Quantcast
[ 3 / biz / cgl / ck / diy / fa / g / ic / jp / lit / sci / tg / vr / vt ] [ index / top / reports / report a bug ] [ 4plebs / archived.moe / rbt ]

Due to resource constraints, /g/ and /tg/ will no longer be archived or available. Other archivers continue to archive these boards.Become a Patron!

/g/ - Technology


View post   

[ Toggle deleted replies ]
File: 222 KB, 718x568, AMD-Ryzen-7-2700X-Tests-03-1.jpg [View same] [iqdb] [saucenao] [google] [report]
65540460 No.65540460 [Reply] [Original] [archived.moe] [rbt]

>> No.65540496
File: 318 KB, 882x758, 1490730560541.jpg [View same] [iqdb] [saucenao] [google] [report]
65540496

>>65540460

>> No.65540617

>>65540460
>how will intel ever recover?
By bribing oems, tech sites and paying idiots to shill for them on the next.
Expecto shiposting as if it's 2004 again.

"Single core cpus are better than dual cores"
--anand pajeet tech circa 2004

>> No.65540624

>>65540617
>next
Net

>> No.65540627

la creatura.....

>> No.65540642

God Intel is shit at multi-core workloads.

>> No.65540665

Will anyone here actually buy this shit? Always wondered that.

>> No.65540680

>>65540665
i want to buy it

>> No.65540694

>>65540680
I do too but I know I'll never buy it. I would like to know if anyone here actually gives a shit enough to go out and get it.

>> No.65540725

>>65540665
No, but only because my SNB i5 still is entirely, perfectly satisfactory for everything that I do. If I were about to build a new system anytime soon, I'd definitely get an R5 2600, but I'm not.

>> No.65540735
File: 40 KB, 720x892, default.jpg [View same] [iqdb] [saucenao] [google] [report]
65540735

>>65540460

call me when amd beats intel in single core performance in NEVER GONNA HAPPEN.

these stupid benchmarks dont even mean anything when intel is always at the top.

>> No.65540742

They'll just fix the benchmarks again and squash this bug right quick.

>> No.65540743

>>65540735
SOPA DELICIA

>> No.65540745

>>65540642
In op's bench intel does better per core than amd due to faster ram.
Amd has been sending out 3600mhz kits with the 2700x and it will end up having more per-core throughout than the 8700k.
Cinebech does depend on mem speed... Not much but a bump from 3200 to 3600 it will give another 2-3% speed advantage.

>> No.65540747

>>65540694
>go out and get it


what's hard about clicking add to cart m8

>> No.65540755
File: 93 KB, 705x740, AMD-Ryzen-7-2700X-Juegos-Full-HD-04-705x740.jpg [View same] [iqdb] [saucenao] [google] [report]
65540755

>>65540735
already has

>> No.65540761

>>65540460
How about both companies produce chips that both have their own pros and cons and consumers should purchase whichever CPU suits their individual needs best? And that both Intel and AMD have a healthy amount of sales and neither are in danger of going out of business?

>> No.65540771

>>65540755

all I see is intel at the top you idiot

amd's best processor of the year cant even beat a pathetic 8400. this is just pure sadness.

>> No.65540807
File: 280 KB, 752x548, 1507216446700.png [View same] [iqdb] [saucenao] [google] [report]
65540807

>> No.65540810

>>65540771
>intel's best processor of the year cant even beat a pathetic 8400. this is just pure sadness.
You know, the 8400 somehow beat the 8700k too.

>> No.65540818
File: 367 KB, 800x800, dat-feel-8-2.png [View same] [iqdb] [saucenao] [google] [report]
65540818

>>65540460
>>65540665
I must admit I had expected a more noticeable increase in clock speed from the 12 nm process. Over the 1800X, the 2700X clocks less than 5% faster, and for an increase in TDP at that. Wasn't the 12 nm process supposed to bring a 10% increase in performance? Is GF being deceptive and lying about things as usual? Is this foreboding something for the """very well performing""" 7 nm process?

>> No.65540828
File: 114 KB, 713x179, fag.png [View same] [iqdb] [saucenao] [google] [report]
65540828

>>65540771
its like you cant read

>> No.65540840

>>65540807
>a piledriver chip beats a 3770k
I was rooting for AMD during the whole Bulldozer era, but let's not get completely retarded.

>> No.65540863
File: 231 KB, 1000x1000, intel pasta.jpg [View same] [iqdb] [saucenao] [google] [report]
65540863

>>65540840

>> No.65540866

>>65540840
It needs to be updated for current gen

>> No.65540872

>>65540810
>>65540828

yes I can read first of all.

the benchmarks show that intel cpu's are at the top and this is amds BEST procesor of the year. the fact that its struggling to beat even coffee lakes mid-low tier cpus is just DISGUSTING.

you do know that intel is also going to be releasing new cpus this year? amd isnt even an option at this point, the 9000 series are going to throw it in the garbage and im going to enjoy watching you all cry for overpaying outdated tech.

>> No.65540875

>>65540863
>piledriver

I repeat, again:
>piledriver

>> No.65540891
File: 31 KB, 650x488, 67b.jpg [View same] [iqdb] [saucenao] [google] [report]
65540891

>>65540875

>> No.65540895

>>65540735

Keep playing your faggy games burger flipper kid.

>> No.65540905

>>65540735
>in NEVER GONNA HAPPEN.
It happened multiple times from 2000 to 2009

>> No.65540913
File: 372 KB, 750x821, ftfy.jpg [View same] [iqdb] [saucenao] [google] [report]
65540913

>>65540872
>the benchmarks show that i5 cpu's are at the top and this is intels BEST procesor of the year. the fact that its struggling to beat even i5 mid-low tier cpus is just DISGUSTING.

>> No.65540921

>>65540872
You are reading a slide where 6700k >8700k
It shows that intel's BEST GAYmen cpu is 7th and amd's unreleased cpu is 2nd with 400mhz less in ram, despite they are shipping 3600mhz with the review kits.
There's a special place for idiots like you in gas chambers.

>> No.65540924

>>65540818
It's about a 10% performance gain. Clocks don't necessarily mean performance.

>> No.65540928
File: 24 KB, 267x297, 7c2.jpg [View same] [iqdb] [saucenao] [google] [report]
65540928

>> No.65540930

>>65540872
lol I do not play games and need multi threaded performance.

Intel is complete shit at multi threaded workloads that it is fucking disgusting.

AMD has the best performance for high workloads.

>> No.65540934

>>65540924
When speaking about the manufacturing process, "clocks" is exactly what performance means.

>> No.65540946
File: 145 KB, 1393x727, intelroadmap.jpg [View same] [iqdb] [saucenao] [google] [report]
65540946

>>65540872
>you do know that intel is also going to be releasing new cpus this year?
Intel isnt releasing shit this year, cry some more intelfaggot

>> No.65540947

>>65540934
It's about a 10% gain in performance regardless of clocks. Latency was improved with memory, so was XFR.

>> No.65540953
File: 84 KB, 653x726, 1497880928990.png [View same] [iqdb] [saucenao] [google] [report]
65540953

>> No.65540964

>>65540930
Even in gsmes amd is better now its why the IIDF has been so active the last few days

>> No.65540966
File: 944 KB, 1228x1502, 1498065865517.png [View same] [iqdb] [saucenao] [google] [report]
65540966

>>65540872

>> No.65540976

>>65540905

FX-55 for life. Single thread beast.

>> No.65540979
File: 864 KB, 1906x1536, 1497840382617.jpg [View same] [iqdb] [saucenao] [google] [report]
65540979

>> No.65540982

>>65540947
But none of those improvements have with the manufacturing process to do. The 10% gains were GF's claims, not AMD's.

>> No.65540984
File: 1.24 MB, 1000x1000, lol.png [View same] [iqdb] [saucenao] [google] [report]
65540984

>>65540979
>>65540966
>>65540953
>>65540928
DELET DIS

>> No.65540989
File: 112 KB, 691x771, 1499963016439.jpg [View same] [iqdb] [saucenao] [google] [report]
65540989

>> No.65540995

>>65540982
Oh. Well, regardless, there's about a 10% performance gain anyway.

>> No.65540999

>>65540946
>Intel isnt releasing shit this year, cry some more intelfaggot
>posts picture where cascade lake release is clearly within this year

>> No.65541008
File: 123 KB, 638x599, 1497916694912.jpg [View same] [iqdb] [saucenao] [google] [report]
65541008

>>65540984
no

>> No.65541014

>>65540999
>Skylake X 2: The Exploding Boogaloo
its nothing

>> No.65541016

>>65541014
You don't know that.

>> No.65541025
File: 2.65 MB, 332x332, 1522105879551.gif [View same] [iqdb] [saucenao] [google] [report]
65541025

>>65540863

>> No.65541029
File: 871 KB, 801x1500, 1492844343742.png [View same] [iqdb] [saucenao] [google] [report]
65541029

>>65540999
what a waste of trips faggot

>> No.65541034

>>65540755
>8400 better than 8700K
Nice test there Ahmed

>> No.65541036

the war between intel and amd will continue going back and forth where one is better than the other for a few years and then vice versa. like it's always been since forever. /thread

>> No.65541045
File: 667 KB, 1548x906, 1515113426097.jpg [View same] [iqdb] [saucenao] [google] [report]
65541045

>>65541016
I know that a overlocked i9 7980XE has a 1000 watt power consumption

>> No.65541051

>>65540999
>will probably change sockets again
TOP KEK
O
P

K
E
K

>> No.65541059 [DELETED] 
File: 246 KB, 882x758, 1498229459814.png [View same] [iqdb] [saucenao] [google] [report]
65541059

>>65541034

>> No.65541064

>>65541025
lol

>> No.65541068

>>65540921
>>65540745
Why are reviewers going out of their way to cuck AMD with less ram speed?

>> No.65541075

>>65541068
to give intel a chance to win something

>> No.65541092

>>65540928
>>65540953
>>65540966
>>65540989

wow congrats on posting shitty memes.

unfortunately it doesnt make amd better than intel in reality. have fun running your pc with shitty silicone faggot. go jerk off to your "more cores".

>> No.65541093

>>65541045
So perhaps that exactly what Intel seeks to remedy with Cascade Lake?

Look, it's not like I want to defend Intel or anything. I'm just saying that it's pretty stupid to claim that they aren't going to release anything this year, when clearly they are. Whether it's good or not is not something any of us knows.

>> No.65541100
File: 136 KB, 682x787, 1503856121352.jpg [View same] [iqdb] [saucenao] [google] [report]
65541100

>>65541092

>> No.65541104

>>65541051
Probably, but that's a different issue.

>> No.65541105
File: 10 KB, 200x200, 1488228868912.jpg [View same] [iqdb] [saucenao] [google] [report]
65541105

>>65541092

>> No.65541109

>>65541068
Well, frankly, because AMD's memory controller is harder to get up to speed. Simple as that.

>> No.65541111

kek intel has had a rough 9 months now with meltdown etc

>> No.65541120

>>65541093
>Cascade Lake
Is on 14++++++++++++^10 like coffee lake, which has been proven to be worse for temps than the normal 14nm used on broadwell E and skylake X, if anything cascade lake will be even more housefire tier than its predecessors

>> No.65541121
File: 43 KB, 630x630, 1497871651906.jpg [View same] [iqdb] [saucenao] [google] [report]
65541121

>>65541111
delet dis

>> No.65541128
File: 67 KB, 1238x846, solarwindws-aws-EC2-Meltdown-patches.png [View same] [iqdb] [saucenao] [google] [report]
65541128

>>65541111
I'm really curious what kind of shoah they're seeing in the datacenter market right now. For sure, if I were running loads on Intel vCPUs right now, I'd be looking pretty hard for an Epyc host.

>> No.65541140

>>65541068
I don't know.
Most of them make benchmarks that don't even make any sense.
Anand pajeet tech is even worse, on ddr4 they use the jedec frequencies which is 2600 tops

>> No.65541146

>>65541120
Did you ever see me sing Cascade Lake's praises? I was only saying that it is a release that is going to happen this year, and that you don't know that it's going to suck.

>> No.65541157

>>65540872
>the benchmarks show that intel cpu's are at the top and this is amds BEST procesor of the year. the fact that its struggling to beat even coffee lakes mid-low tier cpus is just DISGUSTING
>The 8700K is DISGUSTING
>The 6700K is DISGUSTING
>The 8600K is DISGUSTING
>The 7800X is DISGUSTING
Nice logic right there, but if I had to choose an Intel CPU, I will probably still take the 8700. The 8400 have reduced enterprise instructions btw.

>> No.65541163
File: 122 KB, 1500x1054, 1426569813405.jpg [View same] [iqdb] [saucenao] [google] [report]
65541163

>> No.65541165

>>65541051
New sockets are good, because this not only forces the OEMs to dpt and bring out new boards with newer and superior chips, but also allows for more efficient socket design for the current CPU.

>> No.65541167
File: 351 KB, 2122x1415, nuclear-power-plant.jpg [View same] [iqdb] [saucenao] [google] [report]
65541167

>>65541146
It's going to suck power like a thousand Prescotts

>> No.65541176

>>65541165
A socket change a year keeps the soyim in fear

>> No.65541188

>>65541146
>you don't know that it's going to suck
I know its the 6th skylake housefire rebrand so it sucks by default

>> No.65541197

>>65541140
So basically just blatant favourism? But why are normalfag gaymurs not calling this shit out, they should want to go for the best option, shouldn't they?

>> No.65541199
File: 328 KB, 810x587, 1482867301699.jpg [View same] [iqdb] [saucenao] [google] [report]
65541199

>> No.65541216

>>65541111
>meltdown
RYZENFALL REEEEEEEE

>> No.65541221
File: 51 KB, 878x620, amd_retards.jpg [View same] [iqdb] [saucenao] [google] [report]
65541221

>>65541157

oh yeah the new ryzen cpus are so good dude surely they will offer value compared to coffee lake. I wonder how far its going to fall down when intel destroys them with their refresh this year.

>> No.65541225

>>65541140
>Anand pajeet tech is even worse, on ddr4 they use the jedec frequencies which is 2600 tops
They don't do overclocking, quite simply.

>> No.65541231
File: 49 KB, 640x617, 1504438409155.jpg [View same] [iqdb] [saucenao] [google] [report]
65541231

>>65541216
Fake news.

>> No.65541249
File: 213 KB, 1200x629, cpwpdd2eqigx.png [View same] [iqdb] [saucenao] [google] [report]
65541249

>>65541221

>> No.65541264

>>65541197
Favouritism?
Not at all.
The most famous tech sites are financially backed by intel and novidia.
Toms and anand are owned by purch media and they are getting paid to promote intel and novidia products.
One of the hothw gn techspot sites, i don't remember which one, has a page with their cooperations and they claim that novidia is their sponsor.
This is business, there are no fanboys. Intel and novidia has been caught multiple times messing with the sites in order to make them look good.

>> No.65541268

>>65541249
what chart is that?

>> No.65541281
File: 14 KB, 478x523, brainlet.png [View same] [iqdb] [saucenao] [google] [report]
65541281

>>65541268
can't read a chart...
intel shill detected

>> No.65541282

>>65541268
Are you blind? Yahoo Finance 6 month comparison of relative stock change

>> No.65541291

>>65541282
So return percentage?

>> No.65541303

>>65541264
I see, but it really is a big problem that Intel and NVIDIA get away with this shit.

>> No.65541306
File: 123 KB, 240x286, smug-3.png [View same] [iqdb] [saucenao] [google] [report]
65541306

>>65541249
>not using smug lisa

>> No.65541323
File: 62 KB, 400x400, cite-1.jpg [View same] [iqdb] [saucenao] [google] [report]
65541323

>>65541264
>The most famous tech sites are financially backed by intel and novidia.

>> No.65541331

>>65541264
>they are getting paid to promote intel and novidia products
Strange, then, that they keep praising Ryzen in every article they put out about it.

>> No.65541341

>mfw Intel is the one getting shit on for setting people's houses on fire.

Man when did I fall into this bizzaro world

>> No.65541348

>>65541331
>Strange, then, that they keep praising Ryzen in every article they put out about it.
controlled opposition

>> No.65541359

>>65541341
It's the users fault for being too retarded to properly cool their rigs

>> No.65541378

>>65541359
yeah, intel spreading mayonnaise under the IHS has nothing to do with it

>> No.65541381

>>65541359
Intel you can't just take AMD's excuses and apply them to yourself. That's IP theft.

>> No.65541382
File: 120 KB, 964x608, intel.png [View same] [iqdb] [saucenao] [google] [report]
65541382

what did intel mean by this?

>> No.65541398

>>65541378
We re talking enthusiasts here, they should be able to mitigate minor problems like that with cooling, also the coolant isn't so bad. Secondly you should also expect enthusiasts to delid

>> No.65541399

>>65541382
They picked a shit color to represent the countries that have a presence in.

That baby blue blends in with the white.

>> No.65541400

>>65540735
POO

>> No.65541407

>>65541341
What the hell are you talking about, kid? Intel has been in the housefire category for a good couple of generations now.

>> No.65541418
File: 270 KB, 958x336, wtf.png [View same] [iqdb] [saucenao] [google] [report]
65541418

>muh diversity
intel cucks BTFO

>> No.65541431

>>65541323
>i am an idiot who can't into internet
https://en.m.wikipedia.org/wiki/Purch_Group

>> No.65541472

>>65541331
I've seen a lot of shilling but it's under the rag.
Gamernexus shilling
youtube.com/watch?v=-B1OGoVZcUQ
Title: Nearly All CPUs Vulnerable: Explaining Meltdown & Spectre

Facts: meltdown is intel only.
Spectre according to the paper has been confirmed on intel abd some arm chips from samsung and qc. They never confirmed it on amd h/w but speculated it could be done.
It's been months after the meltdown/spectre paper and we haven't seen ONE demonstration on amd h/w.

>> No.65541490

>getting emotionally invested in the jockeying between multibillion dollar companies you have no real connection to

wew lads the autism is real

>> No.65541492

>>65541323
Look up "conflict of interest", fagtron

>> No.65541521

Should I get TR now or wait for TR2?

>> No.65541533

>>65541521
Ofc w8 for tr2.
Better ram support, e.t.c. but you should look into the dates first.
Is it coming soon?

>> No.65541551

>>65541533
I've read it's expected second half of 2018.

>> No.65541574
File: 51 KB, 1000x1000, 1429541406737.png [View same] [iqdb] [saucenao] [google] [report]
65541574

>>65540735
>intel is always at the top
Autistic fanboiism is real, fucking /g/...

>> No.65541584

>>65541574
Its just intel shills they aren't /g/ regulars so to speak

>> No.65541602

>>65541584
>implying /g/ is better
>>65541574
Have you even read this thread?

>> No.65541610

>>65541341
Your newness is showing, since this was the case during the entire P4 era.

>> No.65541634

>>65541348
>media outlets saying good things about intel
>"man, they're just shilling for intel all the time"
>media outlets saying good things about amd
>"man, these shills are just shilling intel by making amd look good"
There's just no way to win, is there?

>> No.65541645

>>65541431
>opens link
>ctrl-f "intel"
>no hits
>ctrl-f "sponsor"
>no hits
Not sure what you're trying to say there, Anon.

Also
>en.m.wikipedia.org
Phoneposters pls leave.

>> No.65541788
File: 162 KB, 1000x1000, 1413913323099.png [View same] [iqdb] [saucenao] [google] [report]
65541788

>>65541602

>> No.65541898

>>65541634
taking the bait this hard

>> No.65542067

>>65541645
http://www.purch.com/experiences/intel/
Connect the dots faggot.

>> No.65542111

>>65541634
What's better marketing?
>amd is dead and finished amd bankrupt, buy intel
Or
>amd is good, they have this and this but you can get a the top product that intel offers.

>> No.65542147

>>65542111
Are you blaming them for not artificially putting AMD on the top of the benchmark charts?

>> No.65542168

>>65542147
He's blaming them for damning with faint praise
>"hurr durr Ryzen is STILL BEHIND Intel in single-threaded perf"
We know that already, why do they need to repeat that in every fucking review?

>> No.65542196

>>65542168
They're also repeating how great value it is, and how good its multi-threaded performance is. They're simply not assuming that every reader has read their every article.

>> No.65542265

>>65540460
Maybe they won't. Maybe they'll go the way of Amdahl, Gateway, Commodore, etc.

>> No.65542266

>>65540735
>being this triggered that the company you are paid to shill for got btfo

>> No.65542321

>>65542265
They'll go the way of DEC

>> No.65542405

>>65540755
this chart is a failure with multiple inconsistences.
Also, we didnt get this already last year with failzen vs coffee lake? Zen literally shit on bed and couldnt even cope with 7700k let alone 8700k.

Zen, Vega, navi, poolaris, the pile of failures just gets bigger. Another failed product

>> No.65542425

>>65540460
Ryzen 5 2600 has 48% better multicore performance per dollar than i7-8700K. With only a 10% drop in absolute performance.

Intelfags btfo.

>> No.65542430

>>65542405
I wonder who's behind this post

>> No.65542436

>>65540665
Really no reason except for cash burners

Zen 2 is going to blow it out of the water soon so no point, they are essentially just releasing it to show intel they can jump right back up to the top again

>> No.65542438

>>65542430
its not me who made a "leak" with 6700k 2 times with different results. Its clearly a joke.

"just wait" like on vega

>> No.65542441

>>65542425
Also, TDP AMD 65W vs 95W Intel housefire.

>> No.65542453

>>65542436
Until AMD dont show decent clocks im not getting any shit from. I mean, i would love to get a cpu with 8 cores, but in the time AMD finally manages to get decent clock and solve those terrible latency and infinity glue issues, im confy on 8700K @ 5.1ghz

>> No.65542454
File: 582 KB, 700x700, 1518839754901.png [View same] [iqdb] [saucenao] [google] [report]
65542454

>>65542425
>2600

>> No.65542475

>>65542453
There is quite literally nothing wrong with latency or any "infinity glue" issues
That was all launch shit

And the lower clockspeeds don't matter if the IPC makes up for it which it does

The "MUH NIGGAHERTZ" meme needs to stop. Hell, one of the primary reasons people buy intel is because "hey man that number is higher"

>> No.65542978
File: 26 KB, 200x200, Holy+i+didn+t+have+that+one+_27f4ea5616d2dbe9b98e1e04f0cfb675.png [View same] [iqdb] [saucenao] [google] [report]
65542978

>>65542430

>> No.65543159
File: 334 KB, 679x702, 1515185367059.png [View same] [iqdb] [saucenao] [google] [report]
65543159

>>65541092

>> No.65543190
File: 117 KB, 800x371, Intel.gif [View same] [iqdb] [saucenao] [google] [report]
65543190

>>65540460

>> No.65543233

>>65540745
um no it doesn't. I have a 8700k with 2666 ram and get a 1420 score like they did with 3600 ram

>> No.65543244

Should I buy a 1600 or pre-order a 2600? The price difference isn't huge, but it's still something to consider for me.

>> No.65543264

>>65543233
Yeah, and I have a 1700X at 4GHz and get 1800 with 3466 memory.

>> No.65543291

>>65543264
Not him but what is your voltage? I kinda want to get 4ghz

>> No.65543310

>>65543291
Sadly it's not stable at 4 for more than benching, at least before my temps get too hot for my liking. I hit 3.9 stable at ~1.32v but can go up to 1.42v without stabilizing at 4.

>> No.65543316

>>65543310
Is there even much of a difference between 3.9 and 4? I didn't really notice much, my 4 isn't stable either.

>> No.65543318

i hate coming to this forum

>> No.65543322

>>65543316
Well, it's a 1/40 clock speed gain. Not gonna make or break anything with that speed difference.

>> No.65543356

>>65543318
>forum

>> No.65543360

>>65543322
it's gonna break his autism.

>> No.65543362

>>65543356
easy bait get fucked /g/ cuck

>> No.65543393

>>65543318
then leave you crybaby bitch

>> No.65543399

>>65543322
Expected as much.

>> No.65543667

>>65543316
>Is there even much of a difference between 3.9 and 4?
yes, huge. Surprised you had to ask.

>> No.65543726

>>65540735
Second generation ryzen matches Intel in IPC. Of course most reviews don't bother to account for clock speed.

>> No.65544271

>>65540460
OY VEYYYYYYYYYYYY

>> No.65544437

>>65540460
>2700X with 3200MHz RAM
>8700k with 3600MHz RAM
Even with this rigged shit coffin lake is shit lmao

>> No.65544505
File: 245 KB, 800x612, 1489160516428.png [View same] [iqdb] [saucenao] [google] [report]
65544505

>>65541490

>> No.65544517

>>65544437
Wait how come there's no Ryzen 2800x's in the comparisons? Aren't the supposed to start at a base clock of 3.8GHZ, turbo boosts to 4.3GHz?

Or should I have posted this in /sqt/

>> No.65544524

>>65544517
They're stockpiling best-binned chips to release 2800x later.

>> No.65545119

>>65540976
then intel paid everyone to not use amd at all, sometimes paying them 106% of what the company made selling intel cpus.

amd literally couldn't give 1 million of their cpus to hp for free as it would cost hp to much money when intel stopped paying them.

Intel did this till amd lost enough money to not have the manpower to make bulldozer what it should have been (there were es's of the cpu doing significantly better but with bugs if I remember right, something the ceo at the time said fuck it ship it now, and the only way to fix it was cripple it)

intel has time and time again shown they are not able to design chips, but they had the money to buy themselves a place, do they still have the money to do that?

>> No.65545165
File: 96 KB, 638x710, 1500260377912.jpg [View same] [iqdb] [saucenao] [google] [report]
65545165

>Multi-core doesn't matter!
>Productivity doesn't matter!
>Price/performance doesn't matter!
>Performance per watt doesn't matter!
>Power usage doesn't matter!
>Temperatures don't matter!
>Soldered dies don't matter!
>Stutters don't matter!
>Streaming doesn't matter!
>Data centers don't matter!
>Locked CPUs don't matter!
>OEMs don't matter!
>Hyperscalers don't matter!
>Upgradeability doesn't matter!
>Anti-competitive business practices don't matter!
>Locked platform features don't matter!
>Synthetic loads don't matter!
>PCI-e lanes don't matter!
>Burnt pins don't matter!
>ECC support doesn't matter!
>*NEW* CPU exploits don't matter!
>*NEW* Games don't matter anymore!

>> No.65545526

>>65544524
>They're stockpiling best-binned chips to release 2800x later.
BS but whatever

>> No.65545623
File: 466 KB, 1200x550, 390.gif [View same] [iqdb] [saucenao] [google] [report]
65545623

>>65540460
And this is just an optimization at 12nm.
Can you imagine 7nm Ryzen 2 next year?
Holy shit it's going to blow Intel out of the water

>> No.65545629
File: 250 KB, 494x700, 1519135751512.png [View same] [iqdb] [saucenao] [google] [report]
65545629

>>65545623
>Can you imagine 7nm Ryzen 2 next year?
prime bully time by then

>> No.65545833

>>65540642

Hyperthreading scales like shit. AMD and IBM do it right.

>> No.65545887
File: 148 KB, 900x1140, 900px-skylake_block_diagram.png [View same] [iqdb] [saucenao] [google] [report]
65545887

>>65545833
Its because intel crams everything onto 1-2 execution ports with a smattering of other functions scattered across the rest of the scheduler crossbar. AMD and IBM spread things out more, and have more of them, which in turn allows for superior SMT scaling.

>> No.65545914

>>65545887

Like I said, AMD and IBM do SMT correctly because they are wide enough that it doesn't cause resource contention and negative scaling scenarios.

>> No.65545938

>>65540460
>comparing 2700X to 1700X
>1800X is the 2700X direct equivalent
SOPA

>> No.65545949

>>65541165
good goy, here is 1 shekel

>> No.65545959

>>65541068
>>65541140
Intel bribes

>> No.65545968

>>65542438
I'm not gonna argue there, somebody definitely fucked something up with the 6700k, but I don't think that you should disregard the entire test. There is some things to take into consideration from El Chapuzos benchmark test. And that is 1700x to 2700x performance, the fact that the entire test is run at 4k high setting, instead of 1080p low and the fact that they are using a 1070ti, I think? I could be wrong, please correct me, if I'm wrong. So with a better card, Intel could still be pulling ahead. So long as the gpu isn't bottlenecked by the CPU and the memory/latencies are just right, we could be hitting a Ryzen sweetspot that allows for AMD to get much closer Intel CPUs, or evens urpass them at 4k. I would like to see more comprehensive tests done with post meltdown/spectre patches on Intel CPUs and Ryxen 1xxx series with latest agesa on x470 boards coupled with high speed, low latency ram vs Ryzen 2xxx series at settings ranging from 720p low to 4k high and with various CPU configurations. Of course, that won't come from a single source, so El Chapuzos findings are definitely relevant and interesting to me, in spite of their fuck ups here and there.

Honestly, was going to wait for Zen 2, Ryzen 3xxx series to replace my FX-8320 and perhaps cheaper DDR4 ram but this might just be the time to upgrade. Hell, it's not like I'm hurting for money, either.

>> No.65545973

>>65541231
>2 scoops
i don't get it, explain plz

>> No.65546014
File: 287 KB, 1170x1974, 1521134686517.jpg [View same] [iqdb] [saucenao] [google] [report]
65546014

>>65541216
>RYZENFALL REEEEEEEE
t.Yahon Eichenblatt

>> No.65546022

>>65545968
>I would like to see
It is 4 days from official release, 4 days from first reviews, and some months before most microcode/BIOS updates and tinkering in the hands of consumers irons out all the details.

I think what we're all really curious about is if AMD changed anything significant, or if this really was just ported to the smaller 7.5track ("12LP") library compared to 14LPPs 9-track

>> No.65546027

>>65540665
I will

>> No.65546109

>>65545973
People were invited to the white house. They got only one scoop of ice cream. Trump took 2.

>> No.65546538

>>65542147
yes, AMD is best at everything, claiming otherwise, you are an intel payed shill

>> No.65546594
File: 123 KB, 659x416, file.png [View same] [iqdb] [saucenao] [google] [report]
65546594

>>65540496
why is the 18 core 7980xe in this "four cores is enough!" meme?

>> No.65546634

>>65540840
>3770k
I had this up until recently when wangblows update caused some fuckery bricking the asus motherboard, CPU itself is probably fine but with the socket a year keeps the goyim in fear its impossible to find a new one second hand
had I gone with the AMD option there are brand new in box old stock AM3+ boards for $100

>> No.65546680

>>65541093
>Intel seeks to remedy with Cascade Lake?
solder or bust

>> No.65546717

>>65543726
there is no practical purpose to normalize for clockspeed. You won't see a 8700k run as slow as the 2700x, or a 2700x run as fast as 8700k.
IPC is nice to know information, but in the end what you want to know is the total performance you get for your money rather than per clock.
And there we will probably again be looking at singlecore going to intlel and and multicore going to AMD (albeit this time by a wider margin)

>> No.65546730

>>65544517
the lineup was slashed down for now, 2700x is top.
Presumably there will still be Threadrippers for the 2xxx generation.

>> No.65546741

Sweet, they only needed 25% more cores for 20% more performance.

>> No.65546748
File: 27 KB, 1353x758, 1493354359032.png [View same] [iqdb] [saucenao] [google] [report]
65546748

>>65546014
>that shutterstock office background
lmao

>> No.65546752
File: 94 KB, 539x538, 13350489050.png [View same] [iqdb] [saucenao] [google] [report]
65546752

>>65540460
Why is the Intel system not running ram like pic?

>> No.65546754

>>65544517
>Ryzen 2800x
doesn't exist this wave is just 2 high ends (2700 & 2700X) and 2 (2600 & 2600X) low ends

>> No.65546767

>>65541140
Jedec updated to 3200

>> No.65546810

>>65546538
I am an AMD fanboy. I'm not gonna ... well, ok, if Intel and nvidia changed business practices and were more consumer friendly than AMD, then I would consider their products, but as of yet, I am a proud owner of a 280x and a not as proud owner of a, FX-8320.

What I keep seeing is that Intel, so far, wins in single threaded performance and there are still multiple workloads that benefit from that single threaded performance as well as AVX ...512? I'm not sure, please educate me on this. Also, a lot of applications have Intel specific optimisations, whereas Ryzen, being a new CPU architecture is not that well supported yet, since nobody the past 10 years or so bought AMD CPUs. This will probably come around, eventually, but not any time soon.

The Ryzen 2xxx series does seem to be closing in on that gap, but there still is a difference. It will come down to personal preference, I guess, to the low end user, but for professionals, it's going to come down a lot towards the applications you are using.

>>65546752
>4133 mhz memory
Because it's $60 even more expensive on a price inflated CPU and an overly expensive chipset, for what amounts to 2-3% better performance? Not worth it, imho. Not that you couldn't or shouldn't, just personal opinion.

>> No.65546876

>>65546810
It's 17-17-17 1T

>> No.65546881

>>65546752

Probably because it will destroy your IMC.

>> No.65546914
File: 369 KB, 1437x874, 13247509457687.png [View same] [iqdb] [saucenao] [google] [report]
65546914

>>65546881
This one is only 4000 12-11-11-240 1T.
8700K Z170

>> No.65546921

Can't Intel just upload an exploit via the Meltdown vulnerability that boosts the Cinebench score to 2000 by editing the config/log file of the results file?

>> No.65546962

>>65546914

>5.4ghz clock on cpu
>4000mhz ram with really tight timings

Yeah, nah. Not for everyday use at least.

>> No.65547062

>>65540460
Why is that so?

>> No.65547082

>>65546914
>ddr4 at 4000 mhz with timings almost as tight as my old 1600 DDR3
where can i get this kind of blue moon unicorn parts?

>> No.65547088

>>65540999
>posts picture where cascade lake release is clearly within this year
Cascade lake is for SERVERS, you retarded piece of shit.

>> No.65547110
File: 51 KB, 283x185, waitasec-1.png [View same] [iqdb] [saucenao] [google] [report]
65547110

>>65547088
>"Desktop Product Roadmap"
>not for desktops

>> No.65547180

>>65547088
Defective cpus that cannot be sold as Xeon Ws become Skylake-X.

>> No.65547228

>>65546810
>I am an AMD fanboy. I'm not gonna ... well, ok, if Intel and nvidia changed business practices and were more consumer friendly than AMD, then I would consider their products, but as of yet, I am a proud owner of a 280x and a not as proud owner of a, FX-8320.
If you want to make your purchase based on what company is less of a jew right now, then that's simple. There is no doubt about if intel and nvidia are scummy.

My biggest gripe with amd fanboys is that take this knowledge and then start shilling for AMD while distorting fact's about products because "hurrr other company bad and lying and jewish, so it's fair if I do some lying myself"

I like facts presented clearly to the consumer, the cons and the pros. And it's up to the consumer to make up their mind what is most important to them and what product to get. For instance, if somebody plays high refresh games and that is their primary interest then suggesting AMD for them, when they could get 8400 or 8700k is deceitful. You can how ever point out that intel is a shit company and that you might not want to support it and if the person thinks that's more important to them, they can make the decision. Not you, not the fanyboys of either company.

I absolutely HATE fanboys, shills, anybody who fucking starts distorting facts just to support their favorite company, for what ever fucking reason. Yet this is how it is it seems. Either your favorite company is perfect and everything they do is perfect and their product is perfect, or it's utter trash and the company is trash and the products are trash.

Fucking all I can imagine is either literal payed shills or immature people who think "it's okay when we use the same tactics, because the other side is evil"

>> No.65547296

>>65546876
Yeah, and at 4133mhz. Those don't come cheap, last I heard.

>> No.65547468

>>65545949
>Everyone I don't agree with is a shill

>> No.65547521

>>65547228
People exaggerating slightly to support their viewpoint is hardly exclusive to AMD vs. Intel threads. Not to say it's a good thing, but if you don't plan on getting used to it, the only thing you can do is stop living.

>> No.65547554
File: 106 KB, 1480x819, 1495060870964.jpg [View same] [iqdb] [saucenao] [google] [report]
65547554

>>65542405
>inb4 you post some benchmarks of some shit setup with 2133mhz ram

>> No.65547633
File: 1.44 MB, 261x253, 1489775419173.gif [View same] [iqdb] [saucenao] [google] [report]
65547633

>>65547554
>Andromeda

>> No.65547695

>>65547554
that image has an alpha hole on the top right where the sun is.
Also what freq does the i7 space heater K runs at?
It is too much washed out to be read.

>> No.65547832

>>65545959
shut up goy

>> No.65547938

>>65540460
If this is real, then that's really good.
8700K is 4.3ghz 6c turbo
2600 is 3.9ghz max 6c 'precision boost 2'

4.3 / 3.9 = 1.103
1277 * 1.103 = 1408.531
1425 - 1408.531 = 16.469 points difference when adjusted for clock.

AMD IPC is literally at CoffeeLake levels.

>> No.65547948

>>65547554
>2.7 FPS lower
FUCKING UNPLAYABLE, AYYYMD FINISHED AND BANKRUPT.

>> No.65548169

>>65547948
7700k is a stutterfest.
Look at the lows... they have 15fps difference from 1700x

>> No.65548172

>>65547938
Just to add though,
We can assume from 2600 and 2700 specs and scores that 3.9ghz is the magic frequency that 100% of dies can attain at all-core turbo/boost at a reasonable voltage.
however
100% of stock 8700K's can do 4.4ghz out of the box - a 12.5% performance increase over 2600X.
100% of de-lidded 8700K's with Kyronaut or liquidmetal can 4.9ghz - a 25% performance increase over 2600X

Intel still has the performance lead, by 12.5% without warranty voiding or 25% with a voided Intel warranty (maybe a SiliconLottery or Derbau8er warranty if purchasing a pre-delidded CPU)

Ryzen 2600 is $199
8700K stock is $349
8700K delid is $419.99 (4.9ghz @ 1.387v -2 avx, from silicon lottery)

8700K stock is 75% more expensive for 12.5% more performance
8700k delid is 120% more expensive for 25% more performance

Buy AMD if you've after value for money
Buy Intel if you need that extra 25% with no expense spared. (although 'no expense spared' would be a 5.2ghz guaranteed part at $659)

>> No.65548251

>>65548172
Is this supposed to be over all performance, or games, or multitask and productivity?
Just saying it's it's for games. You really should compare the 8400 vs 2600 which is priced very similar but still outperforms it as far as I know.

>> No.65548303

>>65548172
>100% of de-lidded 8700K's with Kyronaut or liquidmetal can 4.9ghz
Gee I wonder why silicon lottery sells it them with crippled avx.
100% if you disable half of the chip, delid, repaste and ofc throw a beefy cooler, motherboard and ofc psu.
also don't forget
>Buy Intel if you need that extra 25% with no expense spared.
ofc you could buy a better machine than the shitty 8700k. You can move to hedt with those prices.. but no... inteltards will find always a way to paint a failed intel product as something good.

>We can assume from 2600 and 2700 specs and scores
false assumptions to fit your rhetoric.
Those CPUs are not released yet and you are writing books and giving press conferences about them.

>> No.65548324

>>65548251
This is only comparing the cinebench scores mentioned in OP.
I don't know whether they are real, nor are they a thorough test.

>> No.65548338

>>65548324
Well the only sane thing to do is wait until there is plenty of reviews and people who have it so you can get more info.

>> No.65548357

>>65548338
Whoever said otherwise?

>> No.65548372

>>65548357
I dunno, there is a lot of people here either jumping to preorder the thing, or to condemn it as shit.

>> No.65548395

>>65548303
A lot of salt there, friend.
You sure you don't need to calm down a little?

>> No.65548407

>>65548303
>inteltards will find always a way to paint a failed intel product as something good.
What? You don't like RAID keys?

>> No.65548418

>>65540460
Probably with the next leapfrog iteration such has been happening for the past year or so between Intel and AMD.

>> No.65548419
File: 18 KB, 267x297, 1499956370954.jpg [View same] [iqdb] [saucenao] [google] [report]
65548419

>>65540460
NO NO NO PLS TELL ME THIS ISN'T REAL. STOP POSTING ANTISEMITE

>> No.65548576
File: 246 KB, 960x720, 13294304842523.jpg [View same] [iqdb] [saucenao] [google] [report]
65548576

>>65548303

>> No.65548892

>>65548395
>salt
I am trying since the 1st post in this thread to start a serious conversation or read a serious post and all I see is shitposting.

You must NOT suggest people to buy products with voided warranty, that are impossible to operate under normal conditions(hey summer is coming) and are on a platform that is EOLed in a few months.
There's a certain kind of victims who fall for those things, and I see them circlejerk each other, yet noone dares to admit their shitty purchases.
I've seen intel's sandy bridge and some of the lakes and they are shit.(the 3770k was great though)

If you want to see how unbearable the 8700k at xGHz is, just go and read the review of bitsandchips on 7700k which is supposed to be easier to handle and how it throttles without OC on 50$ coolers.

owning a 8700k is a dangerous sport. either the VRMs are going to blow in a few months or another intel(R) security hole.
If ryzen had half of the defects that the something lake has, everyone would be shitting on it in every website.

>> No.65548953
File: 29 KB, 810x214, hg3h3.png [View same] [iqdb] [saucenao] [google] [report]
65548953

cool they started selling them here already for 400 bucks

>> No.65549170

>>65546914
what does it get on st?

>> No.65549286

>>65548172
>Buy AMD if you've after value for money
Don't forget the upgrade path because Zen 2 will blow past Kaby in just one year. 7+% higher IPC and 28% higher clocks (3.9ghz -> 5ghz)

>> No.65549290

>>65548892
>You must NOT suggest people to buy products with voided warranty,
he's right
that's why i won't buy itnet, it has security issues: meltdown and spectre

>> No.65549408

>>65541398
>"you should expect enthusiasts to delid"
>You should expect the goyim er I mean "enthusiasts" to but specialized delidding tools in order to void their warranty, possibly brick their product in the process.
> All just to apply their own more expensive material in order to mitigate a problem that only exists because we're too cheap to do what we used to do for our entire product stack, and what our only competitor still does.
>Despite still charging a premium for our mainstream CPUs, the unlocked multiplier that is literally just a BIOS switch and the unlocked motherboard chipset you need to use it.
>And after all those extra charges, not to mention the voided warranty, the goyim still fall for our tricks because muh gayms, despite still getting an objectively inferior, overpriced product.
You can literally drop TWENTY DEGREES CELSIUS from your CPU package temperature with a delid. Sometimes more than that.
Kill yourself.

>> No.65549431
File: 436 KB, 1612x829, grizzlyconductonaut.jpg [View same] [iqdb] [saucenao] [google] [report]
65549431

>>65549408
Better than solder on high wattage cpus ( 6950X / 6c drop ) vs 1800X's 1-2c

>> No.65549446

>>65549431
The 6950X was soldered, right? So even when they were using solder, they were using shitty solder.
The absolute state of intel cost-cutting price-gouging.

>> No.65549462
File: 20 KB, 500x375, 1482085473183.jpg [View same] [iqdb] [saucenao] [google] [report]
65549462

>>65549431

>TDP: 1092.0 watts

>> No.65549477

>>65549446
10 core with higher power per core vs low power 8 core

>> No.65549485

>>65540665

I'm waiting for the reviews and benchmarks after the official release, but if its competetive, I'll buy it.

>> No.65549503

>>65549485
It's not. Most programs still only use up to 8 threads, which means that the 2700X is still going to get beaten by the 8400 thanks to its lower IPC and sustained clock speed

>> No.65549515

>>65549503
The IPC isn't lower though.

>> No.65549526

>>65549515
But it is. Ryzen falls way behind Skylake, Kaby Lake, and Coffee Lake for single-thread-heavy games and about 20% worse at Adobe editing software that only scales up to 6 cores.

>> No.65549640

>>65548953
>4.35
I take it we should be able to push it to 4.4

>> No.65549676

>>65549640
Nope, 4.2GHz is a hard limit across all cores
Kek

>> No.65549678

>>65549408
You absolute retard. Tell me, why should Intel put better thermal paste or even solder on their CPUs if they're going to get delidded anyway?

>> No.65549687

>>65549503
Ryzen has superior IPC throughput, not to mention 8400 is locked garbage.

>> No.65549698

>how will intel ever recover?
as usual, by shilling harder.

>> No.65549700

>>65549687
Ivy Bridge<Ryzen+<Haswell<Broadwell<Skylake=Kaby Lake<Coffee Lake

>> No.65549703

>>65549678
I never delidded my 2500k because it never needed you fucking shill.

>> No.65549712

>>65549676
4.4 on one core is viable

>> No.65549716

>>65549678
Because a soldered IHS wouldn't need to be delidded? Are you retarded?

>>65549526
Another retard. Adobe software isn't optimized to anything but Intel.

>> No.65549718

>>65549703
>Never needing to delid
>Complain about subpar thermal paste
You've just invalidated your own argument.

>> No.65549729

>>65549712
kek, isn't AMD all about moar cores? What a stupid "feature", that's like having a Ferrari that has one cylinder providing all of the power

>> No.65549734

>>65549700
Ryzen > shit > ARM > Intel

>> No.65549740

>>65549718
I wasn't that anon you were replying to.

>> No.65549741

>>65549700
>Ryzen+ below Haslel
wew lad, I hope you get your shekels anyway.

>> No.65549753

>>65549716
>b-b-but muh optimization
Classic AMDrone excuse. If AMD made a good architecture for once in their lives, they would be able to take advantage of the same optimizations.

>> No.65549758

>>65549716
>Because a soldered IHS wouldn't need to be delidded? Are you retarded?
Would you rather have everybody pay an additional premium for their CPUs, or just have the autists that want to delid their shit?

>>65549740
Doesn't make your statement any more valid.

>> No.65549759

>>65549729
How hard does the word Epyc trigger Intel shills right now?

>> No.65549773

>>65549759
Not very because EPYC's adoption rate is still in the single digits. No data center worth their weight in gold will ever switch to an unproven, unreliable, and unsupported vendor.

>> No.65549774

>>65549753
kek, you aren't even trying anymore.

>> No.65549778

Yeah 8 vs 6 cores either show me a single core bench or kindly fuck off

>> No.65549789

>>65549773
And yet 1 Epyc > 2 Xeons lul

>> No.65549791

>>65549759
Not at all, because Intel is still superior regarding the server market, just look at the market shares. Would you rather go with known-good hardware or unknown pooware?

>> No.65549792
File: 305 KB, 962x720, 1522081513482.png [View same] [iqdb] [saucenao] [google] [report]
65549792

>>65549758
I never complained about subpar thermnal paste keep on shilling.

>> No.65549797

>>65549774
Doesn't make it any less true. If AMD wants to be as good as Intel, it needs to be able to take advantage of the same instruction sets...

>> No.65549810

>>65549526
IPC doesn't mean single thread performance.

>> No.65549814

>>65549526

>knows more than Agner Fog

Seems legit.

>> No.65549815

>>65549792
Then what are you arguing for in the first place?

>> No.65549819

>>65549789
And yet enterprise software licensing charges per core now, not per socket. EPYC has no reason to exist other than some super niche apllications.

>> No.65549823

>>65549791
Literally everybody is abandoning Xeons because of meltdown shit, not to mention one Epyc beats two Xeons

>> No.65549839

>>65549819
So that's why the entire planet is throwing worthless Xeons into the trash bin because of fatal cpu flaws lul

>> No.65549840

>>65549823
>Literally everybody is abandoning Xeons because of meltdown shit
Source: [my ass]

>> No.65549847
File: 24 KB, 419x420, 1521491188452.jpg [View same] [iqdb] [saucenao] [google] [report]
65549847

>>65549815
>Tell me, why should Intel put better thermal paste or even solder on their CPUs if they're going to get delidded anyway?

>> No.65549853

>>65549797
This is impossible, unless AMD makes an exact copy of Intel's curren uarch. Zen is closer to a RISC CPU than CISC.

>> No.65549859

>>65549823
>Literally everybody is abandoning Xeons because of meltdown shit
Not true. Digital Ocean replaced all of their vulnerable Xeon platforms with Xeon Platinum and Xeon Golds. Amazon upgraded their existing servers with Intels again. Azure did the same, but bought a handful of EPYCs for evalutaion and market testing (it's not going so well for EPYC, fyi... Not enough performance for the money)

>> No.65549868

>>65549847
Keep on being retarded you AMD shill

>> No.65549869

>>65549853
>Zen is closer to a RISC CPU than CISC.
Got an actual source for that?

>> No.65549876

>>65549840
Source: CES

>> No.65549882
File: 132 KB, 354x233, The mouse.png [View same] [iqdb] [saucenao] [google] [report]
65549882

>>65549868
Use a 2500k i'm an amd shill k.

>> No.65549891

>>65549859
>Not true
Ok, 99% of people are abandoning Xeons. The 1% are just delusional fanboys.

>> No.65549901

>>65549882
Even AMD shills know their platform isn't worth shit and they have to fallback to Intel.

>> No.65549904

>>65549859
Intel still doesn't have a proper patch for meltdown so you're lying.

>> No.65549924

>>65549891
>Digital Ocean, Amazon and Microsoft are delusional fanboys
Whatever helps the AMD shill sleep at night.

>> No.65549925

>>65549839
Nope, you've never set foot in a server room before, let alone approve RFPs for large scale purchases of equipment.
It's far cheaper and easier to go from pre-Haswell Xeons to Skylake Xeons since the vendors are the same, the platform has maturity and wide-spread support, and the platform is reliable.
EPYC would lead to vendor changes (as well as expensive support structure and contract changes), higher costs of adoption caused by extra manhours needed to make custom applications work on a different architecture, and even higher costs associated with the downtime needed to test and deploy a virtually untested family of equipment.

EPYC is a horrible idea and does not deserve to be anywhere near a well-managed IT infrastructure.

>> No.65549932

>>65540771
>>65548251
Indeed. This shill is everywhere.

>> No.65549936
File: 115 KB, 1281x720, AMD-EPYC-7000-Tech-Report-slide-15.jpg [View same] [iqdb] [saucenao] [google] [report]
65549936

>>65549924
Stay delusional.

>> No.65549947

>>65549925
All the vendors are the same with EPYC too...
Did you forget HP is the largest server vendor and has been for 12 years running now...
Did you forget HP does **5 version of every proliant...

>> No.65549959

>>65549925
Xeon is worthless garbage, your butthurt won't ever change this.

>> No.65549961

>>65540665
The most someone on /g/ will typically spend on a CPU is 200-300.

>> No.65549965

>>65549904
Intel not having patches for Meltdown is a lie itself. Plus, it is cheaper to replace older equipment that had those vulnerabilities than to take those hardware down in order to wait for patches to release. Downtime is hugely expensive for _AAS providers for a variety of reasons, so purchasing known good hardware is cheaper than waiting for patches.

>> No.65549985

>>65549965
Well good thing nobody would want to get fired so they go with Epyc now since it's infinitely superior.

>> No.65549992
File: 12 KB, 259x194, boo.png [View same] [iqdb] [saucenao] [google] [report]
65549992

>>65549925
>reliable

>> No.65550005

No matter what. Ryzen+ will beat Intel on productivity tests and come damn close on games.

>> No.65550008

>>65549925
Now, that's a legit high level shill.
On a fuckind sunday. Told ya intel would ramp up the shilling.

>> No.65550014

>>65549891
Why would you call billion dollar companies who signed contracts over thousands of CPUs across multiple years fanboys for getting their replacement deal under warranty when they learn the other party fucked up?

>> No.65550016

>>65549947
>All the vendors are the same with EPYC too
Not so. Some vendors do not have 1-to-1 EPYC replacements or counterparts of their existing Xeon product range, which means shopping around for the vendors that do have EPYC in the configuration that they need. Not only that, but support for those hardware is usually contracted to third parties that aren't the vendor. Changing service contracts mid-contract is expensive and time-consuming, not to mention a headache for bug fixes, patching, and maintenance.

But not knowing any of this clearly marks you as a basement-dwelling /g/nome who has never worked in a non-retail position.

>> No.65550017

>>65550005
DELID THIS

>> No.65550025

>>65549992
>This retarded meme again
>What is mitigation

>> No.65550027

>>65549992
Intel shills are the most retarded life-forms in the universe, what do you expect?

>> No.65550032

>ITT: AMDrones pretend they know enough about enterprise IT infrastructure to defend a horrible product

>> No.65550040
File: 64 KB, 691x771, 1515145992968.jpg [View same] [iqdb] [saucenao] [google] [report]
65550040

>>65550008
SHUT UP

>> No.65550056

>>65550014
Meanwhile >>65549936

>> No.65550076

>>65550056
That slide isn't a contract agreement for services and/or products for a particular company. Stay delusional, AMDrone. EPYC adoption will continue to remain in the low teens or single digits for years to come.

>> No.65550081

>>65550025
>spectre
>mitigation
Literally retarded, kill yourself.

>> No.65550086

>>65550081
>Spectre
>AMD unaffected
You're the retard

>> No.65550090

>>65550016
Did you miss the part about HPE and HPE being the largest server vendor
Don't need to change anything, just option out for a DL485 instead of a DL480.

>> No.65550107

>>65550086
Zen uarch is invulnerable.

>> No.65550110

>>65550025
>meme
yes, mr Shill, the absolutely worst design flaw in the entire history of the semiconductor business is nothing more than a meme.
>mitigation
google: intel to halt support for spectre/meltdown. I'll wait.

>> No.65550130
File: 200 KB, 1347x1024, Screenshot from 2018-04-16 02-11-26.png [View same] [iqdb] [saucenao] [google] [report]
65550130

>>65550086
forgot pic

>> No.65550131

>>65550110
>intel to halt support for spectre/meltdown
Braindead intel macacos permanently BTFO

>> No.65550133

>>65550110
That's overblowing it, nothing will be worse than IEEE1394.

>> No.65550142

>>65550090
Switching to a new platform is not as simple as changing SKUs on an order form, dumbass.

Also, even HPE do not have a full line-up of EPYC products to replace all Xeon variants that they have sold over the years. Not only that, but HPE doesn't even provide the same support services for EPYC platforms due to its immaturity. Source: fifteen multi-hour long conversations with HPE and third party sellers.

>>65550107
Zen and every OOB execution capable CPU is vulnerable, dumbass.

>> No.65550157

>>65550142
>Zen and every OOB execution capable CPU is vulnerable, dumbass.
[citation needed]

>> No.65550163

>>65550086
i don't give a fuck about vulnerabilities that are only exploitable when the hacker is literally right there and might as well carry the fucking PC away

>> No.65550171

>>65550157
Did you even read the white papers? Oh wait, no you haven't because you get all your tech news from WCurryFTech

>> No.65550186

>>65550163
>Spectre requires an attacker to sit right next to an AyyMD CPU
The absolute state of the AMD shills

>> No.65550197

>>65550163
You're getting your vulnerabilities mixed up there, kiddo

>> No.65550199

>>65550171
>doesn't even know what a PoC is
See, Krzanich, this is what you get for hiring pajeets to shill your shitty products

>> No.65550206

>>65550142
>Switching to a new platform is not as simple as changing SKUs on an order form, dumbass.
Entirely depends on what your doing.
Windows DC - swap it out
Some web shit - swap it out
vdi - swap that shit out.
filesystem - needs more validation
some science shit - years of validation which Skylake-E won't have even completed yet

>> No.65550210

Whenc omcparing MCE to XFR which way do you swing on which is 'right'? MCE is not something advertised by Intel. It's purely a motherboard branded auto overclock.

On the other hand XFR is advertised by AMD as a feature and is part of the whole spec.

MCE auto boosts 8700K to 4.7Ghz.
XFR should boost to at least 4.2Ghz and possibly 4.35Ghz on good silicon.

Both use more power than the base spec. But which do you feel is a 'fair' way to test performance?

>> No.65550214

>>65550199
Spectre Variant 2 had working PoC on Zen-architecture CPUs. So...

>> No.65550235

>>65550197
>>65550186
it's just going to turn my noscript off before running the javascript code?

>> No.65550237

>>65540460
Where are the high end AMD boards?

>> No.65550250

>>65550237
>>65547580

>> No.65550263

>>65550235
>The only thing I do is browsing the web

>> No.65550325

>>65550210
No CPU overclock at all :^)

>> No.65550341

>>65550263
What else is going to run the spectre code Mr Hackerman sir

>> No.65550365

The problem here is as always. AMD have to brute force in hardware what Intel and Nvidia get for free through developer optimizations. All the while developers are lazy cunts who only bother with specialized Intel and Nvidia instruction sets AMD has to rely on sheer brute force. All things being equal AMD would be on the same level or higher if code was written fairly and balanced.

It's the reason companies like Amazon upgraded vulnerable CPU's to more Intel CPU's instead of switching the entire architecture and codebase of their server software. It's just not as cost efficient.

>> No.65550369

>>65550341
>What else could possibly run malicious code on my computer
>I would never expect an ACE-vulnerability in any of the software I'm running
>I would never run anything in an unprivileged mode and expect it to stay that way

>> No.65550376

>>65550250
I said high end, not budget boards

>> No.65550382

>>65550032
Says the FUCKWIT

>> No.65550418

>>65550382
>not even 9am Monday morning and we already that this week's Friday Fuckwit.

>> No.65550419

>>65550376
X is high end, B is budget...

>> No.65550420

>>65550419
Well, it still looks like budget crap

>> No.65550433

>>65550420

If it added more RGB would that satisfy your needs?

>> No.65550444

>>65550433
No, they need to add more quality parts, better voltage regulators, and more features from the chipset

>> No.65550458

>>65550444
You want to pay for RAID keys?

>> No.65550463
File: 318 KB, 870x1271, r013.jpg [View same] [iqdb] [saucenao] [google] [report]
65550463

>>65550444

>SoC
>more features from chipset

>> No.65550469

>>65550458
yes, master, please change my socket and bleed my wallet

>> No.65550484

>>65550369
You can do anything after gaining access with elevated privileges, spectre fucking around with branch prediction is probably the least of worries after that is already done...

>> No.65550494

>>65550484
You absolute mong
I was saying you can gain elevated privileges

>> No.65550524

>>65550458
>AMDrone bitching about paying for features that 0.01% of the market will ever need
RAID keys only exist on Skylake-X platforms because of VRoC for far better RAID performance with NVMe drives. You can still RAID SATA and NVMe SSDs on LGA1151 motherboards without VRoC

>> No.65550590

>>65550524
You seem upset, intlelcuck.

>> No.65550827
File: 216 KB, 393x391, 1502992496264.png [View same] [iqdb] [saucenao] [google] [report]
65550827

>>65540735

>laughs in Threadripper 1950x @ 4.3ghz on water

>> No.65551005
File: 31 KB, 1000x421, AMD_Interactive_Stock_Chart_Advanced_Micro_Devices,_Inc._Stock_-_Yahoo_Finance_-_2018-04-15_19.21.25.jpg [View same] [iqdb] [saucenao] [google] [report]
65551005

As much as I would like to see AMD winning it's just not going to happen.

>> No.65551460

>>65551005
ahahaha... now the discussion has left the cpus behind.
Ryzen sucks, coz stock market.
ahahahah.

>> No.65551504
File: 35 KB, 453x317, 9-18-10-intel600.jpg [View same] [iqdb] [saucenao] [google] [report]
65551504

>>65550524
hey man, do you know if I can download some performance for my i-cuck cpu?
The keys are expensive. I am looking for a key-gen or something.

>> No.65552344

>>65549925

> Marketing shill who never has been an actual server room or dealt with IT projects

>> No.65552433

>>65552344
Dat projection tho

>> No.65552461

>>65549773

> Completely ignores mid to late 2000s where AMD was dominating 86 server/datacenter space with Operton family

Sorry, shill. Intel knows that their server/workstation divisions are screwed until their server EMIB SKUs are ready. Their marketing press have been nothing but 1990s era IBM-tier FUD campaigns.

Intel gambled too much on their fab tech to overcome the inherit difficulties in making massive pieces of silicon. Ring-bus topology loses its scalability beyond 6 to 8 cores.

>> No.65552512

>>65552433

Nah, the real-world server/datacenter far are more concerned about density, TCO and licensing then raw performance.

Intel platforms ever since Haswell-EP have been coming up short for most server/datacenter tier shit. They only have an advantage in HPC-related shit (AVX2 and AVX512).

Intel marketing has been scared witless ever since first public announcement of EPYC family.

>> No.65552547

>>65552512
>the real-world server/datacenter far are more concerned about density, TCO and licensing then raw performance
Gee, it's a good thing I never once mention the word "performance" in any of the reasons I listed for EPYC being an epic failure.
>Intel platforms ever since Haswell-EP have been coming up short for most server/datacenter tier shit
>aka I have no idea what I'm talking about
>Intel marketing has been scared witless ever since first public announcement of EPYC family.
Baseless speculation. And you expect me to believe you work with a data center?

>>
Name (leave empty)
Comment (leave empty)
Name
E-mail
Subject
Comment
Password [?]Password used for file deletion.
Captcha
Action