTheHomeVideo for helping me record at Micro Center in Phoenix :) Support me on Patreon: https://www.patreon.com/geerlingguy Sponsor me on GitHub: https://github.com/sponsors/geerlingguy Merch: https://www.redshirtjeff.com 2nd Channel: https://www.youtube.com/@GeerlingEngineering 3rd Channel: https://www.youtube.com/@Level2Jeff Contents: 00:00 - Pi vs PC 00:53 - Enough for Jellyfin? Local LLMs? 01:50 - 4 GPUs 1 Pi 02:16 - Comparing costs and energy use 03:07 - Gaming on haitus (for now) 03:39 - The setups 04:56 - ffmpeg and Jellyfin media transcoding 07:31 - 3D rendering with GravityMark 08:39 - LLMs on AMD 09:36 - LLMs on Nvidia 12:27 - Drivers, Vulkan, and CUDA 13:06 - Dual GPU - Setup 15:24 - Sharing memory and PCIe ACS 16:28 - llama.cpp performance - mixed Nvidia GPUs 17:08 - 52GB of VRAM on AMD 17:28 - Intel PC dual Nvidia GPU comparison 17:56 - Who wins?
Not rated yetApple didn't have to go this hard...
· 14 min
RDMA over Thunderbolt support landed in macOS 26.2. Is it good for HPC? For Exo and LLMs? Find out in this video. Resources (some links are affiliate links): - Mac Studio: https://www.apple.com/mac-studio/ - Exo 1.0: https://exolabs.net - Apple MLX PR for Thunderbolt RDMA: https://github.com/ml-explore/mlx/pull/2808 - M5 Neural Accelerator speedup: https://machinelearning.apple.com/research/exploring-llms-mlx-m5 - DeskPi Rackmate TL1: https://amzn.to/4p5ek94 - Tupavco mini-rack PDU: https://amzn.to/4j0DpAq Disclaimer: The Mac Studios used in this cluster are on loan from Apple for testing. They are not sponsoring this video, nor had any say in the contents. DeskPi sent the Rackmate TL1 used in the video for my testing, but they also had no say in the video contents. I always add the "sponsored" flag on my videos if I did not purchase the equipment being shown. Support me on Patreon: https://www.patreon.com/geerlingguy Sponsor me on GitHub: https://github.com/sponsors/geerlingguy Merch: https://www.redshirtjeff.com 2nd Channel: https://www.youtube.com/@GeerlingEngineering 3rd Channel: https://www.youtube.com/@Level2Jeff Contents: 00:00 - A rack-full of Macs 02:01 - Putting together a cluster 03:16 - The downsides of Thunderbolt 05:09 - One is enough 07:45 - Managing macOS in a cluster (vs Linux) 08:37 - HPL and llama.cpp cluster performance 09:30 - Enabling RDMA 09:51 - Massive LLMs 11:34 - RDMA over TB5 is great... when it works 12:10 - EXO community and OSS 12:51 - Apple could go harder 13:30 - Jack of all trades, not just AI
Not rated yetWhy is Raspberry Pi only selling this in China?
· 3 min
Raspberry Pi launched the CM0... in China. Let's take a look at this little stamp-sized Pi. Resources mentioned in this video: - My test data and discussion: https://github.com/geerlingguy/sbc-reviews/issues/98 - ED-CM0NANO from EDAtec: https://edatec.cn/docs/cm0nano/ds/ - Raspberry Pi CM0 discussion on the Pi forum: https://forums.raspberrypi.com/viewtopic.php?t=386639 - Order the ED-CM0NANO-10000 from DigiKey: https://www.digikey.com/en/products/detail/eda-technology-co-ltd/ED-CM0NANO-10000/28263891 - Hackster article on CM0: https://www.hackster.io/news/raspberry-pi-unveils-the-18-compute-module-0-but-only-for-chinese-customers-for-now-913bf59ab6cc Support me on Patreon: https://www.patreon.com/geerlingguy Sponsor me on GitHub: https://github.com/sponsors/geerlingguy Merch: https://www.redshirtjeff.com 2nd Channel: https://www.youtube.com/@GeerlingEngineering 3rd Channel: https://www.youtube.com/@Level2Jeff Contents: 00:00 - Castellated Computing 01:01 - EDAtec's CM0NANO dev board 01:34 - Flashing the CM0's slow eMMC 02:02 - Chromium runs, but it's not for that 02:42 - Why in China?
Not rated yetThis is not a Framework Laptop
· 15 min
It's a Framework 13, but there's no ordinary CPU inside... Resources I mentioned in this video: - DeepComputing DC-ROMA II: https://store.deepcomputing.io/products/dc-roma-ai-pc-risc-v-mainboard-ii-for-framework-laptop-13 - All my benchmarking data: https://github.com/geerlingguy/sbc-reviews/issues/82 - Framework 13 with AMD Ryzen AI 5 340: https://frame.work/laptop13 - SemiTO-V RP2350 GPIO Expansion Module: https://github.com/semitov/rp2350-gpio-card - SemiTO-V MCL Python to MicroPython tool: https://github.com/semitov/SemiTOV-MCL Support me on Patreon: https://www.patreon.com/geerlingguy Sponsor me on GitHub: https://github.com/sponsors/geerlingguy Merch: https://www.redshirtjeff.com 2nd Channel: https://www.youtube.com/@GeerlingEngineering 3rd Channel: https://www.youtube.com/@Level2Jeff Contents: 00:00 - This is not a Framework 01:49 - Hardware 04:02 - Strange architecture, seen before 04:44 - RISC-V in a strange place right now 05:28 - How is it as a laptop? 07:18 - NPU and a missing 16 GB of RAM 08:12 - Benchmarks 09:20 - Why so inefficient? 10:19 - A trip back to 2009 11:49 - It's a Framework though 12:23 - Upgrading to AMD 13:04 - 6 cores demolishes 8 cores 13:57 - Adding 2 more RISC-V cores (via expansion module) 14:57 - Final thoughts
Not rated yetThe RAM Shortage Comes for Us All
· 4 min
Some of the articles and posts I referenced: - TeamGroup 64GB DDR5 Memory Kit I bought for $209.99 is now $649.99: https://www.microcenter.com/product/680461/teamgroup-t-create-expert-64gb-(2-x-32gb)-ddr5-6400-pc5-51200-cl34-dual-channel-desktop-memory-kit-ctced564g6400hc34bdc01-black - PC Part Picker Memory Pricing Trends: https://pcpartpicker.com/trends/price/memory/ - Samsung can't buy Samsung memory: https://www.pcworld.com/article/2998935/ram-is-so-expensive-samsung-wont-even-sell-it-to-samsung.html - Micron stops selling Crucial memory and NVMes: https://investors.micron.com/news-releases/news-release-details/micron-announces-exit-crucial-consumer-business - Libre Computer's quote for LPDDR4 4GB RAM increasing beyond entire board BoM: https://x.com/librecomputer/status/1995592912063922578 - Lenovo Stockpiles RAM for Laptops: https://www.pcworld.com/article/2986266/lenovo-stockpiles-ram-to-hopefully-keep-laptop-prices-down.html - Raspberry Pi memory-driven price rises: https://www.raspberrypi.com/news/1gb-raspberry-pi-5-now-available-at-45-and-memory-driven-price-rises/ - Mono's future production runs impacted by LPDDR4 memory pricing: https://www.youtube.com/watch?v=AyueVGLT7qI Support me on Patreon: https://www.patreon.com/geerlingguy Sponsor me on GitHub: https://github.com/sponsors/geerlingguy Merch: https://www.redshirtjeff.com 2nd Channel: https://www.youtube.com/@GeerlingEngineering 3rd Channel: https://www.youtube.com/@Level2Jeff
Not rated yetNvidia GPU on Raspberry Pi—it WORKS!
· 5 min
It finally works! Blog post with all the details (and how you can try it): https://www.jeffgeerling.com/blog/2025/nvidia-graphics-cards-work-on-pi-5-and-rockchip And here's the open driver PR I mentioned in the video: https://github.com/NVIDIA/open-gpu-kernel-modules/pull/972 Support me on Patreon: https://www.patreon.com/geerlingguy Sponsor me on GitHub: https://github.com/sponsors/geerlingguy Merch: https://www.redshirtjeff.com 2nd Channel: https://www.youtube.com/@GeerlingEngineering 3rd Channel: https://www.youtube.com/@Level2Jeff
Not rated yetAre you sure that's air you're breathing?
· 16 min
It's a neat, portable Air Quality monitor, full of personality. But is it worth $250? Links to things I referenced in the video (some links are affiliate links): - Air Lab: https://networkedartifacts.com/airlab/ - AirGradient One: https://www.airgradient.com/indoor/ - Temtop M2000C 2nd Generation: https://amzn.to/3XGbpbs - CO2 study referenced in the video: https://pubmed.ncbi.nlm.nih.gov/23008272/ Support me on Patreon: https://www.patreon.com/geerlingguy Sponsor me on GitHub: https://github.com/sponsors/geerlingguy Merch: https://www.redshirtjeff.com 2nd Channel: https://www.youtube.com/@GeerlingEngineering 3rd Channel: https://www.youtube.com/@Level2Jeff Contents: 00:00 - A $250 portable air quality monitor 01:14 - Versus AirGradient ONE 02:06 - Setting up an AirGradient with Home Assistant 05:36 - Setting up the Air Lab 07:24 - Air Lab Console walkthrough 11:28 - Stale air on a road trip 13:02 - Air quality at an indoor convention 13:40 - Battery life, Home Assistant integration 14:05 - DIY or Buy 15:18 - Bonus, the Temtop
Not rated yetEvery Intel GPU runs on a Raspberry Pi
· 6 min
And you don't even have to recompile the Linux kernel! I purchased almost everything shown in this video, however, the Intel Arc Pro B50 was provided for testing by Intel, thus I'm marking this video as 'sponsored'. They did not pay me outside of providing the B50, nor did they have any say in the video's contents. Things mentioned in this video (some links are affiliate links): - Blog post with the latest guide for setup: https://www.jeffgeerling.com/blog/2025/all-intel-gpus-run-on-raspberry-pi-and-risc-v - Blog post on Resizeable BAR support: https://www.jeffgeerling.com/blog/2025/resizeable-bar-support-on-raspberry-pi - Minisforum DEG1 eGPU Dock: https://amzn.to/481EHao - JMT eGPU Dock: https://amzn.to/4oUZ4M9 - chenyang OCuLink cable: https://amzn.to/47ZxWWF - Micro SATA Cables OCuLink to M.2 adapter with ReDriver: https://amzn.to/4nTtJse - Lian-Li 750W SFX PSU: https://amzn.to/3JG3Myv - Linux Kernel Mailing List patch for Intel on non-x86: https://lore.kernel.org/lkml/20250715061837.2144388-1-zhangzhijie@bosc.ac.cn/ - Raspberry Pi OS Kernel PR for AMD and Intel GPUs: https://github.com/raspberrypi/linux/pull/7113 - Full testing and results using the latest patches: https://github.com/geerlingguy/raspberry-pi-pcie-devices/issues/764 Support me on Patreon: https://www.patreon.com/geerlingguy Sponsor me on GitHub: https://github.com/sponsors/geerlingguy Merch: https://www.redshirtjeff.com 2nd Channel: https://www.youtube.com/@GeerlingEngineering 3rd Channel: https://www.youtube.com/@Level2Jeff Contents: 00:00 - Every Intel GPU 00:51 - Physical connections and overcoming PCIe issues 01:38 - You don't need to recompile Linux anymore 03:09 - Black Mesa (the curse of the blinking cursor) 03:55 - Resizeable BAR support 04:27 - Performance: Vulkan, OpenGL, llama.cpp AI models 05:26 - Compared to AMD 05:56 - Partnering with RISC-V to go mainline
Not rated yet