Intel Core Ultra performance in Linux is 15% higher than in Windows

BuddyTheBeefalo@lemmy.ml to Linux@lemmy.ml – 315 points –
The Result Of 84 Tests Showed: Intel Core Ultra Performance In Linux Is 15% Higher Than In Windows - Gadget Tendency
gadgettendency.com
15

In my experience, every computer is faster with Linux than with Windows. But if this measures just the processor performance on similar tasks I guess it's news.

I think it comes down to the culture. A minuscule improvement to a file system is big news in the Linux community. There's also lots of academic interest in the performance critical parts of the kernel that you just can't emulate with a closed source model. Is anyone writing papers on how to obtain a 2% improvement in the task scheduler on Windows?

Linux dominates the server market, so even small improvements matter when you're talking about a server farm with thousands of machines or the latest supercomputer. Many, many people care about the scalability of Linux. On Windows, we say: NTFS? It's good enough. The user won't notice on modern SSDs.

A lot of the software components under the hood in Linux are replaceable.

So you have a bunch of different CPU and disk IO schedulers to suit different workloads, the networking stack and memory management can be tweaked to hell and back, etc etc.

Meanwhile Windows Server 2022 has...... ?

Consequently battery life tends to suffer on Linux vs windows. Especially on newer hardware before people figure out how to manage performance and battery life.

Usually, applying the same tricks that Windows does, its not true.

But by default, mostl Linux ditros dont do something special for having performance managing.

But actually. Windows does neither, at least the pure Vanilla form. Its a huge difference when using my Levono Ideapad with the preinstalled Windows versus Windows that is reinstalled Vanilla without drivers. Then Linux is more plug and play and better at this job than Windows.

Maybe they do it differently on ideapads. But on all of the modern thinkpads I own the all install at set up the same power profiles and dynamic tuning that the factory image does. Factory install vs fresh install performance is the same on these machines once windows update has done it's thing. Even the random POS HPs will do the same thing.

Older machines yes absolutely.

I wonder if Linux is 15% better, or Microsoft tracking uses that much processing.

It's most like due to power governor and scheduler behaviors. If there's background activity impacting the test it would more likely be Defender.

I'm not sorry that the CIA is using your closed-source software you mistakenly thought you owned anything because you paid way too much for anyone else to actually control your shit, you ignorant slave.

Edit: You're a bunch of ignorant fuckwads. You can't read shit and know who's on what side because your sensitive to sarcasm and blatant...nevermind. Fuck off.

True, as we just found out, the performance of Meteor Lake is significantly influenced by the BIOS.

What's that supposed to mean?

There was an issue with the BIOS limiting the power available to the processor during benchmarks. It has supposedly been fixed since.

You sure this isn't just anti-rowhammer et al mitigations?