We've gone full circle, haven't we?

We've gone full circle, haven't we?

Attached: EEP4.jpg (259x194, 8K)

Pretty much. Who'd have thought a 5GHz 8-core chip would be a housefire?

Not yet.
I bet on Zen 2 only clocking at 4.5-4.8 Ghz but still defeating the 9900k due the superior IPC, and then intel doing a 9999k that clocks to 6Ghz, with all the expected heat issues.

does the k at the end of Intels CPUs names stand for kelvin?

Attached: 1528524240061.jpg (500x500, 48K)

>I bet on a 14nm to 7nm die shrink only netting 200-500mhz clockspeed increase :P
You're either retarded or a bad shill.

>AMD 7nm is going to better
>FUCKING INTEL SHILL

maximum overdrive

Attached: >95W.png (650x337, 40K)

Shill then.

So he's an AMD "shill" who is making conservative estimates about Zen 2 clockspeeds?

I don't know what's the big deal. It's very obvious that the i9 chips are binned to the most effective yields and then overclocked very high from the factory. 5ghz will cause a housefire for any chip.

Given the signs of a 13% increase on the IPC, i have a hunch that AMD might be making the pipeline shorter.
That limits the clockspeed, but increase the non DSPShit performance quite a lot.
It's literally how the Athlon defeated the Pentium 4.

Yes, he's an Intel shill trying to make Zen2 look bad without knowing anything about it in an attempt to sell more 9900K's today.

>in an attempt to sell more 9900K's today
9900k isn't even in stock anywhere

what shocks me is that amd is now the top dog when it comes to performance per watt

Attached: chomp.png (468x486, 219K)

Learn how a power curve works.

yes why not? give 300mhz upkeep and just slap more L-cache around given native double avx 256
support and watch as intelaviv cries

Well, i just explained on the post above yours my point.
AMD can participate on the retarded nuclear race, or they can go smart, shrink the pipeline and still deliver the same or better performance, but going colder.

This may not be the right way to go sales wise though. MUH GIGGANIGGAHURTZ is definitely a part of why normies think intel is better.

Theres no point in chasing intel in avx, there are other areas they need to improve on that will still give them a performance edge.
I doubt transistor count will increase that much, and increasing cache size tends to make caches slower, the opposite of what AMD needs now.

Then they can revive the PR Rating.

look at 2950x they increased the l1 and l2 a bit more and it crushed intel even on games
it will def do its job but im more interested to see what they will bring on the table in terms of FW and in general software support

Performance per watt never mattered at the high-end desktop level. Well not very much. Clearly you don't want a 220W housefire that you can only cool by summoning eldritch abominations, but it's raw performance that matters on a desktop.
On the mobile market, performance per watt is the only thing that matters. Intel is still doing good there, mainly since AMD's mobile parts aren't easy to find. I really want to see a Ryzen 2 mobile chipset make its way into some high quality laptops.

Stands for kike.

7nm Zen won't be. It'll probably use the same power as the 2700X while doing 5GHz turbo.

This is, as I mentioned a few days and weeks ago leading up to the release, a limitation of Intel's ringbus.
Yeah, the latency isn't too bad for 8 or even 10 cores, but the uncore power consumption fucking skyrockets.

Ryzen's uncore is so unbelievably low that they can duplicate much of it and still be more power efficient than Intel's mesh uarch.

Shouldn't be shocking considering that AMD competing with Intel in laptops on 32nm while Intel was on 24nm.
It was only until Intel was on 14nm while AMD was on 28nm that they could no longer compete.

Process node type matters.
There's still not confirmation that they're using TSMC's HPC process. They might be using... NPP I think it's called? The smart phone process which would likely only be a few hundred mhz increase.
However, worse case I'm still expecting some SKU which has 5GHz turbo on 2 cores even if all core turbo is only like 4.4GHz. At the very least, it'd likely happen on the NPP process due to even more improvements to XFR and SenseMi. From what I've heard, AMD has been trying to regulate voltage on a per-core basis.

Yep. This 5Ghz shit is definitely a marketing gimmick. It's mostly meaningly performance wise.
But we already have people who go to /pcbg/ asking why their 8700k keeps stuttering in games. It's almost always because it's thermal throttling.

well just look the situation now compare the acer nitro 5 intel and acer nitro 5 ryzen 2700u
intel needs all the binned shit they can find while amd just lowers the base clock by 1ghz and has like 20 watts lower power draw..