Windows 11 vs. Linux Performance For Intel Core i9 12900K In Mid-2022

[coder]

In this specific benchmarks. Add I/O and your windows will be on its knees.

Especially with an antivirus scanner!

My Windows laptop for my job has a decent NVMe drive, but feels like it has a mechanical hard disk!

[user1]

Especially with an antivirus scanner!

My Windows laptop for my job has a decent NVMe drive, but feels like it has a mechanical hard disk!

It seems MS itself is aware that Windows experience is crappy on HDD, so now it wants manufacturers to stop shipping pc's with HDD's in the near future. While it's not necessarily a bad thing, this just shows that MS isn't bothered in the slightest to fix the inefficiencies in its software.

[aufkrawall]

I've not heard of a single person on Windows having troubles with it.

I couldn't care less about what you hear or don't. Dynamic clocking of Nvidia Windows driver is mediocre at best, and at least D3D11VA VP9 in Firefox 103 has presentation timing issues. Usually people like you are who claim not to notice anything just lack perception or experience how to test things. Your opinion is that of a clueless and willingly dumb peasant, it doesn't matter at all.

[quaz0r]

the "--omg-optimized trollololol!" crowd must hate seeing clear lunix consistently show measurable performance gains over everything else

[Linuxxx]

the "--omg-optimized trollololol!" crowd must hate seeing clear lunix consistently show measurable performance gains over everything else

Still not worth recompiling your whole system when most of the performance uplift of Clear Linux can be achieved by simply switching over to the performance governor...

(And for anyone concerned about power consumption when doing so:

I'm running my 200+ Watts Intel i7-11700F with intel_cpufreq performance by default, and playing back a 1440p60 AV1 video doesn't increase the already low fan noise, meaning that the CPU is NOT running at its highest frequency all the time, but rather deep-sleeping intermittently during video playback.)

[Volta]

Especially with an antivirus scanner!

My Windows laptop for my job has a decent NVMe drive, but feels like it has a mechanical hard disk!

Yes and even after it "loads" it still uses the disk for a long time.

[tildearrow]

Windows overall is just 4% slower in benchmarks than Ubuntu 22.04 + Linux 5.18. Nothing to worry about for anyone.

At the same time this comparison doesn't show that in Windows all web browsers and most video players support HW accelerated video decoding and proper power management out of the box which is what people will actually appreciate, not some surface-level improvements for various benchmarks most people never run.

What do people actually care about? How fast their web browser works. How fast their system boots. How fast they can perform calculations in Excel. Wait, there's no Excel in Linux. Whether their hardware works without issues. Overall on the desktop this is far more important than performance in benchmarks.

What about disk performance? Windows loses immediately.
Windows 10 takes 3 minutes to boot up on HDD, and then every disk operation takes so damn long.
Not even APFS on HDD is this slow.

[stormcrow]

Not bad. Linux fanboys usually say that linux is way more fast. Linux is like rabbit Windows is like turtle .
You think linux is maybe two times faster? Maybe three times? No, it's only several percent. Linux fanboys always lie .

(Ubuntu) Linux isn't faster if you're opening Firefox... OH SNAP!

I'll show myself ou...... >>>>>>

[arQon]

Kinda interesting how we're 4 pages into the comments and so far nobody has even mentioned, let alone considered why, zstd at low levels is massively faster on W11.

My guess would be that it's predominantly readahead behavior on NTFS vs ext4 rather than anything CPU-related, but that's very much WAG.

[NobodyXu]

Kinda interesting how we're 4 pages into the comments and so far nobody has even mentioned, let alone considered why, zstd at low levels is massively faster on W11.

My guess would be that it's predominantly readahead behavior on NTFS vs ext4 rather than anything CPU-related, but that's very much WAG.

I suspect it to be a scheduler issue.

It might be that the scheduler on Linux isn't optimized enough and placed the zstd thread onto the e cores.

As for readahead behavior, both linux and windows have that.