# How to build Deep Learning Computer **URL:** https://www.compsmag.com/blogs/how-to-build-deep-learning-computer/ **Author:** Ayushi Chauhan **Published:** 2022-04-21 **Updated:** 2022-04-21 **Categories:** Blogs **Tags:** Blogs Daily, Computer Guides, guiderobert, Guides and Tutorials, guiding12, How To Guide, How To Guides, Linux Guides, Tips, Tricks **Reading Time:** 6 min --- This tip is about the how to build Deep Learning Computer. So read this free guide, How to build Deep Learning Computer step by step. If you have query related to same article you may contact us. ## How to build Deep Learning Computer - Guide Recent advances in Deep Learning and its implementation have created chaos in the market. Monster GPUs in the cloud can run massive computations in parallel at lightning speed. In areas like speech recognition, image classification, and autonomous vehicles, mastering the objective requires a lot of training data and running it on 4 or 8 GB of RAM can be a never-ending process on your personal PC. And accessing cloud services for personal model training can cost a fortune. Here, we will study the specifications needed to build a small-scale system for deep learning applications at your location. We'll also see why high-end CPUs can't be compared to low-end GPUs when it comes to parallel data computing. ## CPU (Processor) The CPU is the main processor of our computer, so in reality it is the unit that processes all the information to run the programs we use and to run the operating system itself. ### How important is the CPU for Deep Learning? Considering that all the heavy lifting for Deep Learning will be done by the GPU (graphics card), we can say that the CPU is not that important. At the same time, we want a decent CPU to have a fast computer that is capable of multitasking when we open a lot of applications. A CPU of 6 cores (12 threads) is now a good enough processor to use the computer to work even with heavy programs, and even if you like to play video games, it will be good to play in FULL HD (of course with support up for a good graphics card too). #### AMD or Intel? On CPUs with similar performance AMD usually has a better price, so my choice goes to AMD. ### CPU choice AMD Ryzen 5 2600 (6 cores, 3.4GHz) – $141 ## GPU (graphics card) The GPU is the most important component for Deep Learning as it will take care of all the computing power to run the deep learning libraries when we train/test our models. ### How to choose the right GPU? When choosing a GPU, we need to consider these 2 things: It is compatible with the Deep Learning library we use. The most common deep learning libraries at the moment are Tensorflow, Pytorch and Keras. It has enough vRAM memory to fit the models. in a simpler way words, the larger our training dataset, the more memory we need. If there is not enough memory, we will get an error. #### AMD or Nvidia? Nvidia. There is no real competition on graphics cards for deep learning and Nvidia is the only way to go right now. Nvidia developed CUDA, an architecture that allows parallel computing and is compatible with the most important deep learning libraries (Tensorflow, Pytorch, Keras, Darknet and others). #### What about AMD, is there an alternative AMD version of CUDA? The amd CUDA alternative exists and is called ROCM, but it is not popular yet, there is a lack of support regarding nvidia and the performances are still too low to be considered as a real alternative. ### GPU Options Nvidia GTX 1660 (6GB, 1408 CUDA cores) – $220 Nvidia GTX 1070 (8GB, 1920 CUDA cores) – $400 Nvidia GTX 1080TI (11GB, 3584 CUDA cores) – $1050 ## Motherboard The motherboard is the board that connects all the components together. #### How to choose the motherboard? When choosing a motherboard, we need to take into account: It is compatible with the CPU we have chosen. It has a PCI Express 16x 3.0 slot. This is usually not a problem, as all modern motherboards have them by default. Even better if there are more slots, so in the future we may decide to upgrade the computer by adding a second GPU. ### Motherboard choice: Gigabyte Aorus Elite b450 (ATX motherboard) – 110$ ## RAM RAM (Random Access Memory) in a normal use of the computer is important when we want to open many applications, since once opened, they will be temporarily loaded in RAM). For deep learning I suggest a minimum of at least 16GB and a minimum frequency of 2400mhz. ### RAM Options 16GB (2x8GB) – 70$ 32GB (2 x 16GB) – 140$ ## PSU (Power Supply) The power supply is the unit that supplies current to all components in the computer. ### How to choose the power supply? When choosing a power supply, we need to consider: How many watts does the computer need. The 2 components that consume the most power are the CPU and the GPU. The CPU I suggested takes around 50 watts while GPUs take 100 to 250 watts. The power supply must be of good quality. Don't try to save money by buying cheap power supplies (under $50) because they can easily damage your computer. There is an identification that indicates the quality of a power supply and it is the 80 PLUS mark. You can take a look here to learn more about it: https://en.wikipedia.org/wiki/80_Plus A 600 watt power supply will be good enough to support all the components I suggested in this post. ### power supply options EVGA 600w (80+ Bronze) – $85 Cooler Master MWE 600 Watt (80+ Bronze) – 80$ ## Memory We need fast memory to install and run the operating system (windows or linux) and large memory to save files. It's a good option to get 2 different memories. For the operating system, we can get an NVMe, which is a very fast SSD drive that can be connected to the motherboard's PCI express. I suggest buying at least a 250GB disk to install the operating system. As a second drive, we can use a classic internal hard drive. The price is really low and we can get a big 2TB memory for a little over 50 bucks. ### memory options Silicon Power 256GB (NVMe) – 38$ Seagate Barracuda 2TB (3.5″ HDD) – $55 ### Case How to choose the Case? When choosing the case you need to consider two factors: It's big enough to fit all the components. If you have an ATX motherboard, the case must also be ATX. It has good airflow to keep hardware cool. Normally, nowadays the house has 3 front fans that move fresh air inside the case, and a fan at the back that takes the hot air out of the case. ### case choice Montech FIGHTER 600 ATX (Midtower case, 4 FANS included) – $54 ## Final note I hope you like the guide How to build Deep Learning Computer. In case if you have any query regards this article you may ask us. Also, please share your love by sharing this article with your friends. --- *End of Article*