Student working with local datasets
If you work with Python, notebooks and tables at home, check how much RAM the specific configuration actually has. pandas explains that tables and intermediate copies can use a lot of memory.
Start with the tasks and datasets you actually run locally. pandas explains how tables and intermediate copies can require extra memory, while some scikit-learn operations can make use of multiple CPU cores. Only choose a specific GPU setup if your course or project requires a supported framework and computing platform. Check the complete configuration and connections for your home workspace, too. These are selection criteria, not a performance test of the models mentioned.
If you work with Python, notebooks and tables at home, check how much RAM the specific configuration actually has. pandas explains that tables and intermediate copies can use a lot of memory.
If you run models or experiments, first find out whether your course requires CPU calculations, a specific GPU platform or external computing resources. scikit-learn describes parallel calculations across multiple CPU cores; PyTorch and Microsoft Learn describe different software and GPU routes.
If you want to use a compact computer on an existing desk, make sure the operating system and connections on the specific configuration work with the monitors and other equipment you already have.
| What to consider | What should you check? | Who is this important for? |
|---|---|---|
| Memory for local data analysis | Check the installed RAM, not just the maximum. Also see whether the configuration can be upgraded. | If I run notebooks with local datasets, I want to know whether there’s enough memory for my work and any intermediate copies. Memory for local data analysis is the main priority for around 29% of consumers. |
| Total cost of the workspace | Check what’s included with the specific configuration and whether storage, an operating system, monitor, input devices and any necessary cables are included. | If I still need items for my desk, I want to assess the complete setup, not just the computer. The total cost of the workspace comes first for around 22% of consumers. |
| CPU for local calculations | Check the exact processor model and match it to the software and tasks you’ll run locally. | If I run calculations locally, I want to know which processor capabilities my tasks will actually use. For around 21% of consumers, local processing time is the most important buying priority. |
| Compatibility with GPU tasks | Check which framework and computing platform your course requires, and whether that combination works with the exact hardware and software. | If my course requires local GPU training, I want to be sure the complete hardware and software combination is right for that workflow. A suitable route for GPU tasks is the main priority for around 14% of consumers. |
| Connections and your home workspace | Compare the ports on the specific configuration with your monitors and other equipment, and check which accessories you still need. | If I want to connect the mini PC to my permanent workspace, its available outputs must suit my monitors and equipment. Fitting into a home office is the most important consideration for around 14% of consumers. |
ASUS lists up to 64 GB DDR5 for the ExpertCenter PN54-S1. That maximum doesn’t tell you how much memory is installed in a specific configuration.
Check the memory configuration and upgrade conditions for the exact product before ordering.
A compact computer doesn’t automatically make a complete home workspace. What you already own and what’s included with the specific configuration determine what else you’ll need.
Make a list of the monitor, input devices and cables you need, then compare it with the contents and connections of the exact configuration.
The points below apply to the product ranges and configurations mentioned; they aren’t a general assessment of the brands.
The ExpertCenter PN54-S1 is worth comparing if upgradeable memory and support for multiple screens matter for your permanent workspace.
The ThinkCentre Neo 50q Gen 5 Tiny is a compact business desktop range with several configurations.
The Mac mini may suit a macOS workspace if the software and tasks you use for your studies are compatible with it.
The Pro Mini 400 G9 is a product range worth comparing if you’re looking for a compact Windows workspace.
The GTi15 Ultra is worth considering if you’re thinking about an upgradeable GPU setup as well as a mini PC.
These mini PCs are aimed at local study tasks and a permanent home workspace. For each configuration, check that the memory, operating system and connections suit your datasets, software and existing equipment.
No image available
A compact option if you want to run local study tasks; consider the memory configuration in relation to your datasets.
Windows 11, 8 GB DDR4, 256 GB SATA SSD, Bluetooth 5.2 and Wi-Fi 5.
No image available
An option to compare for local study tasks and a desk setup where you want output to multiple screens.
16 GB RAM, 512 GB SSD, Wi-Fi 6, Bluetooth 5.2 and output to three 4K screens.
No image available
A compact option for study and everyday use; compare the memory and connections with your own workspace.
16 GB LPDDR5, 512 GB M.2 NVMe SSD, Windows 11 Home, HDMI 2.1 and DisplayPort 1.4.
No image available
A configuration with 24 GB memory to compare if you run local tasks and need your workspace to support multiple screens.
24 GB LPDDR5, 512 GB PCIe 4.0 SSD, Radeon 760M graphics and output to three 4K screens.
Match the memory capacity to the size and working methods of your own tasks. pandas explains that tables and intermediate copies can use a lot of memory; processing data in chunks can sometimes be an alternative. There’s no general RAM threshold that applies to every data science student.
Don’t assume that integrated graphics will meet a CUDA requirement. Check the required framework and the exact hardware and software combination. Microsoft Learn describes a specific route for NVIDIA CUDA in WSL, with its own installation requirements.
Check the display outputs and USB ports on the configuration you’ve chosen, and compare them with your monitors and input devices. Also make sure the installation instructions for your course are compatible with the operating system and supported packages.
First check the rules and agreements that apply to your project. The Dutch Data Protection Authority considers storing and sending personal data to be processing under the GDPR. Whether you can use a personal device or cloud service depends, among other things, on the data and the agreements with your course provider.
According to the Dutch Government, when you buy remotely you usually have a 14-day cooling-off period; check whether the rules apply to your purchase. You’re also entitled to a product that works properly, and if there’s a problem you should contact the seller to find a solution. An additional warranty doesn’t replace your statutory rights.