Comment by encom
14 hours ago
Just produce your own numbers. Install whatever flavour of Linux you like (all distrohopping leads to Debian) on a separate partition and benchmark it yourself. It isn't complicated.
In the case of my machine, I haven't observed any difference. And by observe I mean with my eyes, I haven't bothered with actual benchmarks because it seems to work about the same, which is good enough for me. I haven't booted my Windows partition in months, and I'm probably just going to blow it away next time I need storage space.
>It isn't complicated
Getting reliable, consistent, meaningful performance numbers is in fact, extremely complicated:
* You need a consistent way to reproduce the exact same outputs - accounting for things like the game's RNG. You can't just walk around and snap the FPS counter in the corner of the screen and call that good.
* For Windows (and occasionally Linux) you need to ensure nothing is running that will taint the results (updates, AV scans, etc)
* Sometimes individual driver versions work very poorly with a specific game. Just because it ran badly doesn't mean you got good data, it may just be a bug in that specific driver version
* You can't just run the benchmark once. You need to run it many times, establishing run-to-run variance
* There are often a good dozen-to-hundred individual OS settings which can impact performance, and in some cases run-to-run variance. You need to know which to tweak, and which to leave alone.
* Sometimes the result of individual in-game settings differs between driver versions. Just because setting X had a big impact once, doesn't mean it always did
* FPS is not a great metric - it's an average. You need to check and see if there are huge frametime spikes. If there are, the game will have a 'good' FPS but feel horrible to play due to stuttering.
* You need to decide if you're benchmarking more GPU-heavy or CPU-heavy - those types of benchmarks require drastically different settings. If you run a CPU-like benchmark you may see a wildly different gap in framerate compared to a GPU-heavy one for the same game.
Benchmarking properly means accounting for thousands of tiny variables. Only a handful actually do it right.
You are making benchmarking WAAAY more complicated than it has to be. We're talking about some dude considering a switch to Linux, but isn't sure the performance is on par. Just load up your game and hit the benchmark button. No sane and rational person is going to be clowning around with driver revisions or regedit, because those types of people think that is more fun than playing the games.
>nothing is running that will taint the results
No, running background crap IS the result, because that's real world conditions, and not some artificial lab condition.
>You need to know which to tweak, and which to leave alone.
That one is easy. You leave all of them alone. Windows tweakers do more harm than good. Besides, replicating benchmark results is impossible after you do brain surgery on the OS.
>You need to decide if you're benchmarking more GPU-heavy or CPU-heavy[...]
You benchmark the games you play. Benchmarking anything else would be completely pointless.
>Only a handful actually do it right.
Rumors say that Hattori Hanzo used to work for AnandTech. I wonder what he's up to these days.
[dead]