Intel to launch 6 core Coffee Lake-S CPUs & Z370 chipset 5 October 2017

zornyan · 2 Jun 2017 at 23:08

gavinh87 said:
Avx is not used in realbench....
Your realbench temps are higher than my 8 core using full avx testing.
Intel definitely has a TIM problem.

avx is used in realbench......even says so on the asus website (and the forums)

https://rog.asus.com/articles/guides/guide-realbench-hwbot-edition-v2-43/

quoted

" It uses up to AVX CPU extensions."

as said when running a specific non avx stress test/ benchmark my temps are a fair bit lower.

gavinh87 · 3 Jun 2017 at 06:46

zornyan said:
avx is used in realbench......even says so on the asus website (and the forums)

https://rog.asus.com/articles/guides/guide-realbench-hwbot-edition-v2-43/

quoted

" It uses up to AVX CPU extensions."

as said when running a specific non avx stress test/ benchmark my temps are a fair bit lower.

Ah yes it does. I wasn't aware of that.
My point still stands, your chip under the same tests should be a fair bit cooler.

N19h7m4r3 · 3 Jun 2017 at 08:47

gavinh87 said:
Ah yes it does. I wasn't aware of that.
My point still stands, your chip under the same tests should be a fair bit cooler.

Well yes, the average Delid of the 7700K sees a 23-30 degrees Celsius drop in temps when under such stress tests.

One of the reasons I'm apprehensive about Intel using TIM on their Skylake X processors instead of soldering like with Xeons and previous X99.

zornyan · 3 Jun 2017 at 08:54

N19h7m4r3 said:
Well yes, the average Delid of the 7700K sees a 23-30 degrees Celsius drop in temps when under such stress tests.

One of the reasons I'm apprehensive about Intel using TIM on their Skylake X processors instead of soldering like with Xeons and previous X99.

a 20c drop would make my cpu run cooler than the 1500 ryzen which is soldered.

according to tomshardware and legit reviews my cpu is only 10c hotter than a 1500 at 1.375v, which partly as I've said is down to the avx instructions.

so maybe 8c of my stock 7700k vs the ryzen solder?

not sure where 30c comes from...

N19h7m4r3 · 3 Jun 2017 at 09:11

zornyan said:
a 20c drop would make my cpu run cooler than the 1500 ryzen which is soldered.

according to tomshardware and legit reviews my cpu is only 10c hotter than a 1500 at 1.375v, which partly as I've said is down to the avx instructions.

so maybe 8c of my stock 7700k vs the ryzen solder?

not sure where 30c comes from...

So getting a 20 degree drop is now bad? That's fantastic and should be applauded, now imagine if your CPU didn't have rubbish TIM and was soldered. You'd run 15-20 degrees cooler right now with your overclock and during stress tests.

Toms found the 1500X at 1.35V to run at a max of 64 Degrees under AIDA 64 and Prime 95. Done using single tower Noctua NH-U12S air cooler.
http://www.tomshardware.com/reviews/amd-ryzen-5-1500x-cpu,5025-2.html

In the U.S. lab, we recorded up to 64°C (per AIDA) with our Noctua NH-U12S SE-AM4 during extended stress tests. Seeing plenty of available thermal headroom

Where as they found the 7700K to run at 85 degrees at stock under AIDA64, and package exceeded 93 degrees, and they state. This was done on an 240mm AIO Liquid cooler
http://www.tomshardware.com/reviews...i7-7700k-i7-7700-i5-7600k-i5-7600,4870-8.html

With the Tcore reaching up to 85°C, we’re still in the clear. But that's not a good sign for the thermal headroom available when we want to start overclocking.

They also saw the 7700K pushing up to 71 Degrees after 22 minutes of gaming.

The temperatures increase at a different pace, depending on the position of the sensor. They do stabilize after a maximum of 22 minutes at up to 71°C, though.

So according to Tomshardware the 7700K at stock after 22 minutes of gaming runs hotter than an Overclock Ryzen 1500X being stressed in AIDA 64; while the former is also using a 240mm AIO liquid cooler.

Which for someone like myself that does lots of encoding and rendering makes a rather big difference. Having all CPU cores pegged at 100% for hours on end, running cooler is a great benefit.

zornyan · 3 Jun 2017 at 09:18

N19h7m4r3 said:
So getting a 20 degree drop is now bad? That's fantastic and should be applauded, now imagine if your CPU didn't have rubbish TIM and was soldered. You'd run 15-20 degrees cooler right now with your overclock and during stress tests.

Toms found the 1500X at 1.35V to run at a max of 64 Degrees under AIDA 64 and Prime 95. Done using single tower Noctua NH-U12S air cooler.
http://www.tomshardware.com/reviews/amd-ryzen-5-1500x-cpu,5025-2.html

Where as they found the 7700K to run at 85 degrees at stock under AIDA64, and package exceeded 93 degrees, and they state. This was done an 240mm AIO Liquid cooler
http://www.tomshardware.com/reviews...i7-7700k-i7-7700-i5-7600k-i5-7600,4870-8.html

They also saw the 7700K pushing up to 71 Degrees after 22 minutes of gaming.

So according to Tomshardware the 7700K at stock after 22 minutes of gaming runs hotter than an Overclock Ryzen 1500X being stressed in AIDA 64; while the former is also using a 240mm AIO liquid cooler.

Which for someone like myself that does lots of encoding and rendering makes a rather big difference. Having all CPU cores pegged at 100% for hours on end, running cooler is a great benefit.

since we're talking about skylake x let's compare it to skylake

http://www.tomshardware.co.uk/skylake-intel-core-i7-6700k-core-i5-6600k,review-33276-11.html

6700k averages 64c stress test with an air cooler,

6600k is 48c with an air cooler.

so again, skylake x shall be fine, it's built on the same process as skylake (hint not kabylake)

Journey · 3 Jun 2017 at 09:24

zornyan said:
since we're talking about skylake x let's compare it to skylake

http://www.tomshardware.co.uk/skylake-intel-core-i7-6700k-core-i5-6600k,review-33276-11.html

6700k averages 64c stress test with an air cooler,

6600k is 48c with an air cooler.

so again, skylake x shall be fine, it's built on the same process as skylake (hint not kabylake)

It depends on the PCB quality/thickness also, this is the issue Kaby faces. I'd imagine the X299 CPU's will need a better/thicker PCB than Kaby 1151 due to the extra pins.

N19h7m4r3 · 3 Jun 2017 at 09:26

zornyan said:
since we're talking about skylake x let's compare it to skylake

http://www.tomshardware.co.uk/skylake-intel-core-i7-6700k-core-i5-6600k,review-33276-11.html

6700k averages 64c stress test with an air cooler,

6600k is 48c with an air cooler.

so again, skylake x shall be fine, it's built on the same process as skylake (hint not kabylake)

Only if the TIM is correctly applied. That's the issue, there's some big variances in the application leading to high temperatures since Kabylake's introduction again.
Similar to Haswell's 4770K vs the 4790K and against my Haswell-E x99 system.

Also Skylake and Kabylake are all built on the same Intel 14nm Process.

Hopefully it is fine, but considering the TDP, and core count I'll be paying a lot of attention to the 10-12 core Skylake X parts' stress tests before deciding if I buy one.

zornyan · 3 Jun 2017 at 09:45

N19h7m4r3 said:
Only if the TIM is correctly applied. That's the issue, there's some big variances in the application leading to high temperatures since Kabylake's introduction again.
Similar to Haswell's 4770K vs the 4790K and against my Haswell-E x99 system.

Also Skylake and Kabylake are all built on the same Intel 14nm Process.

Hopefully it is fine, but considering the TDP, and core count I'll be paying a lot of attention to the 10-12 core Skylake X parts' stress tests before deciding if I buy one.

right, as I've been saying this whole time....

it's not using TIM itself, as we can see from skylake correct TIM usage can mean there's a very low temp difference between TIM and solder.

but people exploding over this is kind of stupid imo "lolz it's going to run 90c a 0.08v 100mhz"

as long as Intel use the same process for building skylake x, as they did skylakex, temps should intact be perfectly fine.

N19h7m4r3 · 3 Jun 2017 at 09:59

zornyan said:
as long as Intel use the same process for building skylake x, as they did skylakex, temps should intact be perfectly fine.

Skylake / X and Kabylake / X are all built on the same process, so anything can happen with the TIM application

Simon Belmont · 3 Jun 2017 at 18:22

Any new info on coffeelake? Don't really want to go x299. I can see that platform being stupid expensive. I don't want to wait forever though? Don't know what I'm going to do yet?

GAC · 3 Jun 2017 at 18:24

Simon Belmont said:
Any new info on coffeelake? Don't really want to go x299. I can see that platform being stupid expensive. I don't want to wait forever though?

only thing thats seem to have come out is its going to be closer to oct than august by the looks of it. hopefully we'l see or find something out at e3 in a week but after the damp fart that was computex i bet its all x299 stuff being pushed to streamers.

Simon Belmont · 3 Jun 2017 at 18:55

GAC said:
only thing thats seem to have come out is its going to be closer to oct than august by the looks of it. hopefully we'l see or find something out at e3 in a week but after the damp fart that was computex i bet its all x299 stuff being pushed to streamers.

I was really hoping Aug. I guess they don't want to hurt x299 sales

I would think about going x299 if it wasn't for the price.

GAC · 3 Jun 2017 at 19:10

Simon Belmont said:
I was really hoping Aug. I guess they don't want to hurt x299 sales I would think about going x299 if it wasn't for the price.

been saying that myself with a august release it would make the lower end x299 stuff rather pointless. what do you want it to do though as x299 for the price unless you need the top end chips will be rather silly i feel.

Simon Belmont · 3 Jun 2017 at 19:14

GAC said:
been saying that myself with a august release it would make the lower end x299 stuff rather pointless. what do you want it to do though as x299 for the price unless you need the top end chips will be rather silly i feel.

Mostly gaming on my GTX 1080 and some browsing. I'm looking at 7820x, but might just wait for coffelake. Don't think x299 is worth it for me.

DragonQ · 4 Jun 2017 at 10:45

X299 and Threadripper are a waste of time for pure gaming on a single GPU. Even dual GPUs should be fine on the mainstream platforms at 8x/8x. I'd wait for Coffee Lake or go R5 1600/R7 1700.

GAC · 4 Jun 2017 at 11:13

DragonQ said:
X299 and Threadripper are a waste of time for pure gaming on a single GPU. Even dual GPUs should be fine on the mainstream platforms at 8x/8x. I'd wait for Coffee Lake or go R5 1600/R7 1700.

posted similar in the x299 thread but after seeing the linus video im rather really confused to as wtf intel is playing at with x299 and it worries me as to what they will now do with coffee lake to make x299 seem a better option than it is. if coffee lakes top i5 is basically a old i7 6c12t that makes it better than the kaby lake x299 chips. then what will the i7's do 8c16t which then starts taking more sales away from x299. i get a feeling these two chips will be clocked low to keep x299 the performance leading platform for intel and IF they do a K version it will be priced rather more than in previous generations.

Simon Belmont · 4 Jun 2017 at 12:49

I watched that video as well. Linus is right on the money? WTF is Intel doing? Agree I'm a little worried to what coffeelake will be like now?

Journey · 4 Jun 2017 at 12:58

Well it's the next mainstream architecture, and as such will be positioned as an evolution of the current Kaby Lake, keeping the current 2c/4t low end, but adding in a 6c/6t i5, and a 6c/12t i7, they will obviously be a small amount of crossover, but the majority of i3/i5/i7 CPU's sold will be in OEM machines and even with the crossover shouldn't effect the x299 platform sales, since the cost, size, and power usage is not in the same range.

If all you care about is what is relevant to enthusiast, then Coffee Lake will bring two extra cores, and a new chipset, with costing approximately the same as we have now or slightly less.

drunkenmaster · 4 Jun 2017 at 16:30

zornyan said:
since we're talking about skylake x let's compare it to skylake

http://www.tomshardware.co.uk/skylake-intel-core-i7-6700k-core-i5-6600k,review-33276-11.html

6700k averages 64c stress test with an air cooler,

6600k is 48c with an air cooler.

so again, skylake x shall be fine, it's built on the same process as skylake (hint not kabylake)

They almost certainly use the same process, it's an improved process, nothing more or less. Kabylake wasn't a massive leap in process tech, it's a completely standard tweak, every process ever made has improved little by little over time. This is usually just improved design rules allowing narrower margins for error as the more mature process moves towards the theoretically minimums for feature size. As things get smaller you can use the same die size for slightly more performance or lower power or make the chip slightly smaller and make them cheaper. It doesn't make the process more expensive to use, it's the same equipment, the same silicon and the same time to make(in most cases, when not the difference will be marginal). In other words, for Intel not to use the tweaked process design rules would be choosing a more expensive chip for no reason at all.

There was a very small and very standard change in terms of the process for Kabylake in the first place. Back in previous gens where the gap between nodes was shorter, Intel just released the same chips with a new stepping but the same names, Q6600 C0 stepping is one that springs to mind. AMD often brought out a new stepping everyone wanted as well, it's now process nodes work. There is zero benefit to not using the tweaked process on every new chip you produce from that point forward.