Who here's literally shaking in their chair in anticipation?

Eh it's going to disappoint, that much is certain.
Hopefully it disappoints AND performs well.

File: Ryzen-SKUs-768x347.png (139KB, 768x347px) Image search: [Google]

139KB, 768x347px

If all goes as planned intel may become irrelevant in the desktop game soon.

OEMs are dying to put 8 core 65w chips on desktops and make bank.

>>59004342
You seem so sure contrary to the dozen of leaks so far and a full test suite by a very old and reputable source.

>>59004350

>If all goes as planned intel may become irrelevant in the desktop game soon.

You wish.

I bet Ryzen is so supply constrained Intel doesn't even have to drop prices because either you buy Intel or wait for 4 months for a chance to grab a Ryzen if any become available.

AMD being fabless might damage high volume business like CPUs a lot. Even GPUs which are sold in much smaller volume than CPUs are often supply constrained months after their official launch.

>>59004462
Zen chips aren't big, and GloFo has been building up volume for months, the nice thing with building volume was also that they could pull another 100-200MHz higher stepping while waiting, so not only will there be higher than expected performance on launch, there will be plenty of stock.
Unless everyone just buys one SKU and leave everything else alone, then AMD is finished.
The 6 cores will probably be the most sold ones.

>>59004331
I would be glad this development if I hadn't dropped desktops entirely. Especially if it's confirmed to overclock like a mad bitch.

File: amd.png (29KB, 176x174px) Image search: [Google]

29KB, 176x174px

>>59004497
>the nice thing with building volume was also that they could pull another 100-200MHz higher stepping while waiting,


What do you mean by this? Wouldn't that entail a new SKU?

>>59004986
Bugs get ironed out during and yield increases the wait and it allows the same performance at lower power or higher performance at same power, these aren't big changed but 100-200MHz on a 16 thread CPU is a pretty big increase in performance when all cores are running.

why do you think that? everyone wants bang for the buck.

File: 1486781890851.jpg (23KB, 472x472px) Image search: [Google]

23KB, 472x472px

>>59005096
Are you telling me they pump out chips for months, and while waiting for their stock of chips to build up they design changes to the litography, then at some point start pumping out chips with a different (improved) litography? If so, do they name the improved chips something else and give them a new SKU? Surely they don't sell the old and the improved chips under the same SKU? If they did, some people would be lucky and get the improved ones, and some would get the old ones.

>>59005127
They just make new SKU.

>>59005127
IF there is a change in the lithography masks (and this is a big IF, that doesn't happen very often), it will be indicated by a different stepping.
Stepping changes normally only happen if there are serious bugs in the silicon that can't be properly fixed with Microcode.

Examples of changed steppings during production are the Sandy Bridge SATA bug and the Phenom TLB bug. Both of these were only fixed in a newer stepping.

You won't really be able to verify which stepping you get, unless the seller specifically mentions it

>>59004331
It's going to be a whole month of only damage control and shitposting.

>>59004350
those prices look amazing.

how real are they tho?

>>59005248
Lithography masks haven't been used by AMD or Intel since vishera

>>59004331
is ryzen pronounced "risen" or"ry-zen"?

you are over hyping it, please stop

>>59005633
The latter.

>>59005633

Rye-Zen.

>>59005662
this better be a cocktail soon

>>59005248
So if they change the stepping in such a way that it increases the performance, will it be under the same SKU? If so, then some people will get better CPUs than others.. How can you then rely on benchmarks? Have this been done in the past, if so got any examples?

>>59005621
What do they use now then?

File: twisted psychopath miata.jpg (62KB, 1149x332px) Image search: [Google]

62KB, 1149x332px

I've been here since like 2006, I'm ready to troll the fuck out of literally everyone--I don't care which side you're on.. I can't wait for the next-gen woodscrews housefires memes.

>>59006016
Other CPUs working acid and layers directly on the die

>>59004331
I'm excited to finally not have to pay 600+ for a dual-pc set up.

Even if zen is 80% of the performance (clock-for-clock) at HALF the price of a 6900k or even a 6950x with it being on the am4 platform (fuck intel for their stupid overpriced x99 shit), then we're golden.

>>59005999
Only i remember were the last Core2Duo with OC potential, the E0 steppings overclocked more easy while the C1 steppings needed more volts or couldn't OC that high (like 4ghz E0 vs 3,8Ghz C1 with a same and safe vcore on a Q9650/E84000), still the preformance per clock was the same, and even then it wasn't that exclusive (like some C1 special snowflakes that did OC'd better than some E0).

>>59006793
>one SKU got 4GHz while another got 3.8Ghz at the same voltage
>except for when it's different
That's hardly proof that they're even different cpus

>>59006037
Redo this image with the lights flipped up