JEWTEL BTFO

JEWTEL BTFO

Attached: 4F069DE5-3A32-4C30-977F-82DFCCDFE4D3.jpg (1433x668, 290K)

lix ass

goy stop with this antisemitism and buy intel ples

Attached: 1527629778452.jpg (679x758, 54K)

>MT
They hid the single thread for a reason.

>4.8 Ghz
>Beats intel in single thread
I will go AMD all the way.

some user form another thread claimed ST was around 275 whirl intel sat at 240
on a ES...still waiting for confirmation tho so it's BS for me right now but I'm interested

Maybe it is 275 if their clock is 5 Ghz. While Intel's 9900K stock is turbo'd 4.7. So that would make sense. Or if both are 4.7 and AMD has better IPC, in which case, excellent.

But I'm leaning more towards similar IPC but higher clocks.

>14nm++ uses more power than 7nm
Duh. Your brandshilling is fucking retarded.

45W 4-core 3.8GHz when

Intelfags claimed their 14nm++ process is equivalent to their rival's 7nm, so

14nm++??? you mean 14nm+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
if this is a ryzen 5.. intel is finished and bankrupt

>4 cores
4 cores are dead, Jim.

oy vey , competition is bad stupid goyim!

Assuming all other things equal this puts it at around 4.6ghz. The 2700x hits around 1800 at 4ghz

I already mentioned that brandshilling was retarded, why even mention what retarded Intel shills claimed?
Ahh yeah, sorry, i meant the eternal (((14nm++++∞)))

No, competition is definitely good. I'm not the one running around screeching INTEL BTFO, AYYMD BTFO, NVIDIA FINISHED.

You can't hide the ST when both CPUs have the same number of cores.

yet you're here in a thread obviously about shitposting. gonna claim high ground now?

There's still the Athlon

Whatever the SMT implementation of Ryzen does it does it better than HT, so MT is a tiny bit AMD biased.

I'm waiting until we see something other than cinebench to see if that discrete memory controller isn't harming performance.
The 2990WX did well in cinebench too because it isn't a strenuous workload to memory.

Pretty sure a chunk of that has been determined to have been an issue with Window's scheduler being shit.

That's not the point. Even if the windows scheduler was the entire reason the 2990WX was getting memory bottlenecked, it was getting memory bottlenecked in windows, Cinebench and other rendering workload scores were barely affected by this.
Same is true of memory overclocking and latency tightening, cinebench barely budges when you fiddle with that, so if the memory controller is adding 20ns of latency to memory access times it wont show up in a cinebench score but that'll hurt something like gaming

Really? I actually think IPC is gonna see huge gains. The issue with ryzen, and partially with ryzen 2, was memory latency due to the speed of the infinity fabric. Actually, memory latency is where the bread and butter was for Intel. While increasing clock speeds certainly helps, I think their focus will most likely be in the IF which will show significant IPC improvements.

>They hid the single thread for a reason.
No they did not. Same number of cores and threads, and we already know based on previous equal core/thread Cinebench tests that AMD's superior SMT implementation is worth less than 3-7% advantage in multicore vs single core scores.

This means the engineering sample, at lower pre-release clocks was within ~5% of the 9900k in single core. Since it's obvious that release clocks will be higher, you know at a bare minimum AMD can slightly exceed 9900k performance with their mid-range Ryzen 3000 CPUs in Cinebench, both in single core and multi thread.