HN Distilled

Essential insights from Hacker News discussions

Games run faster on SteamOS than Windows 11, Ars testing finds

Original Article

Hacker News Discussion

Here's a breakdown of the key themes from the Hacker News discussion, with supporting quotes:

Proton as a Compatibility/Translation Layer vs. Implementation

A central point of contention revolves around how to accurately describe Proton's function. Some argue it's a "translation layer," while others suggest it's more of an "implementation."

  • Translation Layer Argument: "Might be unfair to call Proton a 'translation layer' because the Win32 API is not defined in terms of system calls but rather a set of functions exported from a DLL. Proton supplies a DLL that implements the Win32 API using Linux syscalls. Windows supplies a DLL that implements that Win32 API using Windows syscalls that you're not really supposed to use directly." - PaulHoule. This describes Proton as providing an alternate implementation of the Windows API.

  • Compatibility Layer Semantics: "https://www.winehq.org/ calls it a compatibility layer that translates calls on the fly. so 'translation layer' is not that unfair." - homarp

  • Evolution of Wine/Proton: "I think that's how it started out, and also how a lot of developers still conceptualize it. Wine has had to massively expand that scope to reach the maturity it has now. I think it's kind of straddling the line between 'Implementation' and 'translation'." - delusional. This acknowledges the historical context and how the complexity has grown.

  • Focus on Practicality: "Philosophically its still a translation layer though. It doesn't really care about correctness if the no apps depend on it. Success is in meaningfully running client software. The implementation of the Windows Libraries are just a way to get there." - delusional. This emphasizes the pragmatic goal of running Windows software.

  • ABI Translation: "Wine is translating Windows ABIs (not APIs) into underlying Linux OS and userland. Translation simply means that normally Windows ABIs are meant to be used on Windows, they aren't native on Linux." - shmerl. This highlights the translation of binary interfaces, rather than just source code APIs.

Criticism of Game Selection for Testing

Several commenters questioned the choice of games used in the performance tests described in the linked article, suggesting more popular or demanding titles would be more relevant.

  • Questionable Game Choices: "Borderlands 3? Homeworld 3? Who chose these games? Why not just use the current top 10 on Steam atm?" - jimbob45

  • Call for More Relevant Titles: "Yes, I'm always astonished at the game selection for these tests. If people want to do a comprehensive analysis of 100+ games that involve niche games, sure. But otherwise, just pick a few games from top 10 from Steam and maybe 2-3 of some cutting edge demanding titles." - antisthenes

  • Rationale Behind Game Selection: "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..." - dabber21. This points out the tests may have been chosen specifically because they had built-in benchmarking.

Hardware and Driver Limitations

Some users suggest that the hardware limitations and/or driver issues are overshadowing the differences between Windows and SteamOS.

  • Hardware Bottleneck: "Seems more like a test of the hardware than Windows 11 and SteamOS since they ran into driver issues immediately. Not to mention those frame rates are terrible across the board. Just not very good hardware." - jajuuka

  • Driver and OS Variability: "It's the same hardware on each test. The only difference are the drivers and OS in question. Lenovo has been slow to officially ship updated GPU drivers for this device, but the exact same SoC is used on a number of handhelds." - vel0city. This response refutes the claim it is just the hardware and suggests the drivers are the differentiating factor.

  • Performance Expectations: "As for the performance, its a 15W handheld trying to play games that 600W PCs and 300W consoles struggled with just a few years ago." - vel0city. This adds vital context about the constraints of the hardware being tested.

Feature Differences and Perceived Performance

The possibility that games may not be rendering the same graphical features across platforms is suggested as a possible reason for any performance differences. Potential differences may not be apparent without a still image comparison.

  • Feature Parity Concerns: "Some of these games are practically neck and neck for performance. I'm wondering if it's a similar situation to early Proton comparisons, where framerates were higher in Proton but when comparing still for still you could tell it just wasn't actually doing certain effects. Are there features that are being attempted in the Windows version that are just not functional and thus effectively disabled on the Proton one?" - vel0city

  • Framerate Preference: "But even then, assuming that is true, if they're pretty much the same would people care about maybe some fog looks a little different but you get an extra 15-20fps in a game? I think a lot of people would still prefer the boost in frames." - vel0city. This points out that not all people may care about graphical fidelity compared to framerates.

Anecdotal Performance Experiences on Linux

One user shared their personal experiences with gaming performance on Linux using Proton, comparing different desktop environments.

  • Performance Ranking: "In my purely anecdotal experience over the last few years, performance ranking is as follows: 1. Steam on Linux via Proton + Wayland (Niri) 2. Steam on Linux via Proton + X11 (Xfce) 3. Steam on Windows 4. Games on Linux launched via other means (it's possible I was missing out on certain flags/optimizations, but this is just about the average experience)" - haswell

  • Improved Framerate Consistency: "The biggest thing I noticed when switching to Linux was an improvement in framerate consistency, i.e. I'd have fewer situations where the framerate would drop momentarily. Games felt more solid and predictable." - haswell

  • Wayland Improvement: "The biggest thing I noticed when switching from X11/Xfce to Wayland/Niri was just an overall increase in framerate. I'd failed this jump many times over the years, so it was notable when I jumped and stayed there earlier this year." - haswell

  • Launch Time Tradeoff: "It does feel like games take longer to launch on average, but this makes sense given the fact that it's launching via Proton/Wine." - haswell