>no thread about the freshest ryzen price leak what did

We don't care about AMD here

File: 1485371381044.gif (1013KB, 245x251px) Image search: [Google]

1013KB, 245x251px

>>58852103
intel shills on suicide watch

ayymd and jewtel are botnet shit
open-source arm cpu when

Why the fuck do 6core 12 thread ryzens compare to 4core 4 thread i5s?
Shouldnt ryzens be stronger?

is that website supposed to read overclock 3d or overclocked?

>>58852103
>We don't care about AMD here
At Intel HQ ?

looks like bullshit.

>>58852146
I wouldn't pay too much attention to their comparisons yet. AFAIK only the prices/models are the leak, the intel-equiv list is just what the site believes.

>>58852098
>>58852098
oh shit, amd. what are you doing??? you want 100 bucks too much. it seems i wasted my life waiting all these years for zen.

>>58852187
those are euros tho

>>58852164
looks more realistic than anything else before it

also 1700 has 65w TDP which would be impressive

>>58852098
>€599.99, which includes tax
So that would be about €480 without tax, which is about $510, likely $500 flat state-side. Not a bad price if it will be able to compete with an intel 8-core, as it would cost as much as intel's 6-core.

I'm disappointed in the cheaper SKUs. they clock lower than the higher core count chips. there must be some intense binning going on right now.

we might end up with some monster HEDT chips but otherwise nothing but sloppy seconds in the lower market segments.

all will depend on motherboard pricing, if it's 2011 all over again it will fail
if it has prices of around $100-150

450 for a 8c/16t doesn't look expensive

>>58852219
it's not 2011. no quad-channel memory. fewer PCI-e lanes. motherboards are going to be cheaper to make than even the mainstream Intel stuff.

>>58852198
But the euro is essentially at parity with the dollar now

There are conflicting leaks
https://www.guru3d.com/news-story/amd-ryzen-4-2-ghz-processor-entry-surfaces-at-taobao-with-28th-feb-availibility.html

>What however is interesting is an availbility date on the 28th or February alongside a ¥ 1999.00 pricetag, which translates to 275 euro and 290 USD. The chip is listed at 14nm and yes, that 4.2 GHz turbo clock frequency is mighty interesting, as it does seem 200 Mhz higher then expected.

>>58852198
>>58852187
And most likely with a ~20% VAT already on top as well

>>58852130
>arm
Never. Because ARM itself requires licensing.

SPARC is your FOSS ticket.

>>58852251
that might be 4c or 6c

>>58852216
Clockrates have never mattered, if you think they do you're a consumer fuckwit idiot.

Instructions per second is the only metric that counts.

>>58852251
All the leaks are bollocks, AMD is holding their cards until release to prevent Intel organising a response as long as possible.

Any 'leaks' are clickbait articles.

>>58852198
the exchange is at 1.06 tho

No suprise at base 8core chips coming in at mainstream i7 prices, it's the biggest upset they can cause and it finally brings a decent performance improvement to lower tier consumers.

>>58852216
>>58852251
This is the thing, that r3-1400x chip is the only one I can see hitting 4.2ghz and there's a market for it, my best guess is it's the highest clocking part but isn't coming out on release so it doesn't cause any complication with marketing the 1800x. Not that anyone on /g/ should care where everything is left at whatever oc you want.

>>58852276
You're not very bright.

Already bitching stfu and wait nerd

>>58852298
Or cleverly designed to gather market feedback

>>58852254
>which includes tax
Can't you read?

>>58852276
They certainly matter less when every chip is unlocked.

>>58852254
According to Guru3D, a Spanish store leaked some prices, VAT included

http://www.guru3d.com/news-story/amd-ryzen-r7-lineup-of-8-core-16-thread-cpu-prices-revealed.html

Spain's VAT is 21%, so considerable.
Europe also gets fucked on PC prices anyway.

>>58852302
Please explain how clockrates matter vs. actual performance metrics.

I'll wait.

File: 1467493940200.png (12KB, 246x200px) Image search: [Google]

12KB, 246x200px

>mfw those prices are roughly the median monthly income in where I live

>>58852276
they matter if we're comparing CPUs from the same uarch, because then we can assume IPC is equivalent.

as it stands, intel's 4c/8t chips clock much higher than the broadwell-e 6+ core models and have superior single core perf as a result. this is all verifiable info.

take your reddit tier knowledge and go back.

>>58852301
hopefully you're right and those boost clocks go through the fucking roof. but if they really did clock that high, wouldn't the base clocks also be higher?

>>58852098
Intel is officially dead.

>>58852098
>R7 1700 being priced at €349
I'm seeing €389.95 here, they changed it or you wrote it down wrong.

>>58852332
>they matter if we're comparing CPUs from the same uarch,
Sure, but that's not what the poster was doing.

>as it stands, intel's 4c/8t chips clock much higher than the broadwell-e 6+ core models and have superior single core perf as a result. this is all verifiable info.
>More cores in a similar thermal envelope results in lower clockspeeds
WHO WOULDA THUNK IT!!!

>>58852276
Your performance metric is as overly simplistic as the one you are currently criticizing

>>58852329
higher clockrates mean more clock cycles can be performed in any given period of time and hence more cpu operations. try downclocking your cpu a ghz or two, see how you like it.

>>58852332
>take your reddit tier knowledge and go back.

why don't you try talking to this guy, he needs a chatting to... >>58852365

>>58852342
VAT is 21% and europe gets worst pc parts prices in the world

>>58852130
ARM is as proprietary as x86. I don't know why people thing getting cucked by a different company is any better. Heck intel is already the cheap commodity processor company. They uncucked us from specialised systems like expensive IBM Mainframes.

>>58852379
So clock rates have absolutely zero indication of performance?

>>58852332
>hopefully you're right and those boost clocks go through the fucking roof. but if they really did clock that high, wouldn't the base clocks also be higher?
Mega boost clocks aren't a bad shout, might have low base clocks to keep their tdp in the 65w range as well. Then again I doubt 3.6ghz is hard to hit for 4c if the 8c does it at 95w.

As long as there's no parts like shitty low end i5 (<3ghz with no smt) the entire range should be fine. Bring 4c/8t parts to i3 pricing seems to being the biggest upset here.

>>58852391
I'm in the UK so I'm going to get hit by full english brexit tax on top of that probably. I was just pointing out the discrepancy in the OP is all.

>>58852098

it is overpriced as fuck
will not upgrade til 2021

>>58852276
Skylake and Zen have roughly the same IPC. That's why clockrates matter.

>>58852401
I don't understand why more people don't get this.
SPARC is the only decent free harvard arch now.

But no one does it because ARM was cheaper before SPARC was free.

>>58852358
>Sure, but that's not what the poster was doing.
I was comparing ryzen to ryzen. can you fucking read?

>WHO WOULDA THUNK IT!!!
that's exactly what I'm saying. but it isn't the case here. smaller ryzen chips also have lower clock rates. that means they're just bad candidates.

>>58852420
>Im a poorfag and a general failure in life

File: 1480438418776.jpg (191KB, 640x633px) Image search: [Google]

191KB, 640x633px

>>58852098

>€

EUROPOOR! HAHAHAHAHA!

File: 2017-02-08_16-27-36.png (213KB, 780x439px) Image search: [Google]

213KB, 780x439px

>>58852342
That's like 30€ above the 7700k.

I don't know why the poorfags are so mad.

They probably expected to get 8 cores for free.

>>58852256

>sparc

Are you serious?

>>58852462
Or same as a 5820k

As long as they overclock well, the base 1700 for around 380€ is a killer upgrade for those of us stuck on Sandy/Ivy K-series.

>>58852403
Across different uarches/arches - Yes. absolutely.

Thought experiment time:
an engine doing 100rpm that produces 1000J energy per cycle vs an engine that does 1000rpm that only produces 80J per cycle.
Which is doing more 'work'...

Cycles/speed doesn't matter, work output does.
In the same uarch, more clocks generally means just means more work, but a also now consider (real world example time!) an FX9590 (5ghz) vs. an Xeon E5-2690 V1 (2.8ghz) both are 8 core, both are x86, The Xeon is much faster, even in single core integer math, than the FX9590.

Clockrates themselves don't matter, work output per X period does.

>>58852475
It's the only recent harvard-like arch that isn't encumbered by many patents and crazy licence agreements.

I didn't say you should use it.

>>58852496
Seriously fuck off with your car analogy
I understand ipc and clock speed you massive faggot

>>58852098

Is the R7 line up was the one that only will be released in March?

Are the R5 and R3 an APU line up that rumored will be out in November?

I need to upgrade my shitty sandy.

>>58852522
That nice.

>>58852524
All of them are CPU only without iGPU and supposed to be released at the same date (at least as a paper launch), APUs aren't expected until this summer.

>>58852276
Instructions per second is instructions per clock times clockspeed. How does clockspeed not matter there?

>>58852533
Faggots and your shitty car analogies. I hope you get throat cancer and choke on your own blood.

>>58852548
That's nice.

>>58852524
R5 is 6core cuts of the 8core chip.
R3 is probably a 4core cut, some here insist that they are not going to soon, but most expect them after release. The APUs will probably be another product line and then R3 ryzen chips will be populated with cut apus and 8cores.

>>58852538

Great, looks like I'm gonna get the R5.

If this rumor true then the cheapest R7 will cost 50% more than 6700K in my country.

>>58852524
6 and 4 cores might be released some weeks later. AMD want their top dog to take the stage for reviewers

>>58852541
Not every instruction takes the same amount of clocks to complete.

Let alone cache misses or other failures.

File: 1486005137414.gif (409KB, 315x318px) Image search: [Google]

409KB, 315x318px

>>58852496
nice analogy fag. here's my food analogy

>making spaghetti bolognese
>have your mince, your vegetables, your spices, your spaghetti etc
>'TIME HAS NEVER MATTERED, IF YOU THINK IT DOES YOU'RE A CONSUMER FUCKWIT IDIOT'
>throw all the ingredients onto the stove and immediately take them out again
>your dish turns out raw
>'Gee whiz how the fuck did that happen???'

>tfw Zen looks like it's going to be fast enough to the point that Intel having a slight IPC advantage won't matter at all for people benchmarking Skyirm on low at 736p
>tfw all unlocked so I might finally end up overclocking
>tfw 6c/12t for midline gaymen
>tfw AMD has VT-d or whatever their equivalent on all their chips
>mfw I could ditch this 3570K for a 1700X and run a Linux host with a Windows gaymen VM using GPU passthrough

I want this now, and I want it bad. Do these come with a shitty iGPU I can use for the Linux host, or is my plan ruined?

>>58852538
>>58852560
APU line is a different silicon called raven ridge. Scheduled for Q4 2017, said to be 4c/8t and graphics performance expected to be somewhere between 750ti and rx 460

>>58852581
I guess AMD are going to have to go back to performance ratings rather than clockspeeds, since idiots like you still exist.

>>58852592
>I want this now, and I want it bad. Do these come with a shitty iGPU I can use for the Linux host, or is my plan ruined?
No iGPU

>>58852617
Unless you want a mini-ITX board you should have one PCIe slot free for a cheap R5 250 or GT710.

>>58852617
Seeing as even a shitty barely-working GPU is a decent chunk of money because used things don't lose much value in this third world shithole, I'm now sad.

These price leaks helped me decide on the Intel.

>>58852712
Buy a shitty gpu form overseas.

File: chrome_2017-02-08_12-56-35.png (46KB, 1064x353px) Image search: [Google]

46KB, 1064x353px

>>58852098

>R7 1800X €599.99 8/16 4GHz
>i7 6900K €1.117 8/16 3.2GHz

INTEL BTFO!

>>58852712
Both R5 230 and GT710 cost 31 Euro here in slavland. I don't know what 3rd world shithole you live in but I already pity you.

Hope these chips can make 4.5Ghz with ease like every skylake-kaby lake out there.
If not I'm not interested, I only care about IPC and single threaded performance.

>>58852738
>go to customs
>"What'd you bring into the country son?"
>"A graphics card"
>"What?"
>"A computer part"
>"Well that's not on the whitelist that consists of literally just books so you're going to have to pay a 50% tax on the total price, product and shipping"
>Alright, I paid 50 dollars total for that thing on ebay
>"Whoa, $50? This isn't right"
>Customs guy goes on the local equivalent of ebay and looks up what I just bought
>"On the local market they go for $150, you're going to have to pay 50% of that plus whatever the shipping was"
>"But I paid $50"
>"Yeah, but that's not what they're worth here, and how do I know you didn't just tell the guy you bought it from to put a low price on the invoice just to avoid taxes?"

This has literally happened before to several friends every time they imported something. They end up paying so much in taxes whatever you imported is more expensive than even buying local.

Because, you know, we gotta protect all those stores that live off importing shit in bulk so they get discounts (or know a buddy in customs who will simply let all their shit go through without question) and then resell for a profit. If we let everybody import themselves, they'd go out of a job!

>>58852130
That's a funny way to spell RISC-V.

>>58852330
Most expensive one is something around that. Stop wasting money on vodka and kebabs and start saving up. How do you think all those chavs get those iphones when they are not pushing drugs?

>>58852808
>single threaded performance.

You could just gem Pentium G OR I3 and be happy with it.

Why bother paid 4 times the price for minimal gain in single thread performance?

>>58852852
RISC-V is good but we still need open GPUs and chipsets.

File: CCFkbLRw0kxh_Nq_93Is089TOZ0WstnFPbFJ1ttnOek.png (381KB, 607x767px) Image search: [Google]

381KB, 607x767px

>>58852098
the new Skylake-X and Kabylake-X are going to rape Zen so bad

no one can compete with Intel, AMD thinks they can but they can't

they will be so absolutely destroyed it will be like Nagasaki after the nuclear bomb(Intels new premier products) has been dropped

i'm already pre ordering buckets to collect all the Ayymd fanboy tears,since there are going to be many

>>58852795
Tbh those are beyond shitty. I think me meant stuff at least in the 750ti range.

>>58852872
There's no iGPU in the 750Ti range even in Intel's offerings.

>>58852845
>take apart laptop
>go through customs
>>woah what's all this then?
>parts...?
>>Cary on

>>58852845
What shitty country do you live in?
>Honestly though that's not meant to be an insult, that fucking sucks. I'll give shit to idiots all the time, but that's just shit you get stuck with.

>>58852845
what kind of backwards country taxes you for repair parts?

>>58852910
I don't see how that relates to what I just wrote. iGPU's are terrible.

>>58852872
But why? He only needs it for 2D shit like displaying the desktop on a monitor and video acceleration since all gaming will be done in a VM with the gaming GPU, as I understand it. A dirt-cheap low-end GPU is absolutely enough for that.

>>58852934
They're fine, people are terrible for expecting more than just basic video out of them.

I'd love a 1cu@400mhz iGPU in Ryzen CPU's proper (not just the APUs) just so we have something to drive a display if a GPU is fucked or otherwise to drive a GPU for a host OS in a VM situation.

>>58852845
But at least you have cheap internet?


:^)

>>58852868
this

the 6900k-equivalent zen will cost 600 fucking euros

the r7 1700 is only a 65W part which will be utterly BTFO by the 100W i7-7740K in real-world applications

>>58852934
this >>58852955

>>58852968
they can just put 1cu gpu in SB to be honest.
No need to pollute CPU with such nonsense.

>>58852971
ebin copypasta from another thread you shill >>58852803

>>58852301
Lisa Su said they were going to be going for a full product stack at launch. I doubt they would hold off on launching their quad-core chips in this case.

>>58852993
kys AMDeluded idiot

>4GHz
It's over, Intel is finished.

>>58853000
waist of trips

File: clockmed.png (26KB, 133x124px) Image search: [Google]

26KB, 133x124px

>>58852920

>>58853006
>waist

>>58852980
Probably, SB/NB doesn't really exist anymore though (it's all on the CPU, Ryzen is SoC, even the big 'cpu only' versions)

However the 'chipsets' that exist connet over pci-e, so you could easy chuck a 1cu GPU there. but desu that's probably worse than just doing it onthe CPU, the APUs are still going to have 'bigger' iGPUs on die, so why have multiple GPUs on a cheap APU system?

2/3cu's is a small as fuck space, do that just so that 100% of all 'cpu only' ryzen dies can have 1cu (binning constraints!)

>>58853015
>implying

File: 1464809870978.png (191KB, 419x398px) Image search: [Google]

191KB, 419x398px

>>58853015
>>58853006

>strapping trips to your waist.

>>58852330
Where do you live?

File: 1486565313962.jpg (123KB, 586x575px) Image search: [Google]

123KB, 586x575px

the r7 1700 is only on par with an i7-7700. it's better to have more powerful cores as in the 100W i7-7740K than to have more, but weaker cores as in the 65W r7 1700

>>58853033
>antifa why

Sounds like something those terrorist cunts would do.

>>58853033
>not strapping trips to your waist

>>58853051
>>58852022
Jesus fucking christ stop your shilling already you retarded samefag!

>>58853048
He doesn't.

>>58853051
>it's better to have more powerful cores
Not for multi-threading it's not.

>>58853051
What is this 7740k everyone is jacking off to lately?

>>58852504
>Oracle

Anon pls

>>58853071
its Kaby Lake X, LGA2066 package of Kaby Lake with no benefits of LGA2066 package.

>>58853071
Intel got scared and plans to release a 7700k with a slight bump at clocks, price, and a different socket.

File: 1486569512698.png (403KB, 616x395px) Image search: [Google]

403KB, 616x395px

>>58853063
you linked to here from that thread and it's the same topic you moron

>>58853070
for things like video encoding it can be better to have MOAR COREZ but single-threaded performance is generally important. see https://en.wikipedia.org/wiki/Amdahl's_law

>>58853071
kaby lake-x. it's a souped up 7700K for LGA 2066

>>58853089
Its 4 corez at 100+ watts? So 5.5ghz or something?

>>58852923
>>58852932

Argentina

>>58852970

Hohoho... I pay the equivalent of $50 (a sizable chunk of cash here) for 10/0.5 ADSL. One provider in my area, the oldest phone company in the country. That's the highest speeds their 50 year old copper infrastructure (literally, they don't repair unless absolutely necessary) could provide. It also cuts out for a few minutes once a day every single day at seemingly random times, and they say it's not a problem and that it must be my computer (even though I can't even ping Google's IP address on any of my computers, or from the router's ping test utility).

There was a cable company rolling out that was going to offer 40/5 for the same amount of money, they even put up poles and a decent amount of wiring. But then between the phone company giving the mayor some money under the table, and the cable company refusing to give him more than the phone company was, he legally cockblocked them (they put up poles and wiring all over town, and then mayor said "lolno", I can't imagine the amount of money they lost) and now I'm still stuck with this shit Internet.

>>58853107
Yes, I really shouldn't have. Also holy fuck on you quoting Amdahl's law, on most things we're not even close to having an overwhelming amount of single-threaded operations.

>>58853123
I'll cry for you then I guess...

>>58852098
>what did (((/g/))) mean by this?
It means you need to stop being a crossboarding faggot for everyone's sake.

>>58853107
So I guess Intel's 10c/20t for $1700 is completely useless according to that link you posted.

File: threads-throughput.png (41KB, 600x371px) Image search: [Google]

41KB, 600x371px

>>58853107
>Amdahl's law
You mean this?

>>58853107
Amdahl's law is not relevant here. If you think it does you should leave /g/ immediately.

100% more cores beat 20% faster cores in 90% of use cases and in every other case like that GTA benchmark the difference is negligible.

>>58853165
it's only useful for certain specific tasks. you only go many-cores and dual-cpu on a desktop if you have a specific application you can use it for or if it's for a server

>>58853193
Then why do they sell it as an enthusiast CPU, anon?

>>58853193
>You only need multi-cores if you use multi-core application

WOW amazing

File: 1485528644867.jpg (84KB, 920x754px) Image search: [Google]

84KB, 920x754px

>>58853123
Im sorry anon. I try not to bully others on their older equipment because I know it actually costs a fortune in their own country.

>>58853200
what the fuck is your problem, kid? of course they wouldn't market a $1700 specialty cpu to plebs.

>>58852342
>tfw getting a $80 ryzen quad core

>>58853218
If you don't know what "enthusiast" means with respect to Intel's marketing strategy you should probably get off /g/ for a while.

File: 1486529091627.png (408KB, 571x579px) Image search: [Google]

408KB, 571x579px

>>58853201
you're not worth my time. enjoy your retarded AMDelusion circle-jerk

>>58853148
>>58853203

Meh. At least college in this country is "free", and since my dad's a handicap we pay basically no taxes, so I can drop the quotation marks in my case. It's basically the reason my parents moved back here, too bad they left before they got our citizenship papers sorted out. The degrees are actually worth the paper they're printed on, too. High school was a laughing stock of political indoctrination and teachers eternally on strike (my sister's still in school and she missed out on more than 30 days of school because teachers wanted more cash), but the universities actually hold themselves to international standards and filter about 95% of first years.

That being said, as soon as I get my geophysics degree I'm getting the hell out of this and doing a Master's in the good ol' US or somewhere in Europe. My college has a shitload of exchange students with German and Dutch universities, basically everyone I know who graduated had the option to go to Germany with some sort of scholarship. If it comes to it, I'm more than willing to marry some desperate hamplanet while I'm there and wait out however many years it takes to get citizenship. I'll do anything to get out, preferably back to the USA because 2nd Amendment.

>>58853246
>>58852863
Looks like the actual retard here is you!

>>58853246
>say something so retarded that you can't back it up by facts on technology board
>I DONT HAVE TIME TO WASTE ON YOU

>>58853277
ooh you got me! le ebin samefag!

>>58853279
>blablabla i really enjoy spewing useless bullshit which no one gives a shit about

>>58853291
>only just finding out about "ebin"
>being this new

File: 1483737017981.png (973KB, 801x1500px) Image search: [Google]

973KB, 801x1500px

>>58853291
Glorious meltdown just like Bentium 4 xDDDDDDDD

>>58853274
You can come to the US and get married to me , Anon ;-)

>>58852845
If you paid 50$ for it and have to pay 50% tax on the 150$, it's still cheaper.
>people on /g/ can't into math

>>58853274
Pro tip: Nothing is free in life

File: 1455406557873.png (186KB, 1920x974px) Image search: [Google]

186KB, 1920x974px

(((They))) are afraid.

>>58853274
A/S/L?

>>58853351
That was just an example off the top of my head. I didn't check if the math worked out. Things on the local market are usually slightly over twice whatever the price in the US is, so you end up paying more once you factor in shipping (they also tax the shipping at 50%). With used things that retain more value here, you're still lucky to break even, and you run the risk of customs just arbitrarily retaining your shit because the guy at the counter is having a bad day. And customs likes to retain used items because "how are local manufacturers (even if in the case of computer parts none are fabricated here AT ALL) supposed to compete!".

>>58853342
I would honestly marry a man and take dick in my ass for 3 years if it meant a green card at the end of it.

>>58853351
I don't get it. My household pays something like 25% normal tax because my dad's legally blind. Tuition is free, I live with my parents like all the other college kids. The degrees are actually good, most seniors I've known went on to work in Europe or the geophysical black hole that is Houston.

>>58853409
22/M/Near Buenos Aires bb how bout u?

>>58853468
Shit, third quote meant for;
>>58853368

What Ryzen is closest to an i5-7600k and what price do you guys predict using this leaked pricing model? Wanted to build a new rig and I am tried of waiting.

>>58853502
One of the R3 models, and it's going to be stupid cheap. Cheaper than the 7600k for sure even for the top end model.

>>58853516
If its under 200 then I might be interested. I'm not that big of an OC'er though, I will just use the auto-OC settings on my MOBO.

>>58853528
There might even be a 4c/8t model for sub 200. Not known yet, but would be an even better choice.

>R* naming scheme with the same Intel numbers
>R5 and R3 have 4 to 6 cores
Nigga why.

On another note, I'm hoping I can finally upgrade this 3570k. More cores would be excellent.

>>58853567
R3 only has 4 cores afaik. R5 is the confusing one.

>>58853567
It's there to cater to intelbabbies and total retards who might get "confused" by numbers other than 3, 5, and 7.

>>58853594
R3 -> 4 core, no SMT
R5 -> 4 or 6 core, SMT enabled
R7 -> 8 core, SMT enabled
I don't see anything especially confusing there. Maybe R5 could have been split in two, but that's all.

>>58853468
>M
Shit, yeah good luck marying someone I guess.

>>58852146
The idea is that Zen out of the box has a significant stock clockspeed disadvantage compared to KBL, and Intel's ~10% Int and fully 256b( x 2) FPU along with superior threading performance make up a large lead over Zen at the same core counts.

None of this we can actually confirm until March.

>>58853734
Sounds like a load of horseshit when you consider the Blender bench.

>>58852098
I can see the 3.6/4Ghz SKU being rated at 125w TDP and bundled with the wraith cooler. People can speculate the real TDP from Canard leaks but until I see a 105-110w rating appear in consumer space I'm pretty sure they will market around standard figures for binning (or lack thereof) purposes.

File: 1486048612952.jpg (29KB, 499x500px) Image search: [Google]

29KB, 499x500px

>>58852098
>no info in 6 and 4 core
Fuck you AMD.

Any info or speculation on what do the Pro and X monikers mean? I think I read somewhere that X is the higher binned/clocked version at higher TDP for OC and Pro is the one with a wraith cooler instead of a stock one, but I'm not really sure where and when I got that.

>>58853760
Yeah well as much as we all wish, a hand tailored Blender build optimized by I assume the best of AMD's talent is not representative of average performance.

I almost guarantee you the Zen design will be 85% of Skylake's ST performance clock-for-clock

>>58853734
Who really gives a shit if Intel is 10% for +60% more

>>58853848
>implying zen won't be overpriced/undersupplied at launch
>implying intel won't reduce prices if zen is competitive

>>58853837
They have the same exact test you can download and run on your machine. Its legit.

>>58853837
Who cares, really. Intel isn't selling $500 8 core cpus

>>58853892
Straight up halving the price of the 6900K? Don't see them doing that.

>>58853892
>implying you aren't bad at shilling

>>58853892
Just like Skylake?
Intel never lowers prices, ever.

>AMD surely won't let us down this time, g-g-guise

>>58853892
small dies on a very mature node.
There's not going to be any sort of shortage.

>>58853927
0.01 rupees have been deposited to your account

>>58853958
Intel has has more utter failures tho. What you suggest is not rooted in history.

File: assblasted.jpg (332KB, 800x1150px) Image search: [Google]

332KB, 800x1150px

>>58853370
>posting faked garbage again
Huzzah for the era of Fake News!

>>58854012
What evidence do you have that this is fake?

>>58854044
What evidence do you have its real?

>>58854063
what evidence do you have it's evidence?

>>58854083
So it's fake?

>>58854100
You broke it, you faggot.

>>58854044
Personal experience
Nothing like this has been handed out to any of the usual circles.

>>58854063
Great argument.
It's perfect legalese, just like you'd expect, it has the correct layout and font and the usual asterisks found all over Intel's webpages usually leading to an "All trademarks belong to their respective owners" disclaimer.

>>58854116
What are "the usual circles"? Also the Athlon64 days (when Intel pulled their anti-AMD tactics last) are long gone, so you might not even be familiar with this anymore.

>>58854143
So it's fake ?

>>58854165
No, faggot.

>>58854178
Yes, it's a fake

>>58853169
Amdahls law is a meme. If you go multicore you have far bigger problems like lack of bandwidth. 25GB/s can only get you so far. Also if you use more threads than the amount of cores to handle multiple connections you're doing it wrong.

>>58854143
>What are "the usual circles"
Retail/OEM partners and outside marketing departments.

>>58854202
>>58854243
There's people on Reddit that took Intel Retail Edge training recently confirming this is real.
>inb4 lol plebbit

>>58854243
Let me assure you.
Large companies do run such "modules" for staff to brief themselves with comings and goings. There was likely a web-based slideshow that staff have to sit through first, followed by the brief we see, and then a quiz, which looks to be available at the bottom of the briefer.
We run shit like this where I work, though the subject is different, the idea isn't.

I've no doubt that something like what we are seeing exists in Intel's Intranet, but I don't think it would find it's way out at such a convenient time.

>>58854280
>We
Stop claiming you're an Intel employee or prove it.

>>58854012

It's called "alternative facts" now, get with the times.

>>58853121
no, it's 200mghz plus

>>58853123
lol get this the only provider that have "unlimited" internet cost 56$ and it only goes up to 4 Mb/s

File: fJeD43p[1].png (5KB, 787x53px) Image search: [Google]

5KB, 787x53px

>>58852098
>100 euro more expensive than i7 7700k
lol are you people fucking retarded?

>>58854542
Man you're shit at reading.

>>58854555
ok explain?

>>58854565
>completely ignoring it says "6800K"
>completely ignoring the fact it's 8c/16t

>>58854590
anon did you see it says it right under "INTEL COMPETITOR" if i7 4770k can compete with this why would I want to pay 100 euro more?

>>58852098
Praise AMD

>>58854606
The categorization is obviously based on price you retard.

>>58854643
but anon i7 7700k is priced 100 euro less

>>58854669
https://www.amazon.es/Intel-i7-7700K-4-2GHz-Smart-Cache/dp/B01MXSI216/ref=sr_1_3?ie=UTF8&qid=1486578955&sr=8-3&keywords=7700k
https://www.amazon.es/Intel-i7-6800K-Procesador-color-plata/dp/B01FJLA8NI/ref=sr_1_1?ie=UTF8&qid=1486578975&sr=8-1&keywords=6800k
Not in Spain it isn't.

>>58854606
stop believing a list made up by some chink that isn't even consistent with itself. AMD already showed a 3.4Ghz 8 core to be as fast a 6900k. A 7700k is 40% slower than that.

>>58852251
Taobao has been proven faked.

>>58852130
Why don't we design our processors. I can make the logo.

File: 489194691.jpg (380KB, 3989x2659px) Image search: [Google]

380KB, 3989x2659px

>>58854874
And I'll provide the pipelines
You think these are wide enough? Or do we need to go W I D E R?

>>58854916
50 stages minimum

File: empty-stage-22543321.jpg (34KB, 400x400px) Image search: [Google]

34KB, 400x400px

>>58854931
Okay, so fifty of these that can reach out in all directions
Are they too small? Shall we go for a much larger linear one instead?

File: cacheammunition.jpg (68KB, 480x360px) Image search: [Google]

68KB, 480x360px

I've brought some fast cache. We might need to bring an index for them... There's a lot to go through

>>58852146
It's a price bracket comparison. Not a performance comparison.

>>58853107
AMDalhl's law is pessimistic case, or worst estimate. Actual benefits are more linear because with increase core counts datasets tend to grow as well.

>>58852098
fucking leaks, release in under 3weeks but no official infos. AMD is so fucking bad in marketing, killed my hype completely. noshill here, just waiting for a cause to upgrade my 1055T

>>58855902
surprising people is good marketing. Not having weeks or moths between announcement and actual release is good marketing.
Why do you think Apple is so secretive about their products?

>>58853548
>There might even be a 4c/8t model for sub 200.
AMDelusion
So it looks like anything less than $500 from the Ryzen line will run at 3.2ghz or less and AMD fanboys will continue to insist that multicore performance is more important than single core. Just like they did for the entire lifetime of Bulldozer.

>>58852098
Sounds too good to be true. Also what's up with those X's? I thought all Ryzen CPU's can OC?

>>58852098
>i7-7700K competitor is $500 dollars
>more than 140% the price
Ha poorfags. This is why i stick with Intel

>MSRP for EU

That literally doesn't exist outside of articles. RX 480 8gb is as fast as 6gb 1060? Fuck it, let's price it the same.

>>58856451

Thats why its a bullshit chart. Ryzen is unlocked - its the motherboards that restrict it.

>>58852219
>>58852239

but with Quad-channal memory you don't have to pay out the wazoo for fast ram. 2800 Quad-channal is like 4000 dual last I checked

>>58852098
>highly competitive
In what world is €469.99 competitive? The i7-7700k is just €360.

>>58856674
BUT MUUUH CORES REEEEEEEEEE! WHAT THE FUCK IS WRONG WITH YOU SHILL? DON'T YOU WANT TO KEEP BEATING INTEL IN CINEBENCH R15 SCORE WHILE PLAYING HONEY SELECT AND RICING YOUR COMPUTER?!

>>58853107
>cylindrical heatspreaders
Intel paving the way.

>>58856674

How much is the 6900K again?

>>58856674
The top end Ryzen is competing against Haswell-E and Broadwell-E.
A lower clocked 8 core chip is not competing against intel's mainstream quad cores.

>>58856674
In the world where it competes with 6900k and 6850k instead of 7700k.

File: 1fx2km.jpg (71KB, 888x499px) Image search: [Google]

71KB, 888x499px

>>58856674

>>58856671
quad channels gives no practical performance improvements. It only makes the motherboard more expensive and you have to buy 4 RAM sticks instead of 2.

>>58856671
lol no

>>58856913

Quad channel does matter but for the sort of workloads it does you're likely to be buying a server focused chip (and ecosystem) anyway.

>>58857027
Ryzen is for desktops. Naples is for servers.

>>58857109

Right and outside of niche cases citing dual channel on ryzen as a negative is a non issue.

>>58857146
And they have Snowy Owl for prosumer market anyway, which has quad-channel memory.

>>58856913
Its double the bandwidth. So if you need that, it matters. Also, those APU are well known to be very bandwidth starved. I always wondered why they didnt move up from 2.

>>58857191
Because if you're using an APU you don't need quad channel. Quad channel motherboards and APUs would cost more, by which point it'd be more cost effective to get a separate GPU.

Same with getting super-fast DDR3/4 RAM too really. Beyond a certain point it becomes not cost effective.

>>58857191

In b4 HBM on die connect via AMD's new memory controller thingy acting as a lighting fast L4.

>>58856417
AMD is binning their processors much the same way Intel does if this leak is correct.

What you're paying for is the rated clockspeed at the rated TDP. Almost all 8 core SKUs will be 95w, binned to reach individual perf targets at the same power draw.

>>58857320
All signs point to things looking incredibly good for the 14nm LPP process, yields are allegedly really high too, like 250 good candidates out of 280 per wafer.
CanardPC overclocked their A0 test chip to 3.6ghz base/4ghz turbo and it was pulling 105w under load. Pretty mild power increase for an early chip with a 3.15ghz factory base clock.
Power consumption looks like it rises very gradually when overclocking, and higher binned chips later in the year are likely to reach even higher clocks inside of their 95w TDP.

We very well could see 3.8-4ghz base clock chips on the process that fanboys claimed would never clock past 2.8ghz.

>>58857319
That would be very good, instantly killing that Iris Pro shit. AMD now just have to implement that.

>>58852330
>$4-600 is the average income
>pull it.

>>58857692
i just herd a rumor that Intel is going to use AMDs APU tech in the icore cpus

>>58852330
>>58857716
They'll have poorfag chips too

These are just top of the line for consumers. No idea what opteron stuff will cost

>>58857717
Actual GPUs or just patents aka the way Intel 'uses' NVIDIA tech now?

File: og9n76.jpg (37KB, 512x372px) Image search: [Google]

37KB, 512x372px

>>58852401
>IBM

>>58857770
https://www.overclock3d.net/news/cpu_mainboard/intel_cpus_using_amd_gpu_hardware_could_launch_in_2017/1

>>58857717
Not using AMD APUs.
But licensing AMD Patents to that Intel can product their own iGPUs based off of it.

If the deal is the same as Intel held with nVidia, then this is a 5 year deal, with fees equating to $1.5b.

It's also quite likely that Intel will not switch back to licensing nVidia IP after the AMD deal expires, relationships soured between Intel and nVidia after Intel sued nVidia some years ago.

>>58856451
Might be for wraith cooler/tdp

>>58857818
Wraith is standard on all AMD chips now.
Even newer batches of FX processors are shipping with wraith.

>>58856451
X just means highest clocked variant I think

>>58857941
It means nothing, since everything on that chart was made up.

>>58853033
>unzips trips

>>58857952
No it wasn't

screencap this post

>>58852300
fucking greece i swear

>>58852489
ivy user. not interested yet.
>>58852592
>3570k
>no oc
/g/
>>58852594
Im hoping for 460 performance and dual graphics that work. a8-7600x250 worked great for my dad and his old sim games.

WHY THE FUCK DOES IT MATTER? IF THEY DIDN'T GOT SOME INTEL LEVEL IPC THEN WHO CARES IF IT HAS 8C/16T. DO YOU AUTISTS REALLY EXPECT AMD TO GO HEAD TO HEAD WITH INTEL. AMD HAS ALWAYS BEEN ONLY TRYING TO CATCH UP WITH INTEL AND WILL STAY LIKE THAT. JUST STOP WITH THE BULLSHIT HYPING AND WAIT FOR THE BENCHMARKS FOR FUCKS SAKE FOR ONCE

>>58858691
Kill yourself intelshill.

>>58858722
I actually own a RX 480 and a 2700k that I'd GLADLY UPGRADE if we ACTUALLY FUCKING SEE THE BENCHMARKS. I just can't hold myself when people hype shit up instead of thinking logically for a second and taking a look back at the past. Please tell me, which time did AMD actually got ahead of Intel since the Athlon 64 release? And why exactly Ryzen should be any different? Oh boy people are going to be dissappointed.

>>58858801
We've already seen benches, CanardPC published a small review of an early engineering sample. Exactly as they did with an Athlon64 many years ago. They showed performance more or less on par with Broadwell-E at a 95w TDP.

>>58858691
>IF THEY DIDN'T GOT SOME INTEL LEVEL IPC THEN WHO CARES IF IT HAS 8C/16T

They did get Intel level IPC.

But, they clock 10% lower.

>>58858863
Their 8c parts clock higher. There's nothing showing a 4c part won't be able to manually clock just as high for cheaper.

File: ap_israel_jewish_intel_sabbath_210_14Nov09.jpg (35KB, 201x210px) Image search: [Google]

35KB, 201x210px

>>58852103
shoo shoo

File: apu.png (31KB, 200x201px) Image search: [Google]

31KB, 200x201px

>tfw still using phenom II 3.0ghz and t7200 on notebook

>>58853048
not him but in the same boat as him. Probably eastern europe and/or balkans.

https://videocardz.com/65747/amd-ryzen-7-1800x-1700x-and-1700-listed-online

X SKUs - WOF or without fan.

File: 1459370252769.jpg (45KB, 443x409px) Image search: [Google]

45KB, 443x409px

>>58853798
this. Couldn't care less about their 8c16t cpu. There are like 5 applications that will utilize all of those threads.

Show me a 6c12t cpu for $200 and we have a deal.

>>58859193
It it looks like they're offering a 65w 8c/16t chip for about $320, so I'd imagine they'll have 6c chips somewhere around $200 to $250.

>>58858691
>AMD HAS ALWAYS BEEN ONLY TRYING TO CATCH UP WITH INTEL
Yeah, I always hated how the Athlon 64 couldn't even compete with Pentuim 4. Why the fuck would you release a CPU that doesn't keep your room warm and cook your morning eggs. WTF were amdrones thinking. It took them 10 fucking years to catch up with Intel. Pathetic company ayymd cuck niggers

:^)

motherfucking pajeets and chinks. i waited for years for this shit and you are going to fuck up the price this much? go fuck yourselves.

because i waited for so long, i'll wait to see the benchmarks but my mind is already inclining towards intel. AND I HATE FUCKING JEWS.

File: 1462422715120.png (127KB, 387x357px) Image search: [Google]

127KB, 387x357px

>>58859226
alright, follow up question: if all chips are unlocked why do they charge more for a cpu which has the same amount of cores as the more expensive one?

The cheaper one doesn't oc as well I presume?

>>58859268
>if all chips are unlocked
But they're not fully unlocked. You can't change the core multipliers on the non X chips, only the BCLK.

The 4 core and 6 core ryzen are the going to be the ones that will make ryzen an success or not.
>7770k performance at $150

>>58859289
nah, they said literally every cpu is overclockable but only like 3 motherboard chipsets support overclocking.

Seems like X chips do not come with the wraith cooler is all.

>>58859268
Because clock binning is a thing.
Its not necessarily the case that all chips don't clock as high, but some of them need less voltage to hit the same clocks.

>>58859317
>they said literally every cpu is overclockable
Yes, and that's via the BCLK as per every AMD CPU since Athlon 64. The Black, K, and now X SKUs allows you to adjust the core multipliers on top of that.

File: 1486414771332.jpg (43KB, 918x749px) Image search: [Google]

43KB, 918x749px

>>58859161
>low end 8c/16t for only $300

>>58859355
Xeon E5 2608L v4 and 2609 v4
Goggle it nigger

>>58853051
>Maximum 20% clock speed difference somehow makes up for having half as many cores

Either Ryzen is witchcraft or that chart is retarded.

File: fucking retard.jpg (20KB, 480x360px) Image search: [Google]

20KB, 480x360px

>>58859390
>1.6 ghz base clock
>motherboard is expensive

HOLY -FUCK- and 8core 16thread cpu for fucking ~400 bucks. Intel on suicide watch as fuck

>>58859390
>1.7 GHz

Wow, it's fucking nothing. Not to mention that you need an expensive motherboard to even use one.

>>58859390
Not even remotely worth mentioning.
8c/16t Ryzen with a 65w TDP makes the Xeon look absolutely shameful.

File: 1447792400301.jpg (232KB, 643x697px) Image search: [Google]

232KB, 643x697px

>>58859390

>>58852256
RISC-V. It already here in limited use, and growing.

File: pys3ubnqvpey.png (44KB, 768x319px) Image search: [Google]

44KB, 768x319px

>>58859161
Add 20% for VAT and that's still an absolutely cracking deal.

For comparison, the 4c/8t 7700k is £330 after VAT.

>>58859496
Its really going to toast their almonds when people find out how good they overclock.

File: 1467055276405.jpg (26KB, 251x242px) Image search: [Google]

26KB, 251x242px

http://www.pcworld.com/article/3167279/computers/amd-sorry-there-will-be-no-official-ryzen-drivers-for-windows-7.html

>>58860029
>tested and validated to be fully working under Windows 7
>but won't get any new drivers

I can't remember ever downloading new drivers for a CPU, or a new BIOS for that matter, at least in the past 5 years. Drivers for an IGP? Sure, but not explicitly for the CPU. Seems like thats something that wouldn't only come about as an emergency fix.

File: untitled-2.jpg (34KB, 700x382px) Image search: [Google]

34KB, 700x382px

>>58858837
and it had worse performance than i5 6600k and non k which is complete joke that 4core cpu beat brand new 8c/16ht zen cpu. Zen will be another failure just like phenom was aka it was much cheaper than intel cpus but much weaker in games so nobody except professionals bought them due core count.

>>58860077
>what are clockspeeds
Thats only 10% behind the i7 6900K, while being clocked at 3.15ghz, significantly slower than final silicon.
Also only one game tested there really uses more than 4 threads. The reviewers themselves explicitly praise Zen for having near Broadwell IPC, and consumer chips offer more performance than the chip they tested.

You're grasping at straws, shill.

>>58860077
Canard did make note that the A0 ES and Validation mobo were throwing errors out the ass during testing.
No harm in speculating. We'll all know for sure in the next few weeks.

All these AMD fags hyped over a chip that isn't out, has no real official details or pricing out, and hasn't been independently tested by anyone.

Meanwhile the 6700K I bought on sale more than 4 months ago is running at a super comfortable 4.6 and has never once gotten hotter than like 40c.

Wake me up when AMD has literally anything on the market at any price that can compete.

File: 1486028322785.png (4MB, 1543x1615px) Image search: [Google]

4MB, 1543x1615px

>>58860108
>Canard did make note that the A0 ES and Validation mobo were throwing errors out the ass during testing.

nice damage control you got there, it would only affect stability not performance

Not interested unless they pull off a 4c/4t for $150 or less. This $300+ muh 8+ cores shit is useless as fuck since it's still expensive with little benefit

Until then G4560 is the best bang for buck

File: a picture of me.jpg (323KB, 777x728px) Image search: [Google]

323KB, 777x728px

>>58860006
I want these prices, but I'm not gonna believe this shit until AMD announces it or I see it for myself.

>>58860151
have fun playing any game in 2017 with 60fps on outdated 2+2 processor with perf barely 5% better than ancient i5 2500k

>>58860077
That's a 3.1GHz engineering sample vs a 4Ghz turbo cpu.

>>58860187
it's $60 though. it's sad that there's literally nothing between it and the $239 7600k that's even worthwhile.
all AMD needs to due is release a unlocked 4c/4t i5 competitor at $150 and it will fucking own the market

>>58860307
If they're selling a 65w 8c/16t part for just over $300, then 6c/12t parts will likely fill the range from $250 to $300. They still have 4c/8t parts, and 4c/4t parts. They'll sell every die they can harvest. Dirt cheap SKUs are coming. Entry level AM4 boards aren't just for APUs.

I should had bought AMD stock when it was 4-5 dollars. Judging by how it goes, it will grow up to 30-40 dollars if Ryzen turns out to be what we are expecting it to be.

The stock is going higher and higher by the day. It's unbelievable. Keller THE MADMAN. HE ACTUALLY DID IT.

>>58860386
>tfw got in at 4.7
Its good

>>58860141
If it has to do operations twice to get the right result then it would affect the performance

File: this is the finger.jpg (202KB, 945x573px) Image search: [Google]

202KB, 945x573px

>>58852130
ARM is same or worse than AMD.
As a bonus, it's all non-standard so to boot any SoC, you need a vendor-specific fork of (old) Linux kernel.

Unmaintained un-upstreamed patchsets ahoy! Frankly, fuck ARM, hooray for the state of the x86 union.

>>58852098
>ryzen price leak
Probably a speculation, not really a leak, so don't hold your breath yet.

>>58852454
>EUROPOOR!
Perhaps only until you need to go to doctor/hospital, then the joke is on the USA cucks. They pay through the nose but think they won somehow kek.

File: 1478804345204.gif (338KB, 538x572px) Image search: [Google]

338KB, 538x572px

>>58852103

>paying hundreds more for the same performance

"bbbbut wait for the benchmarks"

>>58853798
Those will probably take more time to reach market, from what the retail guys say (they only see eightcores coming in the first wave).

>>58854063
It looks like a perfectly make fake, with nothing looking fishy. IT would be great work if not real.

The quiz bit matches what I have seen from Intel before BTW.

>>58860446
hey man those made up numbers are presented in a table, that means it's real

File: wew.png (340KB, 1920x1080px) Image search: [Google]

340KB, 1920x1080px

>>58860431
Odroid shit is mainlined.

>>58852103
if you're gonna shill, don't make it so obvious retard

>>58860596
One out of hundred?
And the booting system is still soc-specific and non-standard.

>>58860666
Its standard it uses uboot.

>>58860748
That doesn't really mean shit, when the SoC & board combination still needs specific support in kernel just to boot. Compare x86 PC motherbords.

>>58861011
x86 motherboards need specific support in kernel just to boot.
Lucky for them, it's all standardized!

>>58860029
Hoping someone "leaks" them.

>>58861043
No hes right, he clearly knows what hes talking about and isn't a dumbfuck with a raspberry pi pretending he can computer.
>>58861011
I'm so sorry you caught me, I was blowing smoke and you are actually correct.

>>58852248
includes a 20-40% vat tax

>>58861394
21 % usually.

>>58852214
the 1700 is also an 8 core if leaks are real

>>58852560
there will be a token 4 core sku for either extremely fucked 8 cores, or just enough to justify a sku launch, and it will remain that way till the apu and fucked apus come out.

File: 25fbe65be23955d7fb5f057d81a5ca2a.jpg (708KB, 1080x1920px) Image search: [Google]

708KB, 1080x1920px

>>58853274
>That being said, as soon as I get my geophysics degree I'm getting the hell out of this and doing a Master's in the good ol' US or somewhere in Europe

You are the type of person I want in the US. Please come here, we need as many people as possible who didn't become lefty cucks in college.

Just do me a favor and really TRY to get rid of any accents you may have. Nothing makes me feel more uncomfortable then asking someone to repeat themselves because I can't understand their accent.

>>58861414
>23 % usually.

>>58861414
looking at 500~$ for top end sku american then, and i7 price matching for low end 8 core 16 thread...

all thats left for me to decided between the two is know what overclocking looks like and if there are any worth while improvements.

>>58859193
true, but headroom isn't something to just dismiss, expeically considering I got a quad core when it was fucking stupid to get a quad core because next to nothing used it, and today on the same quad core, im just starting to feel a need to upgrade.

that 8 core will last you a fucking long time, same with the 6 core, but do you want to hamstring yourself on something that could last you 7-8 years just because its not fully utilized now?

>>58859268
binning. I am sure intel bins their good cpus to be a k sku and their worse cpus to just fit power requirements of lower ones, likely the same with amd, binning to fit power requirements, higher clocking chips at the same tdp means more head room.

>>58856123
>>58855902
amd is drip feeding shit, this is another drip feed, they gave us a speculation point for the price with the new horizon event

they told us the price at ces without telling us the price

they 'leak' the price like this, either on purpose, or they knew damn well it would leak so they planned it.

each one of these was worth a week or so of news articles and speculation, if not longer, and this new leak is yet another news cycle where people pick it up and run with it, without an actual announcement. till the actual announcement of price amd will continue getting more news based on leaks and rumors.

the marketing is fucking genius.

>>58857952
>>58857941
>>58856451
the cpus have no gpu, but amd is makeing a 'run the fucking os' tier gpu, so one of the skus, either x or pro could be bundled with a shit gpu to run the os.

I agree it's fairly stupid and horrible for sku naming, but would you really put it passed amd to fuck that up?

>>58856758
>>58856674
do we just ignore the one below it... you know the 8 core 16 thread that is 389.95 and when you remove the vat comes in at around 350$

we just ignoring that? yes? ok...

>>58852146
price competitive not performance.

the chart is mostly just 'here is the intel sku and for a similar price here is the amd' and when price just is not equal at all in the case of the 6900, they just put it there.

>>58852496
so we know broadwell and ryzen are about equal, so clock rates don't matter, amd could be 1ghz and equal to a 4ghz broadwell?

no, because clock rate fucking matters, if you could push a bulldozer to around 8ghz, it would beat out an i7 in single core

when we know what the chips preform like we can make these assumptions.

clock rate matters faggot get over it.

>>58856451
x= higher bin, no cooler.

>>58852298
The leaks are likely amd just putting some info here and there to get more news cycles, shit is genius.

File: 1482649549863.png (427KB, 599x357px) Image search: [Google]

427KB, 599x357px

Damn that R7 1700 looks attractive as fuck to me. Cheapest octocore with cooler that stands in for an aftermarket one, and it's 65W even.

Crap, I wish I didn't need to save up money.

>>58853104
the different socket makes me laugh harder than I should, every fucking time someone sneezes in a boardroom meeting they design a new socket.

>>58862212
It's not a completely asspull socket, it is the platform for next gen HEDT (Skylake-X).

What is laughable is that with those Kaby Lake-X chips, when you insert them into a LGA 2066 board, half of memory slots and probably also half of PCI Express slots will stop working.

If you'll want PCIE ×16 for your graphics cards, you probably won't even be able to use M.2 slots, unless they are routed from the southbridge.

I think Intel droped the ball there a bit.

Daily reminder that any leak that has less than 3.4Ghz on any sku is automatically incorrect.

>>58862514
https://www.youtube.com/watch?v=4DEfj2MRLtA&t=26m47s

source

>>58862647
>Ryzen processors at launch will have 3.4Ghz or higher

https://www.youtube.com/watch?v=4DEfj2MRLtA&t=27m29s

It's like you guys are fucking retarded, literally anything that says they'll have a CPU with a lower clock speed is automatically wrong

>>58862668
>>58862647
>>58862514
Either all CPUs have a base clock of at least 3.4ghz
or
Some SKUs have a baseclock of 3.4ghz or higher, and those will be available on launch, while lower clocked SKUs come later.

"at launch" is a qualifier in Su's statement.

>>58862694
These are all launch leaks that show CPUs with a clock speed lower than 3.4ghz

>>58862722
A full lineup of SKUs does not mean all of them listed will be available at launch.

File: 1485842230192.jpg (70KB, 720x960px) Image search: [Google]

70KB, 720x960px

>>58862746
>being this delusional

>>58861493
But anon, I grew up in the US. Mummy and daddy just never got us citizenship while we were there, so when we left we had to give up my right to residency.

I still speak more fluently on English than I do Spanish and I've got absolutely zero accent.

>>58862761
The A10 7800 was a Kaveri SKU detailed before release, and it wasn't available for half a year after the A10 7850k came out.

>>58853734
>compared to KBL
Thank you based gaymer for pointing that out.
Too bad ryzen will compete with workstation-oriented broadwells, not your MUH GAMEZZZ kabylakes.

>>58862779
This situation is more similar to the release of the athlon 64 than some extra no one care APU

>>58862884

>>58862884

>>58862884

>>58862774
You are top tier then anon