[ 3 / biz / cgl / ck / diy / fa / g / ic / jp / lit / sci / tg / vr / vt ] [ index / top / reports / report a bug ] [ 4plebs / archived.moe / rbt ]

Due to resource constraints, /g/ and /tg/ will no longer be archived or available. Other archivers continue to archive these boards.Become a Patron!

/g/ - Technology

View post   

[ Toggle deleted replies ]
File: 17 KB, 512x512, A46D2D9D-EF38-42CB-99F9-D648CE5F2048.png [View same] [iqdb] [saucenao] [google] [report]
67435605 No.67435605 [Reply] [Original] [archived.moe] [rbt]

This instruction set is such a security mess/nightmare.

Which instruction set will take over server space in the next decade?

>> No.67435635

RISC-V is the only acceptable future!

>> No.67435641


>> No.67435680
File: 2 KB, 295x171, 664B4009-3CFE-43B7-8D75-06960C0E6245.png [View same] [iqdb] [saucenao] [google] [report]



>> No.67435715

Intel isn't X86

you are thinking about INTEL (22+ confirmed hardware security vulnerabilities (NOT COUNTING; the exploits found by pakistani hackers who report to their government instead of reporting to Intel))

AMD is secure
ARM is shit
RISC-V is shit
VIA is shit

>> No.67435725

Power has already taken over the most powerful servers and supercomputers.

>> No.67435731

AMD64. POWER and friends will continue their slow wind-down into nothing.

>> No.67435741

Intel created x86, and the Intel x86 chips that AMD produce are no more secure than those that Intel produce.

>> No.67435743

That is x86.

>> No.67435752
File: 58 KB, 2000x1543, Untitled-1.png [View same] [iqdb] [saucenao] [google] [report]

this benchmark says otherwise

>> No.67435766
File: 94 KB, 1044x812, 1524481180373.jpg [View same] [iqdb] [saucenao] [google] [report]


Do you have any fucking clue what you are talking about?

>> No.67435772

>This instruction set is such a security mess/nightmare.
You're a retard. The ISA has nothing to do with security. It's the particular implementation of certain CPU techniques by Intel that's vulnerable.
Thankfully, it's September 3rd so we'll be seeing fewer idiots like you from here on out.

>> No.67435773
File: 36 KB, 900x900, unnamed.jpg [View same] [iqdb] [saucenao] [google] [report]

The only acceptable option.

>> No.67435775
File: 61 KB, 711x457, intel-stock-cooler-next-to-motherboard.jpg [View same] [iqdb] [saucenao] [google] [report]

What rock have you been living under, Intel CPUs really do have a lot more serious security problems than AMD chips. With the latest exploits performance on Intel chips is down 20+%. These are hardware-level flaws only found in Intel products.

And it's not really a problem with x86 or x86-64, just Intel's later implementations.

>> No.67435782

Yeah - in the same way your tricycle is a 750cc Harley-Davidson.

>> No.67435793
File: 15 KB, 400x400, 1522363855449.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.67435849

So x86 year olds just want to defend AMD now? CISC scum never learn.

>> No.67435867
File: 85 KB, 500x500, 1522878306485.jpg [View same] [iqdb] [saucenao] [google] [report]

Check out this well-reasoned rebuttal. With debating skills like this, x86 should have died decades ago.

>> No.67435874

This benchmark is extremelly antisemitic. Please delid it at once

>> No.67435888

I wish. Oh well, it's only entry level consumers that use it anyway.

>> No.67435906
File: 623 KB, 1280x910, Processor_families_in_TOP500_supercomputers_svg.png [View same] [iqdb] [saucenao] [google] [report]

And the odd supercomputer here and there...

>> No.67435914

Your little chart is way out of date. Power is taking over. Nobody wants x86.

>> No.67435920

Man I'm so upset about IA-64. Intel should have stuck to their guns when AMD started bolting 64bit support into x86

>> No.67435934
File: 8 KB, 215x149, 1514816800563.jpg [View same] [iqdb] [saucenao] [google] [report]

>implying 450 of the TOP500 supercomputers have been shut down in three years

>> No.67435938


ARMv8.1 and later has some really nice features for improving I/O efficiency with KVM (or other type 2 hypervisors like FreeBSD+bhyve) guests in a saner, more secure way than x86. ARM servers are incredibly dense in cores/U, cores/Watt and IOPS/Watt relative to Xeons, so all that's missing is enough single threaded perf to be "in the ballpark" and a bunch of cloudy things become very attractive to run on ARM. OSes have been ported, open source packaging is nearly up to par with x86, and even big things like Openshift are getting close.

>> No.67435939

Maybe the bottom 400 still run x86 crap, but that's because they can't afford better.

>> No.67435946

I hated Itanic as much as the next guy at the time - but these days I would agree. There's something quite attractive about VLIW.

>> No.67435968

Having spent way too much time looking at dependency data on packages.debian.org, Itanium required most programs to use libunwind to manage call stacks in userspace. I'm pretty sure there were good reasons even Gentoo on Itanic never really took off. VLIW does sound neat, but I think a more modern take on it like the Mill CPU (eternal vaporware...) is the real path forward beyond RISC.

>> No.67436041
File: 165 KB, 1018x753, 1534780264289.jpg [View same] [iqdb] [saucenao] [google] [report]

Hopes: POWER
Reality: ARM

>> No.67436061

It's not smart to rob an armed homosexual.

>> No.67436064

[10 shekels have been deposited into your account]

>> No.67436075

Why would Jews want open hardware?

Name (leave empty)
Comment (leave empty)
Password [?]Password used for file deletion.