No no no, that's impossible!

>+36% in ST over a 2700X
>at 40W less and with a lower max boost
Seems believable.

New node, not low power anymore, New arch

Just because Intel didn't want to do it doesn't mean AMD can't

>3.6GHz

who cares

By all means keep being retarded but it will only hurt more when the embargo is lifted and actual reviews will be out.

>argue with facts
>get replied to with lame ad hominem answer.

If you genuinly believe there is a 40% IPC improvement then you're a lost cause and no amount of argueing is going to make you change those beliefs anyway.

>there was no 40% jump between excavator and zen

You sayin' cinebench is rigged or something?

This. Claims are still claims, nothing more.

would they risc a retraction if this wasn't true? especially since most likelt they have already tested all the ryzens by now under NDA.

maybe youtube tech faggots will feel charitable and run all benchmarks at 3.6ghz or whatever zen 2 will top out at. i've no doubt that hardware box aussie twat will pair zen 2 with 8000000000mhz ram and intel with 2666.

They are both running at stock retard

Cope

cinebench has beed proven to be rigged against amd

>2979
Exactly the same score as my 8700k @ 4.9GHz

Attached: [email protected] (786x593, 64K)

Explain why the 3600 loses to 2700x in cinebench r15 (which is considered best case) yet it stomps the 2700x in multithreaded score in passmark.
You can't.

Considering how many times this thread topic has been posted lately you'd think you'd have known by now the @3.6 is just the title of the cpu. It's not locked at base clock.

iCope

dude, fuck the reviews. we need people that purchase them and give us truly unbiased real world tests.

reviewers never ever show true performance,

Again with this? they only show the base frequency on all cpus on that list, all of their boosts are still active. so the 3600 and 9900k are running at their max speeds.

you can click the cpu name and review run and see the operating freq faggot...

they even manully deleted the best one with 3200mhz memory topkuk

Attached: file.png (924x724, 108K)

to expand upon this:
there was one run with 16gb ddr4 3200mhz mem and 3.6 base clock, that got removed last night it seems

something something optimisation :^)

Too good to be true. Still, I'm looking forward to it, even if it turns out to be a bit behind these leaks, AMD already won this one.

>i9-99
>turn it upside down
>666
!!!

Attached: C9UBb9IXgAE-bsV.jpg (800x450, 30K)

delid dis

Attached: 1561901824689.jpg (719x568, 152K)

>i7-6700K BTFOs a i9-9900K
wtf intelbros

>same latency
>btfos

Meanwhile, a brand new Ryzen has twice the latency.

>AMD is good at not delivering
/thread
If they had delivered what they actually claimed in the past, Intcel would be long dead and gone by now

>beats the 9900k with higher latency.

must be infuriating owning intel cpus now

>give it shitty ram
>get shitty timings
Spaniards are really retarded

They even said in their test that there were issues with the Bios they had on the board.

People just copy-pasted the benchmarks and left that disclaimer out.

>intel is dogshit even with better latency
kek

nah

Attached: 1561878873119.jpg (718x379, 102K)

>he posted it again

You don't like it?

>I'm going to fucking ignore the 2700X bench and how the 3600 destroys it in single core.
>I'm also going to ignore that Skylake/Kaby Lake/Coffee Lake have the same IPC, with the only difference being their clocks.

>The entry level part is BTFO'ing an i7 K SKU, with a gimped bios, shit timings, and it still does good. It's less than 30% slower than a part that's double its price.

Delivering latency, yes.

>hah, My Supra Mk5 outran your Corolla by 12ns
>all while costing more than double
>and sucking more gas

The benchmark is real but the 3600 has expensive RAM and is overclocked to hell and back with LN2.

All the other benchmark sites show the 3600 to be considerably slower than the 9900k and its not even close.

kek, imagine lying through your teeth when the benchmark scores all have footprints that specify what speed they ran on.

someone screencap this tripshit right now

Attached: 1548982916626.jpg (1293x1730, 1.16M)

You know Intel will just come back with an even more pozzed uarch to beat Ryzen. Throw more speed holes cuz Intel fanboys are braindead monkeys anyway.

OY VEY, THE GOYIM KNOW, SHUT IT DOWN

Attached: 76d.jpg (1196x676, 673K)

pretty much. they copy-pasted QPI, literally glued two Pentium 4's and called it a Pendium D(elid).

Pretty much everything that end users in the last half-century need are thanks to AMD. x64, multicore, APU's for consoles.

into the filter you go

I wouldn't buy Intel pozzed housefire garbage even if it was infinitely better for just $1.

this trapfag is still here

>3600 destroys it in single core

nah

Attached: 1561990437318.png (718x379, 84K)

Imagine 3600 getting on par with 9900k pozzed housefire in single core kek

youtube.com/watch?v=prAaADB9Kck

results like this are what happens when the entire test fits in cache

Even if it were on par in single core (it isn't but let's just pretend that it is) its high latency makes it perform like shit in games.

see

stop posting these rigged dogshit graphs. no one cares about fake news, incel marketing team

Attached: cuck.jpg (491x550, 38K)

Something seems fucky here. I mean AMD only managed to (barely) beat Haswell with zen.

They couldn't even get FlareX RAM to 3200mhz.

ma gaymz

Attached: emotion_.jpg (350x350, 16K)

What kind of ram is sopa de macaco website using for those tests?

This this this
Reviewers get the best chips while the ones in the market are always worse

I don't know but there's something clearly wrong with it.
All memory benchmarks are somewhat fucked up.

Forgot to attach image

Attached: AMD-Ryzen-5-3600-X470-Tests-9.jpg (719x568, 203K)

Look what popped up in userbenchmark.
Can't wait to see the proper benchmarks in few days to see how these really perform with proper RAM.
This one seems to have 3600mhz cl14 RAM, but it says clocked at 2400mhz and the latency is kinda fucked.
Who the hell knows what the specs really are, but we'll see the real performance of these chips pretty soon. The single core might actually be equal to the 9900k.

cpu.userbenchmark.com/Compare/AMD-Ryzen-5-3600X-vs-Intel-Core-i9-9900K/4041vs4028

Attached: 5863ad0b94243783bc92dbf0ce505723.png (1439x766, 236K)

Looks like they are using G.Skill FlareX DDR4 3200 MHz witch should be cl14...

Attached: las memorias.png (757x805, 204K)

Attached: 1548337851059.png (440x229, 10K)

SEETHING

AVX nigga

Also see reviewers with kikeripper 1 who needed 1.4v to hit 4ghz meanwhile a forum poll suggested the average range is 1.2 to 1.3. I'm 1.25 llc2 assrook mobo 1920x

Makes sense. The 3600 only clocks up to 4.2GHz.
Apparently they're testing with beta firmware too, so expect the number to go up when you can use not-shit ram.

>using 2666 ram
Yea, there are still issues getting faster ram to work with it (seeing as it's an unreleased chip).

And 2x cache

sopa de macaco benchmarks sure are reliable

Attached: haha.jpg (1421x447, 119K)

what bench is that?
looks like the 3600 wasn't even running at full boost, should be 4.2ghz