2016 already, where is the ZEN?
Summer , just wait patiently, I can't wait to get myself a Central Processing Unit that will shut up the intel shills for half the price of an i7. God help us /g/ will be full of people shilling intel processors for the uneducated gaymers. I see intel similar to razer products unless youre going to buy a xeon.
Pic very related another overpriced placebo product. People should be buying chips based on their Raw Compute power, i can't believe people think the GTX 970 or 980 is better than its blatantly more powerful Radeon Counterparts, even disregarding the landslide difference in memory bandwith and VRAM capacity.
>>52492898
>Citations needed
Also, what was that card, something about reducing noise or something?
I can not wait untill benchmarks proove that Zen is just a little better than Sandy Bridge with need to be watercooled to prevent throttling. But muh lower price than intel.
>>52492898
>Summer
So it should already be out? Or do you mean i'll have to wait until December?
>>52492812
Q4 if we are lucky.
>>52493029
herp hurrr de durp durrr
>>52493119
>mommy wont let me spend more that 500$ so I have to use FX9590 and justify my poor spending
>>52493029
>I can not wait untill benchmarks proove that Zen is just a little better than Sandy Bridge with need to be watercooled to prevent throttling
To be honest, people would have said the same thing about Core 2 Duo before it came out, considering its predecessor was Netburst, the Bulldozer of the 00s.
>>52493025
>something about reducing noise
something yes
>>52492812
Functional untethered 8-core znver1 Summit Ridge A0 engineering samples with an estimated 95W TDP have now taped out on Samsung's 14nm FinFET and are undergoing validation.
It's not yet clear how well they will clock (but Samsung's 14nm FinFET process is slightly better than Intel's - Intel haven't pulled ahead with 10nm yet as that EUV process is very late, hence their Kaby Lake). IPC efficiency is very roughly equal with Broadwell, very slightly behind Skylake, but it's push and shove.
They are currently targeting October 2016 (ish) for release. AM4 motherboards compatible with Summit Ridge have chipsets already shipping and should be coming out around about March 2016 (ish).
Subject to change. They may dual-source on GF14A.
>>52493743
>but Samsung's 14nm FinFET process is slightly better than Intel's
Proof? Because Samsung is notorious for industrial espionage in the tech industry and has targeted Intel fabrications for decades.
>>52493802
Given how dogshit Intel 22 was, and 14 hasn't shown to be any better; I'd believe it.
>>52492812
No Zen based parts are due out any time soon.
>>52492898
Summit Ridge does not launch until the end of the year. The only thing launching this summer is AM4 motherboards and the desktop variant of Carrizo called Bristol Ridge.
Don't spread disinfo.
>>52493802
There is no proof because there are no comparable parts produced on both nodes. You could look at test chips and vaguely extrapolate some details from each, but you could never make a head to head comparison unless it was the exact same IC.
By way of area scaling and density no one is near intel, though they quite clearly have been struggling with the node since mid 2014. Broadwell Core M launched in late 2014 and it was a small disaster. Desktop Broadwell was delayed, and ultimately pulled off market after only a couple months of extremely limited availability. Even desktop Skylake had limited availability compared to Ivy Bridge and Haswell at launch.
Samsung has had no such issue since they brought their 14nm LPE online. Samsung/GloFo may end up hitting a real home run here simply by being able to churn out volume numbers of higher binned chips on their FinFET node.
>>52493802
I'm retired, but it's what I've heard from former colleagues in the reversing field and I believe it. Proof on that costs serious money, which Chipworks et al will gladly take if you want it that badly. It'll cost you six digits for that kind of 'competitive intelligence' espionage.
Samsung don't seem to be violating any Intel process patents. Don't know about so-called 'trade secrets'. Samsung may have a rep, but it's nothing that everyone else doesn't do to everyone else as well.
Samsung's been throwing more money and very talented engineers at 14nm: meanwhile, Intel's been trying to forge ahead as usual, spending their time to get their 10nm working - and failing, instead of refining their 14nm. GF are shipping, but late deliveries, and TSMC's 16nm has been a clusterfuck - there's a reason Apple dual-sourced. At least one AMD Polaris sample has been fabbed on Samsung too.
Hell, Keller moved over there from AMD recently once he was done with Zen. Not sure why: Exynos? We'll see.
>>52493932
FinFETs are great ideas in general, but that's where most of the process improvement comes. The feature size reduction isn't as significant as it sounds.
>>52494156
>The feature size reduction isn't as significant as it sounds.
I'm almost 100% positive you're the very same pathological liar showing up in all the threads recently. You have nothing to do with the industry, you never have, and its painfully obvious.
14nm has greater than a 70% area scaling advantage compared to 28nm. A 100mm2 28nm die would end up about 30mm2 or slightly smaller implemented on Samsung's 14nm process. The differences in FEOL and BEOL feature size are absolutely enormous.
You do not know what you're talking about. Stop spreading bullshit on this board to make yourself feel better. I will call you out for it every single time.
Q4 lmao
>>52492812
It's 2016 already, where are the two unlimited detail games we were promised?
>>52494217
>I'm almost 100% positive you're the very same pathological liar showing up in all the threads recently
Ditto. I thought he was a pro-Intel shill, but pathological liar seems to fit the bill better. Maybe even literally insane.
>>52494299
you need 4chan gold accaunt
2016 already, where is the Vulkan?
>>52495930
Going through legal processes I believe.
>>52492812
Will there be zen in gaming laptops?
>>52496421
yes in 2017
