8 CORE RYZEN BEATEN BY 4 CORE INTEL CPU IN VIDEOGAMES IT�

basically that moar cores meme is still relevant, unlike amd hehe

File: amd1.jpg (71KB, 552x661px) Image search: [Google]

71KB, 552x661px

>>58294701
DELET

>>58294701
>gaymen

>>58294921
that's 80% of the people that would have bought ryzen

nobody will buy this when it's outperformed by a cheap ass 4 core cpu

it's over for amd

finished

done

cUckolded

File: 1410046519529.jpg (6KB, 250x241px) Image search: [Google]

6KB, 250x241px

>look mom i posted it again

>>58294989
>intel takes over gaymen
>amd takes over AI and scientific computing

Good riddance.

>ignores the fact it btfo'd 6800k in compute
>doesnt link to actual benchmark
>review site (canardPC) ran a lower clocked engineering sample at 3.3 ghz, say sretail boost clock is at least 3.6 and can OC to 5ghz on air

Make 3957 more intel posts and you can earn yourself a new pentium, apply at intel social services today!

File: lol.gif (2MB, 200x150px) Image search: [Google]

2MB, 200x150px

>>58295038
>amd takes over AI and scientific computing
You mean nvidia?

>>58294701
>8 CORE RYZEN BEATEN BY 4 CORE INTEL CPU IN VIDEOGAMES

It was a 4 GHz Intel vs a 3.3 GHz Ryzen, and it only won by something like 5%.

That means Ryzen is faster clock to clock.

>>58295072
>nvidia
>AI and science
What?

inb4 pic of elon musk and nvidia CEO

>>58294701
I'm guessing this is because of Intel's Parallel Studio being used in the game industry. Does AMD have an equivalent high-speed C++ compiler? Would be great if they did, enough getting cucked by the intel monopoly.

>>58295112
>what is nvidia DGX-1
as expected of gaymur poojeet

>>58294701
>http://bgr.com/2017/01/02/amd-zen-vs-intel-ryzen-i7-benchmark-comparison/

" 3.3GHz frequency with half of its cores being used, while the 6700k is running at a much faster 4.0GHz"

Good try OP

>AMD is back

Fucking. Finally.

>>58295062
Yeah the Zen looks great if it's really going to be $500. If it's any pricier I might as well get a 6800k for $400 though

File: 363209.jpg (88KB, 640x644px) Image search: [Google]

88KB, 640x644px

AMD is gonna blow the roof of CES this year

>Trump wins
>Vulkan beats DX12
>AMD now holds bark with both Nvidia and Intel

2017 is gonna be good.

>>58294701
did you even read the article?
The 6700k beats the 6900k in games too.

And he even concludes that- clock for clock, the amd chip is faster.
Just crank up the cpu to 4ghz and you'll be gold.

>>58294701
If AMD stopped focusing on the more core meme and actually worked to increase single-core performance, you wouldn't have this issue.

haven't seen I(nt|sra)el this worked up in a long time
it's not like zen was just announced so what the fuck have they've been doing? just letting marketing/jidf shitpost around the net isn't gonna help when your competitor is bringing out new stuff and trying to catch up

the haifa unit don't exactly blind you with their brilliance..

>>58295112
>what are quadro cards?

>>58295460
8-core is going to be pretty standard in the next generation. cpu's can't get much faster per core without getting larger in size at this point.

The problem is just game performance currently maxes out at 4 cores at this point in time. there are some optimized for better performance on multi core cpu above 4 cores but they're a rarity at this moment.

>>58295609
obsolete

>>58294701
you know that the 6900k is 8c/16t right?
and your quote said that it performed better than the 6c/12t 6800k

tell me how the 6700k beat it then

>half of its cores being used
Not a fair comparison. Someone should come up with a list of PC news websites bough out by apple, Microsoft and Intel.

>>58294701
>But remember, the AMD chip in this test is running at a modest 3.3GHz frequency with half of its cores being used, while the 6700k is running at a much faster 4.0GHz

File: amdfag.jpg (3MB, 8624x3896px) Image search: [Google]

3MB, 8624x3896px

>>58296332

>>58294701
nice b8, OP.

I'm reserving judgement until I see how well Summit Ridge OCs on mobos with quality VRMs.

File: Capture.jpg (103KB, 1242x988px) Image search: [Google]

103KB, 1242x988px

>>58296407
>amd users

good thing nvidia sent the kid a 980 though

Notice how all of the (shit)posts about Zen don't acknowledge the difference in clockspeed or that it performs about on par with Intel's own 8 core chip if you take into account the clockspeed difference.

>>58295112

>implying that the fucking compute backbone of industrial, resource extraction, military, scientific and biochemistry firms is not Nvidia Tesla GPUs

>implying that Nvidia quadro cards arent mandatory, unavoidable parts of all design, modelling, machine learning, simulation, and visualization work

>fucking implying nvidia doesnt have an unshakeable monopoly on professional partner certified driver packages for everything from the top discovery and design firms to the daily work of avionics / aerospace applications

Nvidia hardware and software has almost certainly saved your fucking life at some point.

>>58296628
B-BUT MUH BUTTCOIN FARMS

>>58295026

/thread

>>58295296
>>Vulkan beats DX12
keep believing that fag, Vulkan was DOA
>AMD now holds bark with both Nvidia and Intel
kek

File: 1478550885220.png (269KB, 500x472px) Image search: [Google]

269KB, 500x472px

>>58296332
>I perfer Intel
>BUT AMD IS BETTER
>Intel CPUs have better performance in every scenario, particularly single threaded tasks which is particularity important
>FUCK YOU BUY AMD!

File: sad_meme_mittens.jpg (162KB, 1024x992px) Image search: [Google]

162KB, 1024x992px

>The ultimate retardation of this thread

Like the other guy said, Trump winning + my $10k investment in AMD stock sextupling over the year have made America great again. Ryzen is going to BTFO intel in 2017. Vega on the way soon after. What a time to be alive.

>still citing one benchmark with flawed methodology

i7-6700k also beat all the lower clocked 8 core Intels.

IPC results are yet to be seen.

Don't be such a consumer whore that you start identifying yourself with the products you're sold.

>4 core intel beats 8 core intel for gaymin
INTEL BTFO
INTEL ON SUICIDE HOW WILL THEY EVER RECOVER
#AMDREDTEAM
#BETTERRED
kys urself op

>>58294701
>AMD single core performance isn't as good as intel's
>literally anyone surprised, ever
pick one

>>58294701
>LMAOOOOOOO
>laugh my ass off off off off off off off

>>58295625
>mfw switching from an fx 4170 to an 8320 @ 4ghz doubled my framerate in metro last light
>framerate was also significantly more consistent
Only example I can think of where devs actually managed to pull that off. Bravo 4A games.

But yeah dx11 was still centered around single core as it was still bloated with legacy code because MUH BUISINESS SECTOR and from dx9 onward new versions were more like stacking on new features than actual revisions. And the multi core scaling was haphazardly slapped together requiring retarded amounts of work on the part of devs.

At the very least vulkan and dx12 have lower level muticore support natively. Now we just gotta see if developers are up to the task of taking the reigns so to speak.
>looking at you, ubisoft

AYYYMD is a fucking joke

>>58297832

>Ubisoft AlwaysOnline® DRM takes 40% of all processing power thus making efficient use of all your cores.

>>58297841
fuck outta the thread Huang

>>58296407

>these are the sorts of humans the rest of us have to interact with on here

sometimes i wonder why i even bother

>>58297865
I hope low level api's are the death of shitty developers like ubisoft. I seriously doubt they'll be able to make a half decent game on dx12 and we all know those cunts won't bat an eye at vulkan.

Only thing I fear is it might complicate things for smaller developers who are more interested in investing time in the game itself than hardware level optimization. Licensing an established engine is obviously still an option but I'd hate to see the same situation as last generation with endless games using unreal 3 looking and feeling exactly the same.

>>58297958
low level APIs will be used for two things:
> big-name engines that get licensed out
> frameworks for writing portable user-space higher level APIs

It's very hard to see custom engines being written in DX12/Vulkan moving forward, and I foresee DX11-level APIs sticking around for a long time after some initial tentative experimentation by the dev community with low-level stuff.

>>58295103
4790K was tested and that boosts to 4.4ghz

>>58294701
Intel 8 cores also get beaten by intel 4 cores in video games

>>58296628
Preach

>>58298109
Someone will make a wrapper for people. It will just take some time. Engines will get thrown together and then a wrapper will get used. I'm willing to bet money.

Sure, it won't perform as well as an engine built from the ground up taking full advantage of low level access, but it will still have native access to some of that low level stuff, just abstracted so crappy developers can use them without spending months to learn the function calls.

>>58294701
>The tests were performed by French technology outletCanard PC.

And into the bin it goes. Why the fuck do intel shills think anyone will believe this retarded website?

>>58298595
>butmad pajeet

>>58295609
>what are quadro cards?
Expensive motherboard shelves.

>>58294989
>>58294701
>yfw AMD will sell unlocked Ryzen for 350$

File: dog.jpg (42KB, 664x520px) Image search: [Google]

42KB, 664x520px

>>58294701
IT'S ANOTHER FAILDOZZER AHAHAHAHAHA

>>58295062
>5ghz on air
*on one core

I'll grant it was with a shitty buggy motherboard, however it's a large chip and a new arch and a new node. I'd guess all-core 4.4-4.6Ghz at best.

>>58297150
DX12 is just a closed source ripoff of Vulkan that can't even handle monte carlo pathtracing. that's FACT. Vulkan is superior in every single way.

>>58298816
no the pricepoint is going to be 500-800

LMAOOOO

>>58299012
>inb4 AMD does this and crushes 5 years of my hopes and dreams

>>58294701
Entire thread based on some cuck saying

"In a gaming head-to-head, the AMD chip stumbled a bit, "

no benchmarks, no nothing

/g/ laps it up, why do I bother coming here

>>58294701
amd sucks

>>58294701
Disinformation campaign. Intel must be really worried.

>>58295103
>EVERY CPU CAN OC FOR THE SAME AMOUNT

Zen is DoA.

>>58294701
How afraid are Jewlets?

>>58298877
While I agree that Vulkan is superior and probably even better for Nvidia (DX12 gives them no gains, like at all, meanwhile both 1070 and 1080 get a 20-30 fps boost in Vulkan when compared to OpenGL). I can't deny fact that Vulkan API is already dead in the water. Mind blowing library of 6 games 18 months after the release? Wew.

>Inb4 but muh Engines!

Just because a game engine supports Vulkan it doesn't mean the game will.

>>58299554
>gaymer tries to add his 2 cents on graphics APIs/programming/game engines

Not this shit again.

>>58298109
I can see more competent developers that have proven their mettle; such as Rockstar (when they don't offload the pc port to some shitty third party like with gta iv), cdprojekt (memeworks is stupid, yes, but otherwise the game is very well optimized), 4A (2033 wasn't that great in terms of optimization but LL was excellently executed), and even Kojima's old studio with the GTA V port, making good use of low level API's.
Doom's use of Vulkan, and a much earlier example of that Sniper Elite game that supported mantle, have proven already that there is a lot of potential in capable hands.

Battlefield, hitman, and deus ex may have soured things a bit with their pisspoor patchwork implementations, but it's to expected when they tried to cram it in last minute very late into development.

It's a shame we were finally seeing more devs putting the extra work to create brand new engines just to have to start from scratch all over, but I'm optimistic about what becomes possible once we've gotten past the transitionary period.

>>58299587
>A /g/ vermin that does nothing but ricing and shitposting thorough his average day tries to add his 2 cents on what regular/normal people buy and look at when making the choice.

Not this shit again.

>>58299554
>6 games 18 months after launch
You're missing the point here. Vulkan and DX12 are completely new, reworked from the ground up.
It's not the same as what we've seen in the past with DX and OpenGL. New iterations in those cases mostly just added feature level support, general optimizations, and bug fixes.
It was much easier to rework sections of a finalized engine to support those changes, but low level API's pretty require engines being built specifically for them.

The greatest strength to vulkan is that devs generally start development with the ps4 as it is the most popular console by a long shot, and dx12 isn't supported by the ps4.
So it would be easier to use vulkan as the xbone has basically found itself in the same position the ps3 used to be in.
You'd be surprised how often openGL is used on consoles to simplify cross platform releases.

>>58299701
How come that every AMD feature that /g/ uses to shill for their inferior products is always set "somewhere in the near future"? I've heard that for last 10 years.

>>58299730
because amdfags always wait for something

>>58294921
literally the users of top segment. (>= 4 cores)

Kinda love how you guys hang onto the one test that wasn't particularly favourable, over a 1.5% difference against the i5 with a pre-release sample chip.

I'm pretty confident it'll match the i5 in gaming when the full release chip comes out and maybe even beat it once code starts getting optimised for the Ryzen architecture.

>>58294701

Will Zen even be an upgrade over my i5-2400?

I feel like this CPU has no place in this world at all.

>>58294989
>cheap ass 4 core

>>58294701
The 6700K destroys Skylake-E in muh vidyagaymes too, so it's no surprise.

I wonder how the 64w version performs against the 95w chip.

Since i'm an ITX fetishist its probably the one i would go for if i do get Ryzen.

>>58299950
poorfag detected

Have they suggested any prices for Ryzen? If they're selling 6800K performance for $250, great. If they're priced the same as Intel, why fucking bother?

>>58299970
30 watts less, 30% less performance. SCREENCAP THIS.

64 watt Ryzen will be more shit than Excavator

File: 5-1080.2494205094.png (64KB, 1920x1074px) Image search: [Google]

64KB, 1920x1074px

>>58299970
Power curve for Excavator modules in Carrizo:
XVR with AVFS
3700mhz: 22.5w~
3500mhz: 16w~
3150mhz: 10w~
2800mhz: 7w~
2450mhz: >5w~
2100mhz: 3w~
1750mhz: >2.5w~

AMD originally touted Zen as having the same clock scaling as Excavator, but have clearly surpassed it by a significant margin as they're hitting 3.4ghz at 10w~ per core.
Going by Excavator's scaling we'd expect a 65w chip to be clocked at 2.8ghz, but final silicon should be able to clock higher than that. Likely about 3ghz at 65w on all 8 cores.

>>58299730
>vulkan is an amd feature
Vulkan shows performance increases on all hardware. And offloading the lopsided workload from core 1 is benifitial to all processors using more than one core, which is basically all of them at this point.
just because low level api's weren't implemented overnight doesn't mean developers aren't excited to use them. Many large figured have already confirmed they're hard at work on implementing dx12 or vulkan or both, but it's a massive change that requires a lot of work. We've seen some terrible examples so far, but this is because games take years to develop, not months, and neither api can simply be patched in to existing engines as they're completely different from outgoing api's.

stupid thread.

Zen is better than Broadwell E clock for clock.

faggots are in denial that Zen is an actual upgrade.

If it clocks to 4.6ghz on 4 cores, it will beat a 4.8ghz 6700k.


Nvidia fags on suicide alert.

OP on death row for wasting oxygen.

intelcucks in the hospital with cancer in their female penises.

everybody is getting raped.

>>58299981
>2k17
>still thinks poorfag memeing makes overpriced stuff look cheap
good riddance /g/

>>58300711

>faggots are in denial that Zen is an actual upgrade.

If you are the ''''proud"" owner of an AMD CPU then sure.

If you already have an Intel then not so much.

>>58300745

How many characters dit you save typing 2k17 instead of 2017?

>>58300758
he saved a "0".

>>58300747

but it is an upgrade, my 4670k clocked at 4.6ghz is actually slow compared to what i need. and im not putting my money into intels 8/10 core because they wont clock that high.

Zens true 8core if it clocks to 4.6 on all cores will be a beast. hopefully amd doesnt let us down this time.

>>58300747

sadly this, if you already have an newer Intel chances are it won't be an upgrade

It primarily is supposed to have an insane price to performance ratio which is a good strategy

But if you have an really old CPU, Zen is completely fine because of the above

>>58301176
Have they revealed or suggested a price? It will have to be pretty cheap to prise people away from Intel if the performace is similar.

>>58296407
>AMD users are poorfags that can only afford a shitty R9 380 and a fucking shitty Acer monitor

top kek, I want AMD poorfags to fuck off from /g/.

posted from my iphone 7 plus sitting in my herman miller chair

>>58301078
>hopefully amd doesnt let us down this time.
They should put that phrase under the AMD logo

"AMD™, hopefully we won't let you down this time"

>>58301078
Setting yourself up for disappointment there, bud. OC's so far have been on single core, and we have zero data on all core OC as we don't even have examples of fully functional consumer parts.

Just chill and see how it plays out. Chances are you don't stand to gain much from Zen just like you don't really stand to gain much from a newer Intel part.

I for one am happy I can upgrade my 8320 without paying the intel tax.

>>58301284
You could still be a little more pathetic, if you bore a tripcode.

File: dank-memes-for-the-weekend-26-photos-1.jpg (50KB, 600x583px) Image search: [Google]

50KB, 600x583px

>>58301284
>iphone 7
How was it paying the yearly apple tax, friendo?

Posted from my iphone 4

Looks like the perfect Workstation CPU to me.

If it's dirt cheap imma get one, even if it underperforms. 200 for a 6 core 300 or less for a 8 core at max otherwise for gaymen Intel wins.

>>58301284
You sure do love a dick up your ass.

File: 1476848340723.jpg (63KB, 261x281px) Image search: [Google]

63KB, 261x281px

>>58301422
It's probably not going to be that

Unless AMD wants to come out swinging, we'll be lucky to get the slowest 8 core Zen for 350 USD.

Have fun with Intel though
They can game
>PC DOES WHAT?! Intel Ad starts playing

File: 1467954428920.jpg (91KB, 303x357px) Image search: [Google]

91KB, 303x357px

>>58301464
I ain't taking sides, I want a 4 core or preferably more that really is good value and lasts long. The same can't be said about FX CPUs.
>inb4 muh good boy AMD

>2 days till intel novidia cucks are literally hanging from the rafters

>>58296407
Look at the tits on that lass!

would it be worth it to upgrade from an i5-2400 to a ~$400 amd chip? i am sitting at almost full workload 24/7

File: 1453325123698.jpg (17KB, 316x239px) Image search: [Google]

17KB, 316x239px

>>58301575
>would it be worth it to switch from this processor to the other one we dont know anything about?

>>58295389
If the 8c/16t Ryzen chip can reliably hit over 4GHz, you won't see it for less than $900 and what will all of you poor AMDrones do then?

Why do all of you irrational retards think AMD is going to heavily discount their product but still think it will compete with or beat Intel's high tag offerings?

File: 1481490878300.jpg (367KB, 1600x1200px) Image search: [Google]

367KB, 1600x1200px

Wheres the benchmark?

>>58301597
Well we know very little about it. There are a few possibly legit leaked benchmarks floating around but those are using an apparently early engineering sample that may not have all functionality enabled. The leaks looked promising though.

>>58301600
>Another consumerist fanboy
And another one to disrecard.

File: 1455616175723.jpg (70KB, 618x564px) Image search: [Google]

70KB, 618x564px

>>58301612
we dont know anything about pricing. we dont know anything about performance but some leaks that may or may not be legit. we do not know if anon is a millionaire or a poor cunt, therefor "worth it" could be anything.
questions about something "being worth it" in general are absolut bullshit, and even more in this case.

File: 6-1080.4039457724.png (129KB, 1912x1080px) Image search: [Google]

129KB, 1912x1080px

>>58301612
>Well we know very little about it.
The core arch was detailed at length during HotChips last year. Before that ever took place we knew almost everything from various linux kernel patches.
The only details we're lacking are the specifics of their new interconnect fabric, how the turbo core works, the extent of the boost brought by XFR, and how much performance is lost by the bugs CanardPC noted in their ES testing.

>>58301654
There is nothing even remotely questionable about CanardPC. They have an ES and a test board. They've proven it beyond all doubt.
Asserting otherwise is childish shitposting.

File: p1436853280527.png (2MB, 771x1362px) Image search: [Google]

2MB, 771x1362px

>tfw I thought Shitwrecker was a silly meme
Someone explain to me what, generally, this guy does and is known for. What the fuck happened here?

>>58301677
Yeah we know about the differences relative to bulldozer so we can be pretty sure at this point that it can't possibly be slower than the bulldozer line, I just don't know for certain yet whether the performance increase will make it competitive with Intel's products. The demonstrations and benchmarks shown so far have been promising though.

>>58301677
>>58301750
Are there areas in Zen's architecture that would be still bad and under performing compared to Intel?

>>58301744
Hes done tons of things. Hes old and been in the industry for decades.
http://www.anandtech.com/show/6129/apple-a4a5-designer-k8-lead-architect-jim-keller-returns-to-amd

>>58301782
The FPU isn't near as big as intel's latest core i arch.
Theres a chart comparing instruction throughput and latency going around.

>>58301677
>There is nothing even remotely questionable about CanardPC
yet this is useless without a price tag on it.

Is AMD also part of the botnet?

>>58301918
Yes, but not as much as Intel.

>>58301820
What site/s would you suggest for a detailed review of Zen's architecture when it launches?

>>58302071
anandtech
computerbase.de
hardware.fr
tomshardware

>>58302088
Thanks anon.

>>58301677
>Aggressive clock gating with multi-level regions

Why are they listing this as a positive? It literally steals your megahurtz.

>>58302097
It saves power.

File: 1483453984059.png (2MB, 750x1362px) Image search: [Google]

2MB, 750x1362px

>>58301918
>>58301944

HOLY fuck this is relay a meme that needs to die and I'm not using
>X is a MEME

This
>X is a botnet
shit needs to die.
I swear to god MEMEs will destroy humanity, at one point we will lose the capability to have meaningful conversation since all words lost their meaning.

File: 322.jpg (377KB, 1280x1280px) Image search: [Google]

377KB, 1280x1280px

>>58296530
if this is true his brother is a faggot just like OP

>>58302243
Big brothers have been scumbags since the dawn of time.

I'm a newfag in my first year in a IT course. Isn't the outer clock speed (I assume it's on the motherboard) that synchronizes the other parts of the computer also an important factor in speed? What about cache?

>>58302278
I have no idea what they're teaching you.
Clock generators are all on die. The northbridge has been integrated for generations now

>>58302142
>le memes are bad meme
Calm your tits.

>>58294989
There will be 4 core ryzen CPUs for poorfag gamers.

>>58302142
>MEMEs will destroy humanity
>>58302396
>You are a MEME now anon
NOOOO

>>58302142
Memes are a meme.

File: hqdefault (1).jpg (15KB, 480x360px) Image search: [Google]

15KB, 480x360px

>>58302568
memes, memes never change

why do you guys not AMD to succeed, it would be better for everyone.

>>58302601
mainly this
>>58301323

I'd be very happy if they succeed, put some actual pressure on Intel and turned the enthusiast CPU market into a two-horse race. Same with their high end GPUs vs nvidia. They just always seem to fuck it up.

>>58302601
I don't want them to succeed because they support Islam and don't support their own hardware.

>>58302655
>t. Shlomintel Nvidiaburg

>>58302681
t. western hating pajeet rapist

>>58302601
Not for intel

>>58302732
Better for Intel's employees though

>>58302784
Not if they're all layed off...

>>58302913
When a competitor is doing well a company should be less likely to cut employees.

>>58298862
>it's a large chip
Actually it is significantly smaller than Broadwell E per core and of course there will be high clocked cut down 4 cores for the gamer fagots to compete with the 6700k. The full 8 core die is set to go toe to toe with Intel's 2011v3 platform. Ryzen has aggressive clock gating so it might even be able to disable 4 cores and clock the others higher when all 8 cores are not needed.

File: nggghvidia.png (568KB, 750x750px) Image search: [Google]

568KB, 750x750px

>>58294701
GET REKT GREENS <------ lmao that was a dank meme i r89 r9001 390/3.5gb of v for vendetta ram, novidiots be like do a 1080 and walk awayy lmao, 144 wayy sli novideo grillforce gtx 480 clocked at 3.5hz 1080p 3.5 fps gimpworks edition, human eye can see only 3/5 ram anyways, water boiler titanal xp at 3.5p, fryforce shitan rascal 1080shiTi at 87.5% GTX(GetTheXtinguisher) (pisscal xp malware experience thermi burning sillayy bridges edition) + krappy fake $hilltel gore iSemen (cooler not included in a box) end of life edition at 0.90 GHz. 8GiB? What do you need 4GiB for? I'm giving you 3.5GiB! mumble mumble 3.5!?!?!? mumble mumble mumble 390?!?!?!

>>58303032
Intel has been laying off thousands over the past few years but this has more to do with a shrinking market for x86 CPUs in general and other stupid business ventures than anything to do with AMD.

File: 1431038158232.gif (473KB, 297x212px) Image search: [Google]

473KB, 297x212px

>>58303244
You forgot
>woodscrews

File: jews intel.jpg (47KB, 960x674px) Image search: [Google]

47KB, 960x674px

Sorry bluefags but intel is ripping you off.
https://www.youtube.com/watch?v=jy1M0jkRWWk&t=0s

http://www.anandtech.com/show/10968/the-intel-core-i7-7700k-91w-review-the-new-stock-performance-champion

AYYMD IS FINISHED & BANKRUPT!

File: 1436123930726.jpg (16KB, 512x384px) Image search: [Google]

16KB, 512x384px

>>58303667
>want to do a build as soon as the 1080 Ti drops
>settled on 6700K and Asus Z170 deluxe
>7700k turns out to be gud, better than the 6700K
>Shit selection of Z270 motherboards for the forseeable future

fug.

>>58303667
>4.5GHz at 0.8V vcore

not bad

>>58295296
>>Trump wins
In your dreams, dumpchin

File: 85135.png (40KB, 650x450px) Image search: [Google]

40KB, 650x450px

>>58303773
Did you even read the article?

There is about a +/- 1% difference in IPC between Skylake and Kaby Lake.
Kaby Lake uses more power than Skylake, it runs hotter as well, which has been known for weeks since Toms and other places published previews before NDA was up.
An Overclocked Skylake i7 is nearly identical to Kaby Lake.

>>58303773
It's like 0.1% better than a 6700k clock for clock. As long as your 6700k isn't a total dud you'll be fine. Z270 mobos will be overpriced out of the gate, don't forget about that.

>>58303846
What did he mean by this?

>>58303644
>linus
He is the biggest lintel shill there is
>tries to sugarcoat the fact that Intel CPUs are practically stagnating after 2011
I was thinking all CPUs are stagnant do to hitting the limits of silicon like the 4GHz barrier.

Computer progress is practically over.

>>58303883
we will just increase die size

File: 1457890624450.jpg (618KB, 1669x1615px) Image search: [Google]

618KB, 1669x1615px

I wonder how many of you are legitimately retarded thinking this is a bad thing. The 6700k is the king of single core performance, it beats everything on its realm. Zen being that close at lower clocks is nothing short of amazing since it means it doesn't suck in single threaded shit. If you can OC the 8c/16t to 4.0GHz, then you will be golden. I have my shekels waiting, I'd like to finally upgrade from a 3570k.

>>58303936
I wish to see this.
Not happening now.
CPU only meaningless part of the motherboard like build in sound card.
No improvements in power.

>>58295609
Yeah, guess what, they don't work without a CPU.

>>58303847
The gaming results are hilarious. There is less stratification than I even thought. It's insane how viable a 2600k still is in this day and age.

>>58298816
Stop parroting this shit. At least $500 and I'm being nice about it.

>>58303994
CPU advancement are dead.
You need to be a idiot to buy this years CPUs
>>58303990
>Yeah, guess what, they don't work without a CPU.

Yes CPUs are required however there is no point in buying new ones unless you have a relic from the stone ages.
for me the CPu is nothing more then another peace on the motherboard like the sound card and Ethernet card.

I spend my money on getting RAM or GPU, the real improvement is in them.

>>58303976
make 2000 watt power supplies great again

imagine a credit-card sized die with a car radiator attached to it

It would make being a computer enthusiast fun again

File: 1412204503320.jpg (108KB, 736x751px) Image search: [Google]

108KB, 736x751px

>>58304077
A new era of giant everything.

A ripcord power switch.
An airhorn for POST sounds.
A 6.0L diesel PSU.

>>58304124
Trump will make those dreams come true.

>>58303847
Did you read the article? Delta does not equal wattage drawn.

>At 4.8 GHz, the Core i7-7700K passed POV-Ray with ease, however the 1.400 volts needed at that point were pushing the processor up to 95C during OCCT and its mixed AVX workload. At that point I decided to call an end to it, where the CPU was now drawing 122W from idle to load. The fact that it is only 122W is surprisingly low – I would have thought we would be nearing 160W at this point, other i7 overclockable processors at this level in the past.

File: image019.png (59KB, 989x742px) Image search: [Google]

59KB, 989x742px

>>58304169
Aww look at the little spastic with a 5th grade reading comprehension.
The segment you copypasted has nothing to do with what I presented. Taking out of context overclocking results, and making comparison to older 22nm and 32nm parts is pointless.

Stock 14nm Kaby Lake draws more power and runs hotter than stock 14nm Skylake. The increased clocks with Kaby directly correlate to its increased power consumption, virtually identical to overclocked Skylake.
The only difference here is that the average Kaby Lake chip appears to OC 100mhz to 200mhz higher max than Skylake.

There are no IPC uplifts. There is no reason for any desktop user to purchase Kaby Lake over Sky Lake.
At stock clocks it draws more power and runs hotter.

>>58294701
isn't gaming fucked for multy core utilities ?
Will the Ryzen be as expensive as those i7 or as a i5 ?
The power consomation (i can't speak english) is massive, this is actualy the only thing I notice and I take in consideration right now.

Ofc i only buy middle price stuff.

>>58301600
>If the 8c/16t Ryzen chip can reliably hit over 4GHz, you won't see it for less than $900
>source: my ass

>>58304301
lol whatever fag

>>58304342
Plenty of games utilize 6 or more cores. Haswell-E tops benches in a ton of newer titles because of it, and newer Broadwell-E is now taking the crown.

Price of the Ryzen CPUs hasn't been disclosed yet, but people are expecting them to sell for $500 and lower.

Ryzen tops out at a 95w TDP. Its pretty low considering its 8 cores and 16 threads at moderately high clocks.

>>58304431
>Plenty of games utilize 6 or more cores

No they don't

File: w3_intel.jpg (209KB, 545x852px) Image search: [Google]

209KB, 545x852px

>>58304447
Yes, yes they do.

>>58298595
Actualy CanardPC is probably the best French Language Hardware Magazine (because it's more a magazine than a Website) And it's a bit focus on gayming (cause the main CanardPC is for gayming)

File: 81881.png (40KB, 650x450px) Image search: [Google]

40KB, 650x450px

I've been saying this from day 1 , extra cores don't mean shit in gaming and everyday computing.
Even intel's high end CPUs with many cores don't do well over their quad core counterparts in things like gaming.

Why do you think they showcased a rendering demo ? Because that is the best use of all those cores.

>>58304465
Spreading the load across more cores doesn't automatically mean more performance.
pic related

File: w3_proz.jpg (101KB, 549x440px) Image search: [Google]

101KB, 549x440px

>>58304619
>Spreading the load across more cores doesn't automatically mean more performance.
Sure thats true, but its not in this case.

http://arstechnica.com/gadgets/2017/01/intel-core-i7-7700k-kaby-lake-review/

>The Intel Core i7-7700K is what happens when a chip company stops trying.

kek

>>58304653
And its the same story in Hitman, Civ 6, Project Cars, Dirt Rally, Quantum Break, Watchdogs 2, COD IW, Battlefield 1, Gears of War 4, and dozens of other new titles.
All of them favor the more threads provided by Haswell-E or Broadwell-E CPU vs the highly clocked mainstream quad cores.

I made a purely factual statement. Plenty of games utilize 6 or more cores.

>>58304619
Absolute truth.
However if these cores are used to be dedicated to other things like
Core 1 = OS
Core 2 = game
Core 3 = web browser or some shit
you can get significant improvements.
Also multicore is not great, there are tasks that will never benefit from multiple cores.

File: wd2_proz_2.png (92KB, 556x440px) Image search: [Google]

92KB, 556x440px

>>58304653

>meanwhile, in late 2016

>>58304925
>all those i5

How petty do you have to be to save $100 bucks on a once per 3-5 years purchase?

>>58301677
>>Well we know very little about it.
>The core arch was detailed at length during HotChips last year.

We've seen decent detail about the core itself, but horribly little about the L3 (+crossbar?) design and essentially nothing about the IO, inter-cluster or otherwise.

Naples, and likely Zen as a whole, is gonna live or die by how well the Infinity Fabric works.

>>58304671
>Useful for storage fans, however, is the addition of four extra PCIe lanes to the top-end Z270 and Q270 chipsets, for a total of 24. With high performance storage having moved to the PCIe interface over SATA (see the NVMe-based Samsung 950 Pro), PCIe lanes are quickly becoming scarce. Combined with the 16 PCIe lanes on the processor, a Kaby Lake system now sports a total of 40 PCIe lanes.

wait what

Adding four extra lanes to 16 gives 40? I don't understand PCI-E lanes very well.

>>58304964
Allow me to clear this up for you:
16 + 24 = 40

You're welcome.

>>58304964
why do you need all those lanes on consumer chip??

File: tww_pr.jpg (100KB, 523x440px) Image search: [Google]

100KB, 523x440px

>>58304958

/g/ can't see the forest for trees which is why it eternally thinks i5's are enough for any and all vidya, that zen won't compete agaisnt broadwell-e anda host of other things that are not true.

>>58304993
Two GPUs in PCI-Ex16?

but will it be a botnet?

>>58305007
You have to be very stupid or very rich to buy two GPUs these days. Support never was good, now it's terrible.

>>58294701
>muh vdyagames


KYS

>high-performing 4-core beats 8-core in applications that dont scale well to multiple cores

WOW!

will AMD ever design a processor with decent IPC or will they always keep putting 10+ cores into their garbage as if that counts for anything?

File: IMG_0400.jpg (41KB, 533x492px) Image search: [Google]

41KB, 533x492px

>In a gaming head-to-head, the AMD chip stumbled a bit, largely due to the fact that the vast majority of games don’t fully take advantage of the huge number of cores it boasts.

LITERALLY THE SAME SHIT AS BULLDOZER

AMDRONES SAID IT WOULD BE DIFFERENT

>>58301323

all trolling and shit aside. thats fucking true.

>>58301330

thats why i said IF it clocks. but i assume that a 5ghz on one core will equal to 4.5.ghz on four cores.

and hopefully the magic 4.2 on all cores

>>58305109
How does someone as dumb as you exist?

>>58304904
>dedicating cores to tasks

shit solution

File: 14512753255427.png (51KB, 414x248px) Image search: [Google]

51KB, 414x248px

>>58294701
You do realize the same thing happens with the i7 6900k right? There will be higher clocked 4 core variants

>>58305138
Are you retarded? It has better IPC than Broadwell. More cores = lower clocks. If you want higher clocks than get the cut down quad core.

Literally the same thing happens when you compare a 6700k and a 6900k.

>>58304996
Holy shit, a game that's actually hyper threaded? thats some kind of fluke.

>>58304996
Why aren't there skylake cpus in that benchmark?

>>58305356

>a game that's actually hyper threaded?

On a technical level this is complete gibberish.

>>58305356
Not even remotely.
See: >>58304750

Its been like this for years.

>>58305385

Go ask ivan.

File: 1461894534738.jpg (65KB, 525x257px) Image search: [Google]

65KB, 525x257px

>>58303936
1. power consumption is proportional to area
2. ipc increases as the square root of width
3. width increases are exponentially

special rule for x86/64
decoding more than 4-6 instructions per cycle will increase power consumption so much that you die in a fire

In other words the IPC train is over. It's MORE CORES for now on.

>>58305451

POWER9 when?

I know everyone here likes to jerk off to desktop performance, but without a mobile chip AMD might as well be dead.

AMD desperately need a 5-15 watt Core M-series equivalent, and a 15-25w Core U-series equivalent

File: macfag.png (5KB, 217x233px) Image search: [Google]

5KB, 217x233px

>>58305451
And then Intel has another breakthrough, moore's law chugs along, and AMD spends another 10 years talking about paralyzing innovation. See what I did there? I thought it was clever :3

>>58305473
Your assessment of the market is piss poor. AMD doesn't need any presence in the mobile market to remain profitable.
Despite the humor in your obvious ass pulled statement, AMD does have an upcoming APU that will scale from 4 to 35w in the mobile segment.

>>58305473
Zen scales from low to high. There's Zen chips planned for every segment of the market.

File: 1456645624578.jpg (23KB, 381x250px) Image search: [Google]

23KB, 381x250px

>>58305485
moore's law applies to transistor density not speed. Those extra transistors are better spent on more cores. Single core performance increases are dead. Even with Amdahl's law multithreading is still the way to go for now on.

>>58305554
They aren't dead though. Intel and Qualcomm make gains, why can't AMD?

>>58305473
Zen APUs are in the works

There are counterparts for Y, U, and H families I think

>>58305473
8 core Zen @ 3.4ghz is only 95watt. It will scale excellently to laptops and AMD has a MAJOR advantage in integrated graphics.

>>58305590
>intel
>serial performance gains
HA!

http://images.anandtech.com/doci/10944/dxva-checker.png

AYYMD HAS NO 8K HEVC MAIN10 HARDWARE DECODING & NO 8K VP9 PROFILE2 HARDWARE DECODING

>>58305590
>Intel make gains

Nothing significant. All they are doing is fine tuning their core, and eliminating bottlenecks. The Core architecture has reached a dead end. Intel is going MORE CORES with Coffelake.

>>58305451
great, so the next major step in cpu performance relies on a completely new assembler compiler and a new cpu building material?

can't even imagine the throwback in everything replacing x86

>>58305646
>8k media
>mattering
>when 4k is barely used anywhere

Chasing horribly tertiary features like this is actually a very bad sign. It shows a company is stuck in a run, they go for less meaningful small bullet points instead of more important performance uplifts.

And what have we been seeing lately:
Haswell to Broadwell 3% uplift
Broadwell to Skylake 3% uplift
Skylake to Kaby Lake 1%~ uplift
Kaby Lake to Coffee Lake? Still 14nm.

This is not good for the industry.

>>58305646
>8k
>relivent

4k doesn't even have any content yet. What makes you think 8k will within the next 10 years?

>>58305665
>Nothing significant.
So you admit they are still getting faster.

>>58305707
Coffe Lake is literally just MORE CORE Kaby Lake.

>>58305761
Not making the core bigger they are not. It's a dead end. There are no more gains to be made.

>>58305761

The gains anon quoted architecture to architecture often fall to within margin of error.

>>58305707
Dude what?
My i5-3210M was made in 2012 and I can't decode a lot of 4k HEVC content.
If I had known that fucking 50" 4K tvs were going to cost $350 bucks in 2016 you can be damn sure I would have wanted that capability.

File: 1483157195248s.jpg (2KB, 124x120px) Image search: [Google]

2KB, 124x120px

>mfw we are entering the age of diminishing returns negatively affecting the consumer market as we approach the theoretical physical limits of silicone yet people are still hoping for 50% performance improvement in real world applications

lol

I have a 2500k at 4.4 ghz and a 290x with 16gb ram, and it runs everything I play no problems.

Why would I even upgrade? I'll wait to see what the next generation offer

File: 1481678062695.jpg (283KB, 638x800px) Image search: [Google]

283KB, 638x800px

Why do Intel shills look to the gaming benchmark for the zen 8 core? Do they forget that there will also be higher clocked (ie better single/4 threaded performance) 4 and 6 cores meant for gaming? 8 core zen is primarily a workstation CPU, as is the 6900k which also did poorly in the gaming bench

>>58305798
Barely any 4k content exists. The 4K Blu-ray market is just now picking up, and it is still a ridiculously small niche.
What are you going to do, watch 4K netflix? Its low bitrate upscales. A pointless gimmick.

Put your retarded fanboyism aside, triptard.

File: 1483157167666.png (308KB, 492x419px) Image search: [Google]

308KB, 492x419px

>>58294701
Not this thread again, fucking kill yourself retard.

>>58303975
COPE
O
P
E

File: 1455927433863.jpg (64KB, 455x628px) Image search: [Google]

64KB, 455x628px

>>58305786
Haswell to Broadwell 1%
Broadwell to Sky lake 2%
Sky lake to Kaby lake 0%

All other gains were due to increased memory bandwidth, better multi threaded perf, or hardware acceleration.

>>58305840
Gonna watch fucking Xmen Apocalypse on a 50" 4K tv obviously.

I'm just glad its a smart tv or I wouldn't be able to.

File: 1480932135220.gif (107KB, 640x480px) Image search: [Google]

107KB, 640x480px

guess the onus now is on software producers to make their programs more amenable to using MORE CORES since we can't have MORE SPEED.

>Tick-tock
>Process-Architecture-Optimization
>????
>Optimization-Optimization-Optimization (read MOAR COERZ)

>>58302243
How does this guy drink anything? lol

>>58305890
He drinks cum when some dude throat fucks him from the side of his mouth to get further down his throat.

>>58294921
The chink CEO said ryzen was perfect for gaming when they unveiled those benchmarks last year.

File: 1463765651929.gif (312KB, 342x512px) Image search: [Google]

312KB, 342x512px

>>58305889
Humans have trouble coding in parallel. It will be up to our AI machine learning overlords to lead us to our parallel processing future.

So if I'm reading all this correctly, AMD is releasing a $500+ 8c/16t CPU to compete with Broadwell-E and the X99 chipset, but it can get shit on by my 299 5820k? Jesus AMD what happened? Once upon a time (K2 to intel Core) AMD was the obvious choice vs the Pentium 4 at nearly every price point under 500 and now they are relegated to the budget market entirely. I really wanted to see Ryzen do well as it would have meant a new HTPC in the living room running their 4c/8t 6700k competitor, but as of now, it seems like Id be well served getting a Skylake chip after Kaby lake drops and a Z170 mobo.

>>58305812
>290x
Well bought my friend. It pains me to look how much my friends gtx 780 fell behind, while getting much better results at the release. The same way as me looking at my 760 compared to the 280X.

>>58305918
It will be as games are becoming more parallel. Once DX12/Vulkan adaptation is in full swing cores will be king.

>>58305984
Same was said about the FX8350, also multithreading has been a thing since forever.

>>58305984
Surely graphics represents the bulk of parallel processing in vidyajaems? CPUs haven't been the bottleneck for fucking forever. There's barely any vidya benefit to be had with a 6700k over a fucking 2500K.

>>58304993
PCIe SSD, M.2 SSD, NIC.. the list goes on. Standard PCI is nearly dead these days as it sits. Soon, it will go the way of AGP.

File: qSNrpeA.png (39KB, 648x808px) Image search: [Google]

39KB, 648x808px

>>58305997
And it was true. Just most decs were to lazy to do it. DX12 and Vulkan make multithreading easier.

>>58305941
t. brainlet

>>58305984
>this whole thread

I love this /v/kid technobabble. It's like they have absolutely no fucking idea what they are talking about but spout things like "parallel computing", "multithreading", "DX12/Vulkan" and other shit they know absolutely nothing about.

>>58306081
Literally one game.

File: 1726450i03g9jmahjj0mjj.png (63KB, 421x248px) Image search: [Google]

63KB, 421x248px

>>58306097
You're just mad because you will be out of job soon :^)

>>58306110
That isn't even a game its a cryengine cpu benchmark that doesn't use DX12 or Vulkan.

>>58306110
Didn't i already say devs were to lazy to do it? Now the devs have no choice. There will be no more significant IPC or clock increases.

>>58294701
It's still that Canard PC leak, nothing new. When will you stop that clickbait shit.

When can we expect to see Zen CPUs on the market?

http://www.intel.com/content/www/us/en/nuc/nuc-kit-nuc7i7bnh.html

If Ryzen was any good, why AMD can't include GPU within 8 core CPU SKUs and put it in a NUC form factor?

CHECKMATE, POOJEETS

>>58306244
>comparing an 8 core CPU to a 2 core
AMD has Raven Ridge, a 4 core 8 thread APU coming out. With an IGP that utterly decimates intel's top of the line Iris Pro.

>>58296407
kind of sad because that's an autistic kid who AMD tried to hook up with a better rig, but his bigger brother took all the components and got tons of shit for it on reddit

>>58306311
>utterly decimates intel's top of the line Iris Pro
[citation needed]

>>58296530

>Not sending new monitors and GPUs to the whole family.
What kind of a poorfag company acts like this?
It was a bad move on their side.
>Yeaah we can send.. one..that much we can afford.

>>58306360
Iris Pro in Broadwell desktop parts is only neck and neck with Carrizo/Bristol Ridge, and Broadwell has a tremendous CPU perf advantage on its side.
Thats 8 heavily bottlenecked GCN CU, and Raven Ridge is 12CU with significantly faster DDR4, and new arch.

Intel will totally lose out in IGP performance just like they did for most of the last 6 years.

If we're reaching the limits of silicon they'd better hurry up and find out whether InGaAs or graphene are going to be any better.

Still on a 3570K, why do I care if Zen is DOA?

File: face.gif (685KB, 500x500px) Image search: [Google]

685KB, 500x500px

RYZEN STARTING AT 500

DOESN'T BEAT INTEL

LMAOOOO

IT'S OVER FOR AMDORKS

http://en.yibada.com/articles/182844/20170102/amd-ryzen-summit-ridge-cpu-confirmed-ready-to-overclock-at-5ghz-on-single-core-499-intro-price-to-topple-intel.htm

File: IMG_0502.jpg (132KB, 600x697px) Image search: [Google]

132KB, 600x697px

>AMD just barely caught up to Intel as they're jumping ship from silicone and x86

>AMDs solution to this dying tech is to ADD MOAR CORES

You cant make this shit up

>>58306578
AND THE INTRO PRICE POINT IS 500


LEEEEEL

JUST FUCKING LOL

>>58306578
Moar cores are good for encoding and rendering.

File: 1483347482755.gif (18KB, 252x281px) Image search: [Google]

18KB, 252x281px

>>58306577
>using the meme with the nose

>>58306578
Intel has a 10 core HEDT chip.
AMD has an 8 core HEDT chip.

AMD is also releasing a 4 core mainstream desktop chip with moderately powerful IGP.
Intel's response? They're releasing a 6 core mainstream chip with their top end IGP.
This is Coffee Lake coming in 2018, its still 14nm, the sake as Kaby Lake, Skylake, and Broadwell. Then they still have to release desktop Cannonlake in 2019, a 10nm part, very much still bulk silicon.
They have 3 more generations of arch after Cannonlake already planned. At least two of them so far are just refreshes, still 10nm tweaks of Cannonlake.

Intel isn't getting rid of x86 any time soon, and they will not be moving beyond 10nm bulk silicon any time soon.
intel is the moar coarz company now, not AMD.

When intel finally releases 10nm Cannonlake desktop chips they will be facing off with 7nm SOI FinFET Zen+ parts from AMD.

File: IMG_0516.jpg (82KB, 440x406px) Image search: [Google]

82KB, 440x406px

>>58306623
Yeah but theyre more slow and shitty cores while Intel is making CPUs from new materials and architectures that will be a huge leap in performance

Almost as if this is just Bulldozer all over again...huh...

>>58306578
>silicone
nice one you fucking idiot

>>58306679
You mean CPUs arent made out of fake tits?

File: 1475853111506.gif (564KB, 800x430px) Image search: [Google]

564KB, 800x430px

IT'S BULLDOZER ALL OVER AGAIN

FUCKING LMAO

IT'S FINISHED

AMD IS FINISHED

POORFAGS DESTROYED

NERDS BLOWN THE FUCK OUT

BULLDOZER 2.0 BTFO BY 4 CORE CPU

BULLDOZER 2.0 INTRO PRICE POINT OF 500

HAHAHAHAHA

PACK IT UP AND GO HOME BOYS

>>58306667
>When intel finally releases 10nm Cannonlake desktop chips they will be facing off with 7nm SOI FinFET Zen+ parts from AMD.

>tfw it's actually what will happen

Intel should've dropped their fabs long ago, it only soaks up R&D budget without any true returns like TSMC or Samsung does because they can diversify from mobile to huge 600mm2 chips Intel fabs stuck with intel CPUs.

>>58306724
Their Fabs put them at an advantage

The Altera buy makes ABSOLUTELY zero sense without fabs

>>58306671
For parallel tasks, they can be shitty, as long as there's a lot of them. Although TDP also counts.

File: 1323578059001.png (29KB, 640x753px) Image search: [Google]

29KB, 640x753px

>>58305984
>i-i-in 10 years whe-when they program games specifically for my third tier toilet AMD cpu t-they will see its true power...

>>58306743
Unless world goes to hell next 15 years -
design>fabs

fabbing is dirty work, like steel production
steel producers do not make ships

File: 74939.png (32KB, 650x337px) Image search: [Google]

32KB, 650x337px

>>58306475
AMD's superior GPUs are completely bandwidth starved, and if they rely on "significantly faster" DDR4 they will be once again btfo by Intel's eDRAM. It doesn't matter how many CU's they have in Raven Ridge if there's not enough memory bandwidth. I'm glad the CPU performance will be similar this time around though. Iris Pro is also not "neck to neck" with Carrizo/Bristol Ridge.

>>58306846
>steel producers do not make ships
Conglomerates exist that do exact this tho

File: 1480934105428.jpg (41KB, 344x254px) Image search: [Google]

41KB, 344x254px

>>58306577
>>58306578
>$499
>$499
>$499 USD UNITED STATES DOLLARS

AMD, you had one job.

>>58296530
Wow that dude is a dick bag...

File: adln6eXy[1].gif (1MB, 320x173px) Image search: [Google]

1MB, 320x173px

>>58306873
>mfw amd is using dual channels for ryzen and it will bottleneck their already trash cpu

>>58306724
Intel made one really good move which was jumping on FinFETs before everyone else with their 22nm Trigate process.
Then they made a horrendous error with their 14nm Trigate process. They tried to make something significantly more dense than the transitional FinFET nodes from competing foundries, and they paid for it. It took them years to get decent yields with high power chips, and now they've had to carry it forward for another year still to recoup the losses they've incurred. It will be the exact same case with 10nm.
The problem? They're still using bulk silicon. Adding more mask layers, increasing complexity, requiring more doping, increasing the negative effects of variation within the wafer. All the marketing spin in the world from intel can't hide this.

Tick-tock died because intel isn't sure on where to go from their current bulk FinFETs. They should have been investing in this 5 years ago. They should already have a viable option on the table ready to bring online and ramp up to risk production. They have 4 years of 14nm and a probable 3-4 years of 10nm coming up instead.

>>58306873
Gaming benches are not isolated IGP comparisons. You're seeing similar IGP performance coupled with a massively superior CPU in Broadwell.
intel does not have good IGP arch.

>tfw no on die HBM

>>58305828
That Shark thing is actually true.

>>58306972
it's not, couple years back a few tourists and a grandma were eaten by sharks

>>58298611
Under-rated.

>>58307020
Yes it is. That is one case. Bears don't eat people either.

File: YXATYnc[1].gif (3MB, 200x200px) Image search: [Google]

3MB, 200x200px

>AM4 only supports dual channel memory
>AM4 only supports 24 pcie3 lanes

this blunder is even bigger than bulldozer lmaoooo

keep picking your poorfag amd cpus out of the trash when everyone that has BANK gets superior intel engineered technology on the cutting edge

>>58307046
>Sharks don't eat people
Here are some cases where sharks have eaten people.
>SHARKS. DON'T. EAT. PEOPLE.

I guess planes never crash either.

The predictions were true. People keep moving the benchmark of what the new chips should be compared to. Ryzen was supposed to compete with Haswell-E and now people are expecting Kaby Lake performance.

The exact same thing happened with the RX480, people moved the benchmark from the GTX980 to the GTX980Ti. The cards sold well though because of the price point. I expect these Ryzen chips to do the same thing and be priced reasonably.

>>58306577
>>58306903
>Edition will hit the market with a cash damage of only $499 (or $349 for the regular SR7 variant).

$499 for the special binned "lots of GHz" 8 core
$349 for the normal 8 core
$250 for the 6 core
$150 for the 4 core

All with enabled multithreading because AMD doesn't disable features for no reason like Intel.

>>58307060
Seriously AMD, this and $499.

Intel is looking weaker than it has in decades and AMD continue to fuck it up for themselves. One killer chip this gen could force some actual competition.

>>58307075
It has happen, yes. THEY DON'T GENERALLY EAT PEOPLE. Sharks could have been very hungry.

>>58307105
>WCCFTech said the Ryzen SR7 Black Edition will hit the market with a cash damage of only $499 (or $349 for the regular SR7 variant).

>>58306578
Stop using my waifu for shitposting.

>>58307110
fun fact about sharks, since it's more interesting: great white sharks live more than 200 years

File: 1483159487089.gif (4MB, 307x325px) Image search: [Google]

4MB, 307x325px

>>58306903
>>58307105
amd on ... wait what, 349$ for ii7 6900k performance?

>>58307060
Dual channel makes CPU, mainboard and memory cheaper. Quad channel has little benefit.
24 PCIE3 lanes is more than enough for anyone. You don't need more than 8 lanes per card for SLI/Crossfire.

File: amd_32_core_monster_concept.jpg (63KB, 550x502px) Image search: [Google]

63KB, 550x502px

>>58306932
Soon™
It's only a matter of time.

File: Mkay.png (99KB, 632x371px) Image search: [Google]

99KB, 632x371px

>>58307147
Sure thing bud.

Greenland sharks are the ones that can live a very long time, up to 400 years.

https://en.wikipedia.org/wiki/Greenland_shark

http://www.sciencemag.org/news/2016/08/greenland-shark-may-live-400-years-smashing-longevity-record

>>58307171
>You don't need more than 8 lanes per card for SLI/Crossfire.
more precisely you can't use more than 8

intel shills working overtime

I can literally feel the panic

>>58307192
I'm pretty sure I've read a paper on the subject 3-4 years back where they found 180yo great white shark

File: a5Vogp7[1].gif (3MB, 375x160px) Image search: [Google]

3MB, 375x160px

>dual channel memory
>24 pcie lanes (lga2011 supports like 40)
>less ipc than intel
>500 dollars for worse performance than cpus cheaper than it

I'M DYING

AMD LIED AGAIN

AMD POORFAGS ON SUICIDE WATCH

LOL JUST LOL

>z-zen is gonna ki-kick intel's a-ass, yeah... just look at these fudged numbers from AMD... jim keller will pull through, you'll see!

YOU KEKOLDS GOT BLOWN THE FUCK OUT AGAIN

keep coping you poor impecunious nerds- maybe you'll be able to pick an old 10 year old celeron dualcore out of the dumpster at your local goodwill- I mean it would probably still beat AMD in videogames and everything that matters

going to enjoy this salt with my 6800k and gtx1080, going to be so salty, like anchovies

>>58307232
Haven't seen anything saying that.

>>58307060
There is literally nothing wrong with dual channel memory.
The Zeppelin die has 16 PCI-E lanes, same as intel's current desktop offerings, plus additional lanes provided by the chipset. Enthusiast chipsets aren't limited to a meager 8 additional lanes.
The X370 chipsets on launch support at least two PCI-E 3.0 X16 slots.

>>58307246
Damn, anon. Here's a (You). I don't want to shitpost with (You), maybe someone else will will follow.

>>58307247
https://www.sciencedaily.com/releases/2014/01/140109004145.htm

>>58307246
oh no sucks for the 0.1% of people who run quad sli setups

>>58307314
Interesting, doesn't mention the 200 years though.

>>58307249
>>58307297
>>58307328
keep coping you poor, destitute, virginal dork nerds

keep defending the honor of amd

keep cucking for zen that's more expensive for worse performance

I'll be sitting back here with my 6800k and NVIDIA gtx 1080, gaming in 4k, on windows® 10 ultimate™ fucking my hot bitch, while you NERDS are writing your l00nix shell scripts in a basement

>>58307349
there was another one I can't find, maybe it got refuted

>>58307381
Maybe

>>58307373
This is some pretty weak bait. Like you're not even trying. Apply yourself.

File: 4-hour-boner.jpg (23KB, 480x480px) Image search: [Google]

23KB, 480x480px

>>58307105
>those prices

>>58307105
As much as I want to believe, there is no way it's going to be like this.
$200 for the 4 core
$300 for the 6 core
and $500 for the 8 core sounds more like it.

>>58307405
>300 replies
>weak bait

ok

>>58307392
Anon who initially posted the shark picture here, did not realize I opened a can of worms

>>58307537
>can of worms

>>58307373
hi this is Brian Krzanich, CEO of Intel

I would like to thank you for your services

give me your email and I will mail you a brand new Kaby Lake chip, on the house ;)

File: 1481751560559.jpg (187KB, 1280x1080px) Image search: [Google]

187KB, 1280x1080px

>>58306577
>SR3
>SR5
>SR7

>>58307567
Is something wrong?

>>58307618
I don't know what you're talking about

>>58306578
That's a cute picture. I'm gonna steal your waifu onto my SSD.

>>58305812
Literally a lying shill. 3570k here. Bottleneck on so many AAA games. I love how poorfags on /g/, especially AMD drones always use this "it runs everything I want with no problems" excuse which literally means minecraft and hearthstone. KYS. KYS U FILTHY FUCKING WHORE.

POST YOUR BATTLE THEME BOYS

https://www.youtube.com/watch?v=0W7vUwkgp5A

>>58309285
https://www.youtube.com/watch?v=NUC2EQvdzmY

>>58309285
>>58309308
wrong thread. I'm retarded, this was meant for the winnie the pooh thread on /v/

>>58299640
You're an idiot. Keep watching LTT.