It should, with careful and precise setup (all needed modules built into the kernel, everything locally compiled, USE flags for all packages carefully chosen to eliminate unnecessary dependencies), be the most targeted for the specific hardware it’s running on and the specific workload it’s doing—in other words, it would be carrying less cruft around. Fewer libraries to import, fewer branches to check during code execution.
In other words, execution time should be a bit better in return for spending more setup time. Benchmarks like this tend to only measure execution time.
Okay so a few counterpoints. Firstly, it would only be more optimised assuming the user actually knows what they’re doing when it comes to those optimisations you discussed. Considering the Framework 12 is designed to be a replacement for the Chromebook in education, I’d say the average userbase would be better with an alternative.
Secondly, benchmarks are designed to be a reflection of general usage. Gentoo that’s optimised for productivity, or for gaming, or for creative work just isn’t indicative of every users set up. It also can’t be used comparatively in a fair manner with other devices if it’s specialised in a way that is not automatic on the part of the OS. The journalist would need to set it up in a general manner, and then it loses a lot of that optimisation that you speak of. That’s just how impartiality works.
Theoretically, Gentoo should be the most optimized, but they probably didn’t want to take the time to install it.
What makes you say that Gentoo is the most optimised?
It should, with careful and precise setup (all needed modules built into the kernel, everything locally compiled, USE flags for all packages carefully chosen to eliminate unnecessary dependencies), be the most targeted for the specific hardware it’s running on and the specific workload it’s doing—in other words, it would be carrying less cruft around. Fewer libraries to import, fewer branches to check during code execution.
In other words, execution time should be a bit better in return for spending more setup time. Benchmarks like this tend to only measure execution time.
Okay so a few counterpoints. Firstly, it would only be more optimised assuming the user actually knows what they’re doing when it comes to those optimisations you discussed. Considering the Framework 12 is designed to be a replacement for the Chromebook in education, I’d say the average userbase would be better with an alternative.
Secondly, benchmarks are designed to be a reflection of general usage. Gentoo that’s optimised for productivity, or for gaming, or for creative work just isn’t indicative of every users set up. It also can’t be used comparatively in a fair manner with other devices if it’s specialised in a way that is not automatic on the part of the OS. The journalist would need to set it up in a general manner, and then it loses a lot of that optimisation that you speak of. That’s just how impartiality works.