Redpill me on this distro. apparently it crushes in every benchmark imaginable

Attached: clear_004.jpg (750x436, 49K)

Other urls found in this thread:

openbenchmarking.org/result/1711188-AL-MERGED47192
github.com/clearlinux-pkgs/linux
twitter.com/NSFWRedditGif

>apparently it crashes in every benchmark imaginable
ftfy

>coping this hard
lmao

openbenchmarking.org/result/1711188-AL-MERGED47192

Do people really benchmark linux distros? What the fuck are the benchmarking? How many terminal windows and anime pictures you can have open at one time?

just install gentoo instead

>Do people really benchmark linux distros?
yes because they want to weed out amateur distros made by uneducated weebs therefore it's no surprise Ubuntu and Clear Linux won.
this is what happens whenever actual competent people get involved in muh open sores.

You're one big oblivious nerd loser holy shit

>Ubuntu
>Canonical
>competent
Lmao

yet your snowflake subhuman distros are shit in every benchmark. explain please.

It's Intel's distro, so take that as you will.

>No Arch
>No Gentoo
>No Void
>No Manjaro

No distros I use are benchmarked. Worthless test.

user you're not gonna like the truth, once it hits ya

Clear Linux is developed by Intel in Israel

>COPYRIGHT © 2019 INTEL CORPORATION. ALL RIGHTS RESERVED

Can you trust it?

Surprised how slow CentOS is in some tests.

Ubuntu is very close in performance to Clear Linux actually.

I've seen it just second to Clear Linux in most other tests though, it surprised me too. Normally it's the top

Can SELinux slowdown some tests?

I don't think so, it may affect the launch time for bash but once the binary I'd running I don't see it being used. I don't know much about SELinux though, but supposedly it justs tunes better permissions

OH NO NO NO

Attached: Screenshot_20190117-231922.png (1172x2048, 405K)

>not helping developers to improve their code by sharing your usage data (personal info) with (((them)))
go home Sam

It's basically a proof of concept optimization demo by Intel for latest Intel processors only. When you look at the phoronix benchmarks you can basically divide the results into two categories: 1) Those that clear linux leads by few % are due to it using readily available optimizations that are just not used by the other tested distros. I believe LTO and march= are still the biggest hitters in this area. You can enable these with ease on gentoo as well. 2) Those benchmarks that clear linux leads by massive 100%+ are due to bad phoronix test methods testing different things on different distros and/or Intel applying agressive performance patches that are not yet release ready by upstream (but will come to other distros eventually) making the benchmarking rather null as the work done is, again, not same on the different tested distros

You use Ubuntu? Gross. I prefer Gentoo or Arch.

OMG do you even know what a job is? This is for sysadmins to monitor the machines they're in charge of, not sending telemetry to Intel, FUCKTARD

I didn't see Arch or Gentoo on that graph, explain.

>linux 34364643 distros complete shit
>intel goes at it
>suddently 1 decent linux distro
really makes you think, it's almost as if autistic incel weebs can't do something right, while corporate healthy people can.

OH NO NO NO

Attached: ubuntu-amazon-adware.jpg (620x396, 293K)

OMG do you even know what a joke is?
>implying I use ubuntu

>openbenchmarking.org/result/1711188-AL-MERGED47192
I'd almost assure you that Arch or Gentoo would be faster.

Go shill your meme-distro elsewhere. The way you have been talking in this thread and the buzzwords you use remind me of the baiting and shilling back when I still browsed Jow Forums. It just feels out of place here.

>implying I use something that takes advantage of modern hardware&software
nice comeback, Neo

this is my 2nd post in this thread. the way you talk reminds me of my paranoid schizophrenic neighbor.

>people still like Unity
lmao, what the fuck is wrong with those people.

Reviewers can't install arch let alone gen2

Clear Linux really is slightly faster than most other distributions. Gentoo was very popular for a time (not so much these days) because you could compile with flags optimized for your CPUs instructions. Clear Linux uses GCCs multi-versioning feature to compile binaries optimized for multiple paths depending on available instructions. They also do various optimizations with the kernel configuration and apply quite a few patches too. The patches and configuration they use is here,
github.com/clearlinux-pkgs/linux

You can get the same result by compiling Gentoo with -march=native and a similar kernel configuration.

It's primarily optimized for Intel CPUs but binaries don't care and take advantage of additional instructions on AMD CPUs too.

One interesting tidbit to take notice of is that GCC's actually had function multi-versioning (the secret to much of Clear Linux's performance improvements) since version 6. That kind of begs the question as to why Ubuntu, Fedora and OpenSUSE do _NOT_ have fmv binaries. Binaries would be a bit bigger, so there's a trade-off. And this is actually the main reason why I see Clear Linux as a good thing: If it becomes popular because it's faster then perhaps the rest will get a clue and support it.

>Arch, and Gentoo
>Faster
Those distributions aren't known, or even used, because of how fast they are. Gentoo users don't use Gentoo for speed.
Before Clear OS, debian testing would usually be the fastest at most benchmarks, but people would still use other distros.

It's a tech demonstrator. Not meant for desktop/productive use.

>crushes
it's 5-10% better in most benchmarks at best. See Phoronix's benchmarks.
I'd probably use it for servers but they don't distribute ISOs on the website, only IMGs.

>That kind of begs the question as to why Ubuntu, Fedora and OpenSUSE do _NOT_ have fmv binaries
not only the binaries get bigger, but they probably take a lot more time to compile, and that's a pretty big trade-off for distros compiling hundreds (if not thousands) of packages

I really like this
it's minimalist but without being retarded about it
10/10

it's disabled by defualt in the installer and they dont even ask you to enable it

I can tell you this: I started using Gentoo back in 2003 mostly because it was faster - but also because it got ebuilds for newer versions of packages a lot faster (since it's trivial to change an ebuild and distribute it). Speed was of importance. Being able to compile with -Os was particularly helpful (smaller binaries helps on slower spinning harddrives). But you do have a point, there's plenty of people who use/used it for other reasons.

Not sure how that matters to the bigger distributions. They tend to have an infrastructure of build-boxes. It's not like some guy is compiling everything on his laptop. Your point does make me curious as to what kind of compilation time differences we're talking about (as well as binary size differences).

try actually using it. The package manager (fwupd) is terrible and the repositories are bare. If I remember right, you can't even install individual packages, only package bundles.

>Not sure how that matters to the bigger distributions.
not sure either, but maybe that's one of the reasons? or they simply don't care

>highly optimized for intel platforms
>tfw no intel hardware

>Speed was of importance.
Then Gentoo was the wrong choice.

Fuck off kike shill sage

It has performances gains on AMD too.

>I'd probably use it for servers but they don't distribute ISOs on the website, only IMGs.
what

>unbiased CPU dispatch
because linux is literally *impossible* to compile with anything other than GCC, which means the stupid brain-dead dispatcher from their IOCC is useless here

It's shit. Even its name is shit.

Attached: Clear Linux (asstick).png (233x55, 7K)

try it wanker

I did. It's shit.

>init system: systemd
its shit

Friendly Reminder to install Slackware Linux
>As for the relationship between Slackware Linux and the Church of the SubGenius: Patrick "The Man" Volkerding is an ordained Church minister.

Attached: 0f0744539520dbb2f825885b1778cc6c.jpg (849x1079, 289K)

I tried it a few days ago.

Pros:
- I was able to install it and run VMs in virt-manager
- it works
- its based off Redhat/CentOS

Cons:
- Intel's fancy installer instead of a regular debian-style installer
- Intel's fancy package manager instead of Yum or RPM. Not very many packages. Theoretically, you can install Yum, RPM and DNF, however I was not able to quickly figure it out.
- Very few users. It's easier to find support for NetBSD than for Clear Linux.
- I couldn't get KDE to work. It seems like Intel wants you to use Gnome.

Conclusion:
- I installed CentOS and am very happy with the results. 10/10 would install CentOS again. Only use Clear Linux if there is a specific, tangible, compelling reason why CentOS is not good enough for you.

They have an ISO. However I could not install it on a VM. Only a physical machine. Intel distributes a .Img for VM's but you have to use it with a qemu script that Intel provides. I was not able to run a Clear Linux VM in KVM/virt-manager with the ISO or the IMG file.

What DE is that?

>Pros:
>- its based off Redhat/CentOS
AHAHAAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHHAHAHAHAHAHAHAHAHAAHHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHHAAHAHAHAHAHAHAHAHAHAHAHAHH

That means it's stable and you can figure out how to do things by looking up the CentOS/Redhat version.

Neat.

shit as a distro but a great demo of what can be done to improve performance

>Ubuntu: Linux for human beings

>

that's a good thing retard

> uses computers to run computers benchmark
The absolute fucking state of brainlet leenoks retard

lmaoing at this archcope

>defending lennartware
t. street shitting IBM shill