14 DAY TRIAL // The Steam Hardware Survey for December has been released, and it’s not good news for the Linux platform, which doesn’t seem capable of going past the 1% milestone.
Up until a little over six months ago, the Steam for Linux market share hovered around 1.2%, but something changed overnight, and the usage dropped to about 0.8%. Most likely, Valve changed something in the algorithm, but since nothing of what they do is transparent, it’s simply just a guessing game.
The hardware survey is not a precise tool, and it’s probably not used by developers to gauge interest in the Linux platform, but it’s still being presented each month, and the rest of the numbers seem to be correct. We can’t say the same for the Linux percentages since there are some inconsistencies in the details.
These are not the numbers you’re looking for
The fact that the numbers don’t add up is the first indication that something is wrong. The Steam Hardware Survey used to show a lot more details about the Linux distributions, but that’s not happening anymore. We’re left with four entries; Ubuntu 14.04.3 LTS 64-bit, Ubuntu 15.10 64-bit, Linux 64-bit, and Linux Mint 17.2 Rafaela 64-bit.
The Linux 64-bit entry probably refers to all the 64-bit Linux distros that have Steam and that are not included in the list. The problem is that if we add all the percentages, we only get 0.57%, and the total listed is 0.96% (which is a 0.02% drop). Where is the rest? Is it so difficult to have a miscellaneous category?
Secondly, I find it hard to believe that Ubuntu 14.04.3 LTS 64-bit is more used than Ubuntu 15.10 64-bit. Sure, it’s an LTS, and it has a lot of users, but Ubuntu 15.10 has a much newer Linux kernel and better support for gaming, and traditionally newer Ubuntu version did better than the previous ones, even if they were LTS.
The month of December also had a really big sale, so it’s difficult to figure out why the percentages have dropped, instead of increasing. The only reasonable explanation is that the rate of increase for new Linux users was lower than the rate of new Steam users per total, but that’s just conjecture.