Intel's 9th Generation Core Family - Coffee Lake (Refresh)

Polo6RGTI

Polo6RGTI

n00b
Joined
Jun 7, 2018
Messages
34
https://www.intel.com/content/dam/www/public/us/en/documents/sa00115-microcode-update-guidance.pdf

Screenshot (35).png
 
Tiny little bumps, just like the 8086, to try and combat the Zen refresh.

And just like everyone said the 8-core rumor was just bull.

Would be nice if Intel could come out with a core architecture to replace Skylake, but I guess that would be asking too much after three fucking years?
 
So... Skylake v4.0? Or would it just be v3.2 since it is not a node change?

Skylake v1.0
Kabylake v2.0
Coffee lake v3.0
i7 x8086k v3.1
New CFL+ v3.2
 
kirbyrj said:
So basically the Z170 generation boards could be running these chips but Intel likes chipset revenue.
Click to expand...
Yep, the reason I’m not buying another 14nm revamped skylake product from them. New arch, new node, new platform or bust. It’s not even funny at this point.
 
I'm not Intel or even a millionaire so wtf do I know but that seems like a very questionable business decision. I know there's no way I'd have been able to resist going from a 6600k to an 8700k if I could remove four thumb screws, detach a heatsink, take out this tiny little square, and put in a new tiny little square.

Shipping a processor is going to be easier, too. If I have to deal with ripping the mobo out, redoing all my cables, reinstalling Windows (Not sure on this one) and everything else just for a slight performance gain, eh not worth it. Especially when Z370 offers ____________ over Z170.

I don't think I'm even close to alone in this either.
 
Dayman said:
Notice how 9000 series i7s are omitted from the spec sheet? :)
Click to expand...

Well, that's because they already did that. The 8086k is the i7 refresh.
And they really can't bump the 8700k up any higher. That was an emergency Factory Overclock to counter the 2700x before it was even released, and until they actually build a new core the frequencies will not be traveling much above five.
 
Last edited:
Budwise said:
9th iteration, still 8th generation apparently. At least they admit its not new but they'll require a new motherboard to run it!

Meanwhile someone just broke the SuperPi record using an 8700K on a Z270 :D FU Intel

https://www.techpowerup.com/245689/...ghz-3c-3t-on-z270-chipset-bags-superpi-record
Click to expand...


"Half the cores on the i7-8700K were disabled and so was HyperThreading"

So he run a 3-core / 3-thread chip on a mobo that supports (by spec) up to 4-cores and 8-threads. Wow!

Whereas the technical reason why 8700k required a new chipset/mobo is because not all the 270 mobos (verified only to support 4-core / 8-threads per spec) could support 6-cores and 12-threads
 
Last edited:
defaultluser said:
Tiny little bumps, just like the 8086, to try and combat the Zen refresh.

And just like everyone said the 8-core rumor was just bull.

Would be nice if Intel could come out with a core architecture to replace Skylake, but I guess that would be asking too much after three fucking years?
Click to expand...

A replacement for Skylake was planned, but it used 10nm. Piednoel suggested then to port back Icelake to 14nm, but his suggestion was rejected.

No new muarch until late 2019 or early 2020 when 10nm yields solved.
 
defaultluser said:
Well, that's because they already did that. The 8086k is the i7 refresh.
And they really can't bump the 8700k up any higher. That was an emergency Factory Overclock to counter the 2700x before it was even released, and until they actually build a new core the frequencies will not be traveling much above five.
Click to expand...


lol
 
Polo6RGTI said:
MSI BIOS files showing 8-core support.

MSI Z370 Gaming Pro Carbon
View attachment 87266
MSI B360M PRO-VD
View attachment 87267
MSI H310M Gaming Plus
View attachment 87268

Click to expand...

defaultluser said:
Tiny little bumps, just like the 8086, to try and combat the Zen refresh.

And just like everyone said the 8-core rumor was just bull.

Would be nice if Intel could come out with a core architecture to replace Skylake, but I guess that would be asking too much after three fucking years?
Click to expand...

im worried.. so is 8 core on 14nm++ really coming this yr?
 
oleNBR said:
im worried.. so is 8 core on 14nm++ really coming this yr?
Click to expand...

Doubtful.

Intel still has a 20% IPC advantage in most games, and since they're matching AMD on core count and clocks in the low-end, and their overclocked 6-core exceeds AMD's 6 and 8-core in most benches, they're in no need of moar coars.

But the 2700x's overclocked speeds put it within 15% of the 8700k in games, which while uncomfortably close is still a win. But that bump is enough to win in many productivity benchmarks.

But that's what the 8086k is there to combat. So no, it's a tight battle, but not critical enough to rush out another core rev.

If you need Moar Coars, that's what Threadripper and Skylake-E are for.
 
Last edited:
defaultluser said:
Doubtful.




But that's what the 8086k is there to combat. So no, it's a tight battle, but not critical enough to rush out another core rev.

If you need Moar Coars, that's what Threadripper and Skylake-E are for.
Click to expand...

The 8600k will match the 8700k in pretty much all games except maybe Civilization. So what is the point of the 8700k? Same gaming performance with better productivity. Why not something with the same gaming ability with even MORE productivity. It would be an attractive option over simply 15% better gaming.

What will be interesting is if Intel implements an AMD-like precicion boost into their new 8-core. Even with great cooling, it is hard to imagine the 8-cores going much over 4.5 ghz due to thermals. It would be really solid if they have the chips boost closer to 5.0 ghz on lower thread workloads so as not to lose the advantage to the 8700k in those scenarios.
 
Nightfire said:
The 8600k will match the 8700k in pretty much all games except maybe Civilization. So what is the point of the 8700k? Same gaming performance with better productivity. Why not something with the same gaming ability with even MORE productivity. It would be an attractive option over simply 15% better gaming.

What will be interesting is if Intel implements an AMD-like precicion boost into their new 8-core. Even with great cooling, it is hard to imagine the 8-cores going much over 4.5 ghz due to thermals. It would be really solid if they have the chips boost closer to 5.0 ghz on lower thread workloads so as not to lose the advantage to the 8700k in those scenarios.
Click to expand...

Last time I checked, the i5 8600k is a factory-overclocked Intel 6-core. I mentioned this in my post (overclocked 6-cores exceeded all AMD parts in games).

The 8700k only exists for bragging rights, and to give you a reason to not buy the 2700x if you do more than game. But for whiny post going fishing for an Intel 8-core, I had to touch on it as the top-end part.
 
defaultluser said:
Last time I checked, the i5 8600k is a factory-overclocked Intel 6-core. I mentioned this in my post (overclocked 6-cores exceeded all AMD parts in games).

The 8700k only exists for bragging rights, and to give you a reason to not buy the 2700x if you do more than game. But for whiny post going fishing for an Intel 8-core, I had to touch on it as the top-end part.
Click to expand...

It is late here, but everything about this post confuses me.
 
I'd be mildly interested in a 8-core non-HT variant to replace my poor 8600K sample doing 4.75GHz at little over 1.3v.
 
ROG-MAXIMUS-X-HERO-ASUS-1503 BIOS file
Screenshot (13).png
Latest ROG-MAXIMUS-X-HERO-ASUS-1602 BIOS file
Screenshot (12).png
 
defaultluser said:
The 8700k only exists for bragging rights, and to give you a reason to not buy the 2700x if you do more than game. But for whiny post going fishing for an Intel 8-core, I had to touch on it as the top-end part.
Click to expand...

What an odd post.

The 8700k is a beast.
 
RPGWiZaRD said:
I'd be mildly interested in a 8-core non-HT variant to replace my poor 8600K sample doing 4.75GHz at little over 1.3v.
Click to expand...

1.3v is too much for 8 core imho. i'd say 1.23v for 8 cores be good. also i'd get 8 cores with HT just to turn off HT and get the extra cache.

Polo6RGTI said:
ROG-MAXIMUS-X-HERO-ASUS-1503 BIOS file
View attachment 88186
Latest ROG-MAXIMUS-X-HERO-ASUS-1602 BIOS file
View attachment 88187
Click to expand...

excellent find man! now we can almost be certain intel is super tight lipped about this but it is coming soon, especially it is for a 6core bios same mobo to 8 core, also a good chance for 8 core to be plug n play into z370 with just a bios update/mod. looking good for me laptops.
 
oleNBR said:
also i'd get 8 cores with HT just to turn off HT and get the extra cache.
Click to expand...

Please run benchmarks first. Things which can use 8 threads can almost always use 16 threads even better, even if they share execution units. It is workload dependent which is why I say run benchmarks, but it's actually pretty rare HT isn't a very good thing on heavy loaded anything.
 
PhaseNoise said:
Please run benchmarks first. Things which can use 8 threads can almost always use 16 threads even better, even if they share execution units. It is workload dependent which is why I say run benchmarks, but it's actually pretty rare HT isn't a very good thing on heavy loaded anything.
Click to expand...

thing is i might have to disable HT due to extra heat in laptop. 8 core with HT off will allow higher frequency to hopefully make up for the HT portion while being faster on single threaded workloads.
 
You would pay extra for 8 cores no ht over a 6/12 setup? Seems like they would perform similiar at similiar heat levels.
 
Nightfire said:
You would pay extra for 8 cores no ht over a 6/12 setup? Seems like they would perform similiar at similiar heat levels.
Click to expand...

yea the benefit aren't too great but imo 8 core with HT turned off with more cache should perform higher than 6c/12t. threading only gives 10-20% in some cases, where as 33% more cores always gonna be there along with more cache.
 
oleNBR said:
yea the benefit aren't too great but imo 8 core with HT turned off with more cache should perform higher than 6c/12t. threading only gives 10-20% in some cases, where as 33% more cores always gonna be there along with more cache.
Click to expand...

Having spent a nontrivial amount of my life researching and developing SMT - I would just reiterate that you should run benches to validate assumptions. Things are often quite counter-intuitive.

Cache with relation to the cores is very tricky. Cache population is quite good in modern processors. It turns out "more cache" rarely has much of an impact in most programs (emphasis on most - there are some things which are all over the place with memory locality), because the access patterns which blow up the 'smaller' cache still blows up the larger.

Constrast this with having another thread on the core - you avoid stalling the entire core based upon 1 cache miss or similar stall. This is why SMT ends up giving really solid improvements, even though you'd think the cores should be fully utilized by aggressive workloads. Stalls are very expensive.

So if the desire is to have more cache versus more threads, consider that the bargain really boils down to slightly increasing your cache hit rate at the cost of significantly increasing the cost of a cache miss.

If power/heat is a huge concern, I'm not going to have any good advice. Trying to optimize throughput while saying power is the primary constraint is not a world I've had to live in. Well, I guess my advice would still be "Get more cooling!" :)
 
I've flashed my MSI Z370 Gaming Pro Carbon MB today with the latest BETA BIOS.

"Support New Generation CPU!"
Screenshot (17).png

The "8-core ratio" wording is now omitted from the BIOS file with only "Undefined" showing.
MSI BETA BIOS A (2).png

MSI BETA BIOS A (3).png

MSI BETA BIOS A (1).png
 
You must log in or register to reply here.
Back
Top