INTEL WORKERS SEETHING

this

>When did this board become discussing twitter and reddit posts again?
since literal unashamed redditors come here daily to say stuff too retarded even for reddit and not ruin their score or whatever it is

they're even running the generals now, pcbg is infested by them

Attached: 82 - TzpQMAh.png (574x461, 103K)

I'd agree with him. The gaming bit is a little childish, but AMD has been the loser for a long time now. Extraordinary claims require extraordinary evidence.

> Let’s face it, AMD still can’t beat Skylake tuned up, it tells you how behind they where 5 years ago, and Icelake 1st strike will cause them problem, Tigerlake will put them back 5 years behind. Mark my words

Kek this man sounds like a french general in 1939 .

They may have lost in 1939 but they've won the war.

Just my opinion but I feel security is way more important when running servers.

No, the war was won _for_ them by the Americans, English, and Russians.

Death to Intel shills.

>Piednoel
That guy has a serious mental illness, hes 24/7 seething, for absolutely everything. The best part is when someone calls him out and literally tweets "I was only pretending!!!", claiming it was bait to train his neural network, lmao. Some time ago he went on a rampage calling every single YouTuber an AMD shill just for reporting on one of the many retarded decisions taken by Intel since Ryzen came out, he even defended the 1500W chiller madness.

>if you lose, you win
is this prick a canadian by any chance?

Attached: 1520405762023.png (717x717, 546K)

>>Piednoel
>That guy has a serious mental illness, hes 24/7 seething
Kek you wonder why?
This guy was responsible for Netburst all the way to Kaby Lake, and the chief architect of huge ""successes"" like Hyperthreading, Atom, Android on x86 (lol), and in general responsible for every single pozzed shortcut intel took that came to bite them in the ass today.
The guy has the reverse Midas touch, everything he touches turns to shit.

But like a true frenchman he quit when shit hit the fan lmao
If I were him I'd time dilate to seethe 36 hours per day kek

Attached: 73972983280.jpg (1200x1156, 166K)

>Operson
top kek

ahahaha fuck he actually had a major role in Intel during their fiasco years?
really makes you think

>We Wuz Kangs

Is there anything more pathetic than old white people reminiscing about the past when they themselves destroyed the present?

Truly evil vermin.

Attached: C4284ABE-AA4F-40F4-836B-459C9E066FB0.png (685x441, 590K)

The world will never know peace from these vermin until Intel dissolves.

>German invents transistor
>Nothing
>Years pass
>Patent runs out
>Immediately after Murricans patent transistor
>Revolution!

Attached: 1500103161129.gif (1080x1080, 504K)

His name sounds more canadian than french tbqh
>t. frenchfag on anti-seething drugs

>more canadian than french
how do "canadian" names sound like

absolute moron

French Canadian names can be quite different from ours in France.
His name literally means "footchristmas", that's weird as fuck and I can't find statistics about this name in France.

What a bizarre group of people, I've never seen employees actively peeved about competition before.

Like names you'd find in france 200 years ago

>t. other frenchfag

Maybe he's a jew, that could explain a lot actually.

so where are the benchmarks proving and I mean PROVING they have the lead now?
AMD can't be trusted, Intel is the reliable brand.

Jews make actually good CPUs, like Sandy Bridge.
Piednoel is fond of flaunting his participation trophy for his work on Pentium 4.

Intel's benchmark was total bullshit. They basically turned on every fucking compiler optimization and then compared it to AMD's bench at stock settings.
Personally, I'd rather see a bench competition with a TDP limit so Intel can't go 500w housefire mode on AMD.

hey muhammads, he's literally born and raised in france
graduated from the university of le havre

>He's from Bretagne
Oh then he's just a turbocuck.

>AMD can't be trusted, Intel is the reliable brand.
good goy, have more security vulnerabilities as your reward!

My favourite point of amd is that they don't keep changing the fucking sockets. Can't even find a fucking motherboard for my i7 4770k after mine died when I didn't reply to a post saying my motherboard would die in my sleep tonight if I didn't reply to it.

Attached: 1546622461046.jpg (1200x1200, 116K)

>intel employees after watching the amd keynote

Attached: 245[1].gif (1280x720, 1.77M)

Oh, look! It's the Intel Employee of the Month.

Attached: 1528750847859.jpg (1536x2048, 358K)

>We went from a kick-off meeting to a public demo of an overclocked CPU we already produced and worldwide acclaim in the press that is on our payroll in 20 days!!
Truly a marvelous achievement.

> P4 EE
AHAHAHAHA how dare this fag criticize ANYONE'S chips when he was responsible for that abominable housefire?

What a shitty name then
It sounds like a foreigner was asked to create a "french sounding" name and just added two random words together

>giving your real name to the American government

[x] Single thread performance don't matter

Intel shills are mentally ill jewish sociopaths. Kill them all.

AHAHAHHAHAHAHAHAAHAHAHAHHAHAHAHAHAAHAHAHAHHAHAHAHAHAAH

Based

>an engineer behind the infamous pentium 4 talking shit about zen 2

Attached: 1536218061246.gif (320x180, 153K)

>that pic
amd is truly based and redpilled
AMD
>owned by it's majesty queen Lisa Su
>"we have to unite all the cores under one package"
>affordable for working class people
>scalable like a hierarchy instead of special snowflake solutions
>eastern esoteric ZEN spirit
>builds ROME and NAPLES
>literally RYZES to new hights
>Reactionary (RX) vision and outlook
>wages warfare against israshilltel since decades

What does happen to the Frog??

Attached: Screenshot_20190602_164302.jpg (1080x1576, 334K)

Fuck the working class, if you're not a datacenter, you don't matter. And if you're an enthusiast, your either a debt ridden retard or not working class.

>literally seething
HOW CAN INTEL EVER RECOVER

The frog rules the earth

Attached: i_will_notice_you_pol.jpg (2000x2016, 218K)

>the Intel shill fears the frog

The jew fears the frog.

Attached: gmnjku.png (662x505, 161K)

THIS FUCKING GUY. HAHAHAHAHAHAHAHAHHAHAHAHHAHAHA

Attached: 1497658837285.gif (2000x1153, 365K)

>Former Performance Gurus Intel Corp
he isn't even working for Intel anymore. Why does he still suck Intel cock?

>responsible for intel housefire era
>got fired from intel
>PLS INTEL HIRE ME AGAIN

>AMD's bench at stock settings
What's "stock settings" in this context? AOCC compiles as inefficiently as possible?

>responsible for netburst

OOOF

Look at this french faggot lol

Attached: Screenshot_20190602-180248.jpg (1080x1920, 572K)

cool r*ddit post my dude upvoted. now kys and fuck off back to r*ddit, never come back here yhou disgusting onions fag

t.

Attached: 1548678478608.png (1070x601, 816K)

All of this media coverage of the flaws in their products is making them extremely sensitive and probably has many of them fearing the future of their employment both at Intel, and anywhere else if they get the axe. It's a natural response, but remaining quiet would probably be a smarter tactic.

Even the german did more to win the war for them then the frogs themselves.

A Frog shitting on frogs? WTF? This man has no scruples.

First guy was fired, second guy was in a low level sales position and stopped working there a decade ago according to linkedin.

Stock settings as in a stock release build. Which in GCC is like O2, I think. Intel turned on everything, even the architecture specific tuning options. They released their settings somewhere.
Intel, is that you?

Attached: 1528528938552.png (755x952, 126K)

Do you have a source for what exactly they used?

This page has their compiler setups.
hothardware.com/news/intel-amd-misleading-rome-epyc-zen-2-benchmarks-xeon-9282
Funnily, AMD even compiled their Intel setup with Intel's compiler.

Why the fuck can't twitter just put a proper date and time stamp on every post? I have no idea who said something first in this screenshot.
And I'm not going to use twitter extensively just to find out. Twitter is evil and it makes everyone who uses it evil.

>Intel's competition are the imitators
Remind again, what's Intel's 64-bit architecture called? I keep forgetting.

They used some beta version of AOCC which doesn't even appear to be public. According to the documentation in the one I found however

>“AMD Optimizing C/C++ Compiler” 1.3.0 abbreviated as AOCC is a highly optimized C, C++ and Fortran compiler for x86 targets especially for Zen based AMD processors.

That doesn't sound like gcc -O2 esque cflags to me.

It's still not a fair comparison. It's not like AMD is going to use instructions that only work on their processors like FMA4.

Attached: faggin.png (1395x196, 20K)

So your argument goes from compiler optimizations in general to different instruction set extensions. How is this unfair exactly? It seems rather arbitrary to me. The program at the actual level that matters for the people who use it is still the same input and still the same output.

Of course they would be. 10nm was delayed by half a decade and the new arch is still shit. They can't oppose their Israeli overlords so they can only blow steam off at AMD.

Attached: .png (426x146, 29K)

And also, how do you even know that to be the case?

>So your argument goes from compiler optimizations in general to different instruction set extensions
No? My argument never changed. Intel is using -xCORE-AVX512 which is a compiler option to use Intel-specific extensions.
>How is this unfair exactly? It seems rather arbitrary to me.
Well it defeats the whole point of a benchmark, apples to apples.
Because AMD doesn't even expose FMA4 support in CPUID.

You're going to have to define what you believe a benchmark to be first if you even want to begin with the analogy straw grasping.

>Because AMD doesn't even expose FMA4 support in CPUID.
So you have the source code to the version of the compiler they used and know that CPUID is the sole arbitrator on what it does?

I just like the frog and the other Intel dicksuckers complaining like everyone here and elsewhere didn't buy Intel straight for like a decade.

The last couple of architectures from AMD were absolute memes. We and everyone else fucking made fun of them for years and as it turns out the "MOAR COARS" thing ended up working out when they made an architecture worth a fuck. I got a 2700x and that's my first AMD processor I've purchased since an X2 4200 approaching a decade and a half ago. There wasn't some secret legion of AMD fanboys waiting in the shadows; Intel literally just stopped innovating and as we're finding out every trick they used to get where they are is a disaster.

>You're going to have to define what you believe a benchmark
You mean a competitive benchmark? Two products, head to head with no architecture specific instructions. Pretty simple.
For example, if you benchmark games then you shouldn't use any company specific features like PhysX, RTX, etc. If it's just a generic optimization that either can do though then who cares?
>if you even want to begin with the analogy straw grasping.
Lol? It's a pretty standardized way of handling benchmarks.
>So you have the source code to the version of the compiler they used and know that CPUID is the sole arbitrator on what it does?
I'm not saying that AMD is not using architecture specific tricks because AMD is all that is holy. I'm saying that AMD is not using architecture specific tricks because all their exclusive instructions are fucking broken and will definitely cause a crash over the many hours of that specific bench.

>If it's just a generic optimization that either can do though then who cares?
Well, for one, generic optimizations can have effects on performance. If the comparands don't get the same generic optimizations the results can be biased.

>Lol? It's a pretty standardized way of handling benchmarks.
So post your standardized source for this definition that applies to NAMD. SPEC for example doesn't care about ISA much and only prohibits SIMD in general, and not in specific variants, in certain portions of the benchmark.

>I'm saying that AMD is not using architecture specific tricks because all their exclusive instructions are fucking broken and will definitely cause a crash over the many hours of that specific bench.
I'm still not sure how you come to this conclusion reliably. How did systematically probe and discover all undocumented instructions and figure out that they're all broken?

>Twitter is evil and it makes everyone who uses it evil.
True, I hope all these twitter posts on 4channel won't ruin it like it did to Jow Forums

Truly pathetic.

Attached: 6F824FF3-3BA1-47CE-A2A1-00FF0DFFAB68.jpg (634x483, 35K)

never thought I'd see the day when Jow Forums would be trying to make fun of Federico and/or his name.
This place has truly become neo-Jow Forumsreddit

Attached: fede.jpg (912x557, 52K)

correct me if I'm wrong, but wasn't he fired from intel? if yes then why he defends intel like a fanboy? can't he carry on with his life?

Cope64

K10 was actually pretty good, not better than Intel's offerings, but it was passable, and dirt cheap. Then 15h came along, and even after Bulldozer + 3 refreshes, they still never managed to match K10's IPC. It's a little sad to consider that AMD would most likely have been better off if they had done nothing on the design side and just improved their lithography for those eight years.

switch owned to ruled. She is CEO not major shareholder.

Is he the ex intel employee who said that the 14nm+++++++ is an amazing arch that has a lot to provide yet?

kek, it's like being proud of designing Bulldozer

They did. K10 was part of the first generation of AMD APUs on the 32nm node, and it didn't clock any better than on 45nm. Used less power? yes. But it didn't go faster.

They would have had to have reworked the architecture for it to do better.

> and it didn't clock any better than on 45nm
Phenom II x2 could run on 4GHz. Surely 4.2 could be achievable?

See

A good Thuban based setup could hit 4.3, with the golden chips pushing 4.5. The problem was that when AMD ported the arch to Glofos 32nm node for the Llano APUs, the core couldn't clock any higher than it would on the previous 45nm node. It didn't benefit from the new node, as the problem was mainly with the arch itself instead of the silicon it was sitting on.

However, had they chosen to use the available power budget to fatten the core and maybe introduce SMT, they would have been in a much better situation than what they got with bulldozer. 6 and possibly 8 third gen K10 cores with SMT would have served them much better than the CMT shitpile they shat out. Glofo dropping the ball on 32nm wouldn't have shafted them as bad either.

Defeated by Hitler.

Americans are the bugmen of the west.