Finally, we’ve gotten to a point in time when laptops are relatively affordable and easy to buy. Laptops became household items after being a dream for many to own, for many years. However, could laptops ever replace desktops for some of the most complicated uses? We often get this question from data scientists looking for an alternative for their desktop set-ups that occupy a large chunk of space in their homes.
Not everybody is lucky enough to live in a big spacious house that has room to spare to build a study setup for their work.
For all of you data scientists out there who share this problem, we researched how to overcome it and get yourself a laptop that will be a suitable or even a superior replacement for your desktop computer.
But first, we would like to talk about a few things you’ll encounter when choosing a laptop, and how to choose the best option to do the job right.
Choosing the “Brand”
The first thing you’ll ask yourself is: “What’s the best brand for data science?” The answer is incredibly simple – any brand will do. Sure, there are many brands available worldwide, and the prices vary from cheap to ultra-expensive.
However, for the work you’ll do, choosing a brand shouldn’t be an issue. Data analysis can be done with any brand of laptop, no matter how much you pay.
Many laptops can be overpriced just because of the famous brand that produced it. For example, there are many claims that Macs are way too expensive and that you can get a better deal if you choose a less-known brand but with strong specs.
Yes, we claim that any laptop will do the job, but the bottom of the barrel options will hardly make you more productive and efficient.
So put specs before the brand, and you won’t make a mistake.
Are the specs the only thing that matters?
Even if you buy a laptop with specs that would be considered overkill, the connection to a remote server can cast a shadow on its performance. If, like many data scientists, you’re analyzing web data feeds from Webhose, the connection to the server must be solid and fast.
If your desktop didn’t have any connection issues, then there shouldn’t be any if you replace it with a laptop. A fast connection is critical to get the job done on time. Do invest a little more on the faster internet connection.
1. RAM (System memory).
For the job of data analysis, the first thing you should check is the RAM. Don’t even think about getting a laptop with 8GB of RAM – 16GB should be your minimum if you want to have faster processing speed for heavy machine learning algorithms. Even though some might say that 8GB will be enough, with 16GB of RAM, you’ll have a laptop that will be better in the long term.
2. GPU (Graphics).
It’s important to understand that NVIDIA graphics will be the best choice. Anything above NVIDIA 960 series will mean a lot when you’re using deep learning libraries with CUDA processors. These processors only work well together with NVIDIA ones. If you choose an AMD chipset, you might end up writing a lot of low-level codes in OpenCL. Keep that in mind, when choosing a GPU in your laptop.
3. CPU (Processor).
To complement quality RAM and GPU, a processor should be the next thing to look into to round-up a solid specs laptop. We must recommend getting Intel 7th generation and above. The computing power of Intel processors is well-known, and in combination with 16GB of RAM and NVIDIA graphics, you’ll get a machine that will last you longer than you might think.
There is no better option than choosing an SSD to make your machine incredibly fast. An SSD will be a cherry on top when you buy a laptop. However, keep in mind that SSDs don’t come cheap. You might end up paying a lot if you want to buy a 1TB SSD. If you don’t want to spend too much cash, you can get one with 256GB and use it only for your OS. An external HDD can store all your data, but the SSD will run the OS at high speed.
There you have it, a quick guide on what to pay attention to if you’re planning to buy a laptop for data analysis. We hope that the transition will be smooth. We wish you all the best in your search for a quality option because a good one will make a big difference at the end of the day.