it is already 3rd month of 2016, where is Zen?
Stuff is ALWAYS late. Always.
Zen was expected at the end of the year Q4 I believe
>>53430024
If Zen were released now it would be early if anything
AMD has a presentation about Zen next monday.
WAITFAGS ON SUICIDE WATCH
There's no point in waiting. It will be bulldozer all over again. I fell for the meme. Don't fall for it too.
>>53430141
everybody's on suicide watch these days even the people watching the suicidal are on suicide watch.
disclaimer: I am actually testing out a keyboard I got out of the closet, its one of those shitty cheap insignia MUH MINIMALISM keyboards.
>>53430186
but it was pretty good
>>53430222
It was supposed to blow away the 2500k already out for a long time. Instead, it took another 4 years until they had a new CPU that could reach its performance.
>>53430076
Really? Proof?
>>53430367
Sandy was released in January 2011 and bulldozer was October 2011 so not that long.
>>53430961
http://ir.amd.com/phoenix.zhtml?c=74093&p=irol-eventDetails&EventId=5220380
2017
They would be released in time for kaby lake to free the Intel sheeple from spending Dollaroos on refreshed processors
>>53430056
AMD confirmed Q4 2016. OP is probably trolling.
>>53431566
some say Q3
>>53431053
>>53430961
they going to talk about AM4 and some other useless gimmicks
>>53431689
First AM4 Excavator APUs supposed to appear in Q2
>>53430961
>Doesn’t know any technology conference
>GDC is next week
Really /g/? all you know is shitposting and shilling jew products
>>53433013
Most of the silly NEET "posters" here are bots from Virginia EDU or Virginia GOV or Virginia MIL sites.
>>53429970
>>53430024
NO IT ISN'T. IT HAS BEEN AND STILL IS GOING TO RELEASE Q4 2016. IT WASN'T CHANGED
>>53431689
>AM4 & FM4
>gimmicks
>flocks to suck off Intel's 7 different modern sockets
GTFO
>>53431662
Wishfull thinking. If you read between the lines, it is pretty clear that they mean very late 2016. I'd say november earliest.
>>53429970
Zen comes after years of meditation.
>>53431566
I heard Q1 2017
>>53431662
The motherboards are shipping q3, zen is shipping late November.
>>53433716
That is not based on anything, it's just from people counting on a delay happening, or trolling.
>>53433740
Source for november? And mobos should ship together with AM4 Bristol Ridge, teh date of which is not really known ATM but could likely happen on Computex (june).
>>53433740
No. Stop pulling things out of your ass.
AM4 mobos launch this summer with Bristol Ridge aka desktop Carrizo.
According to an internal slidedeck from March 2015 AMD has Summit Ridge on track for a launch in October. Though it was never made clear if this would be a paper launch.
>>53433716
You heard wrong unless you have evidence.
>>53434304
Isn't am4 going to use coherent fabric as a bus or something? I heard some motherboard designs are being delayed.
>"I heard" thread
>>53434494
>http://ir.amd.com/phoenix.zhtml?c=74093&p=irol-eventDetails&EventId=5220380
>>53435919
This is about their upcoming Polaris GPUs. It has literally nothing to do with Zen.
>>53429970
Nowhere.
It will be assraped by nvidia as always
I thought it was slotted for October of this year.
>>53433090
Really? Proof?
>>53434436
>>53434304
AMD said months ago they don't expect significant zen revenue in 2016, implying a paper launch.
also, zen STILL hasn't even taped out, so the idea that we'll get it in Q4 is pretty much impossible at this point. nobody has taken a cpu from tapeout to general availablility in less than 12 months before, and AMD has always taken at least 15 months.
>>53437937
Proof that it hasn't taped out? I remember there being news stories that it has.
>>53437974
you're misremembering. it was big news when AMD said zen still had not taped out
http://techfrag.com/2015/12/09/confirmed-amd-zen-hasnt-taped-out-yet-reveals-chief-financial-officer/
they will make an announcement when zen finally does tape out, but right now it isn't looking good.
>>53438036
No I definitely remember that, and I also remember AMD people saying that it wasn't that Zen hadn't taped out its that they were going to release information about it closer to the launch for marketing purposes.
>>53437937
The statement was that no Zen part would have a *full year of revenue* until 2017.
Literally everything you're saying is a distortion of the truth. Every single thing.
>>53438191
That entire fiasco stemmed from a Currynig speaking broken English and a transcript from SeekingAlpha being done poorly.
The consumer Summit Ridge chips have taped out, enterprise Opterons hadn't yet.
I am not expecting much from Zen CPUs.
The advantage of Intel is simply too big.
AMD should focus on other stuff, searching for niches. They will never compete against Intel again.
In the GPU market on the other hand, they could turn things around with Polaris.
>>53438278
CPUs do not need 1:1 performance to compete with each other. Intel is featuring Xeons with FPGAs to increase their utility and performance. AMD has opted to focus on their interconnect backbone, decreasing latency between each socket in a rack to increase throughput.
AMD wont ever touch intel's FPU dominance, but they don't necessarily have to either.
>>53429970
Its funny but amd hasn't hinted at anything offically
but unoffically we should of had the new
motherboards by now because of Bristol ridge come out after the tax year, wtf is going on why are all motherboard manufactures sitting on their hands.
heck we don;t even have prototype motherboard pictures out.
No leaks no nothing
>>53430076
I doubt its about Zen though but then again i'll be at work by the time this session is Over.
Bristol ridge is strangely quite in a spooky way
common its just excavator it isn't even a secret.
who here next build /zen+polaris/?
>>53438278
You do know CPUs aren't only used for gaymen right?
>>53438782
/disappointment+rebrands/
>>53438782
Maybe. Zen better at least match Haswell or I'm going Braodwell-E. Polaris for sure, I can't wait to get rid of this 970.
>>53436028
*but ZEN APU
>>53438233
>The consumer Summit Ridge chips have taped out, enterprise Opterons hadn't yet.
consumer summit ridge chips are the same as the enterprise ones you idiot, the 16 core opterons are just two of the consumer 8 core dies on an interconnect.
it hasn't taped out yet, if it had we'd have heard something about it. AMD is in dire straights, we'll be lucky to see zen before 2018 at this point.
>>53430367
>It was supposed to blow away the 2500k already out for a long time.
It actually did, in multithreaded apps.
>>53433690
This guy gets it.
Then again, they used an overhyped codename once already...
>>53440559
>look at me pull things out of my ass: the post
Desperate disinfo shill is desperate
AMD stands for Another Month Delay
>>53429970
going on the normal the normal tap out turn around, it should be out the first half this year, but amd has said/alluded to latter part of 2016
>>53430056
nah, thats when the second tap out whould have happened, the market chip would have been q1 this year.
>>53438278
>The advantage of Intel is simply too big.
They've done average 3% increments since Sandy.
If Zen is only 10-20% better than Sandy, then it'll be pretty darn close to latest Intel cpus.
>>53430367
it handed intel its ass in multithread, sadly the bet amd made was single threaded applications were going to die off, and they didn't.
>>53434304
fairly sure amd contracted out to samsung for additional fabs, we still sure this will be a paper launch? or are we thinking it will be a full serve launch and the desktop will be paper?
>>53440729
AMD didn't make any such bet, stop it with this retarded fanboy bullshit.
Bulldozer was designed as a server architecture, its made primarily to handle highly threaded integer workloads because that addresses the majority of the enterprise market. They couldn't tweak the final design enough, and GloFo couldn't hit target clocks so AMD ended up with their hands tied and had to deliver underperforming consumer chips. It took up until the end of 2014 for GloFo to start turning out decent Vishera chips on their broken 32nm PD-SOI node.
>>53437937
no, what amd has said is they don't expect market share to pick up on the home pc front.
it would be retarded to assume otherwise with a more or less stagnant market. what they are expecting is to regain server marketspace.
>>53440559
nope, the 16 cores are a full chip, amd is also launching a 32core chip which is 2 16's on one socket though.
>>53440788
Stop repeating bullshit you read in clickbait headlines.
Global Foundries and Samsung have a working partnership as the only two remaining members of the Common Platform Alliance. GloFo licensed Samsung's 14nm FinFET process, dropping their own internally developed 14XM. Part of the agreement in licensing the process was that they could load share between facilities if necessary, and would run test wafers across all facilities to ensure global uniformity.
AMD did not contract Samsung to produce any of their chips. AMD is bound by a wafer supply agreement with Global Foundries for years to come, GloFo is their dedicated fab. If, and this is a pretty big if, Samsung ends up producing any chips for AMD it will be done as carry over from GloFo.
>>53440693
even the most conservative estimates put it around sandybridge, and if that was true, amd still has an advantage because of the turbo jew intel has been, they could sell the cpus at such a low price it would be impossible to argue against them, all while still pulling in FAR more profit per chip then piledriver do.
>>53440848
yea, they need to give glofo X amount of business, does that contract say they can't contract out additional fab work?
>>53438782
all depends on the price, i'm hopeful to get a system built while using my current gpu, depending on how much the performance from the die shift either jumps or how much the price on performance goes down, i may get get a new gpu, or stick it out one more gen as my 280x gets me by.
>>53440826
>nope, the 16 cores are a full chip,
AMD already came out and said that they will only be making 8 core dies for summit ridge, you idiot.
>amd is also launching a 32core chip which is 2 16's on one socket though.
no, that's a custom supercomputer chip that's going to have a GPU + HBM + 32 cores, it's not socketed and won't be available to the general public.
>>53440925
Nowhere did AMD ever clarify if consumer and enterprise chips would be using the same harvested dies.
Nowhere did AMD ever state that their enterprise APU would be slotted instead of socketed.
Stop pulling things out of your ass, you pathetic faggot liar.
>>53440956
>Nowhere did AMD ever clarify if consumer and enterprise chips would be using the same harvested dies.
that's how they've always done it, do you live under a rock or something?
>Nowhere did AMD ever state that their enterprise APU would be slotted instead of socketed.
read the patent AMD filed for it you retard, it's pretty clearly stated.
>Stop pulling things out of your ass, you pathetic faggot liar.
the only person lying here is you, quit saying shit you know nothing about out of blind fanboyism and inform yourself of the facts before clicking the post button.
>>53429970
>intel pentiums out perform the FX series in both single and multi threaded applications
>continuing to add more physical cores
>more heat
>>53441040
>that's how they've always done it
How they've "always done it" was out of necessity due to large die sizes directly correlating to low yield. AMD only did this apposed to taking intel's approach of dedicated enterprise dies because their foundry could not accomplish the same.
They are not facing any foundry woes with Samsung's 14nm process.
More pathetic faggot child lying isn't going to make you look any less stupid.
