Intel's 8th Generation Core Family - Coffee Lake (LGA 1151, 6C/12T)

Where do you expect Core i7-8700K's Turbo to land?

  • 3.8/3.9 GHz

    Votes: 0 0.0%

  • 4.0/4.1 GHz

    Votes: 3 23.1%

  • 4.2/4.3 GHz

    Votes: 6 46.2%

  • 4.4/4.5 GHz

    Votes: 3 23.1%

  • 4.6/4.7 GHz

    Votes: 1 7.7%
  • Total voters
    13
  • Poll closed .
TaintedSquirrel said:
DICE helped develop Mantle, they were also the first to tease the 290X and Fury X on their Twitter pre-release.
AMD's involvement in the Battlefield franchise is equivalent to Nvidia's w/ GameWorks.

Anyone who attempts to downplay it is in denial aka damage control. Same people who denied AMD's involvement with Ashes of Singularity.
Click to expand...

Mantle was independent but BF supported it. AMD CPU's and RTG GPU's get no preferential treatment in BF. Dice just happened to be more forward looking than most developing houses and favour the future potential of massively threaded games to bring more out their engine which is possibly the most advanced game engine if pure horsepower needed to max out is taken into consideration. In fact Dice are as open to Intel and Nvidia as they are to AMD support, I have played over 12000 hours and nearly 1 million kills form 1942 to BF4 and have not experienced any issues with Intel/Nvidia hardware on Battlefield.
 
akiba said:
Intel BX80684I58600K Core i5-8600K 3.7GHz 6-Core

65W TDP, Turbo 4.2GHz, 9MB L3 Cache, GT2 Graphics, Socket-1151-V2
$314.00

Intel BX80684I78700 Core i7-8700 3.7GHz 6-Core

65W TDP, Turbo 4.2GHz, 12MB L3 Cache, GT2 Graphics, Socket-1151-V2
$368.00

Intel BX80684I78700K Core i7-8700K 3.8GHz 6-Core

95W TDP, Turbo 4.3GHz, 12MB L3 Cache, GT2 Graphics, Socket-1151-V2
$428.00

source -> http://gamepc.com/shop/productscategory?list=CPU
Click to expand...
Anyone else notice the 8600k dropped to 65w tdp
 
tnka7aj7gviz.jpg
 
Crossposting from the Destiny 2 thread.
Steve @ GN claims Ryzen SMT is non-functional in this game, and disabling it might actually improve performance...

Are AMD owners actually just hardware beta testers?
Does "Fine Wine" simply mean fixing broken features down the road?
More news at 11.

 
oleNBR said:
http://wccftech.com/intel-xeon-w-workstation-cpu-family-launched-c422-lga-2066-platform/

intel m!lk fest begin. imagine these new cpu is likely unlocked, better binned and possibly has soldered IHS. even the 4c come with quad channel and full 48 pcie lanes which is what we should have gotten in the first place, just like originaly sandy-E. but have to pay a few hundred more rofl.
Click to expand...
I don't see what workstation processors have to do with Coffee Lake or HEDT, for that matter. And no, none of Intel's products use solder anymore.
 
TaintedSquirrel said:
Crossposting from the Destiny 2 thread.
Steve @ GN claims Ryzen SMT is non-functional in this game, and disabling it might actually improve performance...

Are AMD owners actually just hardware beta testers?
Does "Fine Wine" simply mean fixing broken features down the road?
More news at 11.

Click to expand...



It's a beta for something (the game that is) if it was working 100% it would be release.
 
Armenius said:
I don't see what workstation processors have to do with Coffee Lake or HEDT, for that matter. And no, none of Intel's products use solder anymore.
Click to expand...

im sure their xeon lga3467 still uses solder. the xeon w is same socket lga2066 but i wonder if intel will allow it to be in x299 or has to be server chipset which is absurd. cause if they came out with these xeon w, i'd have gone with it 6 or 8 cores for full pcie lanes rather than the crippled i9 6-8c
 
oleNBR said:
im sure their xeon lga3467 still uses solder.
Click to expand...

There are no Intel chips that use solder anymore and you obviously dont understand why. Yet you keep trying to say something more expensive does every time you got the option as some kind of poor excuse that someone got "cheated". Using solder is a PROBLEM, not a solution.
 
TaintedSquirrel said:
Crossposting from the Destiny 2 thread.
Steve @ GN claims Ryzen SMT is non-functional in this game, and disabling it might actually improve performance...

Are AMD owners actually just hardware beta testers?
Does "Fine Wine" simply mean fixing broken features down the road?
More news at 11.

Click to expand...


Haswell IPC for Ryzen not even close again. A slower clocked 5960X is over 20% faster.

And Fury X performing like a 1060.

http://gamegpu.com/action-/-fps-/-tps/destiny-beta-test-gpu-cpu

8700K is what you want for this and all other games :D
 
oleNBR said:
http://www.hwbattle.com/bbs/board.php?bo_table=news&wr_id=85751&ckattempt=2

well now boys. 5ghz 6c at 1.415v? i guess its NOT BAD but i was hoping sub 1.35v maybe there are better silicon quality ones out there.
also score of 213 at 5ghz thats hella low. we were tricked after al. theres no IPC increase as expected, it is the same architecture as 7700k afterall.
Click to expand...

were you actually shocked?

A 6800 scores 1550~ at 5ghz so this is only impressive from the fact it can do 5ghz other than that its just another small step
 
Last edited:
Shintai said:
There are no Intel chips that use solder anymore and you obviously dont understand why. Yet you keep trying to say something more expensive does every time you got the option as some kind of poor excuse that someone got "cheated". Using solder is a PROBLEM, not a solution.
Click to expand...

i dont need to understand on this part since i wont be buying lga3647. howeverrrrr hedt use to be soldered just like xeons, but this gen hedt is now 2066 and xeon is now 3467. going by that looks like they'll get different production process. do you have proof that xeon will not be soldered? picture pls :)

also shintai, i am not a fanboy of any kind and you can simply look back at my posts i treat amd fanboys just the same as i treat intel fan boys. but looks like by simply saying "intel lied to us" touched you in a wrong spot, tell me where it hurts kay?


OrangeKhrush said:
were you actually shocked?

A 6800 scores 1550~ at 5ghz so this is only impressive from the fact it can do 5ghz other than that its just another small step
Click to expand...

not really that shocked. i have heard people telling me 8700k will be based off CNL design which results in IPC increase but thats just rumour and i dont think i have seen intel mention it themselves.

so in the end, from my understanding CNL was originally a shrink hence not a new architecture. meaning theres prob no IPC improvement. so a 14nm++ based off a 10nm design which was simply a shrink will not yield any improvement via design but opimization ipc improvement should still possible, just very tiny.

well in the end, that bench of CFL is ES show in screenshot, further more it might be able to do 5ghz but might not be a golden chip, you know the kind thats from silicon lottery. 7800x from silicon lottery could do 5ghz at 1.35v iirc. i want a better one if possible given it is 14nm++ not 14nm+
 
Last edited:
oleNBR said:
i dont need to understand on this part since i wont be buying lga3647. howeverrrrr hedt use to be soldered just like xeons, but this gen hedt is now 2066 and xeon is now 3467. going by that looks like they'll get different production process. do you have proof that xeon will not be soldered? picture pls :)

also shintai, i am not a fanboy of any kind and you can simply look back at my posts i treat amd fanboys just the same as i treat intel fan boys. but looks like by simply saying "intel lied to us" touched you in a wrong spot, tell me where it hurts kay?
Click to expand...

Its already posted with delids in one of the SKL-SP/X thread because the topic have already been covered.

If you already know you try to start a drama over false claims. Then why do you do it?
 
Shintai said:
Its already posted with delids in the SKL-SP thread.

If you already know you try to start a drama over false claims. Then why do you do it?
Click to expand...

it wont be false in my mind as im simply not aware. looks like you are very offended if someone touched you in the wrong spot with their comment. in a way you are very similar to the AMD fanboys you hated. you know the ones, not willing to listen and very angry, quick to retort.

its very easy to say, "no xeon has been proven to not use soldered, heres the link", rather than, "what are you talking about, blah blah, cheated blah blah". since you are now upset im guessing you prob wont link me to the SKL-SP thread i'll go look for it myself.
 
Shintai so after i tried searching few pages for threads under intel processor, i can't find skylake xeon lga 3467 (or 3647) whichever it is, using non-solder. so im going to have to say what you mentioned is false. so far, no proof.
 
oleNBR said:
Shintai so after i tried searching few pages for threads under intel processor, i can't find skylake xeon lga 3467 (or 3647) whichever it is, using non-solder. so im going to have to say what you mentioned is false. so far, no proof.
Click to expand...

oleNBR said:
im sure their xeon lga3467 still uses solder.
Click to expand...

Since you mention false and want me to do your homework for something discussed almost 3 months ago. Why dont you show its soldered?

I can tell you solder isn't used because the failure rate is too high due to thermal cycling.
 
Shintai said:
Since you mention false and want me to do your homework. Why dont you show its soldered?

I can tell you solder isn't used because the failure rate is too high due to thermal cycling.
Click to expand...

base on the past xeon are soldered, until theres picture or some lead, i'll have to say it stays soldered. for 2066 its not we already know but 3467 not sure yet. what about you, do u have any leads cause i'd love to know, even though i replied to you in a way thats not very respectful still share the info..
 
oleNBR said:
base on the past xeon are soldered, until theres picture or some lead, i'll have to say it stays soldered. for 2066 its not we already know but 3467 not sure yet. what about you, do u have any leads cause i'd love to know, even though i replied to you in a way thats not very respectful still share the info..
Click to expand...

The past? Is that your entire argument for a "I am certain it uses solder"? Xeon Phi was never soldered. Mainstream/Entry Xeons parts stopped being soldered with IB. And now with SKLSP/X its the same for all Xeons.

No,no, Intel is CHEATING us for soldered chips, right? I mean you say LGA3467 is soldered so that's the drama set. Lets be honest, this was your entry point to the discussion about solder that you tried making up with your BS several times. LGA2066 Xeons was also soldered in your post.

And the mandatory reading:
http://overclocking.guide/the-truth-about-cpu-soldering/
 
Last edited:
Aren't the new Xeons on a raised substrate/bigger PCB to fit onto 3467? Since it was originally designed for 72core Xeon Phi chips, hence why the same chips on 3467 can be brought down to LGA2066/X299.

That would also indicate that unsoldered X299/LGA2066 chips = unsoldered Xeons
 
Dayman said:
Aren't the new Xeons on a raised substrate/bigger PCB to fit onto 3467? Since it was originally designed for 72core Xeon Phi chips, hence why the same chips on 3467 can be brought down to LGA2066/X299.
Click to expand...

Its due to HMC stacks being higher. And then they can avoid the weird IHS Xeon Phi got with a thicker CPU part.

p7193705.jpg

p7193706.jpg
 
nice so finally some picture. now that solves the problem that even xeon will not use soldering.

going back to original topic that xeon w with 2066 will not be soldered, however intel sure is milking hard. traditional sandy-E hedt 4c/8t would get full quad channel memory and a lot more pcie lanes but kabylake-x 4c gets only 16 lanes and 2 channel. after people buy those they release xeon w rofl with full quad channels and 48 lanes basically non restricted skylake-x chips. people who paid just a bit less for earlier batch kinda got screwed or people who paid for more cores because wanted full pcie lanes also got screwed.

imagine these new xeon w 2066 comes with 14nm++. well played intel
 
oleNBR said:
nice so finally some picture. now that solves the problem that even xeon will not use soldering.
Click to expand...

I somehow doubt its the last post you make about unfounded solder claims.

oleNBR said:
imagine these new xeon w 2066 comes with 14nm++. well played intel
Click to expand...

They dont. 14nm++ is called Cascade Lake in that segment including LGA3467. Its shipping to the big 7+1 now as LGA3467, but for the rest its a 2018 product besides for some tier 4 DCs.

One day you may get something right....maybe.
 
Shintai said:
One day you may get something right....maybe.
Click to expand...

the trade off of not putting in too much time and effort and simply make guesses and theories based on what cards i have. also, if you look back in my posts, its assumption. never once said it will be. should be, might be, probably be. one day you may understand where and how im prediction things though.. maybe.
 
Spec codes also up

i7 8700K SR3QR
i7 8700 SR3QS
i5 8600K SR3QU
i5 8400 SR3QT
i3 8350K SR3N4
i3 8100 SR3N5
 
Last edited:
oleNBR said:
not really that shocked. i have heard people telling me 8700k will be based off CNL design which results in IPC increase but thats just rumour and i dont think i have seen intel mention it themselves.

so in the end, from my understanding CNL was originally a shrink hence not a new architecture. meaning theres prob no IPC improvement. so a 14nm++ based off a 10nm design which was simply a shrink will not yield any improvement via design but opimization ipc improvement should still possible, just very tiny.

well in the end, that bench of CFL is ES show in screenshot, further more it might be able to do 5ghz but might not be a golden chip, you know the kind thats from silicon lottery. 7800x from silicon lottery could do 5ghz at 1.35v iirc. i want a better one if possible given it is 14nm++ not 14nm+
Click to expand...

It is not a rumor, just a logical consequence of the technical specifications of the new chips. The IPC improvement is not based in a new microarchitecture, but in the larger shared L3 cache and the faster DRAM. Both elements increase the IPC in memory bound workloads. The increase is low, but not zero. This all was explained about six or seven times.
 
You must log in or register to reply here.
Back
Top