VDR Official X
February 10, 2026
Aleksa Gordić - The AI Epiphany • ❤️ Become The AI Epiphany Patreon ❤️ https://www.patreon.com/theaiepiphany 👨👩👧👦 Join our Discord community 👨👩👧👦 https://discord.gg/peBrCpheKE In this video, I tell you everything you need to know about building your own machine learning workstation! Should you build it? How to do proper research on the components, which websites to use, how to buy components, and much more. ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Disclaimer: below are affiliate links - by clicking & buying I earn a percentage at no additional cost to you! :) Core components: * RTX 3090 https://geni.us/CU4C * AMD Ryzen 9 7950x https://geni.us/mJAJX * ARCTIC Liquid Freezer II 280 https://geni.us/7ec1Z * Asus ROG Crosshair X670E Hero https://geni.us/XxIa * Kingston Fury Beast https://geni.us/FB4a * Samsung 980 PRO 2 TB m.2 https://geni.us/AQuFQtB * SAMSUNG 870 QVO SATA III 8TB https://geni.us/iyDmwX * Lian Li PC-O11DW 011 DYNAMIC https://geni.us/2NrKFf * EVGA Supernova 1600 G+ https://geni.us/WGcJ Peripherals I use (super happy with my decision esp. keyboard!): * Dell U2720Q UltraSharp 27 Inch 4K https://geni.us/eyUy * Corsair K70 RGB MK.2 Mechanical w/ MX brown https://geni.us/bgZw * Mouse: Corsair M65 PRO RGB https://geni.us/ghLb List: https://uk.pcpartpicker.com/list/ryMsBj ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Blogs I recommended in the video: https://timdettmers.com/2020/09/07/which-gpu-for-deep-learning https://timdettmers.com/2018/12/16/deep-learning-hardware-guide https://www.emilwallner.com/p/ml-rig https://nonint.com/2022/05/30/my-deep-learning-rig/ https://www.mrdbourke.com/notes-on-building-a-deep-learning-pc/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ ⌚️ Timetable: 00:00 Intro - announcing the new series 02:00 Why should you build your deep learning workstation? 03:56 Quick skim of the components 04:45 pcpartpicker is your friend 07:20 Cost comparisons - Lambda Labs workstation & laptop, cloud 15:45 Researching the components - useful resources 15:57 Overview Tim Dettmers' blogs 19:20 Do you need 11+ GBs? 22:55 Don't follow someone's advice blindly! 27:08 Upgrade to NVIDIA 40x series? 28:05 Other resources 30:30 How to pick the right components? 32:10 CPU - Ryzen 9 7950x or Threadripper? 33:50 CPU Cooler 34:45 MoBo 37:50 Memory and storage 40:40 Case 42:00 PSU 42:55 Quick mention of my peripherals 43:15 Advice for targeting your particular budget 44:25 Tips for buying components - site aggregators, Amazon, legitness 48:25 Outro ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 💰 BECOME A PATREON OF THE AI EPIPHANY ❤️ If these videos, GitHub projects, and blogs help you, consider helping me out by supporting me on Patreon! The AI Epiphany - https://www.patreon.com/theaiepiphany One-time donation - https://www.paypal.com/paypalme/theaiepiphany Huge thank you to these AI Epiphany patreons: Eli Mahler Petar Veličković ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 📄 Website - https://gordicaleksa.com/ 💼 LinkedIn - https://www.linkedin.com/in/aleksagordic/ 🐦 Twitter - https://twitter.com/gordic_aleksa 👨👩👧👦 Discord - https://discord.gg/peBrCpheKE 📺 YouTube - https://www.youtube.com/c/TheAIEpiphany/ 📚 Medium - https://gordicaleksa.medium.com/ 💻 GitHub - https://github.com/gordicaleksa 📢 AI Newsletter - https://aiepiphany.substack.com/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ #deeplearning #machinelearning #workstation #rig #pc #build
Content Summary
This report is generated from research on the following videos, based on the requirements set in Video Deep Research.
Analyze selected videos,
My goal is 📑 Discover Content Intelligence
My role is 👓 Business Analyst
I need: 💡 Strategic recommendations and action items analysis, 📊 Data points extraction and trend identification with charts, 🔥 Business trends extraction
https:...WTBQ
Summary
1. Financial Optimization and ROI Analysis
2. Infrastructure Performance and Thermal Strategy
3. Risk Mitigation in Hardware Procurement
Knowledge Snap
Strategic Cost Analysis
🎬 Related Clip
(3)
Video Title
03:50 - 07:20
Pre-assembled workstations are expensive due to warranties and the labor involved in assembly.
11:11 - 13:11
Users can save money by installing their own operating systems and necessary software libraries.
09:31 - 11:35
Individuals can save at least twenty-five hundred pounds by building their own workstation.
Raw Performance Optimization
🎬 Related Clip
(3)
Video Title
03:21 - 07:20
Local machines are faster because they lack the input-output overhead found in cloud environments.
03:33 - 07:20
Workstation so Additionally you don't have to deal with preemptions with interrupt and so it's just much easier.
03:21 - 07:20
For the i o overhead so in general those machines are much slower than if they.
Thermal Management Strategy
🎬 Related Clip
(3)
Video Title
06:19 - 14:19
Liquid cooling with large fans is recommended for managing the heat of powerful CPUs.
06:59 - 14:59
Checking dimensions is vital to ensure the CPU cooler does not block the case panel.
07:05 - 15:05
Incorrectly sized components can prevent the front panel of the computer case from closing.
System Architecture Synergy
🎬 Related Clip
(3)
Video Title
26:33 - 30:30
Verify that your chosen CPU and motherboard can support the intended number of graphics cards.
26:47 - 30:30
The number of PCI lanes is a factor when using multiple high-speed storage devices.
30:18 - 33:50
Bandwidth for graphics cards can be reduced if too many lanes are used by other components.
Future-Proofing Hardware Investments
🎬 Related Clip
(3)
Video Title
24:09 - 26:13
Current GPU owners might be better off waiting for future chip architectures before upgrading.
27:35 - 30:30
Scaling performance is becoming more difficult as chip features reach their physical limits.
27:53 - 30:30
Research is necessary before deciding to replace older hardware with the latest releases.
Research and Sourcing Methodology
🎬 Related Clip
(3)
Video Title
28:44 - 32:10
The completed builds section of part-picking websites provides useful inspiration for new designs.
00:13 - 03:56
Online PC builders list all the necessary components needed to complete a workstation.
05:42 - 13:42
Incompatibilities or issues that you might have so let me uh take a particular example to to demonstrate.
Strategy for Building Machine Learning Infrastructure
UCj8shE7aIn4Yawwbo2FceCQ
🚀
Infrastructure Project Launch
00:00 - 02:00
The series begins by outlining the roadmap for designing and assembling a customized machine learning workstation.
💰
Investment Break-Even Analysis
02:11 - 04:45
Initial capital expenditures are compared against long-term operational savings when moving away from cloud-based solutions.
🛠️
Platform Selection for Research
04:50 - 06:53
Utilizing specialized web tools helps streamline the selection process for complex hardware components and ensures system compatibility.
📉
Market Benchmark Comparison
10:53 - 12:53
A side-by-side cost assessment highlights the significant price premium charged by pre-assembled workstation vendors.
📖
External Expert Validation
16:14 - 18:17
Reviewing external technical literature provides deep insights into specific hardware performance for neural network training.
🧠
Resource Allocation for Models
18:44 - 22:55
Specific memory thresholds are identified as critical requirements for handling advanced model architectures like transformers.
⚠️
Critical Risk Mitigation
24:27 - 26:33
Identifying potential inaccuracies in online guides prevents system failures related to power supply and thermal management.
💡
Final Implementation Intelligence
30:25 - 33:50
Aggregating component lists from diverse community sources serves as the final blueprint for the build.
Learning Pathway for Architecting Machine Learning Infrastructure
| Stage | Videos |
|---|---|
1. Defining the Business Rationale | phany Patreon ❤️ https://www.patreon.com/theaiepiphany 👨👩👧👦 Join our Discord community 👨👩👧👦 https://discord.gg/peBrCpheKE In this video, I tell you everything you need to know about building your own machine learning workstation! Should you build it? How to do proper research on the components, which websites to use, how to buy components, and much more. ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Disclaimer: below are affiliate links - by clicking & buying I earn a percentage at no additional cost to you! :) Core components: * RTX 3090 https://geni.us/CU4C * AMD Ryzen 9 7950x https://geni.us/mJAJX * ARCTIC Liquid Freezer II 280 https://geni.us/7ec1Z * Asus ROG Crosshair X670E Hero https://geni.us/XxIa * Kingston Fury Beast https://geni.us/FB4a * Samsung 980 PRO 2 TB m.2 https://geni.us/AQuFQtB * SAMSUNG 870 QVO SATA III 8TB https://geni.us/iyDmwX * Lian Li PC-O11DW 011 DYNAMIC https://geni.us/2NrKFf * EVGA Supernova 1600 G+ https://geni.us/WGcJ Peripherals I use (super happy with my decision esp. keyboard!): * Dell U2720Q UltraSharp 27 Inch 4K https://geni.us/eyUy * Corsair K70 RGB MK.2 Mechanical w/ MX brown https://geni.us/bgZw * Mouse: Corsair M65 PRO RGB https://geni.us/ghLb List: https://uk.pcpartpicker.com/list/ryMsBj ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Blogs I recommended in the video: https://timdettmers.com/2020/09/07/which-gpu-for-deep-learning https://timdettmers.com/2018/12/16/deep-learning-hardware-guide https://www.emilwallner.com/p/ml-rig https://nonint.com/2022/05/30/my-deep-learning-rig/ https://www.mrdbourke.com/notes-on-building-a-deep-learning-pc/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ ⌚️ Timetable: 00:00 Intro - announcing the new series 02:00 Why should you build your deep learning workstation? 03:56 Quick skim of the components 04:45 pcpartpicker is your friend 07:20 Cost comparisons - Lambda Labs workstation & laptop, cloud 15:45 Researching the components - useful resources 15:57 Overview Tim Dettmers' blogs 19:20 Do you need 11+ GBs? 22:55 Don't follow someone's advice blindly! 27:08 Upgrade to NVIDIA 40x series? 28:05 Other resources 30:30 How to pick the right components? 32:10 CPU - Ryzen 9 7950x or Threadripper? 33:50 CPU Cooler 34:45 MoBo 37:50 Memory and storage 40:40 Case 42:00 PSU 42:55 Quick mention of my peripherals 43:15 Advice for targeting your particular budget 44:25 Tips for buying components - site aggregators, Amazon, legitness 48:25 Outro ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 💰 BECOME A PATREON OF THE AI EPIPHANY ❤️ If these videos, GitHub projects, and blogs help you, consider helping me out by supporting me on Patreon! The AI Epiphany - https://www.patreon.com/theaiepiphany One-time donation - https://www.paypal.com/paypalme/theaiepiphany Huge thank you to these AI Epiphany patreons: Eli Mahler Petar Veličković ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 📄 Website - https://gordicaleksa.com/ 💼 LinkedIn - https://www.linkedin.com/in/aleksagordic/ 🐦 Twitter - https://twitter.com/gordic_aleksa 👨👩👧👦 Discord - https://discord.gg/peBrCpheKE 📺 YouTube - https://www.youtube.com/c/TheAIEpiphany/ 📚 Medium - https://gordicaleksa.medium.com/ 💻 GitHub - https://github.com/gordicaleksa 📢 AI Newsletter - https://aiepiphany.substack.com/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ #deeplearning #machinelearning #workstation #rig #pc #build |
2. System Compatibility Verification | phany Patreon ❤️ https://www.patreon.com/theaiepiphany 👨👩👧👦 Join our Discord community 👨👩👧👦 https://discord.gg/peBrCpheKE In this video, I tell you everything you need to know about building your own machine learning workstation! Should you build it? How to do proper research on the components, which websites to use, how to buy components, and much more. ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Disclaimer: below are affiliate links - by clicking & buying I earn a percentage at no additional cost to you! :) Core components: * RTX 3090 https://geni.us/CU4C * AMD Ryzen 9 7950x https://geni.us/mJAJX * ARCTIC Liquid Freezer II 280 https://geni.us/7ec1Z * Asus ROG Crosshair X670E Hero https://geni.us/XxIa * Kingston Fury Beast https://geni.us/FB4a * Samsung 980 PRO 2 TB m.2 https://geni.us/AQuFQtB * SAMSUNG 870 QVO SATA III 8TB https://geni.us/iyDmwX * Lian Li PC-O11DW 011 DYNAMIC https://geni.us/2NrKFf * EVGA Supernova 1600 G+ https://geni.us/WGcJ Peripherals I use (super happy with my decision esp. keyboard!): * Dell U2720Q UltraSharp 27 Inch 4K https://geni.us/eyUy * Corsair K70 RGB MK.2 Mechanical w/ MX brown https://geni.us/bgZw * Mouse: Corsair M65 PRO RGB https://geni.us/ghLb List: https://uk.pcpartpicker.com/list/ryMsBj ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Blogs I recommended in the video: https://timdettmers.com/2020/09/07/which-gpu-for-deep-learning https://timdettmers.com/2018/12/16/deep-learning-hardware-guide https://www.emilwallner.com/p/ml-rig https://nonint.com/2022/05/30/my-deep-learning-rig/ https://www.mrdbourke.com/notes-on-building-a-deep-learning-pc/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ ⌚️ Timetable: 00:00 Intro - announcing the new series 02:00 Why should you build your deep learning workstation? 03:56 Quick skim of the components 04:45 pcpartpicker is your friend 07:20 Cost comparisons - Lambda Labs workstation & laptop, cloud 15:45 Researching the components - useful resources 15:57 Overview Tim Dettmers' blogs 19:20 Do you need 11+ GBs? 22:55 Don't follow someone's advice blindly! 27:08 Upgrade to NVIDIA 40x series? 28:05 Other resources 30:30 How to pick the right components? 32:10 CPU - Ryzen 9 7950x or Threadripper? 33:50 CPU Cooler 34:45 MoBo 37:50 Memory and storage 40:40 Case 42:00 PSU 42:55 Quick mention of my peripherals 43:15 Advice for targeting your particular budget 44:25 Tips for buying components - site aggregators, Amazon, legitness 48:25 Outro ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 💰 BECOME A PATREON OF THE AI EPIPHANY ❤️ If these videos, GitHub projects, and blogs help you, consider helping me out by supporting me on Patreon! The AI Epiphany - https://www.patreon.com/theaiepiphany One-time donation - https://www.paypal.com/paypalme/theaiepiphany Huge thank you to these AI Epiphany patreons: Eli Mahler Petar Veličković ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 📄 Website - https://gordicaleksa.com/ 💼 LinkedIn - https://www.linkedin.com/in/aleksagordic/ 🐦 Twitter - https://twitter.com/gordic_aleksa 👨👩👧👦 Discord - https://discord.gg/peBrCpheKE 📺 YouTube - https://www.youtube.com/c/TheAIEpiphany/ 📚 Medium - https://gordicaleksa.medium.com/ 💻 GitHub - https://github.com/gordicaleksa 📢 AI Newsletter - https://aiepiphany.substack.com/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ #deeplearning #machinelearning #workstation #rig #pc #build |
3. Resource Optimization for Advanced Models | phany Patreon ❤️ https://www.patreon.com/theaiepiphany 👨👩👧👦 Join our Discord community 👨👩👧👦 https://discord.gg/peBrCpheKE In this video, I tell you everything you need to know about building your own machine learning workstation! Should you build it? How to do proper research on the components, which websites to use, how to buy components, and much more. ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Disclaimer: below are affiliate links - by clicking & buying I earn a percentage at no additional cost to you! :) Core components: * RTX 3090 https://geni.us/CU4C * AMD Ryzen 9 7950x https://geni.us/mJAJX * ARCTIC Liquid Freezer II 280 https://geni.us/7ec1Z * Asus ROG Crosshair X670E Hero https://geni.us/XxIa * Kingston Fury Beast https://geni.us/FB4a * Samsung 980 PRO 2 TB m.2 https://geni.us/AQuFQtB * SAMSUNG 870 QVO SATA III 8TB https://geni.us/iyDmwX * Lian Li PC-O11DW 011 DYNAMIC https://geni.us/2NrKFf * EVGA Supernova 1600 G+ https://geni.us/WGcJ Peripherals I use (super happy with my decision esp. keyboard!): * Dell U2720Q UltraSharp 27 Inch 4K https://geni.us/eyUy * Corsair K70 RGB MK.2 Mechanical w/ MX brown https://geni.us/bgZw * Mouse: Corsair M65 PRO RGB https://geni.us/ghLb List: https://uk.pcpartpicker.com/list/ryMsBj ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Blogs I recommended in the video: https://timdettmers.com/2020/09/07/which-gpu-for-deep-learning https://timdettmers.com/2018/12/16/deep-learning-hardware-guide https://www.emilwallner.com/p/ml-rig https://nonint.com/2022/05/30/my-deep-learning-rig/ https://www.mrdbourke.com/notes-on-building-a-deep-learning-pc/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ ⌚️ Timetable: 00:00 Intro - announcing the new series 02:00 Why should you build your deep learning workstation? 03:56 Quick skim of the components 04:45 pcpartpicker is your friend 07:20 Cost comparisons - Lambda Labs workstation & laptop, cloud 15:45 Researching the components - useful resources 15:57 Overview Tim Dettmers' blogs 19:20 Do you need 11+ GBs? 22:55 Don't follow someone's advice blindly! 27:08 Upgrade to NVIDIA 40x series? 28:05 Other resources 30:30 How to pick the right components? 32:10 CPU - Ryzen 9 7950x or Threadripper? 33:50 CPU Cooler 34:45 MoBo 37:50 Memory and storage 40:40 Case 42:00 PSU 42:55 Quick mention of my peripherals 43:15 Advice for targeting your particular budget 44:25 Tips for buying components - site aggregators, Amazon, legitness 48:25 Outro ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 💰 BECOME A PATREON OF THE AI EPIPHANY ❤️ If these videos, GitHub projects, and blogs help you, consider helping me out by supporting me on Patreon! The AI Epiphany - https://www.patreon.com/theaiepiphany One-time donation - https://www.paypal.com/paypalme/theaiepiphany Huge thank you to these AI Epiphany patreons: Eli Mahler Petar Veličković ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 📄 Website - https://gordicaleksa.com/ 💼 LinkedIn - https://www.linkedin.com/in/aleksagordic/ 🐦 Twitter - https://twitter.com/gordic_aleksa 👨👩👧👦 Discord - https://discord.gg/peBrCpheKE 📺 YouTube - https://www.youtube.com/c/TheAIEpiphany/ 📚 Medium - https://gordicaleksa.medium.com/ 💻 GitHub - https://github.com/gordicaleksa 📢 AI Newsletter - https://aiepiphany.substack.com/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ #deeplearning #machinelearning #workstation #rig #pc #build |
4. Data-Driven Market Benchmarking | phany Patreon ❤️ https://www.patreon.com/theaiepiphany 👨👩👧👦 Join our Discord community 👨👩👧👦 https://discord.gg/peBrCpheKE In this video, I tell you everything you need to know about building your own machine learning workstation! Should you build it? How to do proper research on the components, which websites to use, how to buy components, and much more. ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Disclaimer: below are affiliate links - by clicking & buying I earn a percentage at no additional cost to you! :) Core components: * RTX 3090 https://geni.us/CU4C * AMD Ryzen 9 7950x https://geni.us/mJAJX * ARCTIC Liquid Freezer II 280 https://geni.us/7ec1Z * Asus ROG Crosshair X670E Hero https://geni.us/XxIa * Kingston Fury Beast https://geni.us/FB4a * Samsung 980 PRO 2 TB m.2 https://geni.us/AQuFQtB * SAMSUNG 870 QVO SATA III 8TB https://geni.us/iyDmwX * Lian Li PC-O11DW 011 DYNAMIC https://geni.us/2NrKFf * EVGA Supernova 1600 G+ https://geni.us/WGcJ Peripherals I use (super happy with my decision esp. keyboard!): * Dell U2720Q UltraSharp 27 Inch 4K https://geni.us/eyUy * Corsair K70 RGB MK.2 Mechanical w/ MX brown https://geni.us/bgZw * Mouse: Corsair M65 PRO RGB https://geni.us/ghLb List: https://uk.pcpartpicker.com/list/ryMsBj ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Blogs I recommended in the video: https://timdettmers.com/2020/09/07/which-gpu-for-deep-learning https://timdettmers.com/2018/12/16/deep-learning-hardware-guide https://www.emilwallner.com/p/ml-rig https://nonint.com/2022/05/30/my-deep-learning-rig/ https://www.mrdbourke.com/notes-on-building-a-deep-learning-pc/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ ⌚️ Timetable: 00:00 Intro - announcing the new series 02:00 Why should you build your deep learning workstation? 03:56 Quick skim of the components 04:45 pcpartpicker is your friend 07:20 Cost comparisons - Lambda Labs workstation & laptop, cloud 15:45 Researching the components - useful resources 15:57 Overview Tim Dettmers' blogs 19:20 Do you need 11+ GBs? 22:55 Don't follow someone's advice blindly! 27:08 Upgrade to NVIDIA 40x series? 28:05 Other resources 30:30 How to pick the right components? 32:10 CPU - Ryzen 9 7950x or Threadripper? 33:50 CPU Cooler 34:45 MoBo 37:50 Memory and storage 40:40 Case 42:00 PSU 42:55 Quick mention of my peripherals 43:15 Advice for targeting your particular budget 44:25 Tips for buying components - site aggregators, Amazon, legitness 48:25 Outro ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 💰 BECOME A PATREON OF THE AI EPIPHANY ❤️ If these videos, GitHub projects, and blogs help you, consider helping me out by supporting me on Patreon! The AI Epiphany - https://www.patreon.com/theaiepiphany One-time donation - https://www.paypal.com/paypalme/theaiepiphany Huge thank you to these AI Epiphany patreons: Eli Mahler Petar Veličković ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 📄 Website - https://gordicaleksa.com/ 💼 LinkedIn - https://www.linkedin.com/in/aleksagordic/ 🐦 Twitter - https://twitter.com/gordic_aleksa 👨👩👧👦 Discord - https://discord.gg/peBrCpheKE 📺 YouTube - https://www.youtube.com/c/TheAIEpiphany/ 📚 Medium - https://gordicaleksa.medium.com/ 💻 GitHub - https://github.com/gordicaleksa 📢 AI Newsletter - https://aiepiphany.substack.com/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ #deeplearning #machinelearning #workstation #rig #pc #build |
5. Strategic Risk and Accuracy Assessment | phany Patreon ❤️ https://www.patreon.com/theaiepiphany 👨👩👧👦 Join our Discord community 👨👩👧👦 https://discord.gg/peBrCpheKE In this video, I tell you everything you need to know about building your own machine learning workstation! Should you build it? How to do proper research on the components, which websites to use, how to buy components, and much more. ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Disclaimer: below are affiliate links - by clicking & buying I earn a percentage at no additional cost to you! :) Core components: * RTX 3090 https://geni.us/CU4C * AMD Ryzen 9 7950x https://geni.us/mJAJX * ARCTIC Liquid Freezer II 280 https://geni.us/7ec1Z * Asus ROG Crosshair X670E Hero https://geni.us/XxIa * Kingston Fury Beast https://geni.us/FB4a * Samsung 980 PRO 2 TB m.2 https://geni.us/AQuFQtB * SAMSUNG 870 QVO SATA III 8TB https://geni.us/iyDmwX * Lian Li PC-O11DW 011 DYNAMIC https://geni.us/2NrKFf * EVGA Supernova 1600 G+ https://geni.us/WGcJ Peripherals I use (super happy with my decision esp. keyboard!): * Dell U2720Q UltraSharp 27 Inch 4K https://geni.us/eyUy * Corsair K70 RGB MK.2 Mechanical w/ MX brown https://geni.us/bgZw * Mouse: Corsair M65 PRO RGB https://geni.us/ghLb List: https://uk.pcpartpicker.com/list/ryMsBj ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ Blogs I recommended in the video: https://timdettmers.com/2020/09/07/which-gpu-for-deep-learning https://timdettmers.com/2018/12/16/deep-learning-hardware-guide https://www.emilwallner.com/p/ml-rig https://nonint.com/2022/05/30/my-deep-learning-rig/ https://www.mrdbourke.com/notes-on-building-a-deep-learning-pc/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ ⌚️ Timetable: 00:00 Intro - announcing the new series 02:00 Why should you build your deep learning workstation? 03:56 Quick skim of the components 04:45 pcpartpicker is your friend 07:20 Cost comparisons - Lambda Labs workstation & laptop, cloud 15:45 Researching the components - useful resources 15:57 Overview Tim Dettmers' blogs 19:20 Do you need 11+ GBs? 22:55 Don't follow someone's advice blindly! 27:08 Upgrade to NVIDIA 40x series? 28:05 Other resources 30:30 How to pick the right components? 32:10 CPU - Ryzen 9 7950x or Threadripper? 33:50 CPU Cooler 34:45 MoBo 37:50 Memory and storage 40:40 Case 42:00 PSU 42:55 Quick mention of my peripherals 43:15 Advice for targeting your particular budget 44:25 Tips for buying components - site aggregators, Amazon, legitness 48:25 Outro ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 💰 BECOME A PATREON OF THE AI EPIPHANY ❤️ If these videos, GitHub projects, and blogs help you, consider helping me out by supporting me on Patreon! The AI Epiphany - https://www.patreon.com/theaiepiphany One-time donation - https://www.paypal.com/paypalme/theaiepiphany Huge thank you to these AI Epiphany patreons: Eli Mahler Petar Veličković ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ 📄 Website - https://gordicaleksa.com/ 💼 LinkedIn - https://www.linkedin.com/in/aleksagordic/ 🐦 Twitter - https://twitter.com/gordic_aleksa 👨👩👧👦 Discord - https://discord.gg/peBrCpheKE 📺 YouTube - https://www.youtube.com/c/TheAIEpiphany/ 📚 Medium - https://gordicaleksa.medium.com/ 💻 GitHub - https://github.com/gordicaleksa 📢 AI Newsletter - https://aiepiphany.substack.com/ ▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬ #deeplearning #machinelearning #workstation #rig #pc #build |
Detailed Findings and Insights
1. GPU Burst Current Reality
🎬 Related Clip
(2)
Video Title
17:21 - 19:24
Modern GPUs can draw significant power bursts that exceed their official wattage ratings.
24:15 - 26:21
Sudden peak power draws can trip safety protections every few hours if unplanned for.
2. CPU Market Shift Errors
🎬 Related Clip
(2)
Video Title
17:48 - 22:55
Processor performance rankings can change significantly between different release years.
08:28 - 10:35
Choosing the wrong CPU can be a major error in building a high-performance system.
3. RAM Clock Rate Gimmicks
🎬 Related Clip
(2)
Video Title
22:05 - 27:08
Memory clock rates are often highlighted as a gimmick with little impact on training.
22:18 - 27:08
Lower clock rate RAM provides essentially the same performance for much less cost.
4. VRAM Requirements for Modern Models
🎬 Related Clip
(2)
Video Title
18:44 - 22:55
Training complex models like Transformers requires significant amounts of video memory.
20:50 - 22:57
If you plan to train large models, always prioritize purchasing more video memory.
5. Storage Bandwidth Competition
🎬 Related Clip
(2)
Video Title
26:33 - 30:30
Each high-speed SSD uses a specific number of communication lanes on the system board.
27:02 - 30:30
PCI lane management is crucial for squeezing the maximum performance out of a PC setup.
6. Thermal Impact of Case Panels
🎬 Related Clip
(2)
Video Title
07:05 - 15:05
Removing the front panel of a computer case can lead to a significant drop in temperature.
26:01 - 28:05
Certain cases consistently maintain lower temperatures than others during high-load benchmarking.
