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: 38 KB, 740x416, AMD-Zen-2-Feature-740x416.jpg [View same] [iqdb] [saucenao] [google] [report]
67827335 No.67827335 [Reply] [Original] [archived.moe] [rbt]

How fucked is Intel?

>> No.67827357

>>67827335
Yeah but just wait till the 6 NANOMETER chip!!!

HOLY SHIT I GET A BONER JUST THINKING ABOUT IT

>> No.67827392

>>67827335
Pretty fucked.
Intel is going to hold on to their lead in FPU performance with their higher peak throughput and larger data paths, but they won't have much beyond that. AMD makes even a tiny IPC uplift and intel won't be a clear leader in any benchmarks without a huge clockspeed advantage.
On the enterprise front AMD is going to completely dominate.

>> No.67827399

>>67827335
Intel's 10nm is on par with TSMC 5nm. So AMD is the one getting fucked here

>> No.67827409

>>67827335
Intel zoycucks on suicide watch.

>> No.67827415

>>67827399
Intel´s 10nm doesn´t exist.

>> No.67827422

>>67827399
By current roadmaps... 3nm TSMC will be out by then.

>> No.67827433
File: 118 KB, 1500x650, rqaGg9e.png [View same] [iqdb] [saucenao] [google] [report]
67827433

>>67827335
Very fucked. The price of i7 8700K has gone up 100 bucks in Sweden in the last 2 weeks and retail stores are setting limits on how many each person can buy. And Sweden is already cucked by 25% VAT so these new prices are fucking insane for 6c/12t processors. They cost as much as the rumoured prices of the i9 9900K atm lmao

pic kind of related; a retailer in Germany made some graphs

>> No.67827442

>>67827399
50000000000 rupees ($0.00) has been deposited into your account.

>> No.67827443

>>67827399
but it doesn't work

>> No.67827462
File: 262 KB, 959x642, 2018-09-28 10_48_33.png [View same] [iqdb] [saucenao] [google] [report]
67827462

>>67827433
Actually, I just converted the swedish price to american dollar, and the 8700K in Sweden costs 510 USD right fuckig now holy shit

9900K is rumoured to be 450 I believe. So the 9900K will probably be 700USD here xD

>> No.67827497

>>67827399
>tsmc borrowing from intel's playbook making their new node even slower than 14nm

bold strategy cotton let's see if it pays off for them

>> No.67827511
File: 98 KB, 1109x852, Intel-Core-i9-9900K-Core-i7-9700K-CPU-Prices.jpg [View same] [iqdb] [saucenao] [google] [report]
67827511

>>67827462
>9900K is rumoured to be 450 I believe
you're like a little goy, watch this!

>> No.67827519

>>67827433
why is the vat so high in most european countries? welfare state? 7.7% here but they find ways to justify increased prices over US prices anyway. 9900k is going to come in $600+ in europe.

>> No.67827547

>>67827433
>Mindfactory
Fuck off, poorfag. Chads buy at Caseking

>> No.67827559
File: 584 KB, 486x810, 1519219812179.png [View same] [iqdb] [saucenao] [google] [report]
67827559

>>67827357
God I love waiting for tech so much.
I get a massive erection just thinking about not buying anything again next year in favour of 2020's line-up.

>> No.67827567
File: 1.49 MB, 2048x1536, ebyn.jpg [View same] [iqdb] [saucenao] [google] [report]
67827567

>>67827399
meanwhile in reality
>10nm is dead
>12nm in
>even then: intel claimed first and 2nd gen 10nm would be alot slower than 14nm+++++++
can't wait for virgin Cope Lake (Emergency Edition™) benchmarks vs CHAD Epyc 2 CPU

>> No.67827593
File: 60 KB, 800x600, 1460958982825.jpg [View same] [iqdb] [saucenao] [google] [report]
67827593

>>67827399
I've been seeing this a lot for the past few weeks and don't know anymore if it's just trolling or a real attempt by incel shills to get this bullshit into the "common knowledge" domain
I can see it unfolding before my eyes
>be year 2024
>normalfag coworkers discussing cpus
>bro, intel 10nm is comming next year and it will be better than TSMC 3nm
>dude, nanometers is just marketing, except for intel because they always tell the truth
alas, the ride never ends

>> No.67827607

Exactly a year from now Intel is going to come forward saying they are ceasing 10nm research and production due to it being unprofitable and stick to 14nm forever

>> No.67827637

>>67827593
they can't because their shares would get a 50% hit

>> No.67827650

>>67827637 meant for >>67827607

>> No.67827690

>>67827593
It makes for top tier bait posting. Intel fanboys genuinely do post shit like that all over, a famous one is Weitkin or something like that.
For years he spammed Anandtech's forum, EE Times, and a dozen other websites with thread and comments like
>only intel 14nm is real 14nm

He was dead serious about it all too. Conveniently ignoring how poor intel's 22nm node was on density. Not that any of that ever mattered. Shills are immune to logic.

>> No.67827704

>>67827335
Poozen will be so far behind Intel when the 9900K releases that they'll need a 50%+ bump in performance just to draw level. It's not going to happen.

>> No.67827864
File: 59 KB, 1151x809, ScreenShot_20180927115843.png [View same] [iqdb] [saucenao] [google] [report]
67827864

>>67827704
>>67827704
>they'll need a 50%+ bump in performance just to draw level
yes, we all know that's how the intel compiler works. how else would they make people buy newer cpus?

>> No.67827892

>>67827335
intel is fucked by arm, not amd.

>> No.67827918

>>67827864
>how the intel compiler works
I'm not sure how much there actually is to this but it's is interesting that Linux vs Windows benchmarks always seem to show a performance gap and sometimes it's pretty big. everything on the linux side tends to be compiled with gcc. compiler isn't the whole story on linux vs windows benchmarks, though, since there's usually also a performance gap on amd cpus too. it's usually smaller than it is on the intel side but it's there

>> No.67827924

>>67827559
Why though? I just buy it and still get a boner thinking about the next thing.

>> No.67827966
File: 964 KB, 2085x1130, 1538084215624.png [View same] [iqdb] [saucenao] [google] [report]
67827966

>>67827918
have a look at this bench then. why would 7600k score so inexplicably bad compared to the 8600k or 7700k?

>> No.67827968

>>67827519
because it doesnt cost us £40k to see a doctor.

>> No.67827992
File: 2.92 MB, 220x220, 1534616045122.gif [View same] [iqdb] [saucenao] [google] [report]
67827992

>>67827335
Not particularly fucked. There are OEMs coerced and locked-into contracts where Intel can give them stinking dog turds with a BGA on the bottom, they still have to sell them, and normies, shills and corporate will buy them.

>> No.67827996

>>67827559
This is the best (You) I have ever gotten.

>> No.67828002

>>67827966
Because BFV is capable of using more than four threads? Surely that should be obvious.

>> No.67828023

>>67827399
>>67827415
>>67827567
>>67827607
>>67827690
Here's what I'd really like a strait and serious answer to, not that I expect /g/ to be able to actually answer but still: What's up with Intel's 14nm node?

This month we've had several news stories about high-end server manufacturers who are unable to buy Intel chips & are recommend that customers buy servers with AMD chips or older generation Intel chips. And there's also been news stories about Intel being able to deliver laptop chips and this is apparently so bad that many speculate the shortage of Intel laptop CPUs will cause DRAM prices to go down. This at a time when AMD has been taking market share from Intel.

Both these high-end server chips and lower-end laptop chips are made on Intel's 14nm node which has been in use for a very long time. Yet suddenly they have these huge delivery problems?

I would fully understand why Intel would be in this situation if there was some major disaster event like an earthquake taking out one of their factories or something like that. No such thing happened.

>>67827547
paying an extra premium to buy the exact same product doesn't make you rich, just stupid

>> No.67828025
File: 53 KB, 674x800, ipc.png [View same] [iqdb] [saucenao] [google] [report]
67828025

>>67828002
then why would 2600x and 2700x score so poorly? given how filthy and greedy EA is I'd say there's a high likelihood they accepted shekels from intel

>> No.67828033

>>67827966
are you some kind of stupid shill just replying with copypaste without reading what you're replying to?

>> No.67828078

>>67828025
>then why would 2600x and 2700x score so poorly?
Because Poozen is fucking trash for gaming and always has been? It isn't just about core IPC in synthetic tests and clock speed. Poozen is massively hamstrung by its inherent latency due to its architecture. It'll never be good for gaming.

>> No.67828094

>>67828078
but muh high yields, muh high cores

>> No.67828105
File: 1.45 MB, 288x198, 1528035758536.gif [View same] [iqdb] [saucenao] [google] [report]
67828105

>>67828078
How much are Intel paying you? Whatever it is, you deserve more for having to embarrass yourself so disgustingly in front of everyone.

>> No.67828128

>>67828025
>then why would 2600x and 2700x score so poorly?
It's proprietary code, we can't know that. Compiler problem is a probable cause.

>> No.67828164

>>67828023
Because Intel expected 10nm to be working on time. They already moved their chipset to 14nm, they already have some fab ready for 10nm. They're fucked up thanks to bad management.

>> No.67828165

>>67827335
Intel has the cash and it has the market share. AMD has shiny promises.

>> No.67828176

>>67828078
This.

>b-but monolithic design is dead!!
Yeah faggot let me know when developers start to optimize for Ryzen and then I'll buy it. I'm not buying something with a promise that it'll perform better in the future. The architecture itself has latency issues and gets top of the line CPU gets regularly BTFO by low-end Intel lineup, fucking embarrasing.

>> No.67828179

>>67828023
That's easy. Intel didn't invest in more 14nm manufacturing capacity because they thought they could get 10nm working in a reasonable timeframe.
Instead everything they made on 14nm is still stuck on 14nm, plus all the new 14nm orders, which means they have a huge capacity problem that's not going away any time soon because it takes years to build new foundries. Ironically, they might even make 10nm work by that time.
It's also why they're moving auxiliary things like chipsets back to 22nm, and outsourcing to TSMC.

>> No.67828183

>>67828078
t. proud owner of a quad core i5 who closes every program when playing his games like a good boy

>> No.67828195

>>67828176
>The architecture itself has latency issues and gets top of the line CPU gets regularly BTFO by low-end Intel lineup, fucking embarrasing.

Why Intlel cucks being this delusional?

>> No.67828207
File: 86 KB, 400x518, 1464865087183.png [View same] [iqdb] [saucenao] [google] [report]
67828207

>>67828183
s/boy/goy/

>> No.67828218

>>67828195
Rationalisation.
>ZOMG I coulda bought a ryzen that would have gotten within 2fps of my /v/idya gayme for half the money I got off daddy

>> No.67828238

>>67828023
they're trying to manufacture bigger dies (for coffee lake-s and whatever the codename is for their upcoming server shit) and it's affecting yields.

>> No.67828268
File: 10 KB, 514x269, Deus_Ex_Mankind_Divided_average_fps.png [View same] [iqdb] [saucenao] [google] [report]
67828268

>>67828183
>I run benchmarks and encode videos while I play games

>>67828195

>> No.67828286
File: 85 KB, 500x500, 1522878306485.jpg [View same] [iqdb] [saucenao] [google] [report]
67828286

>>67828268
>next ill trot out the gta5 benchmarks thatll teach dem amdrones

>> No.67828305

>>67828268
heh, you dare come here? get ready to be destroyed, you came to the wrong site buddy

>> No.67828308

>>67828286
>I'll post cherrypicked GPU bound benchmarks to cover up AMD's shitty single threaded performance and then I'll tell them it's the developers fault they can't optimize for 16 threads :)

>> No.67828309
File: 48 KB, 658x646, 1536178332204.jpg [View same] [iqdb] [saucenao] [google] [report]
67828309

>next gen is slower than prev gen
best timeline ever

>> No.67828319
File: 13 KB, 1286x144, 1537707457895.png [View same] [iqdb] [saucenao] [google] [report]
67828319

>>67827392
Why does Threadripper at 4.5GHz beat Intel at 4.5GHz?

>> No.67828339

>>67828308
The future is now old man, benchmarks below 4k don't matter.

>> No.67828354

>>67828023
14nm has been pushed to its absolute limits and yields are dropping. Also, bigger dies because they don't have a process that can shrink them as expected.

>> No.67828390

>>67828319
Have fun sitting there running Cinebench over and over. Also:
>Skylake-X
Another architecture which suffers from many of the same issues as Poozen, and lags similarly behind in gaming. Nobody gives a shit about Skylake-X. Coffee Lake is Intel's latest and greatest, and even the 8700K blows Poozen away. The 9900K is just going to be plain unfair.

>> No.67828394
File: 487 KB, 349x199, 1521382426193.gif [View same] [iqdb] [saucenao] [google] [report]
67828394

>>67828268
http://cpu.userbenchmark.com/Compare/Intel-Core-i7-8700K-vs-AMD-Ryzen-5-2600X/3937vs3956
>16% slower
>42% cheaper
Fuck, did >>67828218 ever hit the nail on the head.

>> No.67828413

Should I get a 2700X to support the good guys or support the bad guys by getting a 8600k/8700k for 3D modeling and gayming?

>> No.67828426
File: 164 KB, 1200x800, 1536694146733.png [View same] [iqdb] [saucenao] [google] [report]
67828426

>>67828390
Well it's a good thing monolithic garbage hosuefires are dead for all eternity and Intel gets pulverized where it matters where their $10000 abortions get completely curbstomped by Epyc. Intel can keep their gaymurr market (still getting outclassed there too lol) while AMD gets the HEDT and server market because MUH MONOLITHIC DIES

>> No.67828430
File: 35 KB, 1082x1055, 1483727679475.png [View same] [iqdb] [saucenao] [google] [report]
67828430

>>67828308
>games
>not GPU bound
oh wait, I keep forgetting you incel soiyms only play on 320x240 low

>> No.67828434
File: 359 KB, 1300x599, 1497918643725.png [View same] [iqdb] [saucenao] [google] [report]
67828434

>>67827399
>Intel's 10nm is on par with TSMC 10nm.
>Intel's 10nm is on par with TSMC 7nm.
>Intel's 10nm is on par with TSMC +7nm.
>Intel's 10nm is on par with TSMC 5nm.
>Intel's 10nm is on par with TSMC 3nm.
>Intel's 10nm is on par with TSMC 1nm.
>Intel's 10nm is on par with TSMC graphine prosess
when does the denial of intelfags end?

>> No.67828439

>>67828413
>intel stutterfires
>productivity
Lol no

>> No.67828441

>>67828413
Naturally buying hardware is political, it's a political statement. So you should see which party you want to support and buy their most expensive cpu.

>> No.67828442

>>67828394
I'll bet quids that Ryzen wasn't running RAM at 3200MHz either. Would have brought the gap into the single digits.

>>67828413
Get whatever you like, anon. If you're a gaymer though, I'd get the Ryzen, and plow the hundreds you'll save into an even awesome-er GPU.

>> No.67828462

>>67828268
>>>/v/

>> No.67828471

>>67828442
>but it's the B-die fault
>but it's the drivers fault
>but they didn't even BLCK overclock it! It's basically an underclock
This coming from the people who always quick to point out delidding.

>> No.67828473

>>67828430
>320x240
What we need is 240x144

>> No.67828479

>>67828441
I'm an AMDrone fanboi++++, and tend to buy their cheapest CPUs.
>snuggles up to Piledriver - it's only early spring in here in the antipodes

>> No.67828480

>>67827462
>9900K is rumored to be 450 I believe
its $500 in america and bound to be paperlaunched here too

>> No.67828487

>>67828480
>$500 in america
So $700 in Europe LMAO

>> No.67828489
File: 58 KB, 500x310, 1517189142174.jpg [View same] [iqdb] [saucenao] [google] [report]
67828489

>>67828473
C'mon, at least 4:3 - 256x192.

>> No.67828498

>>67828487
It wont even be in stock in europe, screencap this im calling paperlaunch

>> No.67828499

>>67828487
And might even crack the magic $1,000 here in Austfailure.

>> No.67828503

Intel sure loves NSA

>> No.67828515

>>67828498
Strange deja vu

>> No.67828519
File: 28 KB, 468x360, 11769852_295425147936706_7408115886000701440_n.jpg [View same] [iqdb] [saucenao] [google] [report]
67828519

>>67828471
Hush now faggot child, the grown-ups are talking.

>> No.67828522
File: 109 KB, 2250x2324, 1496480240772.png [View same] [iqdb] [saucenao] [google] [report]
67828522

>>67828471
by not getting a b-die ram for your ryzen you risk a few less frames
by not delidding your Intel® Core™ i7 you risk getting your house burned down
it's literally a difference between life and death

>> No.67828527

>>67828441
Intel lied and pushed false shit on their "independent" PR voices about Meltdown.
AMD never melts.

>> No.67828530
File: 611 KB, 1920x1080, 1534204099106.png [View same] [iqdb] [saucenao] [google] [report]
67828530

>>67828094
>MUH GAYMEN MUH 6 GORILLION FPS

>> No.67828542

>>67828499
Sorry kangaroo

>> No.67828548
File: 33 KB, 604x604, 1528225072270.jpg [View same] [iqdb] [saucenao] [google] [report]
67828548

>>67828530
CS:GO DOESN'T MATTER!!!
you heard it here first, goy

>> No.67828567
File: 85 KB, 406x1024, 1537397259274.jpg [View same] [iqdb] [saucenao] [google] [report]
67828567

>2020
>Zen 3 gets released
>(((Intel))) files bankrupt
>Intlets on suicide watch

>> No.67828570
File: 336 KB, 546x442, Evil-Kangaroo.jpg [View same] [iqdb] [saucenao] [google] [report]
67828570

>>67828542
I'm hopping mad! Or I'll just buy a $400 Ryzen with 90% of the go, and party.

>> No.67828580

>>67828394
>>42% cheaper
>I'm poor
AMD is good for one thing I'll give you that, they made Intel bump up the cores and then i5 8400 put top end Ryzens in their place again.


>>67828442
>hundreds you'll save
How would he be saving hundreds by going 2700X over 8700K? Not to mention he'll need to hand pick better RAM for Ryzen or else suffer another 10% performance decrease.


>>67828413
>should I buy a worse performing CPU just so I can virtue signal on an anonymous anime forum that I support le good guys

>> No.67828583

>>67828522
by not getting a b-die ram for your ryzen you risk performance below 30 frames per second
by not delidding your Intel® Core™ i7 you risk a few extra degrees
it's literally a difference between life and death

>> No.67828596

>>67827335
Not fucked at all. Contrary to popular belief on /g/ reducing die size doesn't mean you're magically better. It's also a feature that is not comparable across architectures, like clock speed, but most of /g/ is too retarded to realize this since you're all retarded consumerists who shill for companies and spout of numbers you don't know.

>> No.67828602

>>67828268
weirrrrrd every site i've looked at shows mankind divided running better on 2700x.

>> No.67828608

>>67828268
>7700K
>Low end
Intlel cucks are really being delusional.
Anyway, why don't you just put a link to a review sites instead of cherry picking a games that have bad optimization? (8700k only had 2 fps advantages over 7700k, even though it has 2 more cores than 7700k lmao.)

>> No.67828609
File: 40 KB, 600x532, 1527881816592.jpg [View same] [iqdb] [saucenao] [google] [report]
67828609

>>67828583
>this is what incucks actually believe

>> No.67828610
File: 50 KB, 1289x594, Untitled-1.jpg [View same] [iqdb] [saucenao] [google] [report]
67828610

>>67828268
AYYMD ABSOLUTELY BTF

>> No.67828626 [DELETED] 

>>67828567
In a perfect world, Intel would not exist.
But this is not a perfect world.

>> No.67828632

>>67828580
>buy a worse performing cpu that lit a fire under intels ass and completely btfo then in productivity
Not everyone that has a built computer only plays gayms
>I have to hand pick ram
3200mhz and I’m good to go. You have to be a class A brainlet to get under 3200 in the first place

>> No.67828634

>>67827519
Taxes work out differently in Europe than it does in the US, simply because of how society is organised (e.g. a federalised country vs. a single state). They also pay for a lot more stuff than the US does.

>> No.67828636

>>67828390
>The 9900K is just going to be plain unfair.
less than 10% improvement in ipc and 2 more cores are considered revolutionary by incels. and don't try to pretend the effective ipc will go up by more than 10%. If it could, it would've happened with 8700k. Gobble up that cum all you want, it's not happening anyway
The real issue is that intel is now selling gimped cores with no hyperthreading on all but the top levels because they chose performance over security

>> No.67828640

>>67828580
>i5 8400 put top end Ryzens in their place again.
back to /v/

>> No.67828642

>>67828394
>Userbenchmark
I will never ever trust """"average"""" benchmark.

>> No.67828646
File: 64 KB, 680x703, 15379575496200.jpg [View same] [iqdb] [saucenao] [google] [report]
67828646

>intel marketers damage control

>> No.67828651

>>67828567
In a perfect world, companies like Intel would not exist.
But this is not a perfect world.

>> No.67828653

>>67828519
>classical shitpost with no point

>>67828522
>classical opposing companie is totall poopoo, our problems are fine though

>>67828583
>and now faced with an exact copy of their shitposting they realize

>>67828609
>that hey it's actually just a shitpost

Imagine lacking the self awareness to essentially shitpost on your own shitpost.

Are AMD shills literal NPCs?

>> No.67828656

>>67828636
>less than 10% improvement in ipc
0% actually

>> No.67828657

>>67828602
Hate to break it to you, but incel streetwalking whore shills cherry-pick, and often completely make up, benchmarks.

>> No.67828671
File: 288 KB, 552x661, 1525362785884.png [View same] [iqdb] [saucenao] [google] [report]
67828671

>>67827335
>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!
>Heat doesn't matter!
>1771w cooler doesn't matter!
>Server space doesn't matter!
>ECC support doesn't matter!
>Free RAID doesn't matter!
>NVMe RAID doesn't matter!
>StoreMI doesn't matter!
>*NEW* IPC doesn't matter!
>*NEW* 7nm doesn't matter!
>*NEW* StoreMI doesn't matter!
>*NEW* HEDT doesn't matter!
>*NEW* Security doesn't matter!
>*NEW* VMware doesn't matter!
>*NEW* MySQL doesn't matter!
>*NEW* Unix doesn't matter!
>*NEW* Linux doesn't matter!
>*NEW* Waffer yields don't matter!
>*NEW* Enterprise doesn't matter!
>*NEW* Stock coolers don't matter!
>*NEW* Games don't always matter!
>*NEW* Hyperthreading doesn't matter!
>*NEW* Benchmarks after full patches don't matter!
>*NEW* Asian markets don't matter!
>*NEW* Own fabrics don’t matter
>*NEW* Litography don't matter!
>*NEW* Chipsets don't matter!

>> No.67828680

>>67828653
>t. classical stupid incel shill
Try again harder, child. Or alternatively, give a few dozen more BJs in the alleyway so you can afford your i7-9999okwzz-FFFF+.

>> No.67828681

>>67828608
>comparing last gen Intels
>i5 8400
>not low end
Nigga you blind, Ryzen got btfo by a LOCKED i5 hahahaha

>>67828632
Don't forget you need the absolute best C14 timings ram otherwise Poozen might shit itself

>> No.67828690

>>67828656
There's no way Intel would have better IPC with a same node and with same architecture too.

>> No.67828692

>>67828580
>Should i pay more for the chance to taste brian krzanich's matured semen under the ihs and pozzed architecture and the ability to virtue signal to my fellow 10 year old gaymers

>> No.67828705
File: 59 KB, 493x493, 1496102500371.jpg [View same] [iqdb] [saucenao] [google] [report]
67828705

>>67828653
that's right fellow intelbro, the G*d's chosen CPU company will always prevail over the dirty AMD infidels
see you on Reddit later today

>> No.67828713
File: 836 KB, 1242x1824, EAFAC8E1-D0A2-43A7-A1C8-94B18A1817AD.jpg [View same] [iqdb] [saucenao] [google] [report]
67828713

>>67828636
Still can’t argue it’s the better cpu. You guys all fucking argue AMD this intel that. When it comes down to the facts. Intel is better atm except for
>much 12 core flex for applications that rarely use it
>muh render rig

You faggots need to quit arguing over cpus. Intel is still in the lead AMD is behind on fpu and clocks. Also people who buy Intel must have enough money to delid, shit costs 30$ to do. 20$ for liquid metal. AMD fags will argue you shouldn’t have to do that and it’s true which is why intel is moving back to solder. Still when there is a problem for intel users they made a solution which is to apply liquid metal. I own a 7700K and shit hits 5Ghz 1.3V ez 65C with a noctua cooler. Now quit with this shit posting threads and actually have a conversation you kissless virgins.

>> No.67828716

>>67828690
Genuinely interested to know how you arrived at that conclusion, given the IPC difference between Bulldozer and Excavator.

>> No.67828718
File: 91 KB, 504x400, 1536612142306.png [View same] [iqdb] [saucenao] [google] [report]
67828718

>>67828705
This, victory will be ours intelbros, we must support jewish people, feminist frequency and diversity.

>> No.67828731

>>67828632
Even any cheap 3000mhz can be oced to 3200 easily.

>> No.67828736
File: 1.60 MB, 228x180, 1517548830941.gif [View same] [iqdb] [saucenao] [google] [report]
67828736

>>67828653
>implying for one nanosecond that your low-grade shillposting by some act of God deserves more

>> No.67828739
File: 112 KB, 915x487, 1493100952111.png [View same] [iqdb] [saucenao] [google] [report]
67828739

>>67828718
go- I mean guy, don't forget to buy your RAID key or you're a literal nazi

>> No.67828753

>>67828705
>>67828718
>thinks people who use intel do so for political reasons, or treat their purchase as some kind of holy war against the infidels
I guess the concept of buying a CPU for your performance needs is really foreign to you.

>> No.67828767

>>67828681
See? Intlel cucks is delusional as fuck. After they said that 7700k is low end, they didn't even know that the picture above is actually Deus Ex on 720p. In 1080p, there's only 1-2 fps differences.

>> No.67828773

>>67828519
>>67828522
>>67828609
>>67828646
>>67828671
>>67828680
>>67828705
>>67828718
>>67828736
>>67828739
AMD drones RISE UP!

>> No.67828776

>>67828753
Yes goy, nothing else matters except the brand name, Intlel™ Core®, stop kvetching and just buy it!

>> No.67828777

>>67828753
>I guess the concept of buying a CPU for your performance needs is really foreign to you.
which is why they will be fine with the next gen hyperthreaded quad co- oh wait

>> No.67828787

>>67828716
28nm vs 32nm?

>> No.67828791
File: 80 KB, 685x570, 1537037697953.png [View same] [iqdb] [saucenao] [google] [report]
67828791

>>67827442

>> No.67828794
File: 27 KB, 600x600, 1537056716945.jpg [View same] [iqdb] [saucenao] [google] [report]
67828794

>>67828773

>> No.67828797

>>67828773
>mass replying
just crawl back to r*ddit or whatever your cum stained lair is

>> No.67828799

Fuck Intel

I'm still running my AMD Dual core processor (The first dual core processor at 2ghz) and its fuking awesome.

When I retire it Im only going to AMD

>> No.67828802

>>67828794
shut up goy

>> No.67828805
File: 54 KB, 679x758, 1527629778452.jpg [View same] [iqdb] [saucenao] [google] [report]
67828805

>>67828802
OYYYYYY VEEEEEEEYYYY

>> No.67828806

>>67828776
>>67828791
>>67828794
>>67828797
Yes, yes don't let up! Shitpost the kike off this board! RISE UP! We are doing it 4chan!

>> No.67828825

>>67828713
You forget
>anyone that actually compiles anything
Which there are a fair few people who do.

>> No.67828833
File: 58 KB, 2000x1543, 1537640035293.png [View same] [iqdb] [saucenao] [google] [report]
67828833

Our office just bought over 450 2990wx systems for our render farm. and it's great.

>> No.67828841

>>67828799
I've still got an Athlon 64 X2 5600+ over there somewhere. Dey were noice.

>> No.67828842

67828806
>if i'm ironic enough on 4chinz people will side with me
did you pull this out of some CIA playbook, nice healthy glow on you btw
no (you) for you

>> No.67828843

>>67828799
oy vey

>> No.67828871
File: 1007 KB, 390x319, 1517387555245.gif [View same] [iqdb] [saucenao] [google] [report]
67828871

>>67828825
I do most of my compilation on a 1GHz Bobcat. Yes - I am certifiably insane.

>> No.67828879

>>67828791
I always laugh when retards post these stock charts that show 1-5 day changes. Meaningless.

>> No.67828897
File: 168 KB, 988x854, 1532756867725.jpg [View same] [iqdb] [saucenao] [google] [report]
67828897

>>67828879
Don't worry, it'll only get worse.

>> No.67828900
File: 119 KB, 320x252, 1494410921788.png [View same] [iqdb] [saucenao] [google] [report]
67828900

>>67828806

>> No.67828916

Will the i7 9900K being soldered have lower temps inherently even without overclocking? I'll be buying it next month and OCing it regardless, I am just curious how much difference it'd make.

>> No.67828938

It's real interesting how when somebody mentions ryzens performance in videogames vs intel, suddenly the thread get shitposted to death by amd-fans. Yet simultaneously, they believe they are fighting off shills or something, while doing essentially what shills would get payed for.
Is this just intel falseflagging as retarded amd drones or something? That's the only thing that makse sense.

>> No.67828953

>>67828897
I honestly don't want them to collapse completely, but they do need something drastic to happen for a deep culture change to take place

>> No.67828971
File: 614 KB, 992x1043, 1532561469544.png [View same] [iqdb] [saucenao] [google] [report]
67828971

>>67828938
Nice try, faggot. It goes more like this.
>some incel shill fags on about AMDrones being btfo by intelaviv
>AMDrones calmly explain that it's 10% at half the price
>incel starts flaming homosexuality about poorfags, it's really 150%, etc.
>THEN the AMD shitposting begins
Of course, you're too blind and fucking stupid to see and understand this.

>> No.67828987

>>67828971
"Play stupid games, win stupid prizes" goes an old saying. Intlel shills are an object lesson on this.

>> No.67829000

>>67828879
>Stocks don't matter

>> No.67829004

>>67828971
looking at kernel amd epyc patch commits i'll just stick with buying intel processors.

have fun having sex with zen or whatever fuck that calls itself a processor

>> No.67829016
File: 425 KB, 250x250, 1520850017292.gif [View same] [iqdb] [saucenao] [google] [report]
67829016

>>67829000
Well-used digits, anon.

>>67829004
Your admission of total defeat and surrender without terms is accepted, anon.

>> No.67829037

>>67829016
wow first reply and already defeated

>> No.67829043

>>67828971
>somebody dares to point out amd isn't perfect for everything
>I present my view points calmly and logically
>my opposition is I AM SILLY
>I logically and calmly shitpost for over 20+ posts with my own I AM SILLY posts, naturally I do this because of my high IQ and because it's completely justified
A frightening look into the mind of a fanboy.

>> No.67829061

I'd rather cut my arms off then buy the shitfest known as AMD
Intel+ Nivida for Life

>> No.67829070
File: 46 KB, 500x383, 1523419754753.jpg [View same] [iqdb] [saucenao] [google] [report]
67829070

>>67829037
>>67829043
>>67829061
Then they cry and scream when they get laughed off the board.

>> No.67829082
File: 56 KB, 621x702, vO7lRZ7.png [View same] [iqdb] [saucenao] [google] [report]
67829082

>>67829004
>kernel comments

>> No.67829102

>>67829000
I never implied that you moron

I said stock charts that show 1-5 day changes. Meaningless.

>> No.67829114

>>67829004
Enjoy your meltdown spectre nemesis tlbleed and all the other l1tf mitigation patches anon

>> No.67829129
File: 69 KB, 668x768, 1523404958749.png [View same] [iqdb] [saucenao] [google] [report]
67829129

>>67829102
I wonder what would happen if you set marketwatch or whatevs to 3 months, then...

>> No.67829134

>>67827399
Yeah, they both don't work yet.

>> No.67829171

>>67829129
>I wonder what would happen if I posted a graph that showed some useful information

The chart would be more meaningful

>> No.67829191

>>67829004
why are intlel shitters so cancerous?

>> No.67829201

>>67828971
>>AMDrones calmly explain that it's 10% at half the price
More like BTFO in every single game with 15% higher price

>> No.67829219

>>67827335
>tfw midrange is going to be 12nm ie 16nm+++
>:)

>> No.67829222
File: 458 KB, 546x1576, D2CDDAFC-A790-4EE4-B372-182CBCF9E0F1.jpg [View same] [iqdb] [saucenao] [google] [report]
67829222

>>67828825
>>67828713
The only woke people on /g/ the rest of you are pic related hands down and fuck pillows you weeeebs

>> No.67829224
File: 348 KB, 1275x714, 8400 btfo.png [View same] [iqdb] [saucenao] [google] [report]
67829224

>>67829201
>muh gaymes

>> No.67829238

>>67829224
Which side of the chart is which?

>> No.67829241
File: 64 KB, 691x771, 1515175316605.jpg [View same] [iqdb] [saucenao] [google] [report]
67829241

>>67829222
>intel is better
>woke
more like (((woke))), get out of here newfag take your faggot reddit lingo with you

>> No.67829247

>>67829238
2600 is the significantly faster one

>> No.67829255
File: 1.08 MB, 276x260, 9342A430-5F7B-4547-AACA-5894821406CA.gif [View same] [iqdb] [saucenao] [google] [report]
67829255

>>67829222
Checked and fucking keked because it’s true. /g/ never works out and probably got man tits. Girls never date em or even gay men because they look repulsive, so they throw their anger out on /g/ have fun dying ALONE /g/ like the fat fucks you are

>>67828713
I regret my Ryzen purchase, /g/ lied to me again

>> No.67829262

>>67829255
Still samefagging?

>> No.67829263

https://www.techpowerup.com/248008/intel-at-least-5-years-behind-tsmc-and-may-never-catch-up-analyst

>> No.67829266

>>67829224
I was talking about 2700x vs 8700k, but in that case:
>twice the threads
>AMD overclocked, Intel stock

>> No.67829268

>>67829241
Found the nerd onions boy cuck with no pussy in his life. I promise if you go change your life you won’t be so angry shit posting

>> No.67829273

>>67829224
Ah yes, the
>AMD has a aftermarket liquid cooler, better motherboard and expensive B-die RAM
>but we wont mention this anywhere instead look at the performance
I remember yesterday somebody actually said they got fooled by this benchmark and didn't buy intel because of that, only later did they find out that you have to pay so much extra to actually get the performance in that screenshot.

>> No.67829275

>>67829263
IT'S OVER INTLEL IS FINISHED

>> No.67829282

>>67829266
>implying games use hyperthreading efficiently
How fucking new are you seriously what year did you start posting on /g/

>> No.67829284

>>67829263
Fucking idiots 7Nm from tsmc is not the same as Intel Nm per Nm. Intel is still ahead. Kys

>> No.67829298

>>67829266
>intel cucked locked cpoo is shit
Water is wet

>> No.67829324

>>67829273
>AMD has a aftermarket liquid cooler
AMD can get 4ghz on the stock cooler it doesnt matter

>better motherboard
possible on a $100 b450

>expensive B-die RAM
$30 more is expensive how fucking poor are you?

Keep debating these facts it makes you look even dumber that you actually think these results aren't possible unless you follow the EXACT testbench, you have aspergers

>> No.67829343
File: 290 KB, 539x722, 1494289777740.png [View same] [iqdb] [saucenao] [google] [report]
67829343

>>67829298
No AMDrone water isnt wet its dry you fucking dumb nigger, who cares if waters a liquid it doesn't matter, all you need to do is buy my $10000 heater to make it dry, what you cant buy it goyim, fucking poorfag

>> No.67829345

Hopefully my 3570k survives till then

>> No.67829349

>>67829324
Found a B die 14Cas 3200Mhz kit at micro center open box for 130$ get fucked AMD drones I got it for my 7700K

>> No.67829358

>>67829268
>nerd onions boy cuck
you are so new you don't know why 4chan has a captcha, fuck off back to /reddit/ now

>> No.67829359

>>67829345
Why not get a 3770K delid and OC that bitch to 4.8Ghz. 3000Mhz ram and call it a day?

>> No.67829366

>>67829324
>only when pressed and exposed will they mention a downplayed version of the costs to get that performance
I mentioned the video of the guys benchmarks and what he used. Not some theoretical shit.

And of course none of this information will ever be mentioned or added to that benchmark screenshot, because it's a conveient cherrypick with no context. Even the guy who made the benchmark says that intel is the better option money wise.

But again my drone friend, just keep posting that dishonest cherrypick and hope somebody falls for it, it literally works after all.

>> No.67829369
File: 220 KB, 1000x750, 1454545810575.jpg [View same] [iqdb] [saucenao] [google] [report]
67829369

>>67829349
> I got it for my 7700K
sorry for your loss

>> No.67829405

>>67829369
50c h100Iv2 faggot 4.8Ghz liquid metal. Kys AMD dick rider

>> No.67829453

>>67829405
>needs deliding to cool properly

>> No.67829463
File: 1.35 MB, 499x654, 1479815475412.gif [View same] [iqdb] [saucenao] [google] [report]
67829463

>>67828833
>over 450 tr2990wx
damn
you guys running the matrix in there?

>> No.67829484

>>67829273
2600 is $165, 8400 is $200, overclocking ram is free, stay mad kike.

>> No.67829495

>>67829453
>Being too poor for liquid metal
It was actually fun to do desu, took me 30 min. 40$ and then high clock speeds. Faster than what AMD can spit out even with solder

>> No.67829502
File: 343 KB, 807x867, 1534595224911.png [View same] [iqdb] [saucenao] [google] [report]
67829502

>>67829284
INTLEL 10NM IS BETTER THAN TSMC 1NM

>> No.67829510

>>67829495
>buying a half baked product and being in the need to fix it

>> No.67829522

Who cares about childish things like games and benchmarks? Smart men buys a product that's solid and trusted. I mean Intel of corse.

>> No.67829536

>>67829522
>solid and trusted.
>intel
>meltdown
>spectre
>trusted

>> No.67829544

>>67829510
So what am I supposed to do? Just stick a finger in someone’s ass? If there is a problem, I find a solution. Heat is an issue so I found a fucking solution you shit poster anime loving nerd. Ez 4.8Ghz never felt so comfy at 50C load. Name one competition that can do that. Don’t think you got anything lad

>> No.67829554

>>67829484
8400 was 10-30$ cheaper than 2600 for like half a year.
>overclocking ram is free
>thinks you can just overlock RAM into making it B-die
I am not sure if you are a shill, or just a pour soul that drank the marketing coolaid and got bamboozled like some many others.

Go look up the actual source on the video, read what hardware was used, listen to the guy himself. Don't take 1 cherrypicked image with no context and conclude from that.

>> No.67829563

>>67829544
>buying a half baked product and boasting about it.

>> No.67829565
File: 1.94 MB, 349x165, 0A230EDA-4A7B-4746-9A0A-CE24A847690A.gif [View same] [iqdb] [saucenao] [google] [report]
67829565

>>67829544
Dubs made my day desu and I just got to this thread

>> No.67829578

>>67829554
7% faster, stay mad. :)

>> No.67829583

>>67829522
10 cpu vulnerabilities have been deposited in your account.

>> No.67829592

>>67829405
>bragging about a shit over clock
>while having a babby cooler
>and delid

>>67829495
>It was actually fun to do
>its fun to delid
>Its fun to correct intels fuckups
jesus

>>67829544
>So what am I supposed to do?
buy things that aren't fundamentally broken

>> No.67829593

>>67829563
Guess you’re the type to not make your own solution when someone else fucks up. You’re sure going to be a big failure at life lad good luck. I don’t cry about Intel’s fuck up and AMD don’t got anything to compete with better cores, fpu or clocks, arm is still a meme and you’re poor ;$

>> No.67829604
File: 324 KB, 882x758, 1506977173618.jpg [View same] [iqdb] [saucenao] [google] [report]
67829604

>>67829536
>>67829583
DELID DIS

>> No.67829606

>>67829593
LOL Intel fanboy mental gymnastics are hilarious.

>> No.67829621
File: 118 KB, 748x900, 453543656362344233255243.jpg [View same] [iqdb] [saucenao] [google] [report]
67829621

>> No.67829628

>>67829592
AMD is fucking broken in performance for 165Hz gaming. I built a Ryzen system with b die cas 14 3200Mhz and shit ran at 124Fps in SLI.
Seriously why are you AMD shill so fucking stupid in hardware? Switched over to 7700K for 200$+mobo for 50$ delid for 40$ total. Boom 240Frames per second. Git gud shitters

>> No.67829638
File: 24 KB, 267x297, 1502488836418.jpg [View same] [iqdb] [saucenao] [google] [report]
67829638

>>67829621
but muh netflix

>> No.67829647

>>67829593
>buying half baked product
>admitting it was a failed product
>fixing with voiding warranty
>boasting about it

>> No.67829657

>>67829628
>240Frames
Nice stutter, Zen gets 400+ fps in CSGO

>> No.67829662

>>67829647
Kys im done giving you my time lad haha stay fucking poor and I’m comfy like I said at 4.8 spent less on this system rather than Poozen system desu

>> No.67829675
File: 36 KB, 704x548, 0 value.png [View same] [iqdb] [saucenao] [google] [report]
67829675

Intel processors have 0 value, they're like dirt compared to platinum value AMD processors.

>> No.67829691

>>67829662
>buying a half baked product
>boasting about how half baked it is
>saying haha to get rid of humiliation

>> No.67829703

>>67827335
>How fucked is Intel?
On the desktop side? Not so much. Judging from what the 9900k can already do its safe to assume that they designed it to match with Zen 2 in mind. Zen 2 isn't some super 10Ghz, 100 core chip. It has to work with old AM4 motherboards so the clocks won't be as high as AMD fanboys and shills would want. My guess is that Zen 2 is still inferior in gaming benchmarks but it will have like 12-cores at 4.5Ghz. The flagship Zen 2 will likely squash the entry level Threadripper.

On the HEDT side? Cascade Lake-X is already DOA since they cannot compete with cores/price against Threadripper.

On the server/enterprise side? Very. Intel has no answer until Jim Keller works his magic once again. I wouldn't bet against the man and a billion dollar company behind his back. It's all slumps and rises and right now Intel is in a slump. It's a good thing though since Athlon winning gave birth to the Core2 and Corei series that made Intel very rich in the end.

>> No.67829720
File: 56 KB, 645x773, 1507211407035.jpg [View same] [iqdb] [saucenao] [google] [report]
67829720

>>67829282
>implying new games cant handle multithreading

>> No.67829721

>>67829691
Dude drop it you’re not even fucking green texting properly and I’m tired of seeing you two idiotd going at it. You suck at debating in all honesty. How old are you seriously you need to be 18 to be on this site.

>> No.67829724

>>67829662
>my ass is so sore from the rape ive received up and down this thread
Good riddance, you deserve to be shitting blood for a month.

>> No.67829725

>>67829593
>I buy faulty products because I blindly follow this one company
the post

>> No.67829732

>>67829522
> Smart men buys a product that's solid and trusted. I mean Intel of course.
>be me
>Buy 4770k because bulldozer is shit
>be told by /g/ and every other intel shill that it can do 5ghz under water
>only does 4.6
>ok 400mhz is fine
>after doing an expensive custom loop
>2 years later need new laptop
>haswell thinkpads suck and have a shitty trackpad so forced to buy broadwell thinkpad
>Wifi (made by intel) has constant connection issues and once in a while bluescreens my laptop
>drivers need to be reinstalled on a near weekly basis because of this
>CPU is worse
>literally a 5200u
>fucking throttles itself to 800 mhz if doing anything more intensive than staring at the desktop
>then maybe 6 months later my desktop cpu starts throttling itself to 2.5 ghz when rendering or doing anything intensive at all
>then on top of that a few more weeks later MCE starts bluescreening me
>have to turn MCE off
>once mighty 4770k at 4.6 ghz is now only 4ghz on a good day and has the single core and multicore of a ivy bridge built for laptops
>tested it and its literally slower in SINGLE CORE then the lowest clocked ryzens
but wait it doesn't end there
>buy a m.2 made by intel
>literally never even worked, just came to me broken

Intel is the lowest quality trash ive ever had the misfortune to use. They are up their in my book as shitty as ASUS, Biostar, Sennheiser. Intel is literally shit and 3 of those 4 companies i mentioned have shills on this board running 24/7 damage control.

>> No.67829736

>>67829725
based and redpilled

>> No.67829742

>>67829721
>I'm an idiot
the post

>> No.67829747

>>67829703
>Judging from what the 9900k can already do its safe to assume that they designed it to match with Zen 2 in mind
you wish

>> No.67829758

>>67829721
Newfag

>> No.67829767
File: 61 KB, 718x568, AMD-Ryzen-7-2700X-Cinebench.jpg [View same] [iqdb] [saucenao] [google] [report]
67829767

You people really like beating dead horses don't you. 2700x will be the best at future proofing since it has 16 threads and intel for todays gayming. I'd rather have a more future proof option that doesn't rip you off.

>> No.67829770

>>67829703
>Judging from what the 9900k can already do its safe to assume that they designed it to match with Zen 2 in mind
AHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAH

>> No.67829780
File: 27 KB, 741x609, 0b5.png [View same] [iqdb] [saucenao] [google] [report]
67829780

>noone in this thread understands that both amd and intel cpus have their upsides and downsidea
yeah, lets pointlessly argue about things that are already widely known, that will surely resolve something that you guys are trying to accomplish.. whatever that is

>> No.67829784

>>67828596
I couldn't give a fuck about CPU manufacturer partisanism, but please don't spout that crap. Clockspeed IS comparable across the board, it's not the only factor, but it's a fucking massive one that's downplayed far too much in recent years.

>> No.67829785

>>67829720
They can but not more than 8 on average which leaves half of Ryzens threads unused, this is why single core performance is still king.

>>67829747
It'll definitely beat Zen 2 on desktop, for Zen 2 to match stock 8700K performance would be an amazing feat. Now for advanced users and servers, I agree that AMD is already better in that field and Zen 2 will solidify that position.

>> No.67829802

>>67829747
The Ryzen 3700X won't be a leapfrog compared to the 2700X.

>> No.67829807
File: 1.04 MB, 1920x1080, 1537708169512.png [View same] [iqdb] [saucenao] [google] [report]
67829807

350 WATTS GLUED TOGETHER ALREADY EXPENSIVE MASSIVE DIES

THIS IS INTEL'S SEKRIT SAUCE LOL

>> No.67829809

>>67829785
>It'll definitely beat Zen 2 on desktop
So if zen 2 is 5ghz, and has a 15% ipc uplift do tell me how intel is faster

>> No.67829814

>>67829809
>5Ghz
>15% IPC
(lol delusional AMDrone

>> No.67829818
File: 162 KB, 633x900, 1491972340310.png [View same] [iqdb] [saucenao] [google] [report]
67829818

>>67829807
delid

>> No.67829832

>>67829767
>FX8350 will be best at future proofing since it has 8 cores
That's not how these things work, the average redditor wants to get the best in game performance for their money, "future proofing" by buying a CPU with more cores is retarded.

That's like using a 7 year old $1000 3960X with 12 threads in 2018 and thinking that it's still a relevant CPU because it has double the threads of an 8600K which wipes the floor with 3960X while having "only" 6 threads.

>> No.67829843

>>67829814
Ryzen 2 which was a literal rebrand on a new (barely better) node and a new memory controller was a 5% ipc gain and 400mhz clockpeed boost
So tell me, if AMD moves to a massively better node (7nm) and the first major revision to the zen arch, How is that not going to result in a 15%-20% ipc gain and a 400+ mhz gain

>> No.67829863

>>67829832
>>67829767
You might want to look at this benchmark Mort

>> No.67829866
File: 375 KB, 1228x1094, 1497506386191.png [View same] [iqdb] [saucenao] [google] [report]
67829866

>>67829814
>zen will be bulldozer 2.0, it will be worse than sandybridge ipc
>zen+ will be a disappointment, no ipc gains and maybe 100mhz if you are lucky
Because intel fags have reliably predicted the future up until this point

>> No.67829867

>9900k is on a level on its own that all the leakers have to overclock a 2700X and it still loses all the time
>AMD shills actually believe that just because AMD moved to a smaller node they'd instantly squash Intel and 2 generations of old Ryzen

Gotta shill dem X570 boards goy, your X370 boards are useless

>> No.67829871

>>67829843
Don't forget that it's also moving from 14nm LPP (which GF's 12nm is based on) to 7nm high performance node as well.

>> No.67829887

>>67829866
>zen+ will be a disappointment, no ipc gains and maybe 100mhz
you are actually correct, its barely an upgrade over zen 1

shameless cashgrab but its okay if amd does it

>> No.67829891
File: 158 KB, 712x1441, 2700xpower1.png [View same] [iqdb] [saucenao] [google] [report]
67829891

>>67829867
>9900k is on a level on its own
2000 points in cinebench, a 4.3 ghz 2700x does that as well while being $200 cheaper and way more efficient

>> No.67829892

(7nm)

(´ヮ´): sour grapes, Intel fanboy. Try to jump a little higher next time.

(Intel fanboy)

>> No.67829910

>>67829887
What's the problem? Zen+ was never advertised as some massive jump, but even then it gave pretty reasonable gains of +5% IPC and +400MHz clocks, fixing a lot of the early Zen problems as well.

>> No.67829919

>>67829871
Thats what intelfags don't seem to get

>>67829887
>you are actually correct, its barely an upgrade over zen 1
Yea but that's not reality in reality they improved the memory controller, ipc, clockspeed and latencies

>> No.67829929

>>67829910
Not to mention intel doesn't even give ipc gains anymore

>> No.67829937

>>67829863
I don't care about multi threaded performance, did you not understand anything from my FX8350/3960X story? Future proofing is a meme. Unused threads are a meme.

>> No.67829953
File: 10 KB, 388x447, 1535198252376.png [View same] [iqdb] [saucenao] [google] [report]
67829953

Can your Intlel CPU do this? This is important.

>> No.67829961

>>67829887
>shameless cashgrab but its okay if amd does it
is it aslo ok when the i3 9100 is just a i3 8100, when the i3 8100 was just a i5 7400, which was just a i5 6400 with 14nm++ and a new gpu no one will use, 4 fucking rebrands

>but its ok when intel does it

>> No.67829970
File: 32 KB, 720x736, 5f3.jpg [View same] [iqdb] [saucenao] [google] [report]
67829970

>>67829937
640K is more memory than anyone will ever need

-Bill Gates

>> No.67829980
File: 217 KB, 615x499, 10nm2021.png [View same] [iqdb] [saucenao] [google] [report]
67829980

>>67829961
Don't forget intel is rebranding coffee lake again next year again and if the i3 is still a quad core then intel will have rebranded the i5 6400 5 FUCKING TIMES.

>> No.67829983
File: 32 KB, 960x521, lol.jpg [View same] [iqdb] [saucenao] [google] [report]
67829983

8 inferior cores are no match for 8 Chad cores

>muh Cinebench
Literally reaches 2100 pretty easily while the virgin 2700X only eats up 1800

>> No.67829988

>>67829961
>new gpu
They renamed it, gpu is same.

>> No.67829998

>>67829983
>HEDT vs Non HEDT

Good job shlomo

>> No.67829999
File: 142 KB, 915x487, 1493100952111.png [View same] [iqdb] [saucenao] [google] [report]
67829999

Stop

>> No.67830006
File: 37 KB, 640x360, 1535623727413.jpg [View same] [iqdb] [saucenao] [google] [report]
67830006

Kvetching

>> No.67830007

>>67829983
>Literally reaches 2100 pretty easily while the virgin 2700X only eats up 1800
>nahah my synthetic benchmark matters more than your synthetic benchmark
how new r u

>> No.67830009

>>67829999
The 9900k isn't HEDT despite the name

It's not Skylake-X or its replacement

>> No.67830015
File: 16 KB, 480x289, 1535623854013.jpg [View same] [iqdb] [saucenao] [google] [report]
67830015

Goy

>> No.67830019
File: 197 KB, 712x455, intel 5ghz btfo.png [View same] [iqdb] [saucenao] [google] [report]
67830019

>>67830009
its shit is what it is, fucking housefire, meanwhile AMD has the fastest cpu ever made

>> No.67830035
File: 65 KB, 731x918, 1535561310892.png [View same] [iqdb] [saucenao] [google] [report]
67830035

Just wait for Intel's 10nm in 2020 or 2021. Probably 2022.

>> No.67830042

>>67830019
>more useless synthetics!

>> No.67830053

>>67830015
how will AMD EVER RECOVER lmoa

>> No.67830054

>>67830019
The top 11 single socket scores on cinebench are all threadripper 2990wxs

intel btfo

>> No.67830064

>>67830042
>nahah my synthetic benchmark matters more than your synthetic benchmark

>> No.67830094
File: 58 KB, 1332x1214, Prodoesntmatter1.png [View same] [iqdb] [saucenao] [google] [report]
67830094

>>67830042

>> No.67830099

>>67829807
wtf?
is that real

>> No.67830103
File: 54 KB, 1332x1214, Prodoesntmatter2.png [View same] [iqdb] [saucenao] [google] [report]
67830103

>>67830042
>>67830094
Pro

>> No.67830128
File: 58 KB, 1332x1214, Prodoesntmatter3.png [View same] [iqdb] [saucenao] [google] [report]
67830128

>>67830099
yes, 10nm was delayed again til 2021

>>67830103
>>67830042
Doesn't

>> No.67830141
File: 65 KB, 1332x1214, Prodoesntmatter4.png [View same] [iqdb] [saucenao] [google] [report]
67830141

>>67830128
>>67830042
MATTER

>> No.67830174

>>67830094
>>67830103
>>67830128
>>67830141
Thanks, these performance figures are exactly what I'm looking for to decide on my next CPU purchase

>> No.67830439

its all about video games for you people isnt it?

>> No.67830451
File: 19 KB, 162x132, 1368561260809.gif [View same] [iqdb] [saucenao] [google] [report]
67830451

>>67830439
Multithreaded Dwarf Fortress soon.

>> No.67830510

>>67830439
For /v/ it is, its helpful to reply to them with >>>/v/

>> No.67830685

>>67830035
you will be blown away when it launches 2023!

>> No.67830717

>>67828713
And compiling. And better perf per dollar.

>> No.67830744

>>67830451
WHEN
H
E
N

>> No.67830786

>>67829732
What's with se(h)nheiser?

>> No.67830800

>>67830451
proofs?

>> No.67830801

>>67829284
INTEL 10 NM = TSMC GRAPHENE REEEEEEEEEEEEEEEEEEEE

>> No.67830820

>>67828833
>28800 threads
>112 kW of TDP from CPUs alone

Never happened.

>> No.67830889

>>67830786
Veiled as fuck and snap like a graham cracker

>> No.67831282

>>67830685
With full production in 2024

>> No.67831287

>>67829349
>tfw only 2866 16-18-18 1.2V

>> No.67831300

>>67831282
in your local store 2025!

>> No.67831303

Intel is going back to 22nm.

>> No.67831478

>>67828413
Do you want your 3D modeling or your gaming to go faster?

>> No.67831657

https://youtu.be/w3g4QQnIqco?t=666

nvidia on suicide watch.

>> No.67832269
File: 130 KB, 500x282, smirky.png [View same] [iqdb] [saucenao] [google] [report]
67832269

>>67829953
I'm a hardcore intel fanboy, but i guess now my next upgrade is a threadripper.

>> No.67832403

>Apple and Huawei both have 7nm SoCs out
>TSMC parts
>both around 6.9 billion transistors
>Apple's 7nm SoC is 83mm2

The density on these chips is impressive, though granted they do both contain GPU logic, and GPUs lend themselves towards high density more so than pure CPU logic.
Still, its quite the feat.
For comparison the original 14nm LPP Summit Ridge die is 192mm2 with 4.8 billion transistors. These 7nm ARM SoCs have about 45% more transistors while being less than half the size.

Unless AMD doubles the core count per CCX, or doubles the CCX count per die, then I don't see transistor count for Zen2 parts increasing too drastically. A die around 100mm2 is a reasonable expectation, and its likely that the yields of such a part, even this early in production, would be pretty good.

>> No.67832433

>>67827335
intel will just release 22nm and amd is fucked ok?

>> No.67832466

>>67827519
because eu goverments is like 10x more crooked than usa, dont believe what they tell you

>> No.67832557

>>67828430
>>67828473
>>67828489
You need 80x25 ASCII text mode.

>> No.67832589

The only thing this thread is telling me is that /g/ should've started banning /v/irgin gamer benchmark posters years ago.

>> No.67832594

>>67832466
American government is so fucked and corrupt they'd never fine Intel even if they sell literal actual turds in an Intel i7 box to Americans.

>> No.67832625

>>67832594
I'd normally agree, but I think Trump is vindictive enough to go after intel as retribution for Krzanich snubbing him after publicly pledging support.
Intel definitely could be open to some fines and regulation under this admin, and I have no problem with that.

>> No.67832672
File: 55 KB, 600x600, CPDFqtlWIAAWUHt.png [View same] [iqdb] [saucenao] [google] [report]
67832672

how hard is it to build a brand new architecture? It seems like AMD did it on the whim with significantly less money for r&d. Why can't Intel just do the same and pull their own Ryzen comeback? forgive my ignorance

>> No.67832734

>>67832672
It took AMD quite a few years to finish the high level arch for Zen, its not something you can do in a span of even 2 years.
Intel has been working on a new core arch for a long time though. It was originally slated for 2021 or 2022. Though they haven't said a peep about it. Nothing but Core refreshes until then.

>> No.67832840

>>67827399
All planned 7nm nodes are ahead of Intel's existing 10nm node in terms of transistor density, albeit by a small margin. However, due to yields issues it is claimed by semiaccurate that Intel is scaling back the density on 10nm to try and solve the issues they're having, making it more in line with being an Intel-standard 12nm node. There is no statement on whether or not this will affect the performance regression vs 14nm+.

However, it's been reported that amd has decided to go with Tsmc's 7HPC node which sacrifices density and power for performance (and reduced cost) by comparison to the node I previously mentioned. Even with this said, the 7HPC node will be at least as dense as the revised 10nm node Intel is supposed to be bringing out, but will likely remain slightly ahead. All-in-all, Intel won't be releasing any performance 10nm parts until late 2019, and going into hedt/server will be even later. Amd will hold a significant process advantage for over a year in the enterprise market, and (likely) 6 months in both desktop markets. By the time Intel is catching up, amd will be moving on to a more refined, and slightly denser node. By the time Intel gets 10nm to where it really needs to be, amd will likely be moving on to the next new node furthering the gap in process advantage.

As long as amd is able to eliminate the majority of the low hanging fruit with zen2, they're going to be sprinting away from Intel for multiple years.

>> No.67832869

>>67832840
Why not make Ryzen and Threadripper on 7HP and Epyc on low power 7nm?

>> No.67832894

>>67832869
Is there any source that EPYC is on 7nm HP? It would make sense for server chips to be on 7nm LP, since you don't want as high clocks and more power efficiency at low clocks.

>> No.67832897

>>67827335
arbitrary scale graph is arbitrary

>> No.67832921

>>67832869
Its most cost effective to have a single die span multiple segments by targeting the right power curve.
A design made on a "high power" process might scale down to 3ghz with very good power consumption. A design made on a low power process is very, very unlikely to scale up in clocks without power consumption exploding.

>> No.67832944

>>67827335
On a scale of 0 to They fired their last CEO and cut his golden parachute?

>> No.67833135

>>67832672
>did it on the whim
They had 10 years after getting shredded by the first i7

Only fanboys bought AMD when the 920 came out, it even made less sense when Sandy Bridge showed up a year later and you have to be most diehard AMD shill around Haswell era

It took 5+ years for Intel to get out of the Netburst era and they spawned the Core 2 series that set them up for the next 10 years

If they had new architecture they would have introduced it by now to combat AMD but they're rehashing Core until its last drop

>> No.67833164

>>67828390
>and lags similarly behind in gaming
>>>/v/ you fucking fag

>> No.67833303

>>67829807
LMAO intel trying to create a second sun at a data center near you

>> No.67833348

>>67827335
Countdown to aZENtion bois

>> No.67833874

10nm delayed

Doesn't matter cause Intel announces it'll meet supply next year and amd stock drops despite 7nm next year. While Intel rises.
Retards on the stock market..

>> No.67833905

>>67827335
If AMD can deliver better gaming performance then they will be king

>> No.67833930

>>67828078
my 2600X works great even with its boring ass looking Stock Cooler, but i'm having to use a spare 720p 59Hz TV as a monitor for a while so it was probably slower on your 1440p 144Hz Monitor.

>> No.67833952

>>67828268
it gets so bad it turns to poopoo

>> No.67833970

>>67829843
It was actually a 3% IPC uplift with consideration of max official memory support (2667 zen1, 2933 zen+). This is according to AMD's own slides. When you match the memory it's actually only 1.5% on average improvement. Which still isn't terrible considering there was no changes made to the actual cores. The big benefit for zen+ over zen isn't in absolute performance, but performance per watt.
>1700 owner who moved to 2700x.
Marginally better overall performance stock vs max oc with significantly lower power draw and heat output. Sold the 1700 which covered half the cost of the 2700x, pretty happy. Will end up doing the same thing when zen2 comes out still rocking my x370 motherboard 2 years after buying it. Unless x570 has some serious benefits to offer.

>> No.67833979

>>67833905
You realize Intelaviv literally only has gaming left, right? In every other category, they've already lost. Why do you think the only thing Intel shills still post are lopsided gaming benchmarks when paired with GPUs that cost as much as entire PC on it's own?

>> No.67833998

explain what this means to a brainlet

>> No.67834005

>>67833998
amd good, intel bad

>> No.67834216

>>67832894
Just some tweets from a writer for the motley fool claiming to have confirmed upcoming 7nm chips will be produced on N7HPC as opposed to N7SOC. This includes the 7nm Vega 20 chip, but there was no mention of Navi in that regard.

For why use HPC on epyc, >>67832921
This. We seen how using a low power node worked for current zen processors. We also know nothing about what architectural changes they're making that would affect power draw, other than that they're continuing to make efficiency a priority and that zen has room for improvement. So it's entirely possible whatever they've done to the cores has offset the increase in power draw on the HPC node vs the soc node. Or maybe they just feel they need the performance / lower cost. They're definitely working the long game, though.

>> No.67834683

>>67827966
>games
Also
>1080p

>> No.67836120

>>67828713
>Intel is still in the lead
Sure. Exclusively at 1080p gaming.

>> No.67836247

Cheap-ass EPYCs on Aliexpress WHEN?

>>67832403
Huawei? Kirin 980 will be used only after few months, in best case (in huawei phones)

>> No.67836253

>>67827335
I can't wait for any kind of x86 7nm product. I just want to see shitposting.

>> No.67836476

>>67827519
I like not to go bankrupt for having my tooth removed.

>> No.67836479
File: 379 KB, 1649x655, intel on suicide watch.png [View same] [iqdb] [saucenao] [google] [report]
67836479

intel on suicide watch

>> No.67836494
File: 402 KB, 1635x557, 2nd page.png [View same] [iqdb] [saucenao] [google] [report]
67836494

>>67836479
second page

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