newsroom.intel.com
anandtech.com
newsroom.intel.com
THANK YOU BASED INTEL
AYYMD IS FINISHED & BANKRUPT
AYYMDPOORFAGS CONFIRMED ON SUICIDE WATCH
THANK YOU BASED INTEL
Wyatt Turner
Ryder Bennett
>Intel estimated results on pre-production HW
What did they mean by that?
Jace Hall
>Performance results are based on testing or projections as of 6/2017
We are almost in 2019, you fucking merchants!
>and may not reflect all publicly available software updates.
>No product can be absolutely secure.
Very funny damage control.
Wyatt Campbell
Gee, thanks I guess
Easton Taylor
What a shit-tier shill thread.
Ayden Rogers
Is it me or all those "UP TO" with a lot of small text and wording like "LEADING" are too desperate show even for retarded wall street investors?
Anyone with a basic knowledge of OpenFOAM compilation knows that < 1.6 is nothing. Same for other benchmarks on previous gen AMD CPUs vs paper Intel one. They even do not compare perf/price or perf/power, not pretty slides? XD
Jow Forumsuys, is Cascade Flake confirmed?
Lucas Rivera
That doesn't sound very impressive at all desu. Most of the real-world benchmarks are on the order of 1.5X, and those that are higher are most likely AVX-accelerated. And hit is for a 2S CSL-AP system vs 2S EPYC 1. Equal amounts of salt applies, of course, but AMD has declared an expected 2X improvement for EPYC 2 over EPYC 1, so a 1S EPYC 2 system should be expected to roughly match this (especially given that it too will now have 256-bit SIMD). And then imagine a 2S EPYC 2 system. And EPYC 2 will probably even hit the market before CSL-AP.
Seems to me Intel is truly finished & bankrupt.
Zachary Stewart
>Cascade Lake
AdroredTV
>Cascading Failure
>Cascade Flake
AT first comments:
>No “Cash Out Of Your Pocket” benchmark? What’s the multiplier versus EPYC on that?
>Translation: Intel is getting worried, like, for reals this time.
And shit is not released yet. EBIN
Parker Butler
>deactivate SMT on EBYN systems
>2x24c with newest Glue™Technology™
>no HT
>security patches not applied despite no infos if they fixed anything at all
>no info on TDP (350W projected)
>rushed grafic one day before anudda shoa named ROME
>14nm++++++++++++++
12billion $ R&D right here, folks.
Austin Edwards
That disclaimer thou. kike/jews
Adam Clark
there's so much bullshit in that disclaimer i got bored parsing it all down.
Mason Cooper
Impressive benchmark results. Let me hook my new cooler for this CPU.
Ian Evans
AMD is already sampling EPYC 64to the big players. They already know how well it performs. They're are under NDA of course. But they will be scoffing at such kikery if EPYC is performing as promised.
Jose Rogers
It's probably doesn't outperform amd as much as they claim, but it will still noticeabley outperform amd.
Levi Smith
Even if it did it would be too late. EPYC 64 is already sampling. CascAIDS Flake is not even out there. It's another paper launch.
Dylan Butler
Rome
Kayden Howard
All roads lead to Rome. Unless you lose your way and end up drowning in Cascale Lake.
Evan Collins
>Even if it did it would be too late
Plausible. This time your claim could eventually be come plausible.
Jordan Taylor
CASCADING FAILURE
>CASCADING FAILURE
CASCADING FAILURE
>CASCADING FAILURE
CASCADING FAILURE
>CASCADING FAILURE
Leo Fisher
Anyone knows if they have patched the www.intelflaws.com/CascadeLeak vulnerability?
Christian Peterson
more like epyc fail rofl
Bentley Green
Cascade Lake AP is just two 8180 dies with 4 cores each disabled, put into a single package, with HT disabled and the clocks turned up to create a 48c/48t thermonuclear house fire.
An engineering sample of Rome beat two 28c/56t 8180s already, so there is no reasonable set of assumptions you can make under which 48c/48t Lake AP has any chance of being faster than 64c/128t Rome.
And since the 8180 28-core is already at the reticle limit for 14nm++++++++++, Intel will not be adding any more cores before they get a new node, either.
This battle is already over and Intel already lost.
William Thompson
>no product can be absolutely secure
My sides are in orbit
Has Intel completely given up on securing its products?
Ryan Carter
Oi vey, didn't you know HT is anti-SMTetic?
Michael Davis
>This battle is already over and Intel already lost.
Intel has been losing all these battles since AMD became a company. I really wonder just how they will recover this time.
Christopher Foster
Literally nobody cares about security or power consumption.
Kayden Bailey
I never thought I'd see a resonance cascade.
Carter Jones
We're talking servers and businesses here, not AWSUM GAYMUR RIGS and the shitheads that buy them.
Jordan Wood
Kys manchild
Jose Adams
It's not funny anymore. Just sad now desu.
Mason Rodriguez
yeah.
Nolan Baker
*yeah, sad.
Christian Rivera
public availability and costs??