Coming from the business side of gaming - this Valve Linux VRAM fix story is genuinely interesting to me, but not for the technical reason. A Valve engineer quietly drops a kernel patch, and suddenly a 4GB GPU that was practically unusable in modern games goes from 14 to 41 FPS in Alan Wake II. No press release. No marketing. Just a fix that triples performance for budget GPU users. From a growth perspective, that’s a massive retained audience - people who were about to give up on PC gaming because their hardware couldn’t keep up. Question for the devs here: how much do you think about low-VRAM users when optimizing your games? Is it even feasible at an indie level, or is that something only studios with Valve-level resources can realistically address?
I generally don’t make games, but I do make other software. Avoiding waste of my users’ resources is one of my guiding principles. It affects my decisions in a variety of ways, ranging from time and effort spent on optimization, to avoiding bloated frameworks like Electron.