Intel i9-12900K Alder Lake Linux Performance In Different P/E Core Configurations

[Sonadow]

Gentoo users re-define the compiler/linker flags to tune code to their hardware.
If you compile from source for debian, without fine tuning flags, you achieve nothing in terms of performance -- surely it remains an interesting thing to do from a knowledge POV.
BTW, Debian is known for been conservative and favor stability. You should try another distro for desktop usage (or heavily modify it with specific kernels)

-march=native not good enough for you?

[Grinness]

Typical remark of someone who has never done so and assumes that takes no skill. Go ahead, build something big like LibreOffice and see everything for yourself. You aren't going to even get past the first compiler error because you don't have the knowledge to do so.

And try telling Gentoo users that compiling their stuff is pointless. Go right ahead. You have shown nothing except your own ignorance and incompetence outside of building a kernel and running pretty benchmarks.

Gentoo users re-define the compiler/linker flags to tune code to their hardware.
If you compile from source for debian, without fine tuning flags, you achieve nothing in terms of performance -- surely it remains an interesting thing to do from a knowledge POV.
BTW, Debian is known for been conservative and favor stability. You should try another distro for desktop usage (or heavily modify it with specific kernels)

[perpetually high]

Sonadow

I compile all the time. If you think that's skillful, it's not. It takes competence, that's all.

Go put on your resume "Knows how to compile software" and let me know how that works out for you.

[Sonadow]

Maybe that was a Debian problem or an I/O-sheduler problem? Especially old Hardware runs generally smoother on Linux (provided there is proper GPU support, but my Debian times are long gone and I can only speak vor Arch Linux.

If it were a Debian problem it would have gone away after building an up-to-date kernel, and up-to-date Mesa.

[Anux]

I have been running Debian with a custom-built kernel and Mesa on my Apollo Lake Atom laptop for the last three years.
Two weeks ago I threw Debian out and put Windows 11 on it. The difference in performance is immediately noticeable. Web browsers and other heavy applications like productivity suites no longer randomly stall for a minute when scrolling through >20 tabs or multiple pages in a docx file loaded with lots of images, photos and tables.

Maybe that was a Debian problem or an I/O-sheduler problem? Especially old Hardware runs generally smoother on Linux (provided there is proper GPU support, but my Debian times are long gone and I can only speak vor Arch Linux.

[Sonadow]

And what skill does it take to compile?.

Typical remark of someone who has never done so and assumes that takes no skill. Go ahead, build something big like LibreOffice and see everything for yourself. You aren't going to even get past the first compiler error because you don't have the knowledge to do so.

And try telling Gentoo users that compiling their stuff is pointless. Go right ahead. You have shown nothing except your own ignorance and incompetence outside of building a kernel and running pretty benchmarks. Some 'enthusiast'.

[Sonadow]

This may be true but it doesn't have anything to do with the scheduler because Windows 10 (which doesn't have the new scheduler specifically for Alder lake) also beats Linux, this seems to be more due to processor support

Which was the point I was trying to make. Windows 10 does not have a scheduler specially for Alder Lake but they have experience on BIG.little architectures because of their work on Windows RT and the ARM64 versions of WIndows 10, where all hardware use BIG.little. It's practically a forgone conclusion that this experience factored into their continuous work on the scheduler to the point where Windows 10 for x64 is able to handle Alder Lake as-is.

[Grinness]

Right, because you have nothing to argue?

Fact remains that for all the benchmarks Michael has done about Linux having better performance over Windows, they simply don't carry forward to real-world computing. Till now nobody can provide a reasonable explanation as to why Windows boots, launches programs and generally respond to application inputs faster than Linux on the same hardware, especially on low-power hardware like Atoms.

????

They do not carry over .... uhm, then why are you here commenting/reading benchmark articles? move on, nothing of interest for you
Go and buy whatever cpu you want/you get hooked to.

[perpetually high]

And unlike a certain person who claims to be a power luser and an 'enthusiast' yet doesn't even know how to compile the X server, Mesa or a web browser and its dependencies and only has enough intelligence to use prebuilt binaries, I have been building my own kernels and drivers and recompiling the applications I use on Linux for the past 13 years for maximum optimization.

Because I don't *need* to compile the X server. LOL! This is what I meant when you guys are doing all that extra shit just to say you can do it, with no real performance impact to show for it. Bravo?

And what skill does it take to compile? Are you even writing the code? That's where the complexity is. Not figuring out how to compile it. I really just have no respect for you or your existence. Take the best of care. I decided to quote you to expose you, because you are a sham and a fraud. Having said that, take the best of care. And do way, way better.

[Sonadow]

That sounds like an OOM problem that the multigeneration LRU patches might alleviate. Did you try that?

It's not an OOM problem at all. The laptop has access to 8GB of memory and 4GB of swap, and even when the applications were stalling free never reports more than 5GB in use at any time. And it was a 5.15 kernel, not the dinosaur 5.10 kernel that got bundled with Bullseye.

Lastly, I never perform in-place upgrades. The upgrade from Buster to Bullseye was done with a full format and install with Debian's netinst image.