[dabber21]

"To test the performance impact of this operating system choice, we started with the SteamOS version of the Legion Go S (provided by Lenovo) and tested five high-end 3D games released in the last five years using built-in benchmarking tools..."

those games come with benchmark tools

[jimbob45]

Borderlands 3 was 2019. Homeworld 3 has a 38% on Steam and sold poorly. I highly doubt it ever received patches or optimizations. Again, these feel unbelievably arbitrary, as if someone just wanted to push a narrative.

[energywut]

Steam favorability percentages are famously vulnerable to review bombing. HW3 got swept up in culture war nonsense around LGBTQ representation.

I'm a hardcore Homeworld fan. I've run campaigns of their TTRPG, modeled their ships, played the old games to death. I found my own experience with 3 to be "mixed", it's hardly the best entry in the series, but the reviews absolutely are artificially low due to brigading.

Aside, an unoptimized game is actually one I'd want included in my benchmark. Games that have the teams and budgets to really polish will likely perform well no matter what. But how does OS level changes affect those other games, games where the developers didn't put in the care? Does one OS make those games worse? Or does it help with the shortcomings? It's valuable to have entries like that in your dataset.

[spartanatreyu]

Homeworld 3 received a terrible rating because its story and delivery was a massive departure to what users wanted, not because of it's gameplay or tech.

It would have been chosen for the same reason ashes of the singularity was chosen as a benchmark for so long: because it looks good, it comes with a benchmark, and it's really good at stressing out a particular part of the computer (for AoS: async rendenring, for HW3: CPU).