HAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAH AHAHAHAHAHAHAHAHAHAHAHAHA

File: amd3.jpg (66KB, 568x612px) Image search: [Google]

66KB, 568x612px

>>58896930
DELET DELET DELET

>desperate shills really have to grasp at straws this hard
https://forums.anandtech.com/threads/summit-ridge-zen-benchmarks.2482739/page-201

>Ryzen is running with DDR4 2400 17-17-17-39 probably 2T. Explains the poor results, those timings suck for that memory speed.

>>58896947
XDDDDDDDDDDDD

>>58896930
>intel getting this desperate

No rupees for you, Pajeet

File: MWSnap_2017_02_10_23_03_35.jpg (33KB, 764x156px) Image search: [Google]

33KB, 764x156px

>>58896947
Also this
https://forums.anandtech.com/threads/summit-ridge-zen-benchmarks.2482739/page-200

Original uploader of these cherry picked tests posted this as well:
>Note: Actual clocks are not listed on these graphs. The true clocks are as follows:
A Ryzen with unnecessarily slow RAM vs OVERCLOCKED INTEL CHIPS

Ryzen's performance looks ungodly astonishingly good. Jesus Christ.

File: sad_snow.gif (1004KB, 500x281px) Image search: [Google]

1004KB, 500x281px

>>58896930
T O P P E S T K E K

Ps
I posted something like this on /v/ but with only 6 HA's and got a week ban, why are the mods on /g/ so cool and /v/ a bunch of fucking pussies, perhaps redditors in disguise?

File: 1484617501639.png (308KB, 492x419px) Image search: [Google]

308KB, 492x419px

>>58896930
>Different clock speeds

What the fuck? Of course a current gen i7 clocked 1+GHz over the Zen will btfo it in single thread performance.

This is like OccupyDemocrats tier fake news.

>>58897027
The guy who posted this on Anandtech did it specifically to stir up shit. OP is either sm625 himself, or he actually is retarded.

File: 1435411968922.png (777KB, 1280x720px) Image search: [Google]

777KB, 1280x720px

Thank you based Shitwrecker.

>>58897021
>/v/ a bunch of fucking pussies
What can you expect form nintendo central?

File: 2500k vs Ryzen.jpg (100KB, 701x852px) Image search: [Google]

100KB, 701x852px

>>58896947
Never doubt the Certified Shit Wrecker.

>>58896930
So it's over. That's why they're priced so low. It's fucking Bulldozer all over again. 8 core CPUs being ass whooped by 2c/4c Intel alternatives in your daily use normie pc.

>>58897096
Read the thread you inbred fucking retard.

>>58897112
He did, AMD fucked it again. What a surprise

>>58897121
Epic bait.

File: FGM2.jpg (46KB, 300x419px) Image search: [Google]

46KB, 300x419px

>>58897112
>2400MHz DDR IS THE CAUSE! I SWEAR GUYS!

Pottery.

>>58897133
Not bait fuckwit, this is the 480 hilarity bullshit all over again

>>58897142
Its stock speed Ryzen with shit tier RAM vs heavily overclocked intel chips.
The perf/clock shown is destroying Haswell-E and Broadwell-E there.
sm625 cherrypicked the results, hid the clocks speeds, then didn't show them until he got a mod warning and edited his original post.

The i7 5960X is running up to 4.7ghz.
Haswell-E at 4.7ghz vs 3.4/3.8ghz Ryzen.

>>58897156
Stay in /v/ shitposting little kid.
Pathetic.

>>58896930
Ryzen running at 3.4 GHz with no turbo. All other CPUs overclocked to 4+ Ghz. See >>58897019

It's actually Intel finished and ded, not the other way around

>>58897170
How do they know it? Chart clearly shows stock speeds for Intel CPUs. You aren't trying to trick me you little AMD shill huh?

>>58896930
ANOTHER MASSIVE DISAPPOINTMENT

File: you_need_to_upgrade_every_gen_goyim.png (98KB, 1234x482px) Image search: [Google]

98KB, 1234x482px

whew boy.

good thing kaby is so good.

>>58897183
Yes lets turn it back onto intel, quick hide how shit our new cpu's are.

>>58897170
>kid
Haha right. When you move out of mummys house let us know.

>>58897181
I linked the source twice. The results are from passmark's database browser, you can search through submitted results and compare anything.
You're actually getting trolled by some faggot from the Anandtech forums. Good job.

I can't wait to see you dipshits repost this for weeks now.

>>58897201
You are a little kid until proven otherwise, little kid.

Intel shills are truly desperate. We already know ryzen kicks ass and Intel is scared shitless. They started sampling new CPUs already. They are actually on suicide watch.

>>58897019

Don't waste your breath anon - /g/ has decided memes are more important than actual fact finding so even after it has come to light the OP's chart is deceiving /g/ will take it as gospel.

Amusingly in his posts over on semiaccurate Fottemberg (dude behind bits and chips who has been pretty on point for zen based leaks) anticipates a huge swathe on Intel favouring reviews/articles to start popping up right around Ryzen's release.

File: Intel 1.png (186KB, 1920x974px) Image search: [Google]

186KB, 1920x974px

>>58897212

> Intel is scared shitless

The rumoured new kaby lake chips and pic related do seem to agree with this.

>>58897209
Fucking hell you autistic twat, how do you prove otherwise on an anonymous forum.

>>58896930
>>58896947

How is this poor results? The results are quite good and comparable to the best i7s and the price is 50% cheaper. It's like watching a faggot buying a lamborghini and he asks for a drag against your nissan and he doesnt come close to flawing it passed you.

>>58897239
Have they officially announced the prices?

>>58897234
You act like a dumb little kid. You shitpost instead of carrying on conversation like a normal mentally healthy adult. You're a little kid.
Little kids like you belong on the little kid board >>>/v/

>>58897170
>sm625 cherrypicked the results, hid the clocks speeds, then didn't show them until he got a mod warning and edited his original post.
Link?
Not that I don't believe you, just want to read the whole thread.

File: 1483632489490.jpg (6KB, 240x240px) Image search: [Google]

6KB, 240x240px

>>58896947
>>58897019
gamers can't read
>>58897183
>>58897193

>intel will react with upclocked skus on already hot packages
>coffee lake confirmed for 14nm
>only difference will be soldered ihs and new backdoors
>intel going down in literal flames

>>58897256
Scroll up.

File: sperg.jpg (35KB, 600x399px) Image search: [Google]

35KB, 600x399px

>>58897253
Now wipe all that spittle off your screen, you seem to have anger issues, has mummy not given you the meds this morning?
Why does amd producing more below standard parts make you spit your dummy out?

>>58897290
Continuing to shitpost only further reinforces the fact that you're a faggy dumb little kid.
Stay in /v/ with your pathetic bait shitposts. Little worthless faggots like you ruin this board.

>>58897298
Autism. You are a freak.

>>58897310
>faggy little kid just keeps shitposting
Pathetic. Absolutely childish. You need to be at least 18 to post here.

>>58897319
Mum, mum the man on the internet keeps saying nasty things about amd.
Id get outside more freak, quit the kiddy porn

File: 1353011947269.jpg (62KB, 340x310px) Image search: [Google]

62KB, 340x310px

>>58897319
>>58897310
>>58897298
>>58897290
>>58897253

HOW AYYYMDPOOJEETS EVEN RECOVER

>>58897340
When /v/'s shitposter clashes with /g/'s autistic AMD drone.

>>58896930
>i7-5960x faster than i7-6900k
shows you how worthless that benchmark is.

>b-but muh lower clocks

RYZEN CAN'T OC

>>58897352
The bench is real, it's just that AMD's CPU isn't oced when compared to 5ghz 7700k.

>>58897019
Hahha Ryzen has the same single thread performance like Kaby Lake.

File: XeonD_scheme.png (281KB, 1562x924px) Image search: [Google]

281KB, 1562x924px

>>58897349
I'm not even an AMD fan. I've tried to make threads about Xeon-D, new Atom lines, the nuances of the CoreM lines, but everyone here is too fucking dumb to know anything about them.
This board is full of /v/ spill over trash that doesn't understand anything but gaymen. Hence why the little faggot kid instantly resorted to shitposting after I posted the source to the OP trollbait.

File: 1.jpg (13KB, 564x423px) Image search: [Google]

13KB, 564x423px

>>58897369
No because they have already wrung the bollocks off them as it is.

>>58897391
Hey look its the autistic amd drone again, shilling a lost cause.

>>58897422
Such an AMD drone that I own an intel powered Toughbook, an intel powered tablet, and make threads about newly released intel hardware.
I like technology, I don't buy into the fanboy mindset. Thats strictly /v/ mentality.

>>58897442
Did you buy those with your disability benefits or did you get them from mummy & daddy

>Single thread benchmark
Everyone expected Ryzen to be slower than a G4560 in single threaded applications but the people who buy Ryzen do it for multi threaded stuff like video rendering.

SHHHH SHUT THE FUCK UP
LET THE RETARDS BUY INTO RYZEN SO INTEL GETS PRICECUTS
REEEEE STOP RUINING IT FOR THE REST OF US

AYYMD IS FINISHED & BANKRUPT

AYYMDPOORFAGS CONFIRMED ON SUICIDE WATCH

File: spec.png (77KB, 819x593px) Image search: [Google]

77KB, 819x593px

Quality thread. If those price leaks for Ryzen are real I'd still recommend everyone else to buy it. A slight oc from 3.8 boost to 4.2 Ghz will probably mean a score of ~2600. Intel can choke on a dick with those taxes.

>>58897507
>slight OC
>Implying it'll OC at all

OCING WILL FIX THE SHIT STOCK CLOCKS!

>>58896930
Jesus christ DELETE THIS!!!!!!!!!!1

>>58896930
Even if that graph wasn't amusingly biased, you still get double the amount of cores for less than an i7

File: 1398835321369.jpg (41KB, 650x320px) Image search: [Google]

41KB, 650x320px

>My G3258 is running at 4.4 GHz.
Are we seriously going to believe some Intel fanboy on Anandtech forums who can only afford a $70 CPU?

>>58897532
Anon CPU in OP's bench doesn't even use turbo. 0.8 GHz isn't a "slight" oc. Compare it to stock 7700k score of 2300 points and that's with turbo.

>>58897578
Apparently, yes.

>>58897183
>good thing kaby is so good.
It's great at heating homes. They didn't want Nvidia holding that crown so they decided to bring it old school ala the Pentium 4. Helps the Nords and Canadians warm in winter.

>>58897583
>this isn't even my final form

Poo in loo

>>58897021
/v/ are cancer and all under 18. that's why

>>58897578
Yes, it's /g/.

YES! THANK YOU BASED INTEL GET READY FOR AMD SHARES TO DROP TO $4 AND FINALLY BANKRIPT THIS SHIT POO COMANY

Could someone rename this board to Shitposting?

>>58897391

This guy has serious issues, do you spend your life on this website literally raging at people and thinking nobody is as intelligent as you?
example> >>58889756
That threads hilarious, do you scream at your shitty 15" screen when replying?

File: 1469883273122.jpg (73KB, 960x960px) Image search: [Google]

73KB, 960x960px

>>58897610
It overclocks itself.
>The Precision Boost feature works alongside Pure Power to optimise performance and enact on-the-fly clock adjustments
>(XFR) which automatically scales the frequency based on the thermal loads and cooling apparatus while permitting frequencies beyond the standard Precision Boost limits.
http://www.eteknix.com/amd-ryzen-architecture-features-revealed/

File: 1483514094254.png (234KB, 800x612px) Image search: [Google]

234KB, 800x612px

>>58897019
DELET

>>58897671
Intimidated by the fact that I'm smarter and more educated than you'll ever be? You won't get a degree by sitting around fixing on all the times I make you asshurt.
You're a pathetic shitposting little kid. Shitbrained children like you are the reason why this board isn't a nice place. Civil and informed discussions don't take place here any more because cretins like you rather spew memes, act like petulant miscreants, and cling to petty fanboyism.

Try to act like an actual adult for once in your life. Stop talking out of your ass and you won't get called out for spreading misinfo. Refrain from shitposting like a 9gagger and you won't get called a child.
Its not hard to understand, kid.

>>58897710
This is like shooting fish in a barrel.
You have a mental disability, basically a mong, you will never be normal, the amount of threads you are in calling people kids is hilarious, its all you can post as a comeback and its shyte!

>>58897710
lmao @ the reddit spacing

>>58896930
Another Major Disappointment

>>58897742
>I can't post a single thing of substance
>literally all I do is shitpost
>p-p-plz stop making fun of me!
Pathetic.
Being mentally unwell is sitting around and shitposting for no reason, rabidly defending a company you have no ties to, regurgitating the same childish memes day after day. That isn't something a mentally sound adult does. Its called being emotionally under developed, stuck in the immature mindset of a kid.

Though I'm sure you're just going to keep crying. Its what a little kid does.

There's nothing that can convince me to not buy a Ryzen.
Jewtel is kill.

File: op.png (328KB, 718x354px) Image search: [Google]

328KB, 718x354px

>>58896930
>no clockspeeds

>>58897710
>trying this hard
you dont have to look cool for us anon, just be yourself

>>58897788
You still live at home with your parents, you dont have a pot to piss in. Literally you spend your life raging on here.
The dump I took this morning has more of a life than you fatty!

>>58897678

There are so many unknowns about the nuances of XFR but at face value it basically invalidates Intel's strategy of locking chips. I would imagine the motherboards control the upper limit that XFR will let a chip boost to but it does mean (in theory) if you slapped a large(r) cooler on your ryzen chip even on a mid or low end board it would still autoverclock by some degree.

This could also mean (ignoring any other variables) buying the ryzen 8320 equivalent chip would be a killer deal because who gives a fuck about base clocks when you have decent cooling and the chip auto overclocks?

If the x370 boards let you disable any TDP limits there could be some rather high boost clocks (naturally manually overclocking will ensure high base clocks as well as the ability to lock at a specific frequency).

Most of /g/ doesn't grasp how (potentially) powerful XFR is.

>>58897810
I'm already looking into pelter plates for sub 0 cooling. That's how excited I am by it. It's a massive game changer.
This will make overclocking so much more stable if it's a carryover function in manual settings.
Normally I'd have to pre cool before booting a serious overclock, but with this it should mitigate any need for that.

>>58897810
the 3.6/4.0 sku will be a higher bin and will probably ship with a water cooler but yeah.

>>58897710
>>58897810
you try to talk reason with neo-/g/ dont do that, there are other boards for serious tech discussions. neo-/g/ is for consumertech/helpdesk or company/fanboyshills only nowadays. sadly

>>58897884

A so-called Black Edition with AIO could be veeeeeeery interesting on the prebuilt front. OEM just slaps it onto a cheapo board (because thats what OEMs do) and sell it off as ultra l33t 360 noscope gaymen pc.

>>58897884
they shipwithout any cooler, you gonna use a costum cooler anyway

>>58897801
Projecting isn't going to fix anything for you. I'm really well off.

>>58897810
XFR lets the chip exploit remaining thermal margin, on face value its a simple concept, but the implications are pretty big.
Under a nominally heavy workload most all modern CPUs won't run at a completely fixed clockspeed. They'll bounce around a bit in their top pstates as they hit power limit. Caveat here is that power limit isn't a single absolute thing, chips are allowed to exceed their rated limit for short amounts of time. AMD's method of calculated temps give them per cycle accurate metrics on power usage characteristics, leakage, which units on the die are under what level of load, a ton of other metrics.
This exposes greater spans of time to the CPU where it would allow it self to operate in a higher clock state as remaining thermal margin allows. Another caveat to this is how temperature effects the performance of a transistor. Granted this effect is less pronounced in a FinFET device, but a colder transistor has less leakage. Keeping a chip colder can allow it to clock to a certain level with slightly less voltage, or it can allow it to reach a slightly higher clock at a set voltage level.
So combining this you have increasing cooling capacity giving the chip potentially significantly more headroom in thermal margin, and that opens the window wider and longer for higher clocks within a certain threshold.

>>58897772
man I fucking love the reddit spacing autist and the plain autist who doesn't get shitposting
is this neo/g/

>>58897905
neo-g stopped being neo like 5 years ago

>>58897914
100+ more dollars for less than 1% better power characteristics? I'm skeptical, unless AMD go the 6850k route and intentionally gimp pci-e lanes or something on the lower sku. A water cooling loop makes more sense.

>>58896930
>>58897019
2046/1925*5.15/3.4 = 1.609 (Piledriver)

2046/2539*4.7/3.4 = 1.114 (Haswell)

2046/2435*4.2/3.4 = 1.038 (Broadwell)

2046/2889*4.8/3.4 = 0.999 (Haswell)

2046/2916*5.0/3.4 = 1.032 (Skylake)

Pretty obvious AMD has reached all IPC goals they said they would. Comparing overclocked to under clocked benchmarks isn’t going to save Intel now.

>>58897933
it's the same autist silly

>>58896947
Want to know what else explains those poor results? AMD.

When will Intel finally get hit with an antitrust suit?

>>58897959

/g/ will be rushing to defend shekelberg soon enough.

File: 1485013232205.jpg (17KB, 363x363px) Image search: [Google]

17KB, 363x363px

>>58896930
>>58897801
>>58897310
>>58897336
>>58897422
>>58897462
>>58897464
>>58897469
>>58897532
>>58897571
>>58897658
>>58897671
>>58897742
>>58897784
>>58897801
>>58897974

As much of a pretentious fedorafag he is, he properly adds substance to a discussion about hardware on this TECHNOLOGY board.

(You) on the another hand, are just another /v/-tier shitposting faggot(s) gobbling up benchmark bars rushing one-liners without any further analysis because you probably don't know better.

I'll take him over cancerous fucks like you any day.

>>58897952
A Noctua NH D15 only $85. Pretty sure there are other huge tower coolers in the same range for $70 and less. It doesn't cost a ton for super high end cooling, sans custom water loops. AIO units only have the appeal of offering tower cooler performance without the bulk, but have the added drawbacks of limited lifespan, and requiring eventual maintenance.

>>58897961
Nope.

>>58898023
This board is less about tech discussion and more about shitposting.

>>58898027
https://warosu.org/g/thread/S58721346#p58724586

>>58898023
>AMD puts out garbage
>people call it garbage

>guys you're being shitposting bullies

I'm starting to believe more and more every day that companies pay people to shill their garbage. When the product is downright indefensible, you resort to calling people shitposters. Sad!

File: inteldestroyer.png (105KB, 255x345px) Image search: [Google]

105KB, 255x345px

BASED AMD

File: 1469693257470.jpg (24KB, 308x302px) Image search: [Google]

24KB, 308x302px

>>58898047
>Sad!

You almost had me.

>>58898027

The only good AIO is the EK predator.

File: 1484172321665.jpg (425KB, 1680x1050px) Image search: [Google]

425KB, 1680x1050px

A M D
M
D

>>58898046
https://warosu.org/g/thread/S58870848#p58873635

Looking through that archive only a single post with the phrase "reddit spacing" is mine. I didn't coin the term, nor do I frequently use it. Tons of other people are there using it and arguing about it, none of them me.

>2 weeks from release
>still no independent benchmarks
>all the leaked ones show Ryzen underperforming

>>58898087
Gotta keep that stock high to the latest moments.

File: h100i cooler and 8350.png (82KB, 735x744px) Image search: [Google]

82KB, 735x744px

>>58898070
The h100 i V2 on my 8350 disagrees.
70c on 100% load with the stock cooler.
38-39c under 100% load with the h100i V2.

>>58898100
Fuck you, I want Ryzen to be good. I have a shitty i5 4670 and I want to upgrade it.
I just refuse to hype, specially after those low retarded prices

>>58898087
>Same IPC as Broadwell
>underperforming

File: self intoxicated, twisted detonator.png (314KB, 1131x1006px) Image search: [Google]

314KB, 1131x1006px

>>58898108

>stock clocks
>watercooling

Nyet. Get that shit 4.7ghz+ and come back to me.

Pic sort of related, my chip being air cooled under torture loads.

>>58898085
because they're laughing at you for unironically using such a retarded phrase to call out people you don't like you fucking autist

i'm convinced you're a legit aspie

>>58898127
Show me an independent benchmark where it isn't under performing
>inb4 excuses
I don't want to hear for AMD marketing team I want actual results

>>58898127

>sandy vagina is all you ever need
>oh shit AMD has caught up to broadwell
>Yeah well if its not kaby lake IPC any given cpu is useless

That is how /g/ is currently playing out.

>>58898144
OP benchmark proves Broadwell IPC. >>58897959
Dumb Intel shills just didn't do the calculations before posting it.

>>58898108
You need to use AMD Overdrive.
AMDs recent CPUs and APUs don't have any sort of real temp sensor on die, they don't output an actual temperature figure.
They output a metric called TCTL/Thermal margin. It shows you how many *units* of heat you have remaining before the chip starts to throttle, but the units displayed aren't an actually tangible temperature.
You have more remaining thermal margin at idle, less under load. Pretty much nothing else displays this metric correctly.

http://www.tomshardware.com/faq/id-2122665/understanding-temperature-amd-cpus-apus.html

>>58898141
Learn to read, moron. The thread I linked doesn't have a simple post from me in it. I'm not going around throwing out wild accusations at random people. Though I can tell you have a personal vested interest here. Undoubtedly you're butthurt that I put you in your place for talking out of your ass like a little kid.

>>58897710
HAHHAHAHAHAHAHAHAHAHHAHAHAHAH

>>58898160
I said no excuses I want pure benchmarks

>>58898087
>all the leaked ones show Ryzen underperforming

Kill yourself you shitposting faggot. Canard's leaks had Zen looking very solid, and even the benchmarks from this Anandtech thread in OP's image aren't bad if you consider clock speeds as listed in >>58897019

>>58898144
There isn't. Only leaks, which mostly looked rather promising so far.

We won't know until it's out.

>>58898137
Only put it on a few days ago. Just 'running it in' to ensure I don't have to RMA. Will overclock soon. Not sure if I'll hit 4.7. I'm on a 970 board AFAIK after 4.5 I start risking blowing VRM's but that's based on posts from when these mobos were new.
Also gotta get some extra cooling for south bridge and VRM's ect.

Looking at ya numbers, you did great. You got extra cooling of the board?

File: 1386434101411.jpg (10KB, 192x245px) Image search: [Google]

10KB, 192x245px

>>58898176
>>We won't know until it's out
They're not sending it to reviewers ahead of time?

>>58896930
>>58896940
>>58897011
>>58897096
>>58897121
>>58897156
>>58897182
>>58897348
>>58897469
>>58897464
>>58897571
>>58897658
>>58897784
>>58897772

Read this shit, fucking dumbasses!

https://imgur.com/a/6XUIw

All Intel cpus is overclocked, while AMD is in stock with the turbo disabled.

>>58898199
changes nothing, show me a benchmark where the AMD is actually performing better than intel

>>58897176
> with no turbo
We don't know about this, maybe software just can't detect it.

>>58898197
NDA.

>>58898166
Benchmark compares 10% underclocked Ryzen to 13% overclocked Broadwell.
Broadwell is only 19% faster than Ryzen.

This is like 9th grade math.

File: HUqFUFS.jpg (98KB, 1089x700px) Image search: [Google]

98KB, 1089x700px

>>58898199
based

>>58898211
>3.4ghz Ryzen
>near equal to 4.2ghz 6900k
>hurr doesn't mean anything

>>58898225
Thanks for proving that you have nothing
>>58898235
(Estimate.)

File: url.jpg (5KB, 251x201px) Image search: [Google]

5KB, 251x201px

>>58898211
>It literally lists Zen as a decent performer

File: aspect su.jpg (10KB, 125x264px) Image search: [Google]

10KB, 125x264px

>>58898235
>graph by [email protected]

>>58898247
>(Estimate.)
Try again pal

>>58898244
>make shit argument
>get disproven
>OMG THESE BENCHMARKS MEAN NOTHING ANYWAY

>>58898244
That isn't an estimate, its taken straight from passmark. The only estimates there are the anon's overclock Ryzen scores.
Fuck you're stupid.

>>58898256
Marvin seems like a great guy.

>>58898213
Even if Turbo was enabled it would put Ryzen 5-10% behind Broadwell for literally 1/3 of the price.

>>58896930
>>>/v/

>>58898176
Hurrr no you lying anon its performing fine, reeeeee im right and your wrong you know nothing,kill yourself hurr durr

>>58898261
You need proof to disprove anything, you didn't offer any.
>>58898267
It says fucking estimate on the fucking image you retard

>>58898261
>estimate
I can estimate that coffee lake will hit 6Ghz and performance will scale too..

File: marvin1.jpg (385KB, 910x800px) Image search: [Google]

385KB, 910x800px

>>58898256
he's /ourguy/

File: anti-housefire fan resized.jpg (485KB, 997x748px) Image search: [Google]

485KB, 997x748px

>>58898195

>Looking at ya numbers, you did great. You got extra cooling of the board?

To be fair my mobo is only rated for 140w chips and under that sort of torture test the chip is probably pulling near 200w.

>>58898280
>>58898283
We already know Ryzen stock clocks are higher than Broadwell stock clocks. Benchmark in OP proves IPC is identical.
AMD has won. Just admit it already.

>>58898273
The OPN of that Ryzen chip in question reads 3.4ghz base and 3.8ghz turbo. Listed turbo is max single core, all core turbo is lower so the discrepancy isn't that large. Per clock Zen is outperforming Broadwell in integer performance here, by a significant amount.

>>58898280
The only estimates are the overclocked scores for Ryzen, which clearly state estimate next to their values. None of the other ones are estimates.
How buttmad are you right now?

>>58898316
The only graphs which show AMD beating Intel are estimates
Again you have nothing but marketing bullshit on your side

File: meme.png (118KB, 1907x628px) Image search: [Google]

118KB, 1907x628px

>>58897183
>6 years apart
Intel really jewed us good

File: muhdrills.gif (570KB, 291x400px) Image search: [Google]

570KB, 291x400px

>>58898314
>AMD has won
>CPU isn't even out yet

>>58898314
If I hadn't be over this exact same mentality on /g/ with the Bulldozer I might actually take your word for it

>>58898329
Fuck, I knew you were retarded, but this is ridiculous.

39,672 points @ 3.4ghz. - Ryzen
40,642 points @ 4.2ghz. - Broadwell-E
Both are 8c/16t chips.

Guess which is showing better performance per clock? Retard.

File: trashman6.png (591KB, 778x1018px) Image search: [Google]

591KB, 778x1018px

>>58898307
>kancerkolle
>seasonal moeshit

>>58898365
Broadwell-E
next question

File: Better than expected.png (223KB, 1899x785px) Image search: [Google]

223KB, 1899x785px

>>58898161
Cheers.
It's actually doing a better cooling job than I thought. Running the stability test and tripcode explorer for funzies.
Pic related.
>>58898308
?? I thought all 990 boards were rated for 200w.
Gotta love getting the stock AMD coller. That little fan is handy as fuck. The heatsink is also very useful on smaller chips.

File: singleCore.jpg (138KB, 1716x962px) Image search: [Google]

138KB, 1716x962px

This is something that surprised me. I dont mean just the Ryzen, but others too. Isn't the single thread performance something that is the most important in some games, like Fallout 4?

File: 81f18e186a8881eecd1ccf4548dc0bd47d10d392_s2_n1.jpg (432KB, 1722x2048px) Image search: [Google]

432KB, 1722x2048px

all things considered it looks like intel is going down in flames lmao

File: M5A99X evo.png (17KB, 1058x419px) Image search: [Google]

17KB, 1058x419px

>>58898387

> I thought all 990 boards were rated for 200w.

You thought wrong.

I just want AMD to crash and burn already so /g/ can go back to normal.
Tired of 5 billion AMD vs Intel threads every day

>>58898397
Generally, yes. Single threaded performance is going to be very important for gaming because the engine will have the bulk of the rendering addressed by a single CPU thread, even if additional cores and threads are well utilized by secondary and tertiary features. Some games and engines have enough work for additional threads to do to the point that a quad core will actually be disadvantaged compared to a 6 or 8 core chip despite its higher clock speeds. In such scenarios the higher core counts tend to prevent minimum frame rates from dropping as low. Other instances a game won't utilize additional threads well, so they'll greatly prefer higher clocks.

The best possible scenario is to have very high serial performance, and have many cores also. Only downside is you pay for it in power consumption and heat.

>>58898440
>i want a monopoly so i can pay more money for worse product

>>58898387
https://github.com/meriken/merikens-tripcode-engine-v3

>>58898463
They will crash and burn anyway, I just rather this happens sooner rather than later

Which part is being tested?

>>58898331
DELET

>>58897578
I mean, I have a g3258 in my plex box.
Its a decent little chip for 50 bucks.

>>58898377
???

>>58898424
Well I'll be.
Makes me happier clocking on this board. Thought I screwed up heavily after seeing the 9650 plugged into a 970 mobo and insta blowing the VRM's.

>>58898397
Cores/threads are starting to matter a lot more now in the most recent games, Including fallout 4.

See https://youtu.be/ezwv4Onr8HQ?t=123

I own a G3258 @ 4.4 Ghz, and a 3570K @ 4.7Ghz

Thier single thread performance is roughly equivalent but the 3570K system still gets much higher FPS in modern games, with the same GPU obviously.

In fact, in some games (GTAV is a major example) the amount of stutter induced by only having 2 cores makes the game borderline unplayable, Subnautica is another game I enjoy playing that is just hampered by towns of stutter and pop-in on the G3258, which doesn't happen on the 3570K

On the other end, Most games don't benefit that much moving from 4 cores/threads to 8, with only GTAV and BF1 being the ones I remember that had a noticeable difference (though it was till only a 10% performance increase or thereabouts.

>>58898517

A 9590 IS a board killer - they run at 1.5v+ out of the box. My little 8320e at 4.7ghz is drinking "only" 1.428v under torture loads and thats pushing the board right to the fucking limit. 70c is the cutoff point for that board (anytime I push my chip harder and the socket goes over 70c it hard locks). I understand if my chip was under constant load for hours on end the motherboard would eventually fail - its not so I can (sort of) run the risk and walk the knife edge.

The lesson is not get greedy with your overclocking and ensure you aren't frying your shit.

File: 01 - Gains over Sandy.png (186KB, 2699x1737px) Image search: [Google]

186KB, 2699x1737px

>>58898331
http://www.anandtech.com/show/9483/intel-skylake-review-6700k-6600k-ddr4-ddr3-ipc-6th-generation/9

This is a far more compelling proof to make your case. Eliminate the Dolphin bench from the line up and at most you have Skylake show a 38.2% improvement in POV-Ray. Some performance uplifts are as low as 5.5% for single threaded 3D particle movement. Thats raw IPC. No SMT gains, equal clock speeds, equal memory speeds. Thats the performance uplift delivered each generation.

Intel's core arch has gone through a lot of refinement, and when you continually tweak the same base design you reach a point where you really have to stretch for meager gains. Every fraction of a percent uplift means more invested man hours than the last. The only saving grace thus far has been FPU ops still seeing big throughput improvements, but on the integer side its very close to stagnating. FPU gains won't continue on forever though. It'll eventually be trying to squeeze blood from a stone.
Its not going to radically change until their new ISA in 2021/2022 on their 7nm node. Coffee Lake, Cannonlake, Ice Lake, Tiger Lake, they're all going to be moderately small uplifts in single core perf/clock.

File: Screenshot_2017-02-11_08-02-13.png (190KB, 973x670px) Image search: [Google]

190KB, 973x670px

>>58897183
Incredibly cherrypicked result you have there

File: gta-v-cpu-4k-vh.jpg (52KB, 806x532px) Image search: [Google]

52KB, 806x532px

>>58898528

>get the pentium /g/ said
>2 cores is plenty /g/ said
>moar coars is a meme /g/ said

Now I know pic related isn't telling the whole picture (its just the pic I have saved) my main point is the meme pentium is a shit chip for da vidya.

Why are we assuming any ryzen chip will be able to reach 5ghz? Or even get over 4?

>>58898576
Simplistic bench charts really don't cut it for gaming. Actually playing a game where shit is happening on screen, things are loading in, there is tons of variability. Min frame rates in real life are going to be wildly different from what you see in charts online. A big part of the problem is that these reviewers aren't playing the game at all. They're running through one consistent part of a map over and over, trying to make it sterile as possible. Its totally unrealistic.

The only solution is a video review where frame rates and frame times are shown overlaid on top of the screen. Digital Foundry does it for console games. Big tech outlets need to be pressured into adopting that format.

>>58898615
True dat. Even memory speed factors heavily now.

>>58898550
Thanks for the info. I'll probably aim for 4.55. I'll be going Ryzen as soon as I can, so going hard isn't much of a need for me even though this PC is less than three months old.
>>58898467
Thanks. I'll keep that in mind. I've got the Jap one because I'm planning on running a CPU racing thread sometime and tripcodes give clear evidence that someone won. Yes I know there's luck involved. But hashes per second is nice too.

Ryzen's performance per watt is pretty great, a substantial upgrade over Bulldozer.

I'd rather wait for comprehensive testing with final parts though. I'm leaning towards getting a 1700 though. Not sure what GPU to pair it with however... Might just "settle" for a 480

>>58898615

The other side of the coin is if one has an agenda you could test product A explicitly in an area where the game chokes (could eassily be something on the software side) and product B in as different area where the game does not just to make product B look better.

>>58896930
Lol, who wouldn't have seen this coming.
>>58896947
>poozen shits itself with a common type of ram
It only gets worse.

>>58898651
17-17-17-39 CAS timings equates to 14.161ns-14.161ns-14.161ns-32.487ns latency. Its horrendously slow memory that would impact the performance of any system you installed it in, genius.

>>58898720
Whats the formula for calculating that?

Why is everyone going mad. I thought ryzen was expected to have haswell like IPC at a lower price. This is fine.

>>58898737
Cycle time x CAS timing.
Cycle time is easy to extrapolate from its data rate.

File: 1448977414617[1].png (27KB, 900x342px) Image search: [Google]

27KB, 900x342px

itt maximum advanced micro dildo damage control

>>58898755

Only those with Intel stocks are getting mad.

>>58898755
It sounds like more "poorfag" memery from intel shills. Sure, ignore that ryzen is bringing high end desktop performance to mainstream prices if the rumours are true.

>but muh 4.2ghz
It's like everyone forgot two weeks ago when babylake was a crap TIM overheating piece of shit and there's still no reason to upgrade from a 3770k let alone an 6700k, but now it's the best thing in the world.

>>58898755
>Why is everyone going mad.

usual 4chan, people like this shitpost, no real science or discussion possible here, don't know why people even try

>>58898755
No, you don't understand you shitbucket.

I own an i5 and thus I am part of an exclusive group of patricians that own the best CPUs in the world. Over MY DEAD BODY will AMD ever be better than Intel, I will literally shoot up their headquarters if they take this away from me. I've already lost most of what little I have.

>>58898778
mad vs envy

>>58898833
Saved.

File: ayy.png (92KB, 1860x239px) Image search: [Google]

92KB, 1860x239px

>>58898574
Not him, but here's my 3570K @ 4.7Ghz (1.275 Vcore) vs a top 33% 7600K @ 4.2Ghz

You know what's worse? That the top 5% of 7600K's can only perform 15% better, after SIX YEARS

>>58898574
>4.2 Ghz vs 3.5 ghz

20% alone on clock speed difference

>33% - 20 = 13%

13% IPC from 5 years

K E K

>>58898463
Problem is theres only two companies. Why should I have to buy an inferior product?
AMD are as bad as intel/nvidia, theyre only in it for profit so why do people defend them so much

>>58898314
>implying this matters
X99, even Broadwell-E is a huge overclocker. Maybe not huge by AMD's previous standards (Intel was huge in the past too) but you can't just assert that it's definitely going to do a bunch of catching up with zero proof and use it as an argument that one is better. So basically >>58898283

It's purely making shit up.

File: 1367107700390.png (96KB, 329x313px) Image search: [Google]

96KB, 329x313px

>>58898929
>Broadwell-E is a huge overclocker

>>58898965
How much exactly do you expect, percentage-wise? Do you want a 50% overclock or something?

File: Go you little celeron go!.png (20KB, 457x395px) Image search: [Google]

20KB, 457x395px

>>58898929
>(Intel was huge in the past too)
But what good is a celeron.

>>58898973

Some of the FX chips can hit 50% overclocks. Naturally this totally ignores the fact they were designed and built to hit silly high clocks (thanks IBM) but didn't due to housefires.

>>58897121
I predicted this like 2 days ago. People told me I was a shill and that I was blowing everything out of proportion.

>>58898988
Liquid nitrogen is a ridiculous metric unless you actually plan to buy these chips specifically for a brief competitive overclocking event.

And absolute clock rate is as well. Does it matter if AMD previously hit 5GHz if it was performing worse than Intel was at a much lower frequency? Clearly percentage is what you're looking for when you've already been provided base clock benchmarks.

>>58899017
What I'm getting at is that just because AMD made them at the time doesn't mean they're going to do well this time too. Overclocking headroom depends purely on speculation until it's provable. What's to stop them from just setting the stock speeds higher themselves and working from that? Not much, considering they already did that not too long ago.

I'm not even saying AMD isn't going to do it, just that there's no good enough reason to believe it's happening with currently available information.

>>58898929
I took an E2160 to 3 GHz with stock cooling. That's nearly a 50% overclock on stock cooling.

Just ordered me some Intel i5. Thanks for head ups, OP!

>>58898108
>AiO Watercooler
>FX-8350

cringe

File: 1476332231066.gif (2MB, 320x433px) Image search: [Google]

2MB, 320x433px

>>58899237
Care to explain?

>>58897810
less then 5% of people overclock, imagine if xfr is good, just imagine, a normal person does some heavy lifting and all 8 cores are fireing at 4.0+ghz

they stop and play a game, now 4 cores are firing on all cylinders going 4.5ghz and they down clock the other 4 down clock to give more power to the cores

a friend comes on and they move to a less optomized game that only uses 2 cores and the cpu now has 4.8 ghz on them while the rest idle

to end the day they they use a program that has no threading, so one core is used, and that one core is now screaming at 5ghz+

canard pc said ryzen hit 5ghz on a single core with no issue but the motherboard was validation only so they could not push it further due to power, not heat, not an architectural fuck up, just power to drive it was not sufficient.

What if you could push this thing like a fucking 9000fx cpu if given a good enough motherboard and the cpu can take it?

xfr is literally the second most interesting thing about the cpu, the first being 3.4ghz minimum base trading blows with a 6900 for 1/3-1/2 the price.

going to have to read the benchmarks, as I was planning on getting a D15 for it, but if xfr is really just that fuckin good, i may get a triple rad aio, possibly a custom loop.

>>58898027
Any idea on how fast noctua ships out new socket mounting supplies?

I want to get a d15, but they also made a d15 with am4 mounts for ryzen, and I am wondering if the 15-25 extra I will likely have to pay would be worth it or if getting a normal d15 and waiting is viable.

AMD not even once

>>58898576
>>58898528

got quad core when quad cores were retarded for gaming, and people were seriously talking about dual core being bad but what are you going to do there are no single cores anymore.

7 years later, and only the most recent 2 I want to upgrade, not because of cpu speed, but because i'm on ddr2 and want 64gb of ram so ram is never a fucking factor in 'can i start up this program now or do I need to close a fuckton of shit first so things don't go to page'.

while I'm doing that, may as well pick up an 8 core and have it serve me well for another 7+ years, and possibly have upgrade paths if amd keeps the am4 for a few years.

>>58898897
because they deep dick you FAR less than intel/nvidia

Part of it is necessary, but even when amd was better, they only dicked you on parts that went above and beyond what intel COULD make, and on the gpu side, even when they were 50/50 or near that with nvidia, they still put out the 4850 and 4870 that were better than nvidia's best but a sizeable margin, and put it out for less then half the cost of nvidia.

if shit flips around and amd is the best, we will know for a fact if they are shit bags or not by how they handle things, is most of their tech that can be open source open or is it black boxed?

>>58899533
True, I think though in reality this will never happen.
I dont understand in such a valuable market theres only 2 competitors really

File: AMD-Ryzen-Chart-5.png (12KB, 1219x744px) Image search: [Google]

12KB, 1219x744px

FINISHED AND BANKRUPT

>>58899655

HAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHA
*breathes*
HAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHA

File: C4Xf4YgUcAALBE6.jpg (103KB, 1024x684px) Image search: [Google]

103KB, 1024x684px

oh my

>>58899707
/guts/

>>58896930
CPU hardware is attaining a performance plateau anyway. You have to be a spoiled kid who has no idea about this field to buy into this market and banter on the dick-measuring contest.

It's as stupid as people buying >50000 bucks sportscar or SUV just to drive 10 miles a day between home and work, or between home and the local mall. It's a waste of resources, as much these threads and these subjects on /g/ are a waste of time.

Hmm that looks like they're overclocked and using low end mobo/ram to test it.

>Ryzen 3.4 Ghz
>5960X 4.7 Ghz
>6900K 4.2 Ghz
>5820K 4 Ghz
>9590 5.1 Ghz
>7700K 5 Ghz

Lets look at 6900K 4.2 Ghz vs Ryzen 3.4 Ghz since that's the obvious comparison.

4.2/3.4 = 23.5% clock difference and since these are purely synthetic benchmark we can assume the clock/score difference are roughly linear increase.

The numbers are 2435/2046 = 19%

Per clock basis, Ryzen is stronger. If the Ryzen were 4.2 Ghz it would net about 2527 score.

7700K @ 5Ghz is 2916. Most would probably get around 4.7 Ghz average. Calculating the 6% difference, the score would be 2741. Now if Ryzen were to rise 4.5 Ghz (hopeful, not guaranteed), 4.5/3.4 = 32% increase in score is likely. So, 2046+32% = 2700. Now it becomes margin of error. 1% difference. Still IPC wise, Ryzen wouuld be stronger it would seem.


The only question now is whether or not Ryzen will overclock.

>>58899762
>no fun allowed

>>58899822
INTEL ON SUICIDE WATCH, DUBS CONFIRMED

>>58896930
So you're telling me Ryzen at 3.4ghz is as fast as the previous generation at 5ghz?

>>58896930

Please do well AMD and get intel to stop being so cancerous.

>>58897019
Hahahaha, overclocked Shitel cpus can barely defeat a Ryzen engineering sample. What a joke.

>Intelcucks on suicide watch.

>>58899864
Its faster than 9590 @ 5.1 ghz .

So if the numbers are right and not lies, its a very good CPU.

>>58899866
Oh sweet child, Intel never abandons its jewish antics.

>>58899864
Yep. Pretty much a flat 60% increase across the board vs Piledriver.

when can we expect real benchmarks? i mean all the tech reviewer got them now and some of them are shilling for intel, right?

>>58899946
Even if they got all these sweet SKUs, there's still NDA.

>>58899822
https://www.cpubenchmark.net/singleThread.html

Just so people aren't confused about the numbers I estimated, here's the CPU Mark's single thread scores.

The top score for 7700K is @ ~2600 and the CPU turbos to 4.5 Ghz. The score from OP's benchmark has 7700K @ 5 Ghz and score at 2916. 2916/2600 = 12%. 5/4.5 is 11%.

Each 100 mhz increase the score about ~1%.

So again, if the benchmark is legit and not fake, you can expect the Ryzen IPC to perform roughly same as Skylake clock/clock, if not better.

>>58899946
The reviewers don't have it just yet, maybe next week so they have 2 weeks to benchmark

>>58899964
WCCF is going to try to find and leak something out

OP PIC IS FAKE WITH NO SOURCE. REAL THREAD

>>58899949

OP PIC IS FAKE WITH NO SOURCE. REAL THREAD

>>58899949

OP PIC IS FAKE WITH NO SOURCE. REAL THREAD

>>58899949

OP PIC IS FAKE WITH NO SOURCE. REAL THREAD

>>58899949

OP PIC IS FAKE WITH NO SOURCE. REAL THREAD

>>58899949

>>58899983
Not put in the rest of the tests, not just the ones the Ryzen comes on top.

File: 2960429.jpg (59KB, 1084x481px) Image search: [Google]

59KB, 1084x481px

>>58896930
you are all getting trolled, OPs image is fake, this is the real one. /g/ is full of autists

>>58900000
nice digits

File: AMD-Ryzen-Chart-6.png (13KB, 1214x745px) Image search: [Google]

13KB, 1214x745px

>>58899983
Its not fake, OP's picture is simply not listing the actual overclocks on those CPU.

Here's the non-overclocked (stock+turbo'd) performance from your source.

The problem here still is that Ryzen is base without turboing, meanwhile other CPU are on stock+turbo.

>>58900005
and also, the AMD cpu was at 3.4Ghz and no turbo, in a A320 chipset. So, GG intel is fucked to heaven and back. The 7700k is at 4.2Ghz and has turbo of 4.5Ghz in single thread. The Ryzen chips looks really nice. It all depends on price and max OC

>>58900030
>A320
Couldn't they find a better mobo to bench?

>>58896930
AMD CPU's and Drivers are ass
this is old news
not worth the money

File: Untitled.png (24KB, 1214x745px) Image search: [Google]

24KB, 1214x745px

>>58900013

>>58900042
they'll use whatever they can get their hands on in a pinch.

>>58900051
>drivers
Their drivers are better than ever since king Pajeet returned home.

>>58896930
>Ryzen CPU the same as FX

Who believed this?

>>58900058
Anyway, it looks good. 1700x without turbo trading blows with top Intel chips.

>>58900069
Only morons like you

>>>/v/
>>>/lgbt/
>>>/out/

>>58900069
see >>58899898 and >>58899822

The FX CPU is overclocked, so are all others.

>>58900042
these are leaked benches, also they used ddr4 2400mhz with shitty cas 17 timings, Ryzen is looking like a champ, if the price is right, theres no reason to buy intel this gen.

>>58900090
And there's no reason to buy Intel next gen because it's still Skylake, but with moarcores.

What time is it in israel? They must be working overtime for this level of shilling.

>>58900098
i7 7740K! .2ghz higher clocks, also btw we replaced the god awful TiM on the Kaby Lake chips so your temps are instantly lowered by 20C

>>58900069
It's more like
>5.1Ghz FX the same as 3.4Ghz Ryzen
5.1/3.4 = 150% clock increase for (almost) the same performance

>>58900116
And it's not Socket 2066 where mobos cost a fortune. I'll pass, gimme that sweet 1700 AMD.

>>58900116
I can never forgive Jewtel for not soldering their chips just to shave 0.01 cents off the cost per chip.

>>58900079
>I believe it
>literaly say the opposite in my post

>>>/v/
>>>/pol/
>>>/out/

>>58900126
Not, 150% increase, but 50% increase.

>>58896930
What's bad about this?

It's significantly faster than the fx-9590 while running at significantly slower and cooler

It's retarded because they don't mention the frequency of the AMD cpus

>>58900165
apparently it's not turbo again, basic 3.4

They eiethr hide turbo because it's low, or because it's way too high auto overclock like advertised dependent on your cooling

File: chekkem.gif (4MB, 488x291px) Image search: [Google]

4MB, 488x291px

>>58900000
holy fugg

>>58900126

It also more or less confirms AMD's claim of over 40% IPC increase (yeah that claim wasn't against piledriver but still).

>>58900262
Last time I heard they got 58% IPC increase over Piledriver.

This is tested on a fucking A chipset that can't overclock while everything else tested is overclocked lmao

File: 1485576476450.jpg (17KB, 542x540px) Image search: [Google]

17KB, 542x540px

>>58896930
>Brought to you by Intel(R) Compiler
No thanks
http://www.agner.org/optimize/blog/read.php?i=49

There is no reason to believe non-free benchmarks, especially not the windows ones

File: AMD-Ryzen-Chart-7.png (14KB, 1219x753px) Image search: [Google]

14KB, 1219x753px

These tests are near bogus or something is being hidden/disabled on both or one side of the fense, did nobody mention how Zen's FPU performance is excellent. That should tell you everything.

>>58900324
in 256-bit OPS it's going to hit very close to Intel's own FPU.

File: 31699.jpg (28KB, 600x338px) Image search: [Google]

28KB, 600x338px

>ITT stupid people believing whatever is being thrown at them over the internet
kill yourselves, dipshits

>>58900370
It still needs twice the registers and cycles to do a 256bit op than Intel.

But then again in normalfag workloads like gaymin nothing really hammers the FPU even close to that so these synthetics are pretty shit indication of anything.

A 20% difference in pure synthetics usually shows a 5% difference in gayming.

>>58896930
and this is why I pay extra for intel

>AMD: my le sides lets just stack a bunch of cores and hope for the best xDD

>>58900414
Ryzen has Skylake IPC.

File: 1482994958043.png (157KB, 1008x854px) Image search: [Google]

157KB, 1008x854px

>>58900431
I don't give a shit. I just want my PC to not catch fire during my gayming experience

>>58900449
Ryzen draws less power and emits less heat than any Intel offering.

>>58900449
>how_non_free_software_works.png

>>58900431
ryzen has nothing untill it's out and benched properly

>>58900465
Two weeks left.

>>58900455
>160W
>less than 93W

>>58900473
8core Ryzen part draws ~105w under heavy load compared to ~140w of Broadwell-E.

>>58897310
>>58897336
Shut up child.

File: 1450193493181.jpg (169KB, 720x720px) Image search: [Google]

169KB, 720x720px

>>58896930
Wonder who's behind this post

>>58900491
>comparing ayymd cores to intel cores
Loaded in what way? FPU computing? including iGPU?
>Ryzen part draws
Nice, you're a clueless idiot
read
>>58900401

File: 1470712905701.png (5KB, 727x432px) Image search: [Google]

5KB, 727x432px

AMD btfo

>>58900491
>not understaing what TDP is
kill yourself immediately

>>58900511
>Broadwell-E
>iGPU
Nice damage control though.

File: AMD-Ryzen-7-1700X-CPU-Integer-840x374.jpg (26KB, 840x374px) Image search: [Google]

26KB, 840x374px

Hey what the fuck people isn't the integer performance of Zen fucking phenomenal?

This is by far the most important metric for gayming, FP isn't all that used heavily outside of HPC, which isn't even 15% of the server market.

>>58900576
it's not

>>58900622
Clarify.

>>58896930
Nice shop OP. Fuck off now

>>58899949

>>58899949

>>58899949

>>58900576
>may may non-free benchmarks
>implying any value
kek
Come back in 2 weeks, noob

File: AMD-Ryzen-7-1700X-CPU-Physics-840x374.jpg (23KB, 840x374px) Image search: [Google]

23KB, 840x374px

Meanwhile if the thing about the shitshow RAM being used it would explain the physics scores as that's pretty much a memory benchmark

File: Sandwich.webm (3MB, 480x270px) Image search: [Google]

3MB, 480x270px

>>58900509

>>58900692
Ram wouldn't factor in much. The main factor would be clock differences.

File: images.jpg (8KB, 260x194px) Image search: [Google]

8KB, 260x194px

>$499 CPU
>Single Core Thread speeds 33% slower than a $300 CPU

>>58900750
You're uninformed about this particular subject about this particular workload, please refrain from commenting.

>>58900760
>$1000 CPU
>Single Core Thread speeds 33% slower than a $300 CPU

>>58900657
It's funny how all the Intel scores gained a extra couple hundred points

>>58900764
You're retarded, please neck yourself to save your embarrassment.

>>58900783
It's only 15% slower, you fail at reading charts. Also they are both Intel CPU so its okay.

>>58900808
What's the point of reading a fake chart?

>>58900851
Its not fake for reasons you believe it to be.

The difference is because all other CPU are overclocked and AMD is not.

Read the fucking thread.

>The Stilt that was a year ago shitting on Zen and GloFo is now singing its praises on Inteltech forums

Lmao

>>58900877
>>The difference is because all other CPU are overclocked and AMD is not.
That makes no sense, why would anyone test a bunch of overclocked CPUs against stock? Kill yourself stupid AMDshill

>>58900902

Thats because AMD sent him a sample.

>>58900917
Best goy^Wguy

>>58900917
A320 doesn't support overclocking :)

>>58900917
potentially because Intel paid them to do so, or simply to troll people.

nevertheless, if you know how to interpret these results at all you know they are bogus. the OP's motives don't matter.

>>58900917
Because tester is retarded.

If you check the leak site again, they fix their chart. It now shows the stock cpu's benchmark instead of overclocked cpu's benchmark.

The problem of Ryzen not using turbo mode still stands however.

>>58900917
>That makes no sense, why would anyone test a bunch of overclocked CPUs against stock?
Because the person who posted it is an attention whore fanboy?

Read the thread m8, it really is a comparison between a stock Zen chip with the turbo turned off against a bunch of OC:d intel chips.

>>58900972
> using le :d meme
>le welcome back from le reddit

File: AMD-Ryzen-7-1700X-CPU-Encryption-840x374.jpg (26KB, 840x374px) Image search: [Google]

26KB, 840x374px

Kek, and this isn't even the top end SKU that's clocked higher and has XFR, I can see why Intel is in maximum damage control these last few days.

If anything comes out of this it'll be some good laughter.

File: AMD-Ryzen-Chart-6-1000x614.png (52KB, 1000x614px) Image search: [Google]

52KB, 1000x614px

>>58900808
ITS FAKE HERE THE REAL TEST
>>58899949

>>58900983
Learn 2 read m80... "OC:d" is shorthand for "overclocked"

>>58901029
>le using le 2 instead of le "to"
>le redditor faggot

>>58899655
wasted 8 bilion $ on macafee
Wasted even more $ on trying to force x86 to mobile/ultraportable market by greasing OEMs.
laid of 12k of workers recently.
Anounces "new" CPU as knee jerk reaction to AYYYYMD having something that is competetive in 80% of x86 cpu market............
Qualcomm sampling server grade CPUs..........

Without decent manager your favorite big blue is on the road to loosing significant marketshare. It's like they thought that amd is never going to recover from fuckup that was beting on CMT.

>>58901181
They have a lot of marketshare to lose. That said, even if AMD went out of business another competitor would show up and get them eventually so long as the government didn't intervene to stop them somehow, ISP-style. Nothing lasts forever and that's for the best.

>>58901352

Nobody who buys AMD is going to bother fighting intel at x86 - its too costly for little return unless you can stomp Intel flat (which nobody can do).

File: IMG_8967.jpg (63KB, 540x960px) Image search: [Google]

63KB, 540x960px

>>58896930
Why this Pajeet-test and not the Spec suite? I don't get it. Can't Pajeet compile the Spec tests?

So which one of you is tp4tissue?

File: local-blockbuster-final-days.jpg (233KB, 1306x979px) Image search: [Google]

233KB, 1306x979px

>>58901392
Sorry, my wording was unclear. I meant that if AMD completely disappeared from the world someone would eventually show up with their own company and start trying to compete. One group can't run things forever. They'll get lazy, they won't be kept in line, they won't know what they should be doing because no one forces their hand so they won't always make the "right" innovations or see things coming. They'll make mistakes.

>>58901529
>I meant that if AMD completely disappeared from the world someone would eventually show up with their own company and start trying to compete

And we're telling you that's impossible because only companies like Samsung or Apple have the money to do that and they don't want to.

>>58901529

>would eventually show up with their own company and start trying to compete

No, they wouldn't. Intel would try and block as many patents ass possible and simply make life difficult for this new company which somehow has even deeper pockets. What you want and think should happen is the unlikely scenario.

If it was as easy as you think Samsung would've got in on the act but Intel's stranglehold is so strong nobody can muscle in. Its why the mobile space is such a bloodbath - big companies could (and did) get in early and are fighting tooth and nail because their competitors aren't that far apart from themselves.

AMD (and VIA) really only continue to exist in the x86 space only because Intel allows them - thats how far ahead Intel is right now.

>>58901546
I'd say we should wait and see. I do get where you're coming from but time does interesting things. As it stands people already hold CPUs for 5+ years, so it's not like you're in a rush anyway.

>>58900692
if the physics benchmark was generally bandwidth limited, wouldn't the 4-channel Broadwell-E trounce Ryzen by an even larger degree?

>>58901575
It's not bandwidth limited but latency, seems the Zen setup is running with 17x3 timing and 2T memory.

>>58900799
>>58897019
Intel chips in OP are overclocked
Videocardz did it properly and used stock benchmarks

>Ryzen will save us
>literally bulldozer rehash edition

HAHAHA

>>58901572
>AMD (and VIA) really only continue to exist in the x86 space only because Intel allows them - thats how far ahead Intel is right now.

lol no
Intel lost a large portion of its technological lead by botching 10nm several times and slacking with their architecture development. Zen looks promising enough for the March launch to be the equivalent of the Hundred Days Offensive.

Intel still has the benefit of obscene amounts of money and virtual monopoly on server market, which will take a while to recapture.

>>58901871
You are very dumb

>>58896930

top kek

AMD is so done

>>58901950
>which will take a while to recapture.
Well, not as long as one would think. OEMs got royally fucked over by Intel's shoddy handling of the latest Atom debacle. I've heard that Intel basically threw them under the bus by not admitting fault with their SoCs until the very last minute, and even then said that the manufacturers would have to bear the brunt of the client-end responsibility while they sprinkle money on the situation.
I think Netgear and some other network device company sent out a PR release to their customers pointing the finger at Intel for their $1000+ devices dropping like flies.

>>58898886
Just imagine if those new i5 were clocked at your speed.

>>58901871
>clean sheet design
>bulldozer rehash
what?

>>58902154
>replying to /v/-tier bait

>>58897389
No it doesn't
These are multithreaded tests anon

>I-it's just the memory
>t-the intel chips have a higher clock speed
>y-you'll all seeeee
Fucking lmao are we bulldozer time again

Anyone know of an ITX AM4 that's not fucking made by ASrock or Biostar yet?

>>58902258
Skylake didn't get a legion of ITX boards on launch either, so wait or get a skylake.

Good ITX boards take a while to make, with them being denser and harder to engineer than pleb-tier ATX

>>58902285
>tfw AMD finally realizes what a waste ITX is with regular chipsets and makes their own ITX only "chipsets"
I've heard that some models might not even feature a chipset at all since most of the basic functions are already provided from Ryzen's integrated Southbridge. That would drive down the cots of ITX boards by some degree.

>>58902285
Bioshit said they will have one at launch, are you saying that bioshit is better than ASUS, MSI or Gigabyte?

>>58902338
All of them make a bunch of crap, but they're generally not better.

I'm saying expecting a lot of choice for ITX at launch isn't realistic unless it's a launch of some 25W part.

>>58902338
Don't knock Biostar so much. I have their A4-5000 motherboard and it's been on 24/7 operation for almost a year now with ZERO problems.