Core i7-8700K benchmarks released

File: Mysterious Merchant.gif (4KB, 452x523px) Image search: [Google]

4KB, 452x523px

DELET pls. You can't do that.

>50% more cores than 7700k, 28% better result

fake as fuck

File: 1504275173842.png (120KB, 480x521px) Image search: [Google]

120KB, 480x521px

>>62365850
>guru3d
>fake

If it keeps kaby lake singe thread(really 4x thread is what's important) then the 8600k will be a great gayman chip for high refresh. Outside of that I'd buy and suggest ryzen for literally every other use case.

>>62365856
look up your sources you daft retard

it's a literally who with a potato camera so photo evidence is unreadable

File: haghahahahha.png (75KB, 704x475px) Image search: [Google]

75KB, 704x475px

>>62365816
better get a i5 7640k while it's *HOT* fellow consumer

>>62365850
7900x has 66% more cores and gets around 70% better result

Jesus. What is Intel doing...

File: Core-i7-8700K-Cinebench-Graph.jpg (344KB, 1920x1080px) Image search: [Google]

344KB, 1920x1080px

>>62365871
It's real.

File: Core-i7-8700K-CPUZ.jpg (236KB, 1920x1080px) Image search: [Google]

236KB, 1920x1080px

>>62365911

>>62365882
skylake-x is a different architecture than kabylake, they're not exactly comparable

coffeelake on the other hand is literally kabylake with 2 extra cores glued in

even if we consider slightly lower clocks, 8700k would never perform as poor as 1230.

File: 1503929937770.jpg (26KB, 629x638px) Image search: [Google]

26KB, 629x638px

>>62365911
>>62365816
bravo intel..bravo>>62365816

>>62365931
>skylake-x is a different architecture than kabylake
Skylake and kabylake are literally the exact same shit.

>>62365944
I said Skylake-x not Skylake.

>>62365944
>Skylake and kabylake are literally the exact same shit.
true. but skylake-x got a mesh bus instead of a ringbus,at least some difference

File: Core-i7-8700K-Cinebench.jpg (274KB, 1920x1080px) Image search: [Google]

274KB, 1920x1080px

Good single thread, perfect for gaymers.

>>62365944
Name one post-Sandy Bridge Intel architecture that isn't the same shit with 2% better performance and housefire temps.

>>62365925
>>62365911

I was actually really looking forward to buying a Core i7-8700K to finally upgrade but these scores are total trash, I'm not paying to be insulted.
They either lower the price to that of a R1600X which will never happen or I wont buy.

>>62365816
> Ryzen 5 is better than the newest i7
Black magic.

>>62365958
>true. but skylake-x got a mesh bus instead of a ringbus,at least some difference
And coffee lake doesn't?

>>62365965
Modern games require powerful 6-10 threads, not just single thread.

>>62365986
nope

>>62365965
> Single core
>> Tops everything
> Multi core
>> Average results
Intel still didn't solve anything with their multicore perf degradation.

File: 2.jpg (703KB, 2048x1536px) Image search: [Google]

703KB, 2048x1536px

HOW THE FUCK CAN IT NOT BEAT A 1600X HAHAHAHHAHAHAHAH

>>62365972
Anon, it's time to let it go.
Just get the fucking Ryzen.

The video

www facebook com/MrTechQc/videos/1985486751696443/

You can see the HP Omen PC they drag a monitor over to it and hook it up.

>>62366021
It's not degradation, it's just that they love to put absurd turbo clocks on a single core while the all core turbo is trash

>>62365816
6 physical cores with IPC on level with their gaymen i7s from previous generations. It all depends on temperatures now, if it's an electric furnace I'll pass. If they match temps of my 6700k Intel instantly gets a generation ahead of AMD just like before Ryzen.

File: cinebench.png (775KB, 1024x1024px) Image search: [Google]

775KB, 1024x1024px

>>62365816
>/g/
> single leak from a nobody.
Intel BTFO
AMD BTFO
X86 BTFO
pc on suicide watch, how will consumerism ever recover?

>>62365816
And this is going to cost $400. Judging from results all-core is sitting at 3.7 GHz. Which seems really, really low. Was expecting 4 GHz all-core. What happened? Is this OEM PC they ran it on junk?

>>62366066
> facebook.com

C'mon anon you can do better than that.
At the very least use a video host that 3/4ths of /g/ doesn't have blocked at the firewall level.

File: 1497658837285.gif (365KB, 2000x1153px) Image search: [Google]

365KB, 2000x1153px

>>62366084

>>62366071
>Intel instantly gets a generation ahead of AMD
That's not how it works, fanboy.

>>62366093
That is the original source video you big baka.

Fine I uploaded it here https://my.mixtape.moe/zymvwe.mp4

THIRD
FUCKING THIRD
DISASTEROUS LAUNCH
IN A FUCKING ROW
Somebody please, I beg you, bring Otellini back.

>>62366099
>It's another episode of subhuman niggers from India live their fairy tales where everyone buys CPUS to run benchmarks and render videos.

Consumer market is all I care about and if Intel manages to transfer their 4c/8t ipc to 6c and 8c CPUs without burning my house down AMD is fucked.

File: india inside.jpg (387KB, 1500x1164px) Image search: [Google]

387KB, 1500x1164px

>>62366132
>My needs are the only ones that matters, therefore AMD is doomed and Intel wins lol BTFO!

>>62366162
Yes, exactly. Your point? Thank god the market itself isn't pandering to your obscure shit taste.

>>62366125
so he can bribe dell to sell these shitty chips too?

Intel will probably switch architecture AND country making the chips again.
I just hope its not a brazilian chip.
"Intel announces the SOPA architecture based on making the itanium actually not suck, with the variants TARTARUGA for mobile, MACACO for the desktop and TUCANO for the server space, all running out of the DELICIA north bridge."

File: intel ceo JUST.jpg (119KB, 1000x1000px) Image search: [Google]

119KB, 1000x1000px

>>62365850
>>62365871
>>62366084
>>62366132
>>62366173

>>62366132
Hoo boy, hungry for (You)s?
>>62366175
So he can start throwing money at R&D teams so maybe in 3-5 years Intel would be competetive again.

>>62365816
>barely higher than an R5 1600
this can't be real right?

>>62366132
>Consumer market is all I care about and if Intel manages to transfer their 4c/8t ipc to 6c and 8c CPUs without burning my house down AMD is fucked.
Not when their chips cost twice as much as AMD's CPUs that have comparable performance.

i call dibs on now

intel will solder few binned chips to send it on reviewers
and they will sell TIMshit tier to consumers

File: 1498256245652.jpg (30KB, 216x250px) Image search: [Google]

30KB, 216x250px

>>62366191
SOPA?

>multithread peformance is on par with Ryzen 5 1600, but it is also twice as expensive
>singlethread peformance barely beats 7700k
Might as well just get a cheap i7-1770k, if all you're do is gaming.

>>62365925
24MHzx154?

>>62366162
I-I'd buy that CPU. It'd be sick for emulation.

>>62366229
24.154= 3.696 aka 3.7ghz

File: blips and chitz.png (970KB, 1124x629px) Image search: [Google]

970KB, 1124x629px

>>62365816
well let's extrapolate from the data
7700k 960 pts
8700k 1230 points
28% improvement seems low but even if..
7700k was already 5-10% faster in most games and single thread applications.
8th gen Will beat the pants off red team for fun stuff. Ryzen will probs be better for productivity.

Can AMD anwer with an updated 14nm Ryzen+ and pull a phenom<phenom II?
Or will intel Crash and burn?

Benchmarks of the neckbearded and lonely
episode 42 Flying too close to the sun.

>>62366270
Well, yeah, but isn't other core cpus have 100mhz bus? Why they've changed that?

>>62366294
You forgot that this is going to cost $400.

>>62366294
And it will cost twice the price of 1600.

>>62366302
>>62366302
yes. fixed in the new cpu-z version. Mult: 37, Bus 100.

>>62366294
1600x costs almost half that price and unless that chip can offer 100-150% more perf then no only braindead idiots will buy it

Well at least 8100 might be decent assuming pricing is not bonkers.

File: 20170821175833_1.jpg (680KB, 2560x1600px) Image search: [Google]

680KB, 2560x1600px

>>62366368
>>62366345
>all this benchmark wankery and you don't have, or wont be getting a 1080ti and high refresh display?
What's $200 when the whole build is done.

>>62366487
Yeah, the 1080 ti that has a whopping 0.56% marketshare on Steam ..

>>62366518
let's just say you don't fit into the target demographic.

Must...Release..Skylake...Again

>>62366589
If the only people buying Intel's flagship are in that teeny demographic, they're probably in trouble.

>>62365816
Jesus fucking christ. We all knew it was gonna be MOAR COARS but not even a 5% increase in IPC? Fuck man, intel has gone to absolute shit.

>>62366652
They're running out of ways to improve the core architecture.
It's time to come up with something new.

>>62366652
>IPC increase
There was no IPC increase in any Intel CPU, all they had was clock increase and now that it capped, expecting better performance out of Intel Arch now, is just foolish.

>>62366487
>Y-you don't need cheaper CPUs, goy
Oy to the vey, rabbi.
>>62366518
It's a halo product.
>>62366652
It's a refresh.

>>62366652
Remember that Coffee Lake is just a stopgag because their 10nm process is crap.

Truly the Coffin Lake.

>>62366700
Speculation is that their next gen arch (ca 2020 should they not fuck up further) will be a "streamlined" x86 with all the inconvenient bits stripped from the ISA.

History shows us that this will be a mistake, since most enterprise customers care far more about not rocking the boat than anything else.

>>62366652

not even 8 cores, just 2 more....Intel really likes its money

>>62366752
>Intel tries to pull an AArch64 in x86
Oh boy.
OH BOY.
It's not gonna end well.

>>62366752
I don't think this speculation makes any sense.
The space that the x86 translator takes from the die is TINY, and you already can skip it with extensions like MMX/SSE/AMD64 etc..
But you probably can strip the main RISC CPU from 16bit capabilities etc.. and offload that to a tiny CPU.

>>62365816
wow. this isn't merely a 7700K with more core. it's a direct regression core-for-core. at least there's a minor gain in single threaded Cinebench. shit better shine elsewhere or Intel is done for.

>>62366837
>Intel is done for
Intel survived 6502.
But they will lose marketshare
A LOT of marketshare.

>>62366871
they've already lost half of the enthusiast market. X299 is basically dead and buried. if Epyc is half-decent it will be a complete and utter disaster for Intel's bottom line.

this is what happens when you sit on a lead and do nothing for 5 years.

>>62366930
EPYC is more, much more than half-decent.
Three hyperscalers are doing rollouts of EBYN H2 2017.

>>62366797
It's less about die size than simplifying design and validation. Virtually nobody gives two fucks about 8b hi/low views of registers, yet recently some Haskell compilers or whatever hit some erratum under hyperthreading with this shit. There're probably are actual paths in the pipeline that do need to get lengthened because of these constraints, and there are probably several thousand similar things Intel would gladly be rid of given half a chance.

>>62366871
For now they can drift on the brand name alone.
And before AMD overtakes them in single core performance Intel still will still be the top choice for gayming.

>>62366871
Intel survived the i432, Itanic, and Hyberbipelines.

It may require some profit losses and/or bribes, but they'll survive their recent stagnation as well.

>>62366987
>brand name
Intel is not nVidia.
They are THE devil in eyes of many.
And there's also Datacenter and Mobile where AMD basically already won.

>>62366974
I wonder if in this case would be possible to simple add a 16bit compatible x86 coprocessor there, and when the system does the unthinkable and jumps to a mode where it is accessible, the tiny CPU does the job.

>>62366930
>>62366963
You don't steal the internet with a half-decent product

>>62367021
These were bad designs.
6502 was like 6 times cheaper and two times more performant than 8080.
And Intel still survived that.

>>62367037
For years Intel had basically a monopol for CPUs that aren't garbage. It will take time before common people even consider buying something with AMD inside™.

>>62367177
Ryzen sales tell the whole other story.

Lack of memory bandwidth is holding it back.
Not enough bw for the cores, it was perfect for the 4 cores.

>>62367229
So... trichannel again when?

>>62367229
And yet Zen is doing mighty fine with 2 channels per 8 cores.

>>62367246
Because Zen has a high bandwidth controller that gets close to the bandwidth of a 4 channel Broadwell-E

>>62367209
Sure, show me the statistics where Ryzens consistently outsell Intel processors by a large margin.

>>62367246
Zen is a fresh core, it's also smaller than Intel cores, technologically it's quite ahead until Intel designs a new core somewhere in 2021

>>62367286
Zen also doesnt have fuckhuge 512bit AVX pipelines and all that that requires.

>>62366068
This, any K-series intel bench should be using a 4.5GHz overclock with turbo disabled, and include another sandy/post-sandy i7 equivalent with the same overclock in the comparison table.

>>62367298
Coffin Lake lacks 512b wide FMA units.

>>62367209

The way I see it, prebuilt market will still be intel's for years to come, and won't lose the supply contracts anytime soon. I'm pretty sure they're shruging off the fact that amd is outselling intel in europe on standalone cpu's

>>62367351
No, that's also the case. Ryzen Pro (the one that matters in OEM sales) got several design wins.
Quite an achievement for a platform so fresh.

File: Special-Edition--Keep-Calm-I-Work-At-INTEL.jpg (74KB, 1010x1010px) Image search: [Google]

74KB, 1010x1010px

OP here.

There is one thing to keep in mind:

1. the other processors might have been tested on a different machine with higher RAM speeds. The HP Omen has only 2400 MHZ ram

2. in the french source video, there is a mouse driver installation popping up during the MT test. Not sure how much effect it could have, but it might make the difference

3. there are now several other leaks like from https://videocardz.com/72471/first-intel-core-i7-8700k-benchmarks-leaked
which put the performance more towards the same level as a default clocked R1700. Still not great but it would be okay(ish) to justify buying the Intel if the price falls to the same level as the R1700 (imo). Heck, even slightly more would be okay because Intel is "trusted" but more so: IGPU included. That has value, sorry but it has especially for me.

> pic unrelated

>>62365965
must be throttling hard on all core loads.

>>62367371

Problem is design wins don't automatically mean sales, the Intel Inside™ name brand still holds enormous value among casuals, even if the design itself is inferior, and they don't know what it means. I've lost count to how many builds I recommended to normies that got the reply word by word "does it have the intel inside processor"?

File: 1488062011543.png (222KB, 449x401px) Image search: [Google]

222KB, 449x401px

>>62365991

>Modern games require powerful 6-10 threads, not just single thread.

lol

>>62367407
Geekbench and CPU-z are very Intel biased.
So far R15 hasn't been compromised.

>>62367497
?

>>62367470
Casuals buy craptops.
And they don't reeeeeeally care about SoC inside.

Did they fix the bangbus yet or is it still killing performance?

File: 1500639922851.png (107KB, 2250x2324px) Image search: [Google]

107KB, 2250x2324px

>>62365816
This is the $420 cpu, folks. Barely above r5 1600.

>>62367515
I'm pretty sure anybody who accidentally bought something with shitdozer inside before will be filled with disgust seeing AMD's sticker on their potential device for foreseeable future.

>>62365850
Lower clocks, non linear scaling.

>>62367586
Cinebench scores linearly.

>>62367578
Don't worry, RR will be more than persuasive.
That leaves faggy OEMs with their favourite "shitty chassis" and "single channel memory" memes.

>>62367578
Ryzen has a good reputation.

>>62367673
It has less than one year. Casual people don't even know it exists and bulldozer's memes will cling to it for long.

>>62365816
Red Rooster Shilling , Inc®™©

>>62367725
That's like saying pentium 4 memes used to cling on core architecture.

>>62367765

Anon, normies STILL think P4 was good.

>>62367810
Nah

>>62367765
From what I read Intel backtracked from it pretty hard and also employed monopolistic practices to damage control before that turd was gone.
Pretty different from AMD's underdog situation.

File: when-the-macaco-soup-almost-kills-you.jpg (51KB, 329x329px) Image search: [Google]

51KB, 329x329px

>>62365816
Delete this right fucking now, caralho!

>>62367810
Later P4s were good, more heat and power isn't a big deal for home users.

>>62367984
Nope, they were still shit. My P4 524 used to shut down during summer due to overheating.

>>62366741
Just like kaby Lake.

>>62365850
> wat is GHz

>>62365911
Why is the 5820k better? They're both 6 coarz and the 5820k is Haswell.

>>62367997
my Northwood 3.0 HT was in a case that I never cleaned for years inside a wooden desk pc compartment and it worked fine during 30C summers for 9 years and didn't die. Although now that I think about it, it might have been throttling, I wasn't aware of such thing back then.

>>62366197
The king has fallen. I didn't think it would ever happen. Definitely not this fast. But they're dead. Three launches in a row now that have been utter trash. How long can their past reputation keep them afloat?

>>62365816
Time to stick with my 5820K. I guess its still better.

>>62367425
This must be what's happening. They avertise a crazy turbo clockspeed that it can't maintain for more than a few seconds before it drops back down to stock to maintain TDP and power draw. They are doing the exact same thing with Kabylake Refresh on laptops.

>>62367407
>The HP Omen has only 2400 MHZ ram
This was never a huge issue for Intel CPUs, only Ryzen CPUs due to the Infinity Fabric link.

>>62367265
Ryze: $$$$$$ ()
Intel: $ ()

Statistic

>>62368149
Just a difference in niggahurtz

>>62368291
The non Extreme chips are designed from a marketing perspective rather than a performance perspective. They can boost one core pretty hight, but when a bench taxes all cores, they throttle like a motherfucker.

>>62368189
Three?

>>62368360
I think he's talking about Kaby, Skylake-X e Coffee Lake

>>62367578
Doubt it, gf has a faildozer CPU and loves it to bits. Convinced it's a high end gaming laptop. Consumers will believe whatever is marketed to them. AMD never lost face with anyone who doesn't know what they're talking about, Intel just had the better reputation.

>>62368360
SKL-X (killed by jewripper), SKL-SP (EBYNed out of existence), CFL-S (killed by $200 CPU).
>>62368370

>>62365816
GET READY FOR MORE CPU CIRCUMCISION

>https://twitter.com/momomo_us/status/896530606321946625
>jewjizz in fucking Xeons
But why.

>>62368464
Delid/OC seems like the only way to get your money's worth out of it.

File: 1504267718201.jpg (84KB, 1024x683px) Image search: [Google]

84KB, 1024x683px

>>62365816
>>62365911

>barely faster than a 1600
>beaten by a 1600X
>€389

Intel wtf are you doing

>>62367298
the way intel segmented avx 512 not a single cpu has the whole set anyways...
intel doesnt have big vectors at all inside and that is hilarious that they are actually perfoming worse

They will sell boatloads to Twitch gaymers.

>>62368665
The 7700K will continue to outsell 8700K for gamers.

>>62368525
>$420 cpu + delid kit + mobo for OC + good cooler
Toppest of keks, complete shit value

>>62365816
>Intel's 6-core 12 thread is as fast as AMD's 6-core 12 thread
>Costs $120 more
Wahahahahahahaha WAHAHAHAHAHA

PROCESS SUPERIORITY
R&D SUPERIORITY
BRAND NAME MEMEING

WHERE IS YOUR GOD NOW

>>62369063
MUH SINGLE THREAD
MUH GAMES

>>62365976
Aint nothing black bout the contract keller

>>62365816
reported for anti-semitism

>>62369063

AMD doesnt have an iGPU you asshole

>>62368665
how? everyone that wanted only one system to stream with obs its probably already with a 1700

>>62366084
>disable SMT on an arch specifically designed to use SMT
>surprised when performance is lower

>>62369119
You say that like it's a negative

File: download.png (5KB, 324x156px) Image search: [Google]

5KB, 324x156px

>>62369119
oh its cumming and its cumming at full force

>>62369063
DON'T YOU FUCKING *DARE* TO ATTACK

MY

VIDEO GAMES

MY FRAME RATE

MY SHOOTERS YOU FAGGOT

>>62368525
Too bad delidding it and over locking it once again pushes it out of the Good Value zone and back into the trash heap. Is this really the end for Intel?

>>62369119
Oh no how will I live without a trash-tier iGPU I'll never use for any purpose

Looks Like I Got Cucked Again™

>>62369063
I thought 8700k was $420 and r5 1600 $200

File: untitled-2.png (44KB, 682x811px) Image search: [Google]

44KB, 682x811px

>>62365816
Looks fine to me OP
6 actually good cores for consumer level

>>62369325
>post single thread
>talks about multithread

>>62367229
Is there any actual use in Quad-Channel for gaming and rendering?
>>62367298
For what do you need these?
>>62369181
I really wish AMD would bring a Zen iGP series. I need a cheap quadcore with 4K output capability. I'm not willing to pay hundreds for outdated Intel platforms. Currently I'm on a 3550, which performs similar like a Core i5-7400, but it cant output 4K. And graphics cards with HDMI 2.0 are expensive as fuck.
>>62369282
Maybe when you only run one gaming PC. I have a second HTPC, I won't be buying an expensive GPU for it. I play my games on a 1080 Ti, but can't playback 4K videos or even use native 4K resolution because my GT 530 doesn't allow it.

>>62368504
Aww, did you expect intel to actually spend money?

>>62369325
6 good cores that shit the bed when all 12 threads are loaded.

>>62369391
For an HTPC you aren't going to be using any of these CPUs anyway. Wait for AMD's lineup with the iGPUs, they'll have a better iGPU that can handle your 4K videos and won't burn your house down.

>>62369366
>>62369410
It's a K processor, which means you can set the turbo to all cores without even adjusting voltages since the cores are guaranteed to run at that speed, not to mention additional overclocking.

>>62369458
we found the delusional fanboy that think intel cpus will run all core boost with mayo tim

>>62369487
Found the guy who thinks sandy bridge tier single core performance is acceptable as long as you have "moar cores"

>>62369515
shits all over the lineup
cant produce enough fps gets behind 10-15%
sandybridge tier

i must say the level of tech illiteratism here is superb

Napkin math time.

i7-8700K
(((4000*1.125)*6)*1.275)/25.4
1355

(((3700*1.125)*6)*1.275)/25.4
1254

These are exactly the results we'd expect if it throttled down to it's base clock.

>>62369458
Even if it's true, it's going to be a real housefire requiring some sick cooling and possibly delidding.

>>62369515
It is acceptable (more like haswell level really) for what is basically a demo version of a new architecture delivering dirt cheap multithreading and that is capped to around 4GHz by the process.

>>62365816
Jesus, how shit the i5 and i3 gonna be?

File: 8700k.png (128KB, 870x869px) Image search: [Google]

128KB, 870x869px

>>62368716
Not anymore. 8700K Delid Dis™ Edition will sell like hotcakes.

>>62369582
>cpu circumcision becomes a normal practice
You people are hit in the fucking head.

>>62369603
From the looks of it, the number of completely deluded Intel fanboys is going down, not up. Every say I see more and more people becoming disillusioned.

File: Capture.jpg (282KB, 1378x746px) Image search: [Google]

282KB, 1378x746px

>>62369582
All core turbo doesn't require voltage offset increase therefore shouldn't cause much overheating.
>>62369533
Wtf are you talking about.

>>62367510
It's more like 4-6 threads

>>62365816
now we're fucking talking. 12 threads with the single core performance of the 7700k. the ultimate gaming cpu. day one buy if it's not over $350

>>62369968
More like $400. It's possible this is throttling and is only at base clocks. >>62369573

>>62369968
>it's not their 10nm refresh on the 570 chipset
>first day buy
fucking retard.

>>62369968
desu I rather wait™ for Intel's Icelake or Cannonlake because it's on a new 10nm architecture. Coffee Lake is just a refresh of Kaby Lake with more cores. On the other hand, Zen 2 is also releasing next year, which is 7nm.

File: [GJM] Urara Meirochou - 08 [65E5361D].mkv_snapshot_14.12_[2017.02.24_11.10.34].jpg (181KB, 1280x720px) Image search: [Google]

181KB, 1280x720px

>>62369968
>ultimate gaming cpu
>may end up running games slower than the 7700K

File: cinebench.png (75KB, 759x370px) Image search: [Google]

75KB, 759x370px

>>62365816
This is my score. I didn't even bother to turn off the torrent client or my IDE.

>>62370070
Wasn't the 6800k worse in gaymen than 6700k?

>>62370055
I'm more interested to see if either of them will be supporting DDR5 or not. Not that it'd make a huge difference but it makes more sense to me to upgrade to a new standard at release than to buy a chipset and keep using years old DDR4. And who knows, with Ryzen's performance against RAM clocks it could be that much better.

i7 cores are now worthless.

>>62370055
zen 2 isn't out till the end of 2019. they've already said they're going to re-release current zen on 14nm+ before then

>>62370214
No one said anything like that.

>>62370129
DDR5 won't be coming to mainstream platforms for a couple more years at least.

>>62370242
>This evening, AMD held a conference call to follow up and provide additional clarification to some of the announcements it made during the Financial Analyst meeting on Tuesday this week. During that event, AMD unveiled the Radeon Vega Frontier Edition, alongside its forthcoming Threadripper and EPYC processors. However, GPU and CPU roadmaps were also updated, with AMD showing plans for Zen 2 and Zen 3 architecture. Before we get to future generations of Zen though, there will be an incremental update to the 14nm Ryzen CPUs currently on the market.

>During the follow-up call this afternoon, AMD confirmed that it would be building another generation of Ryzen processors. However, next time around, they will be based on refined ’14nm+’ technology, likely paving the way for efficiency improvements and perhaps better clock speeds.

>After that ‘incremental’ update, AMD will begin focussing on ‘Zen 2’, which is confirmed to be built on the 7nm process at Globalfoundries. Current estimates point to a 2019 date for 7nm. This will then later be followed up by ‘Zen 3’, so AMD is set on the CPU front for the next several years.

>>62370242
>At FAD last week AMD CTO Mark Papermaster revealed that work is well underway on Zen 2 & Zen 3 on the CPU side as well as Navi and it successor on the graphics side. According to the latest comments from Su the first of those new designs will tape out this year at which point post silicon work can start to bring the products up to speed and move towards product ramp and launch.

>GlobalFoundries, AMD’s 7nm foundry partner, announced earlier this year that it will begin 7nm risk production in the second half of 2018. This puts AMD’s 7nm Zen 2 and Navi based product launch time-frames at some time in 2019. At which point Intel is only expected to roll out its second generation 10nm parts. And while GlobalFoundries’ 7nm and Intel’s 10nm aren’t directly comparable, it still goes to show how much the foundry gap with Intel and the rest of the industry has closed.

File: excel wonders.jpg (42KB, 640x480px) Image search: [Google]

42KB, 640x480px

>>62365816
>barely higher than R5 1600
>despite Ryzen having gimped AVX
wat

>>62370311
>>62370333
These appear to be the sources:
https://www.kitguru.net/components/cpu/matthew-wilson/amd-will-build-another-round-of-zen-cpus-on-14nm-before-jumping-to-zen-2-and-7nm/
http://wccftech.com/amd-confirms-7nm-products-will-tape-year-zen-2-navi/

And they are both from May, so it's relatively old.
The 7nm process has been confirmed to be well ahead of schedule so late 2018 for Zen 2 is reasonable.

>>62368189
It's not their past reputation that keeps them afloat. It's the bribing they do so OEMs like Dell don't sell computers or servers with AMD processors.

>>62370428
no. just no.

they're releasing a 14nm+ zen before zen 2. nobody is retarded enough to think they'd release zen refresh and zen 2 in the same year.

zen refresh: 2018
zen 2: 2019

>>62370474
There are no Zen refresh.
14+ products are APUs.

>>62370265
I've read suggestions it'll drop 2018-19, and earlier adoption makes more sense for AMD all things considered.

>>62370474
inb4 my dad works at AMD

>want to upgrade 4.7GHz 4790K
>get the itch for more cores, want a CPU which is an all-out upgrade in every use case
>let's see what's on the market
>Ryzen has similar IPC and clocks 700MHz slower
>Threadripper is the same, of course
>6700k/7700k are marginal upgrades, still 4C and not worth the platform switch money
>Skylake-X is a nuclear disaster exclusion zone
>Coffee Lake is apparently shit too, if these results are real
What the fucking fuck, Haswell is a 2013 architecture and I cannot find a solid, all-around upgrade actually worth spending money on.

>>62370503
once again, no. there is more evidence to suggest they're re-releasing zen with a new name (2xxx series maybe) and with higher clocks to combat intel's upcoming offerings than just releasing apu's.

https://www.techpowerup.com/233495/amd-ryzen-2000-series-processors-based-on-refined-14-nm-process

this^ is probably what will happen

>>62370518
Except AMD has already said that they'll be sticking to AM4 until 2020 at least. And Intel is likely to adopt DDR5 first on the server and HEDT platforms like they did with DDR4. So don't expect DDR5 on mainstream until 2020 at the earliest, and 2021 is more likely.

>>62370441
They're still doing that? I thought they got slammed by antitrusts or something and had to stop.

>>62366085
Boost might not be working or something.

>>62370567
agreed, we will see the xx50 and maybe xx70 on 14nm. possibly not all, but r5< or x sku will.

>>62366487
A 1tb ssd or 16gb extra ram or a nice chink phone or maybe just a lot of beer. And then you're still forgetting you can upgrade to Zen 2 or 3 if you want to without buying a new mobo.

You tell me what you want to do with $200. And if you don't feel like spending just save up a bit so you can maintain high quality of life if you have a bad period.

Of course a consumerist sheep wouldn't get this.

>>62370575
AMD wouldn't be forced to change the socket, they'd use a new chipset supporting the new standard. You're right though, deployment is scheduled for 2020, but I'd speculate that it'll be accessible prior to 2020.

>>62371040
AMD will probably change the socket a little unless they have a CPU dealing in both DDR4 and DDR5. The chipset doesn't need anything, it's just another PCIe device like your video card.

>>62366741
Their yields are bad. The process is probably gonna be equivalent to whatever glofo calls 7nm at the actual feature level.

>>62371731
https://semiaccurate.com/2017/09/11/semiaccurate-digs-intels-10nm-process-problems/
https://semiaccurate.com/2017/09/06/intel-foundry-customer-bails/

File: H11DSi-NT.jpg (547KB, 1600x901px) Image search: [Google]

547KB, 1600x901px

>>62370563
You need double Epyc.

>>62366294
Zen+ in February 2018
2% IPC Improvement
200mhz clock speed improvement

Zen 2 in May 2019
8% IPC Improvement from Zen+
2ghz clock speed improvement

How badly is Intel fucked

File: party.jpg (127KB, 1024x768px) Image search: [Google]

127KB, 1024x768px

>mfw replaced althon with phenom
>then phenom with 6700k
call me in 7-10 years when i'll consider buying a new cpu


cpus are stagnant tech

>>62372007
2-5% IPC + hardware fixes.

>>62372007
> in February 2018
Too early.

>>62371972
that's a big chip

>>62372044
zen+ is 14nm++.

>>62370563
Zen2. Wait a year.

>>62366652
Intel needs a new architecture, their *lakes are tapped out. Just rehashing the same arch over and over.

They didn't invest in a new uArch because they thought they had the market in the bag. There will not be a magical competitive CPU from Intel. Not until they get a new uArch out - and that'll be ~2020. Maybe 2019 if they can really push it, and that doesn't seem likely with the issues they've had on their upcoming 10nm process.

AMD has the CPU market (and Intel's balls) for the next 2-4 years.

File: 1409184332902.jpg (18KB, 512x384px) Image search: [Google]

18KB, 512x384px

>>62371894
>their articles are locked behind a paywall
I want to read about these supposed problems but I can't

>>62371731
>The process is probably gonna be equivalent to whatever glofo calls 7nm at the actual feature level.

From what I've seen, they're really close, but the 10nm process is still behind GloFo/Samsung/TSMC 7nm.

>>62372048
4U

>>62365972
>only within striking distance of 1600/x
>still going to cost more than a 1700 in my country
Welp that's settled, Intel is fucked, FUCKED!

File: Screenshot from 06-07-2017 19_12_35.png (36KB, 571x556px) Image search: [Google]

36KB, 571x556px

>>62372097
It's about as dense. Lower performance. And the mask count...
...let's not talk about it.

>>62372179
Zen on GF and VEGA on TSMC?

>>62372204
Only Vega20 is TSMC, and that's mostly due to GloFo's reticle limit.

>>62372007
There is no Zen+ any more. Just Zen 2 in 2H 2018 on 7nm.

>>62372179
I don't know what the mask count means

The mask is basically a light filter / lens they use to project the chip design onto the wafer or something like that? How does this 1.12 vs 1.00 vs 1.23 translate to that?

>>62372302
I haven't seen anything from AMD indicating Zen+ was canceled.

>>62372353
1.00 can be considered the baseline amount of masks, whatever it is, required to print a complete design onto a wafer from start to finish. 1.23 i believe translates into requiring 23% more masks relative to the samsung process to achieve the same results. So if the Samsung process needs (lets say) 17 masks to print a full design, Intel's process needs ~21 masks. This in turn means more things that can go wrong during the manufacturing process, and decreases yields.

>>62372378
Zen+ is Zen apus.

>>62372465
says who?

>>62366132
>>62366071
At this point they're guaranteed housefires seeing as it's the same uarch as Kaby Lake, where Intel told 7700k owners not to overclock their processors because of heat issues. Just imagine the throttling with two more cores.

File: 5GHz.png (129KB, 1562x332px) Image search: [Google]

129KB, 1562x332px

>>62372378
Then take your head out of the sand, dumbass. There has been no mention of Zen+ in a long time, and Lisa Su recently spoke about Zen 2 coming in 2H 2018. They're not releasing a refresh and then a whole new product line all inside the next 12-15 months. That'd be retarded, and they'd be accused of doing the exact same thing Intel did, releasing the 7700K and then replacing it with something much better in the same calendar year.

There's literally no reason for there to be a 14nm refresh of Ryzen when GloFo's 7nm process is ready and will be ramping to volume production by mid-2018.

>AMD is winning
The best timeline.

>>62372302
AMD won't go for most of 2018 without some kind of refresh of Zen, even if it was only 200MHz+

>>62372566
Nah, you're dumb if you think they are.

>>62372566
>AMD won't go for most of 2018 without some kind of refresh of Zen
RR is late 2017/early 2018, they'll be heavily pushing those APUs

>>62370587
Only in Europe and it's still in appeals although they have to stop until the court bullshit is done.

>>62367470
Only for laptops. When people ask me what kind of laptop to buy, they usually reference i3/i5/i7 and nothing else. When talking about desktops, they don't even know that Intel uses the same branding for desktops. At best they might think that Nvidia makes good graphic cards.

>>62372601
Mainstream segment, they need a 8 core refresh until Q4 2018

>>62372553
>There has been no mention of Zen+ in a long time
That doesn't mean it was canceled

>Lisa Su recently spoke about Zen 2 coming in 2H 2018
Post source; and this also doesn't mean Zen+ is canceled

AMD said Zen+ was coming. Things could have changed, but you aren't presented any actual evidence, just baseless inferences with no evidence to support them.

Until you can provide AMD saying, "yeah we're skipping Zen+, ignore our last roadmap" I'm going to hold them to their published roadmap. They said Zen+ is after Zen, so until AMD states otherwise, that is exactly what I expect.

>>62373404
The most recent roadmap from FAD has no Zen+.

>>62372627
Intel was apparently paying a billion dollars a year to OEM's not to stock AMD

The $800 million fine they got slapped on the wrist with was such a fucking crock of shit. That fine should've been 20 billion.

>>62373420
Can you post it?

>>62373404
There is no Zen+ and there never was. It's fucking RR on a more mature 14nm

File: Capture.png (194KB, 995x717px) Image search: [Google]

194KB, 995x717px

>>62365816

>using 1 solitary benchmark

File: 1504811041382.png (84KB, 253x239px) Image search: [Google]

84KB, 253x239px

>>62373525

>>62373525

oh and x299 mobos start at 100 dollars less than TR4's

also you can buy the cheaper 2100 DDR4 with zero performance reduction

File: 1502143993622.jpg (81KB, 444x444px) Image search: [Google]

81KB, 444x444px

>>62373538

>>62367497
He's not wrong, all modern games know how to use 4c/8t now thanks to the i7's.. and now with all the new gaming platforms moving to 8c CPU's the developers will move onto expanding their core usage.

>>62373579
Pretty much the only people left focusing on singlecore are the incompetent devs who don't know how to multithread their programs.

>>62373525
>>62373538
>muh fox and chicken analogy
Fuck off, retard.

File: destiny2-cpu-bench-1080p-highest.png (36KB, 912x556px) Image search: [Google]

36KB, 912x556px

>>62373606
:^)

>>62365850
>Monolithic dies scaling linearly

File: Mysterious Architect.png (87KB, 386x495px) Image search: [Google]

87KB, 386x495px

>>62365816
>current $230 AMD 12-thread CPU faster than new upcoming $350 Intel 12-thread CPU
I wonder who could be behind these results.

>>62373912
If I delid that off, will he die?

>>62366132
>JewJizz™ as a thermal interface

I guess the real question is, should I wait for desktop 10nm CPU (which aren't scheduled until 2019) or should I go ahead and upgrade for an 8700k coming from sandy bridge 2500k stock? I will mainly use the PC for gaymin, devving, and watching 4k stuff.

>>62374347
>buying intel when amd is a option

my 6600k 4.5 runs absolute dog, feel like worse than it _should_.
haven't formatted pc in years could this b why

>>62372007

Not enough. If AMD is putting as many resources as they can (which they should) on Zen 2, they NEED a 5%+ IPC, and since its on 7nm, at least 4.6 Ghz Turbo. They also need to reduce the latency between CCXs.
If they can manage all that, then Intel will have a couple very bad years until they manage a new architecture. I'm talking absolutely huge layoffs, and possible sell off of subsidiaries or IP. Intel HAS to maintain their margins as they are if they want to avoid that. They have a lot of overhead, plus the shareholders will be out for blood, as they've become too used to the juicy profits Intel has been raking in since AMD flopped with Bulldozer.

>>62373579
they will move to 8cores when games need it. i don't see what features will need it. there are not >200players games, physics is done by the GPU or some other CPU intensive improvement. we just don't need 8cores now

>>62366752
>Speculation is that their next gen arch (ca 2020 should they not fuck up further) will be a "streamlined" x86 with all the inconvenient bits stripped from the ISA.
So making it incompatible with legacy shit? Last time they tried that they were forced to adopt AMD64 and ditch their efforts.

>>62369428
AMD tends to suck for hw video decoding, so unless you have space problems in your HTPC case just pair whatever CPU you want with a GTX 1030 with passive cooling.

File: 1503327587857.png (228KB, 489x394px) Image search: [Google]

228KB, 489x394px

>>62365816
This is sad

Aren't top of the line CPU's already pretty good? Why doesn't intel come up with something completely new. I want integrated computer in my brain so I can just see the internet in my thoughts when I need to know something.

>>62376793
And you'll get cornholed by the NSA in an instant, good job retard.

>>62369968

This will only fly if it is soldered and offers good temps

>>62377201
Well you know that won't happen

>>62370428

Why would they rush it. Zen is a good product, people are buying it. You release a refresh that will not have the same problems as Ryzen had at start and people will be happy.
If rather they take their time and make sure Zen2 is perfect, instead of rushing it out earlier.

For once, AMD can play the waiting game, what with Intel not really going to one up them.

>>62374560

The 7nm process is designed to have a base clock of 5Ghz and the cross ccx latency has been reduced already on TR and EPYC in comparison to Ryzen.

Intel will be in trouble and if anyone in that company has an iota of economic sense they will give up their insane margins and instead put everything they have saved into R&D. Otherwise there may be a very real chance they could go the way of the dinosaurs.

>>62369181
Can't wait for Ryzen APUs. I hope they can drive R7 360/370/RX 460 performance.

>>62377310
6502 failed to make Intel go the way of dodo.
Don't worry, they'll be fine.

>>62365816
>[x] RELEASE KABY LAKE AGAIN

>>62377349
That was 40 years ago before all of their horrible rebate programs, compiler bias nonsense, absolutely insane price markup in the server premium line, "diversity" hires, etc.

It's a completely different world now, I have no idea why you keep harping on the 6502 situation.

>>62377548
Because during 6502 times Intel's products were straight up uncompetitive.
At least their current product stack can compete on performance (and sometimes price).

File: big_ryzen-chart-1.jpg.jpg (79KB, 1080x463px) Image search: [Google]

79KB, 1080x463px

>>62367265
k

File: Bench.jpg (542KB, 1645x947px) Image search: [Google]

542KB, 1645x947px

It's like doing a re-release of Haswell-e.
Here's my 5820k at 4.2Ghz
This "progress" is pathetic at best.

>>62367952
>glorious portuguese language used as mock because our retarded monkey offsprings can't stay quiet
I don't know if I should be sad or happy.

>>62377236
To leave intel in the dust by turning their 2020 release obsolete.
Once you're in the lead you have to grasp it firm, or else it escapes and you're fucked again.

I'm pretty sure AMD learned their lesson and are currently going full SHOA with their roadmap releases on Intel.

>>62369573
That's true I remember seeing it some where getting 1400 at 4.4 ghz.

>>62370563
Wait for the 7nm chip, I'm in the same boat, we bought the last decent Intel CPU, everything else is not worth upgrading, hell...I even tell friends that are running good 2nd gen i7s to not upgrade and wait for Zen2.

That's what happens when a company has a monopoly and just sits around releasing minor improvements to milk the market as much as they can...

File: amd_ryzen_wraith.jpg (51KB, 730x411px) Image search: [Google]

51KB, 730x411px

>>62373473
1850X get blacked edition w/ bundled wraith max is coming Q1 2018. screencap it.

>>62365872
>End of 2017
>Getting ANYTHING besides RyZen R5/R7/Threadripper
Yeah, nah

>>62365956
>Same shit, just a little bit more COARZ connected via a LITERAL MESH OF BINGBUSES
Mmkay

>>62365958
>kylake-x got a mesh bus instead of a ringbus
It's actually a mesh OF bingbuses, lol

File: 1490875292753.gif (875KB, 250x231px) Image search: [Google]

875KB, 250x231px

>>62378095
>Ryzen Blacked

>>62365965
>Good single thread
>>62366021
>tops everything
Just a daily reminder my 6-years old Sandy i7 2600K gets those 1200-per-core when OverClocked to JUST 4.5GHz on air. Just Saiyan.

>>62366052
Most funniest thing is that it actually cannot beat even non-X 1600, because you can OverClock 1600 just fine.

>>62369391
>I really wish AMD would bring a Zen iGP series
They are, they're called APUs and their APUs have always had infinitely superior graphics to Intel.

>>62378151
There is no way your 2600K is hitting 1200.

File: Intel Nazi Swastika.png (194KB, 443x390px) Image search: [Google]

194KB, 443x390px

>>62377355
Ahckchyually...http://i.imgur.com/K92QwvT.gif

>>62378129
Cache sized are vastly different.

>>62378261
My bad, I've derped out and 'dun goofed, and read numbers on >>62365965 pic wrongly. 8700K gets roughly 2000 per core (200 by CineBench). My Sandy i7 2600K gets 1600 per core on 4.2GHz, 1800 per core on 4.4GHz, and beats 8700K when OverClocked to 4.7GHz

>>62378261
if you OC it 4.5 on water it can
intel multicore scaling hasn't improved a bit, zen is amazing due it it's almost 100% scaling madness

>>62370427
There's not gimped AVX on RyZen. It's just not 512. Why the hell do you need 512? It's a literal meme. Standard AVX/AVX2-wise, there's absolutely nothing that RyZen lacks any. And please, don't ever throw PcPercucks links at me, that shit is downright imbecilic at how autistically shilling it is. They literally bake up "yellow press"-tier made-up fairy tales in regards to Zen.

P.S.
http://www.youtube.com/watch?v=5eD639QNuF4

what do you guys think of getting the 1900x for gaming?

>>62378458
There's no need for water on i7 2600K at 4.5GHz, and not even all the way up until 4.8GHz, for as long as you have top tier air. I have exactly that. Sure, temps are high, but they never go above ~82c when fully loaded and it doesn't throttle at all. I also have very good airflow chart in my case/chassis with a set of best case fans (not Delta, lol) you can buy out there, so that helps a lot too.

>>62378512
Will be awesome in ~3 years from now, but right now it's best to go cheaper solutions. 1600X and 1700/X are currently best choices for /g/oyming for a reason, ya know. No need to overpay right now. Threadripper is not for /g/oymers, but for those who work a lot AND /g/oyim from time to time. If you're just /g/oyming all the time and almost never do anything else on your system - go RyZen R5/R7.

>>62368341
5820K is only 3.3ghz base clock with 1bin all-core turbo (3.4ghz all-core turbo clocks...)
8700K has 3.7ghz base clock with 6 bin all-core turbo (4.3ghz all-core turbo clocks...)

That's 400mhz at stock clocks and 900mhz (~25% of the entire clockspeed!) difference.
Yet the 5820K smashes the 8700K.

Intel really fucked up bad.

File: 536e016a4f91915502a6d0e7b6d199ba.png (566KB, 998x846px) Image search: [Google]

566KB, 998x846px

>>62367351
>prebuilt market will still be intel's for years to come

>>62378566
WOW WHAT THE FUCK ahahhaha intel is dead

File: generic drinking man reaction image.jpg (34KB, 413x395px) Image search: [Google]

34KB, 413x395px

>>62365816
>barely beats the 1600

File: 1489122742869.jpg (49KB, 565x486px) Image search: [Google]

49KB, 565x486px

>>62367351
>>62378576
>Dell, HP, Alienware, Maingear, iBuyPower, Lenovo, and many many others - are all on AMD's side already

File: ryzen1700-r15-3750-on-allcores.png (1MB, 1234x746px) Image search: [Google]

1MB, 1234x746px

I hope this chip at least OC's well, because lordy those be some shit scores.

File: 14836006936380 44.png (389KB, 512x259px) Image search: [Google]

389KB, 512x259px

>>62378593
>and it's not even OverClocked, you can OverClock 1600 just as easily as 1600X
>Literally 80~100MHz OC to beat i7 8700K out of existence

>>62365816
It'll still perform better than AYYYYYMD in CS;GO at 640x480/Very Low, so that means Intel win.

>>62378621
>153 per core
What frequency? My Sandy 2600K does 160 on 4.2GHz, 180 on 4.4GHz, and over 200 on 4.7GHz

>>62372674
Nigga 4c IS the mainstream.
That's what normies REEEEEEEEEEEE buy in laptops, that's what HP puts in desktops, that what Dell puts in AIO units, 8 core would be ideal in the DIY market sure, but RR is what will tackle most of the market, and more importantly, the mobile market.

File: 14825037260090.jpg (248KB, 1000x662px) Image search: [Google]

248KB, 1000x662px

>>62378641
http://www.youtube.com/watch?v=oMNFoNtKCR8

>>62373525
>Gaming 108%
>Desktop 105%
>Workstation 113%

vs

>Gaming 107%
>Desktop 105%
>Workstation 171%

>Somehow the Threadripper is 1% 'worse' than the i7
Hmmmmn

File: 14910505504130.jpg (109KB, 540x679px) Image search: [Google]

109KB, 540x679px

>>62378654
>4 cores IS the mainstream
In 2010.

File: R15.png (1MB, 2498x1440px) Image search: [Google]

1MB, 2498x1440px

>>62378647
3750mhz
My 4820K does 152 (single thread) on 4.5ghz - so I'm not quite sure I believe you, post some proofs.

>>62374347
Intel's 10nm node is looking to be slower than their current 14nm++, and will continue to be so until Intel get their 10nm+ online, Intel said this a while back to their investors.

>>62377310
>if anyone in that company has an iota of economic sense they will give up their insane margins and
Not going to happen.
Brian promised shareholders that he would happily surrender marketshare to preserve their 60% margins.

>>62378671
4c8t is sufficient even for most workstations
Plenty of RAM and a good SSD is what usually make a difference then

>>62378696
>this is what Incels actually believe

File: BOLIEVE IT.png (4MB, 869x600px) Image search: [Google]

4MB, 869x600px

>>62378676
Here's 4.2GHz, for example:

>>62378696
>4c8t is sufficient even for most workstations
H-h-h-h-how i-i-is y-yo-you-r-r stu-t-t-t-tt-t-ttt-tutter-ing g-gg-oin-g?

File: cinebench.png (125KB, 1920x1080px) Image search: [Google]

125KB, 1920x1080px

>>62378730
What does CPU-Z have to do with Cinebench? A 2600K doesn't get anywhere near 200 points single core. Mine at 4.6GHz gets 166 single core and 820 multi-core.

>>62378730
Use the latest version of CPU-Z you faggot

>>62378760
>What does CPU-Z have to do with Cinebench?
CPU-Z's test is exactly same shit, dude. Just without stupid rendering in the background. They differ only in that CPU-Z shows in 1000's, while CineBench shows in 100's. Just remove one 0 and you'll get exactly same score in CineBench that you get in CPU-Z. They're LITERALLY same shit, score-wise.

>>62378671
You look at what is mostly sold.
Laptops.
When was the last time you seen a HP laptop on a shelf with 8c/16t?

>>62378763
There's no reason to. It's just a copy of CineBench. All they did is remove one 0 from the score table, so now it just mirrors CineBench 100%. I prefer to count my scores in thousands, not hundreds.

File: 1fec6f7de86d1a55b2becf3d1b418af2.png (560KB, 997x845px) Image search: [Google]

560KB, 997x845px

>>62378576
Oops, wrong one. That shows sales revenue.
Here's the sales figure.

As a note, these are from Germany's largest eTailer, so a single store.

>>62378788
>When was the last time you seen a HP laptop on a shelf with 8c/16t?
With RyZen introduced to the market, they'll be everywhere soon.

>>62378779
You are dumb.

>>62378804
>Laptops
Nigga, we're talking about Laptops here now.

>>62370123
Week one results even showed the 7700k losing to the 6700k in several games.

>>62378760
>A 2600K doesn't get anywhere near 200 points single core
At 4.7+ GHz it does pretty easily.

>>62378779
>>62378792

They actually changed the scoring and the code of the benchmark for one thing. Secondly they are entirely different benchmarks that perform different tasks. Saying the scores are the same is fucking stupid and also completely wrong.

>>62378822
No, it doesn't. I don't know what you're trying to achieve with this shitposting really. Maybe you were hoping nobody else with a 2600K was online to call you on your bullshit.

>>62378812
Is this all the "arguments" you have? Nice try, kid.

>>62378814
I'm too talking about notebooks, you down syndrome-enriched imbecile. HP does what now? That's right - prebuilt systems AND notebooks. And they have long-term partnership established with AMD on RyZen.

File: 14849308452730.png (182KB, 480x480px) Image search: [Google]

182KB, 480x480px

>>62378824
>Saying the scores are the same is
>CPU-Z shows 1600 per core
>CineBench shows 160 per core
>"NOP, 'dem benches not same at all!"

File: cpu-z.png (16KB, 403x402px) Image search: [Google]

16KB, 403x402px

>>62378835
>Is this all the "arguments" you have? Nice try, kid.

No, I've already posted my Cinebench score and made you look dumb, so here's my CPU-Z one as well. Literally no correlation to anything. This really hasn't worked out for you, has it? Time to drop the act, kid.

>>62378847
Cinebench shows 166 single core at 4.6GHz. You said your sub-1600 CPU-Z score was at 4.2GHz. You've shitposted yourself into a corner here, friend.

>>62378835
And HP having a long term relationship somehow means they'll produce 8c/16t laptops how?

Constrain your imagination to the realm of possibilities else you lead to undue hype.

>>62378847
>duh numbahs ah duh same
>CPU-Z's test is exactly same shit, dude

Hahaha jesus christ you're the stupidest person in this thread, congratulations.

File: BELIEVE IT.png (4MB, 1240x414px) Image search: [Google]

4MB, 1240x414px

>>62378848
>>62378860
>>62378887

Sigh...yare yare...okay, maybe it's not "2000" exactly, but pretty close nonetheless. Given that my CPU is 6 years old, degradation of crystal is pretty apparent. It's purely on top tier air and not even simple AIO water, though. Back when this CPU was only just bought, I could do 2000 just fine, but it looks like rigorous usage and time did take their toll on it by now. So there's that. Either way, my point stand solidly.

>>62372007
>2ghz clock speed improvement
This is what AMD fanboys actually believe.

>>62378864
>And HP having a long term relationship somehow means they'll produce 8c/16t laptops how?
What part of "SOON" you didn't get, wigger?
They're already working on gen1 Zen APU notebooks that are going to be released this year, next year it'll be full sized RyZen in there.

>>62378952
Fuck off

>>62379163
You were just BTFO, so I guess I will admit your defeat.

File: 5gigacocks of holocaust.png (21KB, 445x230px) Image search: [Google]

21KB, 445x230px

>>62379063
>REEEEE IBM nodes for consumers is cheating
t. Rabbi Shekelshtein

>>62379229
>5ghz
>2ghz higher than 3.8ghz
AMD users can't even do maths lmao.

>>62379296
5ghz is average operating frequency. >>62372553

>>62371972
>Epyc
>not Ebyn
come on anon... the joke was there and all you had to do was grasp it.

>>62365831
*Delid dis

File: ebyn.jpg (1MB, 2048x1536px) Image search: [Google]

1MB, 2048x1536px

>>62371972
>>62379725