This is your 6 core Intel Coffee Lake scheduled coming out.

File: 1478301501393.png (145KB, 800x1200px) Image search: [Google]

145KB, 800x1200px

>>61072350

>>61072350
bump

>>61072350
Where'd you get this?

>>61072862
What is URL for $5, Tim?

>>61072350
Wow Intel is gonna sell 1600 for $330?

>>61072930
Don't forget to delid and put a $200 watercooler goy

>>61072972
Thank you Mr. Merchant, will do.

>>61072350
Is it just me or has Intel lost their minds in the last few months?

>>61073018
Yes, they went completely nuts. Why? Who the fuck knows.

>>61073018
Releasing a hexa core on LGA 1151 is a decent move in and of itself. People with Z270 boards will in theory be able to get a decent upgrade without needing to buy a new motherboard. No way they'll make it compatible with Z170 as well though, just like Broadwell only worked with Z97 and not Z87, despite being the same socket.

>>61073150
I don't think it's going to be compatible with either, to be honest.
>>61072930
>$330?
More like $350-$360.

>>61072350
> Decide to build new computer
> AMD was still shit at this point
> End up getting a 6c/12t i7-5930K CPU
> Worry that I fucked up and that next gen processors will blow mine out of the water
> Lol have an i7 with 4 cores and less power use
> Lol have an i7 with 4 cores and less power use
> Lol have an i7 with 4 cores and less power use
> Oh fuck AMD! Here have a $2000 i9 with a million cores that burns your house down
> Coffee Lake will have 6c/12t and more or less just less power use than my current CPU

Welp.

I guess I should say "Thank you, Intel" for making sure I feel confident in my purchase and feel no immediate need to upgrade even years later.

>>61073450
And everyone says moar cores isn't futureproofed.

>>61073450
>complains about less power use
>complains about more power use
stick to your ps4 kid

>3.2 GHZ
Seems kinda low considering CL is just another 14nm refresh.

>>61073604
Gotta keep that 91w TDP rating. Core is no Zen, it's ages behind in efficiency.

>>61073612
Seems like Intel is going to keep the full retard TDPs for their HEDT platforms.

>>61073150
>a hexa core on LGA 1151
that's intel we are talking about.

File: 1364081368050.png (23KB, 558x641px) Image search: [Google]

23KB, 558x641px

>>61072350
(((14 nm++))) is a scam. 14nm is 14nm, performance per watt hasn't changed since fucking broadwell.
Nothing is going to change until cannonlake, and even than performance improvements will be minimal.

>>61073777
They have no choice really. They're barely clinging onto their lead over mainstream Ryzen, despite needing so much power that the 7900X throttles at stock on a 240mm CLC. Skylake-X is already a gigantic embarrassment, but at least they can say that they still have the absolute best performance.

>>61073790
It literally says this engineering sample was running on LGA 1151 in the image in the OP.

>>61073150
intel releasing a new aka same old cpu on the same socket?

ARE YOU MAD?

Those clock speeds...

If Intel knew they were against a wall, why the fuck didn't they start preparing for a new architecture?

Just more reason to shit on that crooked ass company.

File: sock it to me baby.jpg (118KB, 10000x758px) Image search: [Google]

118KB, 10000x758px

>>61073818

>>61072350

no new socket? WTF

>>61073837
v3 omitted.
Not backward compatible.

>>61073837
yeah wait till 3 months before the release i bet my ass virginity that they will create a new socket even with 1 pin more

or going to lock the previous ones to not being able to use it unless you buy a pcie card to unlock

>>61073823
They bet on Zen being a failure. Zen turned out to be extraordinarily good.

>>61073844
Unlikely.

>>61073859
They don't need to do anything except tell motherboard manufacturers they can't add compatibility to Z170/Z270 boards if they want to go full kike, just like mainstream Broadwell and Z87. No need for a new socket.

>>61073860
amd quite literally trolled and lied a lot about them..

remember the engineer q&a when a intel eng asked of the ryzen will have the uncore speed tied to ram and paul said NO?

or how good the branch prediction will actually be?

they trolled the entire industry thats why they send samples and eng samples so late to the mb manufacturers... they didnt want to intel to know anything

File: 1235.jpg (61KB, 604x499px) Image search: [Google]

61KB, 604x499px

Same socket.

>>61073843
Intel changes sockets for arch design changes, Cannonlake is a die shrink. I'm not sure if Z170 or Z270 boards will be able to use Coffee Lake though, they weren't designed for 6 core processors.

>>61073972
>what is a microcode update

Some shitty cheap Z170/Z270 boards' VRMs might fry if you overclocked it, but the vast majority could handle the increased power draw with no issues.

There's absolutely no reason for Coffee Lake to exist or be in an LGA 1151 package if it can't be dropped into existing motherboards. It'll be Intel going retard on a scale even bigger than Skylake-X/Kaby Lake-X.

>>61073972
>what is 7740x

How are there so many people on /g/ who don't understand how Intel's socket changes work yet? It's been the same thing basically since the Core 2 series.

Arch improvement -> socket change
Die shrink -> same socket, although sometimes the the new CPU's won't work on older boards.

So for desktop processors, you have:
1156
>Nehelem (45nm)
>Westmere (32nm)
1155
>Sandy Bridge (32nm)
>Ivy Bridge (22nm)
1150
>Haswell (22nm)
>Broadwell (14nm)
1151
>Skylake (14nm)
>Kaby Lake (14nm)
>Cannonlake (10nm)

KL kind of confuses things in that it's just a refresh, also Coffee Lake is another refresh due to low 10nm yields. (Cannonlake is mobile/low power only) But it's still LGA 1151.

>>61074051
It was a move from the consumer line to the HEDT line, they run on a different schedule. Haswell-E and Broadwell-E were LGA 2011-v3, while Skylake-X and Kaby Lake-X are LGA 2066. Coffee Lake-X will also be LGA 2066.

>>61074073
>three sockets used the same memory type
A socket change a year keeps goyim in fear.

>>61074097
>Intel changes sockets for arch design changes
>I-I-It was a move from the consumer line to the HEDT line
Well kiked

>>61074142
Your argument doesn't make any sense. Skylake-X and Kaby Lake-X are arch changes from Broadwell-E, that's why they use a new socket. Putting 4 core CPU's on HEDT was a silly decision but it still follows the same trend.

>>61073807
yes. The only difference is better transistors, because of yields, that can clock higher.

>>61073812
>this engineering sample was running on LGA 1151
yeah, skylakeX also can run on 2011

>>61074206
You'll have no problem posting evidence of a Skylake-X engineering sample running on LGA 2011-v3 then.

It's funny looking back at Nehelem and Westmere, Intel's product segmentation was a complete cluster fuck. There's a ton of overlap between LGA 1156 and 1366, and really no consistency on what name means what product. Sandy bridge was much less confusing. It's sort of what Intel's going through now where Kaby Lake-X blurs the line between Z270 and x299.

>>61073150
It's not backwards compatible with LGA1151. This is for the new LGA1151v2. The pin-outs on the socket are completely different.

>>61074338
Source: Your ass

There is zero confirmation one way or the other at this point in time.

>>61074355
http://drmola.com/bbs_free/168958

>>61074338
AFAIK it will be the same 1151 socket.
I did read rumors however, that CFL would be firmware limited to the upcoming chipsets.

>>61074380
Oh, I overestimated you. It was actually an even less credible source than your ass.

>>61074425
Why don't you translate what they had to say about X299 before going full shill, you cowshit-eating poojeet retard?
These are memos from a closed door meeting with Intel's product development team.
Coffee Lake-S will not be backwards compatible with 200-series or 100-series chipsets for a reason.

>>61072350

are you sure that's coffee lake, sauce

>>61074446
Sure they are. Post a source with confirmation from Intel that Coffee Lake won't be compatible with Z270 or get out, and take your Korean blog posts with you.

Regardless of backwards compat, I can't see a lot of Intel users switching over, simply because the i7-7700K will still be better for muh 1080p 144hz gaymes, and that's most of the reason people are buying them still.

>>61074447
It says it's a 6-core for LGA 1151, what else would it be?

>>61074481
>waaah waaah it can't be true! even though everything the memos said about X299 turned out to be true!

>>61074518
Are you going to post a credible source, or are you going to continue shitposting?

File: dumb frog.jpg (12KB, 238x192px) Image search: [Google]

12KB, 238x192px

>>61074518
>Currytech posted something that actually turned out to be right once, so I know all of their articles are true. :)

>>61074504
It depends what the price is. The 7800X is $389 at 3.5 GHz, so I could see the 8800 being about $350 at 3.2 GHz and the 8800K being $370 at 3.3 GHz.

>>61074560
There won't be any credible information until Intel or their vendors release details on the 300 chipset.

Oh and Threadripper's 16 and 12 cores will use more power than Skylake-X. AMD's back into housefire territory once more
>>61074583
It's not currytech, that's the difference. Look at who runs the blog.

>>61074589
That is a sidegrade at best.

>>61074606
Yeah, because a 1700X doesn't perform well for the power it uses, two will definitely change that.

>>61074606
>Threadripper's 16 and 12 cores will use more power than Skylake-X
Even if they slap two 1700's together, that's not any higher than 130 watts. And I'm sure they can do better than that thanks to aggressive binning and higher stepping (B2).

Skylake-X 16-18 core will be a literal housefire. If you thought the 10 core was bad, wait until you see that shit. They might have to put out another motherboard version just to handle it.

im sick of these constant chip changes intel makes like they think theyre making cars or some shit

>>61072350
>coffin lake

>>61072350
I've been saying for months Coffee Lake S wasn't going to be anything more than an i7-6800K with jizz TIM under the lid. 1600X performance for $350, can't OC it without DELID and watercool. Sounds like a steal.

>>61075317
>i7-6800K with jizz TIM under the lid
that's i9-7800k, anon

>>61074970
TDP has nothing to do with power consumption. Please stop posting you summer cancer

>>61076319
>TDP has nothing to do with power consumption.

That is incorrect.

TDP is measured in units of power. The rate of heat flow (which TDP specifies) is measured in watts. Power consumption (i.e. input power, in watts) has a functional relation to heat rate (i.e. output power, in watts). Thus, to say that TDP has "nothing to do with power consumption" is to deny that such a functional relation exists.

Quoting https://en.wikipedia.org/wiki/Heat :

"The standard unit for the rate of heat transferred is the watt (W), defined as one joule per second."

Quoting: https://en.wikipedia.org/wiki/Thermal_design_power

"The thermal design power (TDP) ... is the maximum amount of heat generated by a computer chip ..."

where "heat" in this case is referring to rate of heat flow (measured in watts), not heat energy (measured in joules). This interpretation of "heat" can be easily inferred because TDP is measured in watts.

>>61072350
6 cores means 6x scalability. They can make dual core that performs the same under max load but at min load ay least one core would still be active.
On laptops having 1/6 of cpu active instead if 1/2 is better (saves power and shit)

>>61072350
3.16GHz
That's pretty much useless to me.
Thanks Intel.

I think when I upgrade my PC next year, I'll opt for AMD.
First Time in my life but wanna try it.

File: IMG_31337.png (38KB, 935x452px) Image search: [Google]

38KB, 935x452px

oh noes, flawed cpus... again!

>https://lists.debian.org/debian-devel/2017/06/msg00308.html

>https://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/7th-gen-core-family-spec-update.pdf

>known issue for a month
>no fix
nice one (((intel)))

>>61073450
Haswell-E was, and still is, Great.

>>61077659
Skylake and Kabylake have been out for this long and it took this long for anyone to notice it?

>>61076549
TDP =/= power consumption. This is the truth.

>>61072350
~a socket change a year keeps the goyim in fear~

>>61073088
>Why? Who the fuck knows.
Ryzen.

>>61079046
He's just saying that they are linked even its its not direct. The tdp is an engineered number for pens (and users) to follow when making cooling for these systems. A chip is generally not designed to exceed their tdp in power use, but top spec chops seem to throw that idea away, it wasn't so common until turbo clocks started to be abused.

>>61074104
That doesn't rhyme.

>>61082357
Bright colors on the box
Makes the goyim spend lots

File: aHR0cDovL21lZGlhLmJlc3RvZm1pY3JvLmNvbS9JL1EvNjg0OTYyL29yaWdpbmFsLzA0LVBvd2VyLUNvbnN1bXB0aW9uLVRvcnR1cmUucG5n.png (26KB, 711x533px) Image search: [Google]

26KB, 711x533px

>>61076549
Yes what you say is correct, TDP however, TDP value of CPUs stated does not reflect the amount of electrical power a CPU draws.
For example take the i9 7900X which has a TDP of 140W, yet various reviews have shown the cpu alone consumes over 200W of power (stock settings). http://www.tomshardware.com/reviews/intel-core-i9-7900x-skylake-x,5092-10.html
R7 1800X has a 95W TDP, yet it draws up to 110 - 140W of electrical power in workloads
http://www.tomshardware.com/reviews/amd-ryzen-7-1800x-cpu,4951-11.html
Thus you can not correlate TDP with the actual power consumption.
Besides, the TDP numbers are "ratings" used by manufacturers so that an adequate colling solution is used.

>>61082357
>year
>fear
>doesn't rhyme
wut?

>>61084600
learn to stock vs overclock

stock settings will throttle the chip until it reaches its designated TDP

>>61084626
>These numbers are generated using stock motherboard settings
Learn to read

File: 12124423412344.png (240KB, 1394x1102px) Image search: [Google]

240KB, 1394x1102px

>>61076749
1600X has 28% higher clock speed and only 10% faster in single thread

>>61084742
What is turbo, timmy?

>>61084742
Probably way less power used too.

>>61073150
Implying intel won t come up with an even more expensive chipset than z270 to overclock the hexacores

>>61087907
Screencap this: The rub is there is going to be no K sku on 1151.

>>61088464
My jaw would drop pretty hard if they removed OCing from mainstream CPUs entirely. Maybe the K series will remain, but still have a locked multiplier.