A Wild 40-Way Linux System Benchmark Comparison

[VikingGe]

Probably a cheap board that isn't able to fully power the FX, and/or APM in action.

Funny how terrible Sandy/Ivy Bridge perform in the compiler benchmark and C-Ray. Don't often see my six-year old Phenom II beating an Ivy-i7 by like 20%...

[totoz]

No Skylake CPU in the comparison?

[bug77]

So depressing seeing AMD trying to compete in the last (more than a) few years
It's even sadder to think that even if Zen ends up being a great thing, intel is still going to kill them on pricing. Just because they can.

[juno]

Somehow the FX 8150 is outperforming the FX 8350 on every test here. That's a bit odd.

Why is the Xeon E3-1280 v5 often outperformed by slower E3-12xx v5 Xeons? Maybe the heatsink is loose and you're encountering thermal clock throttling?

Maybe it has ECC RAM and you're getting lots of single-bit errors?

IMO, it's worth looking into. Benchmarks are only valid if the systems are working properly.

Just forget this and please let's pretend it never happened

Most of the tests done were focused on the CPU performance.
[...]
Due to one of the faster SSDs being present in the Pentium G4400 box, it came out slightly ahead for the SQLite disk benchmark.
[...]
http://www.phoronix.com/image-viewer...e=40_way_4_lrg
[...]
40 distinctly different Linux systems

WTF is this?
Sorry for getting displeasing but what did you target with this article? What did you want to test? Benchmarks are for comparability, not just for running completely different hardware and software setups for fun. How do you want to be taken seriously as a reviewer with this kind of runs?

No Skylake CPU in the comparison?

Xeon E3 V5 are Skylake

So depressing seeing AMD trying to compete in the last (more than a) few years

They are not. Otherwise there whould have been FX CPUs with steamroller and excavator.

It's even sadder to think that even if Zen ends up being a great thing, intel is still going to kill them on pricing. Just because they can.

Why should they? Losing a bit of market share won't hurt Intel much. Lowering prices will much more. It might actually be a good thing for them to have some competition if you think of cartel offices

[totex71]

So depressing seeing AMD trying to compete in the last (more than a) few years
It's even sadder to think that even if Zen ends up being a great thing, intel is still going to kill them on pricing. Just because they can.

Amd already outperforms Intel price/performance wise. Buy a cheap fx cpu and you can easily run everything you need. No need to waste tons of money for imaginary benchmark numbers. Also i would not really trust a company like Intel that keep's getting accused for cheating in performance tests.

[bug77]

Amd already outperforms Intel price/performance wise. Buy a cheap fx cpu and you can easily run everything you need. No need to waste tons of money for imaginary benchmark numbers. Also i would not really trust a company like Intel that keep's getting accused for cheating in performance tests.

No, it doesn't. Here's what a $150 intel CPU can do: http://cpuboss.com/cpus/Intel-Core-i...vs-AMD-FX-8350 (feel free to link articles proving otherwise).
There are only two reasons to buy an AMD CPU these days: you do stuff that doesn't actually need much computing power, thus you need the absolute cheapest you can find or for some reason you need a better IGP (albeit one that still can play anything recent without sacrificing IQ).
Also, AMD hasn't released an FX CPU for a couple of years now, the FX line never went below 32nm.

[totex71]

No, it doesn't. Here's what a $150 intel CPU can do: http://cpuboss.com/cpus/Intel-Core-i...vs-AMD-FX-8350 (feel free to link articles proving otherwise).
There are only two reasons to buy an AMD CPU these days: you do stuff that doesn't actually need much computing power, thus you need the absolute cheapest you can find or for some reason you need a better IGP (albeit one that still can play anything recent without sacrificing IQ).
Also, AMD hasn't released an FX CPU for a couple of years now, the FX line never went below 32nm.

No wonder you think that when you trust websites like cpuboss.. Have you even ever owned an Fx chip yourself? Sry but i can't use these "numbers" i need real demonstration for this to be any way near trustworthy.

[bug77]

No wonder you think that when you trust websites like cpuboss.. Have you even ever owned an Fx chip yourself? Sry but i can't use these "numbers" i need real demonstration for this to be any way near trustworthy.

That's why I said "feel free to link articles proving otherwise" - I'm at work, no time to dig through review sites right now.
At least I made an attempt to support my claim. You, on the other hand, just made a claim and now ask me to provide credible (according to your criteria) data to disprove it. That's not how a constructive debate happens.

[totex71]

That's why I said "feel free to link articles proving otherwise" - I'm at work, no time to dig through review sites right now.
At least I made an attempt to support my claim. You, on the other hand, just made a claim and now ask me to provide credible (according to your criteria) data to disprove it. That's not how a constructive debate happens.

Here you go. Read the first comment not the post itself: https://np.reddit.com/r/pcmasterrace...visual/cwukpuc
It all makes perfect sense.

[bug77]

Here you go. Read the first comment not the post itself: https://np.reddit.com/r/pcmasterrace...visual/cwukpuc
It all makes perfect sense.

Yeah, I knew all that. Are you telling me to disregard everything on the Internet and take your word AMD is better?

Edit: You're also ignoring that testing on OpenBenchmarking is actually open: the source code is available and it's not being compiled using Intel's compiler.