AMD 1950X OCed To 4.1GHz On All 16 Cores - Delivers Otherworldly

Zen2 chips with higher base clocks in the range of 4ghz~ are going to really rock intel across the board.

File: sunshine-movie_00388545.jpg (819KB, 1920x1080px) Image search: [Google]

819KB, 1920x1080px

Just you wait for the OCed 18 core intel flagship

>>61811027
>mad incel
Even a moderate IPC uplift combined with higher baseclocks is going to eat up intel's marginal serial performance lead. Intel is losing ground and management is crumbling internally.

>>61810976
can't wait to see it in laptops

Voltage!!

File: 1497565895125.jpg (112KB, 691x771px) Image search: [Google]

112KB, 691x771px

>>61810976
NO
BUY INTEL
PLS

For gayming coffee lake or wait for zen2? At least coffee won’t be mesh based

File: 1475211353488.jpg (108KB, 782x782px) Image search: [Google]

108KB, 782x782px

>>61810976
OY VEYYY

DELETE THESE BENCHMARKS NOW!

SHUT IT DOWN!

Nvidia + Ryzen master race?

>>61811383
Just get something now. Games are just as GPU limited as they were 2 years ago.

>>61810976
3300 points with 16c @ 4.1GHz? That's kinda shit IPC. My 18 core haslel @ 2.8GHz does 2600 points.

>That alienware box scores around 2850~ with shit ram
>this result is 3300 with fast ram

Looks pretty legit to me.

>>61811570

>haswell now has higher IPC than broadwell and skylake

File: 1474524874553.png (207KB, 640x400px) Image search: [Google]

207KB, 640x400px

Does this look like a face of mercy?

>>61811383
Zen2 is first months 2019, there will be a Zen refresh with higher clocks before that.
Btw, zen with the latest updates is basically on par with Intel even in games.

>>61811625
I want to cum on her pretty little face!

File: 1502263097811.jpg (17KB, 286x380px) Image search: [Google]

17KB, 286x380px

>>61811696
please don't

So in real numbrrs 50% worse than Intel

>>61810976

>otherwordly

first time I've heard this word holy shit Intel is getting curbstomped

>>61810976
>1.248V
What?

>>61812112
What?

File: fmax-vmin.png (55KB, 1509x905px) Image search: [Google]

55KB, 1509x905px

>>61812146
Thats a low vcore for those clocks compared to what we saw with Ryzen at launch.
If its real it shows either crazy hand picked dies in Thread Ripper, or an substantial refinement of the 14nm LPP process.

>>61812219
Or CPU-Z shitting the bed again

>>61812245
Could be, but its not displaying some out of left field value. Its just a ball hair below 1.25v. Thats just a .15v reduction for 4ghz from what the average Ryzen chip needs.

>>61811949
>first time I've heard this word
becaue you're an illiterate amdpoojet

>>61812270
This is some grand irony.

>>61811383
1700oc seems like the sweet spot but if u want to wait Zen 7nm next year will be amazing

>>61811487
i want to pat Amada-chan on the HEAD

File: amd-starship-processor-feature.jpg (788KB, 1920x1080px) Image search: [Google]

788KB, 1920x1080px

>>61812321
>Zen 7nm next year will be amazing
I'm excited.

>>61812429
Hardware is genuinely exciting again

>>61810976
I'm waiting till coffee lake to make my decision desu

>>61812447
You forgot to say no homo.

>>61810976
I thought ryzen didn't support quad channel memory?

File: 1211068199062.jpg (406KB, 2288x849px) Image search: [Google]

406KB, 2288x849px

>>61812429

>AMD's next flagship server chip, Divine Retribution

>>61812552

Ryzen doesn't.

>>61812557
I was under the impression that threadripper was still ryzen.

>>61812565
>wot

How did you think that worked out, anon?
Threadripper is a repurposed MCM enterprise part. It isn't a single die chip like Ryzen.

>>61812565

You are incorrect. Ryzen, Threadripper and Epyc all come from the zen architcture made up of various amounts of zepplin dies but they are all seperate entities.

>>61812576
Well for one thing, ryzen is still in the CPUs name.

>>61812591
Ryzen is just the marketing name of the product family. Like Vishera for the FX 8350. Or Zambezi for the FX 8150.

>>61812410
>with muh dick :^)

Spread over 4 ccx, real perf number comes from dividing this score by 4, that's as far as AMD goes without bottlwnecks and latency problems!

>>61812867
Brian pls

>>61812867
t. Brian "JUST fuck my fabs up" JUSTnich

>>61812867
Actually, the reason why AMD did split like this is that chip price don't scale linearly with the size as one may guess due yield failures with larger dies.
We can't shrink the transistors as easily anymore, so you either use multiple smaller dies to make a larger chip, OR you can try your luck with mammoth low yield monstrosities.

>>61812995
Intel doesn't have this problem.

>>61813018
Intel is making graphene chips?

>>61813018
That's why competing Xeons cost 3 times as much, have less memory and I/O and use more power for equal performance.

>>61813018
Oh, they do.
That's why cheap and scalable Zen is the bane of their existence right now.

>>61810976
look at how much of a fag this guy is. let's laugh at him and his gay tryhard desktop
haha fag

File: 1498772642789.png (562KB, 443x615px) Image search: [Google]

562KB, 443x615px

Should I wait for ripper to come down in price or just buy ryzen now for muh blender rendering?


Got an old and quad core

>>61813392
You won't regret ryzen. It'll feel like you're working with a supercomputer by comparison. You'll need an graphics card tho. But any $100 job will be plenty.

>>61811383
wait for i3-8350k

>>61813469
Wouldn't that just explode from thermal expansion on boot?

>>61813489
no

File: Intel engineers working on Kaby Lake-X.jpg (69KB, 433x640px) Image search: [Google]

69KB, 433x640px

>>61811040
>OCed 18 core intel flagship

Overclocked 7980XE will blow this babycore out of the market

>>61814315
You got the blow part right.

>>61814315
>overclocked 7980XE
i wonder if it breaks the 1000W barrier

>>61814359
9001W here we go

File: Intel-Core-i9-7900X-Power-Consumption_12[1].png (18KB, 599x657px) Image search: [Google]

18KB, 599x657px

>>61814378
oh god with ln2 1000W could happen

>>61812607
You are wrong, ryzen's vishera is summit ridge.

>>61814634
That much voltage for 100mhz, I wouldn't even OC a damn 16 core.

>>61814684
>1.408v
>High
Nigga you some kind of inteltard?

>>61814716
It's high when 4.0G needs 1.25V

>>61814733
That's just how Ryzens OC a the upper limit fampai, my 1700X does 3.9 at 1.33 and 4.0 at 1.43.

>>61814766
Too high a jump for thaz frequency.

>>61811041
Would love to hear some stories of the shitstorm at Intel HQ right now

>>61812447
My software became hardware

>>61812867
lol faggot. http://www.anandtech.com/show/11544/intel-skylake-ep-vs-amd-epyc-7000-cpu-battle-of-the-decade

>>61813469
>Wait for CPU we know nothing about outside of rumors

Ok.

File: watercooled i9-7980XE.jpg (104KB, 1280x720px) Image search: [Google]

104KB, 1280x720px

>>61814315
Kill yourself BR glueeater.

When is 7nm 5GHz ryzen 2 released?

>>61816768
Q1 2019.

>>61814634
That's a 54 watt increase for a 3% boost to performance, completely not worth it.
>180w * ( 4090 / 3990 ) * sqr( 1.408 / 1.248 )

16-core LN2 results would be nice though, everyone loves bragging about a big dick

I'm literally crying right now. This is the only nice thing we had in a decade.

>>61815120
le upvote for innuendo xD *tips fedora*

>>61811041
>management is crumbling internally.

do you know this or are you just inferring?

>>61817133
Considering they were pretty much verbatim shitposting on their marketing slides it's very likely they're having some coordination issues.

>>61810976
>The only issue are the gaymers
Use Process Lasso to pin the game to one die, and it'll perform basically the same as an 1800X. No cross-die latency to worry about.

>>61817133
>They have internal sources at Intel and the climate there is currently awful apparently, employees are very discouraged and some feel like "they have nothing left to lose". They are making big profits now by totally screwing up the future. According to the magazine, Intel is "probably in the most delicate situation it has had to face to this day". The CEO is reducing cost so much that he's managing engineers like "supermarket cashiers", he doesn't care about taking the time to train them.

>Krzanich is very impatient and eager and keeps changing his mind about projects. If a new architecture isn't created in like a couple weeks, he gives up and cancels the project... he keeps sending contradictory instructions to the teams.

>"Fab Hell": Intel is likely going to have a 6 month delay on 10nm. Worse, even Cannon Lake is not expected to feature any significant architectural improvement. Basically Intel was just hoping AMD would keep not competing with them.

>In the picture before the last picture I posted, it is said that Kaby Lake has exactly the same IPC as Skylake. No improvement as to perf/watt but there seems to be more room for overclocking.

http://imgur.com/a/qo9pH

CannardPC, December 2016

>>61817208
>Krzanich is very impatient and eager and keeps changing his mind about projects. If a new architecture isn't created in like a couple weeks, he gives up and cancels the project... he keeps sending contradictory instructions to the teams.
>They have internal sources at Intel and the climate there is currently awful apparently, employees are very discouraged and some feel like "they have nothing left to lose". They are making big profits now by totally screwing up the future. According to the magazine, Intel is "probably in the most delicate situation it has had to face to this day". The CEO is reducing cost so much that he's managing engineers like "supermarket cashiers", he doesn't care about taking the time to train them.
HAHAHHAHAHAHHAHAHAHAHAHA
THIS IS ACTUALLY WORSE THAN RUINZ
The dude and his gang of beancounters are literally KILLING Intel right now.

File: emoji pillow.jpg (23KB, 640x559px) Image search: [Google]

23KB, 640x559px

>>61817208

In-kernel temperature driver when?
Fix for the segfaults when?
B2 stepping retailing for AM4 chips when for sure?
Decent mobo's with good VRM, 2 x PS/2, pcie_1x on top, no gayming led shit when?

>>61817404
>2 x PS/2
but why

>>61817404
>Fix for the segfaults when?
http://phoronix.com/scan.php?page=news_item&px=Ryzen-Segv-Response

File: poorfag.png (234KB, 800x612px) Image search: [Google]

234KB, 800x612px

>>61810976
>INTEL IS FINISHED
INTEL IS FINISHED

>>61817432
But the segfaulting has been confirmed on opteron-zenver1 chips already? And no definite fix without new stepping and no free exchange to one? How about no then.

>>61817426
Isn't it obvious? Even if I use ratpoison, I do play q2/q3a from time to time and for that I need pointing device and I only have trackball using ps/2 connection.
Besides, exploitable USB on a typing or pointing device? No thanks.

>>61817404
why not just use the it87 module?
did this segfault ever happen to you? its a linux specific problem only happening at max load all core usage. cap usage at no segfaults

suprise the mobos are made by ... wait for it... manufacturers! what does amd have to do with them? you gotta ask them to deliver.

be honest are you a retard or just a really stupid troll?

>>61817498
Autism.

>>61817498
>But the segfaulting has been confirmed on opteron-zenver1 chips already?
It hasn't. Phoronix called that statement back, and the tool is trash to begin with because it segfaults on other CPUs too. There is a bug on some early desktop ryzen chips, none that were produced after march seem to be affected. Threadripper and epyc are not affected at all

>>61817508
IT87 doesn't provide cpu (and especially not specific core) temps, only vrm, sb and unstandard 'cpu' temp and without correct marginals (which are not yet available for latest it87 revisions), they are useless.
For example, correct marginals for the the it87 io chip on 990FXA-UD3 rev.4 are included (at least kernel 3.10 and newer), as are the ones for the it87 io on GA-PA67-UD7-B3.

Yes, it is happening to me on 1700x@stock with C6H and Taichi (with 2x8gb Samsung b-die ddr4 modules). And yes, I've connected AMD support and they're not co-operative when all basic points are checked (cooling, latest uefi/w latest agesa, etc.)
Either way, tomorrow we know for sure, if B2 stepping used on threadnigger chips provides fix.

61817557
>Phoronix called that statement back, and the tool is trash to begin with because it segfaults on other CPUs too. There is a bug on some early desktop ryzen chips, none that were produced after march seem to be affected. Threadripper and epyc are not affected at all
This was posted on anandtech forums and .....amd subreddit, not phoronix. That 'trash of a tool' doesn't cause segfaults on my FX-8350, 2500k or 3930k for example. And neither does any workload either. No matter which gcc version, what kernel version, segfaulting happens on my during multiple make operations randomly.

>>61817642
Now that's what i call shitposting.

>>61817508
And where exactly did I point that trashy mobos originate from AMD? And they do in part, because of AMD flashy 'gayming' marketing, mobo manufacturers are (fairly logically) complementing it with flashy 'gayming' trash.

>>61817498
>Update [5 August]: As a result of feedback, currently working on some updated results. As some have pointed out, the conftest segmentation faults aren't specific to Ryzen, so updating the tests to avoid confusion.

It segfaults on everything, not just Ryzen. Phoronix got a lot of flak for spreading this FUD w/o any serious testing.

http://www.phoronix.com/scan.php?page=article&item=ryzen-segv-continues&num=1

>>61817719
https://www.phoronix.com/forums/forum/hardware/processors-memory/967382-continuing-to-stress-ryzen?p=967468#post967468

>I just tried it on an Intel 2600K at the office... same segfault messages.
>Leaving out the build-php test case makes the messages stop happening.

>>61817719
https://github.com/suaefar/ryzen-test
This is tool and it causes segfaults on my 1700x, but not on my FX-8350, 2500k, 3930k etc.

And that is besides the point: the segfaulting happens at random during multiple make operations (in other words: compiling, encoding with ffmpeg) on my 1700x (both on Crosshair and Taichi, 2x8gb ribjaws5, latest uefi with 1.0.0.6 agesa, smt enabled or not), but not on my fx8350, 2500k or 3930k (on 990fxa-ud3 rev4, sabertooth r2.0, 2500k on p8p67 deluxe and 3930k on p9x79-e ws).

>>61810976
And at what wattage? 400+

>>61817880
The point is the conftest was a red herring, AMD replied and said it doesn't effect Epyc or TR. Contact AMD customer service if you have the issue.

>>61817880
And what the fuck, AMD 'confirms' that 'some minority bullshit' problem exists (and quite likely it happens on the first stepping, which are probably shipped to retailer: I bought my 1700x on mid-June and it atleast isn't b2 stepping) and yet you're damage controlling here to the fullest?
I don't give a shit if if it's fixed on new stepping, if I don't get one exchanged for free. Intel fixed the P67 sata problem by recalling all P67 mobos and paid full expenses resulting in free exchanges for customers. If this is not fixed by microcode, what will AMD do?

File: 1502267904926.jpg (146KB, 1600x1424px) Image search: [Google]

146KB, 1600x1424px

>>61816768
Just wait©

Ryzen was a semi good product made to fool idiots into amd.
Don't really believe they will deliver.
Even if they manage to release a new CPU it will be much lower than 5 ghz

>>61817932
I have already contacted and after they made sure that cooling was in order, latest uefi in and stock clocks used, they didn't offer anything else.

>>61817947
>If this is not fixed by microcode, what will AMD do?
Who knows? Last i heard there were rumours about some manufacturing fuck up and CPUs after march not being affected. Also didn't they offer replacements to some guys that were affected?

>>61817191
Having to resort to such manual fixes when Intel beautifully works out of the box
Also segfault

>>61817985
>Having to resort to such manual fixes when Intel beautifully works out of the box
By that you mean even after you OC the mesh and the clockspeed it still loses to an OC'd 1600? lol

>>61817970
That some guy is not me, and I've asked for replacement to no avail.

File: intel battle tested.png (87KB, 927x527px) Image search: [Google]

87KB, 927x527px

>>61817985
You do know even intel fans even know they don't work "beautifully" out of box at launch.

>>61817985
>Also segfault
>>61817432
>>61817719
>>61817745

File: intel ceo yells at clouds.png (232KB, 800x449px) Image search: [Google]

232KB, 800x449px

>>61817963

>>61810976
Holy shit.
I mean, I'm taking this with a mountain of salt, but... Fuck man...

>>61817968
Try contacting them again? They should be offering something at least.

>>61817963
>Ryzen was a semi good product made to fool idiots into amd.
i like to call it competition and a wakeup call for intel to finally stop jewing us into buying 4 cores in 2017 , ffs the Q6600 is 11 years old now

>>61810976
>cheaper
I fucking wish. A 1700x costs 30€ more than a 7700k in my country.

>>61818277

>has never heard of the 1700

>>61817426
EXTREME autism

>>61817498
>>61817404
AUTISM ALERT

THIS IS NOT A DRILL

GET THE TINFOIL HATS EVERYONE

>>61814435
Jesus christ, makes Pentium D's look low wattage

>>61811383
if you want to game just get r5-1600, but be aware for good 7600k performance it needs 3200mhz ram.

>>61812429

>Using a ship from a series that was so bad it killed an entire franchise into feminist reboot territory to market your products

Ouch

>>61820275
That's fanmade you dummy.

>>61820324
why didnt they use an EVE online ship tho , missed opportunity

>>61820994
Because every salty bittervet that misses the old Raven and Scorpion would respond by not buying AMD chips, selling off AMD stock, and otherwise resorting to good old fashioned EVE Online style economic warfare.

>>61820994
Pretty sure WCCFtech made that one. They aren't very bright.

>>61811583
>haswell now has higher IPC than broadwell and skylake
The 4790k did pretty well with lower speed ram, so maybe this is true.

>>61813035
Everyone dropped graphene a long time ago.

>>61821717
It'll come back around once nanotechnology is advanced enough and actual mass manufacturing is possible

File: okay.png (45KB, 277x175px) Image search: [Google]

45KB, 277x175px

>>61821614
i havent played eve in a bit , i was mostly a humble miner and trader , when """code""" came along i got myself a SOE BS , cloak that fucker near busy systems and bump those freighter bumping wankstains , nothing scares em more than a massive BS suddenly uncloaking and bumping em a few km away :P , got plenty of tips from the freighter pilots in return , was fun

is code still around or did they die off? , they got pretty close to my home system (vorsk)
also pic related, i did some stupid shit in wormholes wich payed out fortunatly

>>61822958
CODE unfortunately is still a thing, and they bully the shit out of miners.

File: 252470ae41f58175b80499e91f8d06d0--chernobyl--chernobyl-disaster.jpg (353KB, 736x802px) Image search: [Google]

353KB, 736x802px

>>61814315
>7980XE

intel test labs when they tried to 3.0 OC

>>61812607
Ryzen is like the FX branding. Summit Ridge is like Vishera, the generational code name. Zen is like Bulldozer, the architecture name. Unlike Bulldozer to Piledriver, Zen won't become Zazen or Satori or some shit, it'll just become Zen 2.

>>61822998
is there a spreadmap for that cancer? if im coming back to eve i wanna avoid code cancer

>>61825076
I've seen a map of systems they operate in on the /vg/ thread for EVE, but I don't check it very often. Might try asking there.

>>61825116
thank you , also one question , what is your name in eve? might add your and hang out a bit , if you dont mind a rockmuncher