Top Three Intel 9th Generation Core Parts Detailed

[Valantar]

@Valantar well, seeing native 10GbE & having more native M.2 slots on the board to really take advantage of the CPU & chipset PCIe lanes would be something cool to look out for. Hopefully there's that to cater our needs later down the road.

Yeah, that would be sweet. I'm not really very hopeful for 10GbE on ITX, though, due to the "niche within a niche" nature of this and the board space needed for the controller (and cooling it). Still, hopefully it'll show up at some point. Maybe someone can manage to stick the controller and a thin heatsink on the rear of the motherboard? More m.2 is possible though, as Asus has shown with their newest ITX boards. Vertical m.2 at the edges of the board would also be a decent solution.

 

[B-Real]

To the less and less people telling that Intel doesn't need to be scared of AMD: why the hell are they raising the core numbers for the last 2 gens now?

BTW, according to Techpowerup's nearly 23000 votes, there are equal % of Intel's last 3 gen and AMD's last 2 (1,5) gen owner's: 28-28%.

 

[efikkan]

Im very surprised at those clock speeds, thats alot higher then I was expecting considering how hard it is to keep the 8700/K at full all core boost speed without thermal throttling. Thats a whole 300MHz higher on all cores over the 7820X. Your going to have to spend $100-$200 ontop of the retail price for a good air/water cooler to keep these 8core CPU's from thermal throttling and at 95W TDP I find that hard to believe. Time will tell I guess how these really fair in the wild, yes there going to fast thats for sure but at what cost.

The 95W TDP might be stretching it a little, but it probably wouldn't be too much higher. Also remember Ryzen 2700X is also boosting way out of its TDP with XFR 2. And even a real 95W CPU does need a decent cooler and sufficient airflow, the Intel stock cooler is not good enough for their 65W CPUs under load.

But if you want to compare it to i7-7820X; 3.6 GHz base, 4.0-4.3 GHz boost, remember that this CPU will have a small memory controller, no AVX-512, simpler core infrastructure, less cache, fewer PCIe lanes (yes, the interface is not free), etc., so a lower TDP than i7-7820X and slightly higher boost should be possible.

 

[StrayKAT]

To the less and less people telling that Intel doesn't need to be scared of AMD: why the hell are they raising the core numbers for the last 2 gens now?

BTW, according to Techpowerup's nearly 23000 votes, there are equal % of Intel's last 3 gen and AMD's last 2 (1,5) gen owner's: 28-28%.

Didn't want/need to. People were buying anyhow.

Besides, they rolled it out pretty quick. Shows how much they weren't blindsided or something.

 

[trparky]

I don't get it. What is Intel trying to do here? What it seems like is that they don't want to answer AMD at all. Ryzen has been out for over a year and this is the best Intel can muster? WTH Intel? Get the fricken lead out. AMD is cleaning your clock in just about every market and all you answer with is crap like this?

I don't get it either. It's like Intel didn't even try here. There's no question, AMD will be thoroughly kicking their asses with Zen 2 if this is all they can come up with.

Hindsight has revealed to me that perhaps going with the 8700K was a bad idea.

 

[StrayKAT]

My new board came with a 10g port btw.. never even heard of the brand (Aqunatia AQC107).

Not that I'd get much use out of it.. :\

 

[Tsukiyomi91]

@StrayKAT that's SuperMicro board for ya.
@Valantar that would be really cool for an mITX board! But for practicality, having all those features on mATX would be more realistic, on top of cramming everything on the table.

 

[Valantar]

Didn't want/need to. People were buying anyhow.

Besides, they rolled it out pretty quick. Shows how much they weren't blindsided or something.

Yeah, design lead times even on "simple" refreshes like CFL are not much less than two years, so they definitely weren't blindsided. The upcoming 8-core, on the other hand, bears quite a few hallmarks of being an "oh s***" response (accompanied by Intel's most chaotic chipset launch/refresh ever, rumors swirling for the better part of a year, questions regarding VRM compatibility, and so on and so forth) - probably due to somewhat underestimating either the performance, core count, price (or all three) of Ryzen.

For all my complaining in this thread, I have no doubt the upcoming 8-core Intel chip will be great if you have a motherboard and cooler to match, and can afford it.

 

[ToxicTaZ]

5GHz With Intel is here to stay!

8086K there first factory 5GHz.

Coffeelake has very good 14nm+++ yield quality that it allows them to join the 5GHz club.

Not Delidded!!
My 8700K @5.1GHz @1.41v with 75c load Cooled by EK

All 6 cores (AVX - 2)
Cinebench 15 single core score is 222

I'm curious about 9900K Cinebench 15 scores

 

[Tsukiyomi91]

inb4 Ryzen hitting full 5GHz when? xD

 

[efikkan]

Yeah, design lead times even on "simple" refreshes like CFL are not much less than two years, so they definitely weren't blindsided. The upcoming 8-core, on the other hand, bears quite a few hallmarks of being an "oh s***" response

8-core Coffee Lake is their backup plan (or the third backup plan to be exact), since they did see the potential for more delas of 10nm.

One important note is that Ice Lake is ready and has been for quite some time, so if it's delayed all the way to mid/late 2019, Intel have had two extra years to fine tune it.

 

[Valantar]

8-core Coffee Lake is their backup plan (or the third backup plan to be exact), since they did see the potential for more delas of 10nm.

One important note is that Ice Lake is ready and has been for quite some time, so if it's delayed all the way to mid/late 2019, Intel have had two extra years to fine tune it.

That's true, but one has to question how many optimizations of value they're able to squeeze in with the on-again-off-again development cycle they've been on for the past two years (while they obviously knew about the delays before the public, they didn't know they had two years - and delays have been announced on a more-than-yearly cadence, meaning the deadlines have likely been quite short). Still, if AMD could squeeze out 3% in a year on a barely-changed node update with what seems like very little effort while simultaneously working to get Zen2 on 7nm sampling to enterprise customers, I'm willing to bet Intel's army of engineers have been able to increase performance some, yeah. Then again, who knows if Ice Lake was slated to be another one of the ~5% IPC increases of Skylake and before?

 

[Konceptz]

Sounds like that i7-9700K will be the new gaming chip to have, 8 physical cores, no hyper threading.
I'm looking forward to the i9 8c/16t.....back to intel I come

 

[efikkan]

That's true, but one has to question how many optimizations of value they're able to squeeze in with the on-again-off-again development cycle they've been on for the past two years (while they obviously knew about the delays before the public, they didn't know they had two years - and delays have been announced on a more-than-yearly cadence, meaning the deadlines have likely been quite short). Still, if AMD could squeeze out 3% in a year on a barely-changed node update with what seems like very little effort while simultaneously working to get Zen2 on 7nm sampling to enterprise customers, I'm willing to bet Intel's army of engineers have been able to increase performance some, yeah. Then again, who knows if Ice Lake was slated to be another one of the ~5% IPC increases of Skylake and before?

No, I was more thinking of maturity, bugs, etc. All newly released CPUs have an errata of 30+ bugs, many of which may affect performance slightly.

If it were up to me, I would have made the third backup-plan a port of Ice Lake to 14nm rather than two extra cores on Coffee Lake. It's of course way too late now, but a slighly modified Ice Lake (no AVX-512) on 14nm would have been a more interesting and innovative product for 2H 2018 than a "patched" Coffee Lake.

 

[GoldenX]

That is a patched Kaby Lake, that is a patched Skylake...

 

[Valantar]

No, I was more thinking of maturity, bugs, etc. All newly released CPUs have an errata of 30+ bugs, many of which may affect performance slightly.

If it were up to me, I would have made the third backup-plan a port of Ice Lake to 14nm rather than two extra cores on Coffee Lake. It's of course way too late now, but a slighly modified Ice Lake (no AVX-512) on 14nm would have been a more interesting and innovative product for 2H 2018 than a "patched" Coffee Lake.

Ah, I see. Are they able to find and eliminate those bugs or mature the design without near-final sample silicon available? No doubt they have test silicon by now, but if the 10nm "i3" they "launched" is anything to go by, that's not nearly production-grade even if you get a defect-free die (shipping an iGPU-less i3 at 15w with lower clocks than KBL and CFL counterparts points to both defects and power consumption issues). They have probably run far more and more detailed simulations than usual, but I don't think you can achieve what you achieve by dabbing actual chips that way.

I completely agree with your backup plan option, though. 14nm Ice Lake would definitely have been preferable. But there are probably quite a few reasons for it ending up this way - multiple 10nm delays; an Ice Lake redesign taking far more time than adding cores to CFL; ICL likely having more transistors and thus requiring a bigger die on the same process. It's even possible Ice Lake continues the hunt for clock speeds at the cost of power, which 10nm would have alleviated/negated, but a 14nm redesign might have been more power hungry than CFL. I guess we'll never know, but it'd be nice if someone serious got an exclusive in-depth interview after I've lake launches.

 

[efikkan]

Ah, I see. Are they able to find and eliminate those bugs or mature the design without near-final sample silicon available? No doubt they have test silicon by now, but if the 10nm "i3" they "launched" is anything to go by, that's not nearly production-grade even if you get a defect-free die

The complete design was taped out in June 2017, with the first engineering samples of these being tested since last November. Yields are much more complex than just defects, it also comes down to thermals and reliability of various areas of the die. The problem with Intel 10nm is not defects, but getting enough dies which can operate on desired clocks. The launch is delayed because they can't make enough dies of good enough quality to make it profitable, that doesn't mean they have no such dies for testing.

You saw the minimal tweaks AMD did in Zen+, similarly, Intel can do small tweaks in their refinements. Massive delays to Ice Lake will allow Intel to do more of these ahead of release, they are certainly not sitting there doing nothing.

Intel haven't given any performance figures for Ice Lake yet, but if it brings similar cache improvements like Skylake-X/-SP in an even more refined form, that alone can give 3-4%. And if so, Ice Lake will be a larger overhaul than Skylake and Haswell. Overhauls can of course be more or less successful (just look at Bulldozer), but at least it's not a "Skylake 2". I would argue that your estimate of 5% IPC for Ice Lake gains is relatively conservative.

 

[Prima.Vera]

5GHz With Intel is here to stay!
8086K there first factory 5GHz.
Coffeelake has very good 14nm+++ yield quality that it allows them to join the 5GHz club.

All I need now it's some standard cooled DDR4 sticks at 5GHz too.

 

[ToxicTaZ]

All I need now it's some standard cooled DDR4 sticks at 5GHz too.

Personally I would go for the 4800MHz CL17-17-17-37

The 5066MHz is CL21

 

[Valantar]

The complete design was taped out in June 2017, with the first engineering samples of these being tested since last November. Yields are much more complex than just defects, it also comes down to thermals and reliability of various areas of the die. The problem with Intel 10nm is not defects, but getting enough dies which can operate on desired clocks. The launch is delayed because they can't make enough dies of good enough quality to make it profitable, that doesn't mean they have no such dies for testing.

You saw the minimal tweaks AMD did in Zen+, similarly, Intel can do small tweaks in their refinements. Massive delays to Ice Lake will allow Intel to do more of these ahead of release, they are certainly not sitting there doing nothing.

Intel haven't given any performance figures for Ice Lake yet, but if it brings similar cache improvements like Skylake-X/-SP in an even more refined form, that alone can give 3-4%. And if so, Ice Lake will be a larger overhaul than Skylake and Haswell. Overhauls can of course be more or less successful (just look at Bulldozer), but at least it's not a "Skylake 2". I would argue that your estimate of 5% IPC for Ice Lake gains is relatively conservative.

Yeah, you're entirely right that there's obviously something wrong with the power and clock scaling characteristics of Intel's current 10nm dice. I'm interpreting launching an iGPU-less SKU as compensating for defects as well, but of course it's entirely possible that that is also due to power issues (or even something as simple as not wanting to work on launching a driver for an updated GPU arch when there's a single, limited-run SKU out there).

As for me mentioning a 5% IPC increase, I didn't mean that as an estimate at all, I was simply saying that that has been what Intel has managed over their last arch updates (Broadwell and Skylake, at least). Without anything to go by, I'd be expecting anywhere between 5-10% (less would be weird for putting the effort of an arch update into it; more would be the biggest jump in quite a while), but given how long this development cycle has been, I agree that 5% would be unexpectedly low. Still, throwing numbers out like this is at best a guessing game, and purely speculative. Also, I think we're in agreement on the potential for improvements during the delays; nothing ground-breaking (like they probably could have done if they'd been told "hey, you've got two more years to finish this, go nuts!"), but they've definitely had time to iron out bugs and tweak various parameters. You sound right in saying that ES silicon fabbed on a semi-broken process with power/frequency issues can still be used to iron out arhcitecture errata - at least unless those errata only show up at high clocks, which I don't think is very common. Not to mention that power issues probably aren't very problematic in engineering situations, as long as they have a few hefty heatsinks or AIOs lying around.

Btw, have you heard/read anything about whether Intel will be moving to a mesh interconnect with Ice Lake? I don't know if they can match SKL-X's cache hierarchy without one (given the changes in how caches are shared), but from what I've read the mesh comes with a noticeable power penalty. Of course, at least with a 4-core die, there's zero real difference between a mesh and a ring bus

 

[enxo218]

Will there be a "Brandy lake" after then?

 

[Valantar]

Will there be a "Brandy lake" after then?

Coffee, Whiskey, Ice, Brandy? That doesn't sound too tasty. I'd keep to the first three, or skip the whiskey. Though I do see a new market opportunity for Intel here, selling official pre-mixed "codename series" drinks.