Monday, June 17th 2019

Intel "Ice Lake" IPC Best-Case a Massive 40% Uplift Over "Skylake," 18% on Average

Intel late-May made its first major disclosure of the per-core CPU performance gains achieved with its "Ice Lake" processor that packs "Sunny Cove" CPU cores. Averaged across a spectrum of benchmarks, Intel claims a best-case scenario IPC (instructions per clock) uplift of a massive 40 percent over "Skylake," and a mean uplift of 18 percent. The worst-case scenario sees its performance negligibly below that of "Skylake." Intel's IPC figures are derived entirely across synthetic benchmarks, which include SPEC 2016, SPEC 2017, SYSMark 2014 SE, WebXprt, and CineBench R15. The comparison to "Skylake" is relevant because Intel has been using essentially the same CPU core in the succeeding three generations that include "Kaby Lake" and "Coffee Lake."

A Chinese tech-forum member with access to an "Ice Lake" 6-core/12-thread sample put the chip through the CPU-Z internal benchmark (test module version 17.01). At a clock-speed of 3.60 GHz, the "Ice Lake" chip allegedly achieved a single-core score of 635 points. To put this number into perspective, a Ryzen 7 3800X "Matisse" supposedly needs to run at 4.70 GHz to match this score, and a Core i7-7700K "Kaby Lake" needs to run at 5.20 GHz. Desktop "Ice Lake" processors are unlikely to launch in 2019. The first "Ice Lake" processors are 4-core/8-thread chips designed for ultraportable notebook platforms, which come out in Q4-2019, and desktop "Ice Lake" parts are expected only in 2020.
Source: WCCFTech
Add your own comment

153 Comments on Intel "Ice Lake" IPC Best-Case a Massive 40% Uplift Over "Skylake," 18% on Average

#26
phill
I'll wait for the reviews and see what comes :) As pretty and nice as the slides are, I'll believe it when we see the reviews :)
Posted on Reply
#27
Xzibit
londiste, post: 4065914, member: 169790"
This is an old roadmap slide. Newer one shows 10nm quickly replaced with 10+nm and then with 7nm less than a year after. Product roadmap at the same time only has 10nm CPUs for mobile by end of this year and for servers early next year. Desktop does not have anything at 10nm on roadmap.
Intel investor day was May 8, 2019. Your confusing Intel with the Dell Leak Roadmap.
Posted on Reply
#28
medi01
FordGT90Concept, post: 4065873, member: 60463"
18% over four years is pretty terrible, especially if you consider how much performance was lost because of security mitigations.
I wonder what is used as a base here.
Posted on Reply
#29
Fabio
my 8700k has a 565 single core score for cpu z, now at 5,2 magically rise at 635?
Posted on Reply
#30
delshay
I can bet by the year 2023, 3nm will be here as 5nm is more or less already here.
Posted on Reply
#31
Manu_PT
This is so funny!!

AMD shows the most biased and misleading benchmarks -> "yes!! Intel is done!! AMD will conquer the world!!"

Intel has leaked benchmarks where 5,2 ghz = the new 3,6ghz, wich is amazimg -> "lol intel. Lol this is fake. Lol only 18% improvement. Lol this is never coming to desktop"

Seriously you guys are annoying. If this IPC improvements are true, oh boy, they gonna rip AMD apart on mobile and they will on desktop once they cam do it. I am for sure gonna be waiting.
Posted on Reply
#32
Imsochobo
Manu_PT, post: 4065937, member: 168799"
This is so funny!!

AMD shows the most biased and misleading benchmarks -> "yes!! Intel is done!! AMD will conquer the world!!"

Intel has leaked benchmarks where 5,2 ghz = the new 3,6ghz, wich is amazimg -> "lol intel. Lol this is fake. Lol only 18% improvement. Lol this is never coming to desktop"

Seriously you guys are annoying. If this IPC improvements are true, oh boy, they gonna rip AMD apart on mobile and they will on desktop once they cam do it. I am for sure gonna be waiting.
I expect Ice lake to either dominate or be highly competitive in laptops, but Intel is quite doomed on desktop(because of 10nm performance).

londiste, post: 4065914, member: 169790"
Intel does not need help with performance. Skylake (and all the minor modifications to it) are competitive if not faster than Zen. Sunny Cove based CPUs seem to be competitive enough with Zen2 based on early information and the architectural changes in both that seem to mirror each other.

What Intel needs help with is power efficiency that 10nm/7nm does bring to the table. Smaller dies may be helpful or harmful but we do not know that yet for sure.
Intel CPUs with Meltdown and L1TF fixed in hardware are already shipping. No doubt things like MDS will get fixes soon given Intel has been aware of these for almost a year now. It takes about year or year and a half to get fixed CPUs into mass production and out on shelves if this is done in a hurry.
This is an old roadmap slide. Newer one shows 10nm quickly replaced with 10+nm and then with 7nm less than a year after. Product roadmap at the same time only has 10nm CPUs for mobile by end of this year and for servers early next year. Desktop does not have anything at 10nm on roadmap.
There is about 6% from faster clocks, making the increase by IPC raise 19%. That is feasible enough even if it would mean CPU-Z bench is pretty much the best case scenario for Zen2 (benefitting directly from massive L3 cache maybe).
Intel needs PERFORMANCE on 10NM, performance on nodes means frequency!
Skylake, CFL and sunny cove is 14NM , 10NM = low clocks hence the reason why they will be on 14NM for desktop because performance is LOWER for 10nm!.
On the bright side for Intel!
I see no issues with Intel with 10nm for laptops, quite the contrary if they can get the chips out the door and I see no way amd's design will create big wins in the 7-25W space Intel wins, 25->45W Amd wins.

I may be mistaken by amd on low power laptops.

Bottom Line, Intel will struggle till 2021 on the cpu side but carried by low power cpu's and datacenter for AVX512 implementations, but I'm optimistic about their gpu and I hope they are back in 2021 with promising new designs :)
Posted on Reply
#33
londiste
medi01, post: 4065931, member: 158537"
I wonder what is used as a base here.
Skylake.

Imsochobo, post: 4065941, member: 66457"
Intel needs PERFORMANCE on 10NM, performance on nodes means frequency!
Skylake, CFL and sunny cove is 14NM , 10NM = low clocks hence the reason why they will be on 14NM for desktop because performance is LOWER for 10nm!.
On the bright side for Intel!
Intel has a slide where they admitted 10nm does not clock as high as 14nm(++) does. Their approach to that problem is two-fold. This shortcoming is one of the big reasons for skipping desktop SKUs at 10nm for now. The other thing is architectural changes for increased IPC in Sunny Cove cores - larger caches, improved frontend etc. I am not even sure 10nm on desktop would have a noticeable performance problem, the problem is much more likely in the size and yield side of Intel's 10nm chips.

This is not that different from what AMD is doing by the way. AMD does not have the frequency problem as Zen - or rather, GF/TSMC 14nm/12nm process - never clocked that high. At the same time Zen2 architectural changes have a lot in common with Sunny Cove changes.
Posted on Reply
#34
Xaled
Manu_PT, post: 4065937, member: 168799"
This is so funny!!

AMD shows the most biased and misleading benchmarks -> "yes!! Intel is done!! AMD will conquer the world!!"

Intel has leaked benchmarks where 5,2 ghz = the new 3,6ghz, wich is amazimg -> "lol intel. Lol this is fake. Lol only 18% improvement. Lol this is never coming to desktop"

Seriously you guys are annoying. If this IPC improvements are true, oh boy, they gonna rip AMD apart on mobile and they will on desktop once they cam do it. I am for sure gonna be waiting.
Yes there is a huge difference between "Showing" official slide and "leaking". İf these leaks are true it'd be amazing indeed. But if not would Intel be punished? No the Chinese leaker would be blamed
Posted on Reply
#35
londiste
Xzibit, post: 4065929, member: 105152"
Intel investor day was May 8, 2019. Your confusing Intel with the Dell Leak Roadmap.
This slide was also in investor meeting (via Anandtech):

as well as this one:


10nm++ will exist but effectively will be overcome by 7nm in 2021. The second slide's "10nm Client Systems on Shelf for 2019 Holiday Season" matches Dell leak's mobile 10nm - and there was a passing mention of mobile in Intel's meeting - as well as servers in 1H'20 matches what I said above (and Dell's leak).

Whether Intel will be able to make their roadmap a reality is a separate discussion but they are quite clear about their plans.
Posted on Reply
#36
Vya Domus
Xzibit, post: 4065908, member: 105152"
Intel has other plans. At least thats what it told its investors this year.



Intel will be using Arizona and Ireland for 7nm. Expansion at those fabs is expected to be completed late 2021 for Arizona, Ireland est sometime in 2022.


I too love a graph with no numbers marked on the y axis. This is hilariously bad.
Posted on Reply
#37
HwGeek
Indeed, since we know that new 10nm for mobile at 15W can only clock upto 3.9~4.1Ghz while current 8th gen Wiskey 15W U models boosts upto 4.8Ghz, even with IPC uplift it's doesn't look so great.
Posted on Reply
#38
medi01
londiste, post: 4065942, member: 169790"
Skylake.
With or without security patches.
Posted on Reply
#39
londiste
medi01, post: 4065968, member: 158537"
With or without security patches.
Without.
Keep in mind though that both compared CPUs are Intel's and security patches' performance hit is definitely bigger on Skylake with mitigations enabled.
londiste, post: 4065874, member: 169790"
Yeah, Intel's testing without security mitigations is shady.

On the other hand, these new cores definitely have fixes for Meltdown/L1TF in hardware and at least hardware-assisted mitigations for other Spectre-likes. And they are comparing these against Skylake which has none of that. When this is tested without security mitigations, Skylake should get a bigger boost than Ice Lake, making the uplift from new cores even bigger when taking security mitigations into account.
Posted on Reply
#40
InVasMani
stimpy88, post: 4065899, member: 178509"
Nothing more than Intel (unofficially of course) trying to steal thunder from AMD. Especially when you take in to account the complete butchering of Intel CPU performance over the last DECADE. You would need a lot more than a 40% improvement to make up for the loss of Hyperthreading, as well as the Spectre and Meltdown "fixes".

Some Intel CPU's will have lost more than 50% of their performance under some circumstances, so a poultry 18% (average) improvement on average is simply not enough.

Too little, too late Intel.
Fixed for clarity.
Posted on Reply
#41
Octopuss
Ice Lake (microarchitecture)
From Wikipedia, the free encyclopedia
Ice Lake is Intel's codename for the 10th generation Intel Core processors based on the new Sunny Cove microarchitecture.

WTF????

someone explain this shit to me, because I am completely perplexed.
Posted on Reply
#42
ZoneDymo
Manu_PT, post: 4065937, member: 168799"
This is so funny!!

AMD shows the most biased and misleading benchmarks -> "yes!! Intel is done!! AMD will conquer the world!!"

Intel has leaked benchmarks where 5,2 ghz = the new 3,6ghz, wich is amazimg -> "lol intel. Lol this is fake. Lol only 18% improvement. Lol this is never coming to desktop"

Seriously you guys are annoying. If this IPC improvements are true, oh boy, they gonna rip AMD apart on mobile and they will on desktop once they cam do it. I am for sure gonna be waiting.
yes it is amazing.... hence fake.... and what misleading/biased benchmarks has AMD shown (also biased...can anything be biased when its representing itself...what?)
Salty Intel fanboy is salty seemingly.

also if nothing else, I would think you would be pissed off knowing Intel was comfortably sitting on that "5.2 - 3.6 ghz" improvement but the bosses saying "no dont push that out now, spread if over many years of full price products, more money for us".
Its would be only now that Intel is being pushed by some competition that they get off their lazy greedy selfish behinds.
Posted on Reply
#43
Manu_PT
ZoneDymo, post: 4065989, member: 66089"
yes it is amazing.... hence fake.... and what misleading/biased benchmarks has AMD shown (also biased...can anything be biased when its representing itself...what?)
Salty Intel fanboy is salty seemingly.

also if nothing else, I would think you would be pissed off knowing Intel was comfortably sitting on that "5.2 - 3.6 ghz" improvement but the bosses saying "no dont push that out now, spread if over many years of full price products, more money for us".
Its would be only now that Intel is being pushed by some competition that they get off their lazy greedy selfish behinds.
Sure, only AMD leaks are accurate. Ive seen the picture now :)

As for the question about AMd misleading benchmarks, ask Steve Burke from GamersNexus.
Posted on Reply
#44
londiste
Octopuss, post: 4065980, member: 74316"
Ice Lake (microarchitecture)
From Wikipedia, the free encyclopedia
Ice Lake is Intel's codename for the 10th generation Intel Core processors based on the new Sunny Cove microarchitecture.
WTF????
someone explain this shit to me, because I am completely perplexed.
Ice Lake is CPU, Sunny Cove is the core.
Posted on Reply
#45
Bwaze
Octopuss, post: 4065980, member: 74316"
Ice Lake (microarchitecture)
From Wikipedia, the free encyclopedia
Ice Lake is Intel's codename for the 10th generation Intel Core processors based on the new Sunny Cove microarchitecture.

WTF????

someone explain this shit to me, because I am completely perplexed.
You can look at it this way. Sunny Cove is the core architecture. Ice Lake denotes architecture (Sunny Cove) + process (10 nm).
Posted on Reply
#46
Enterprise24
Fabio, post: 4065935, member: 175032"
my 8700k has a 565 single core score for cpu z, now at 5,2 magically rise at 635?
635 for 5.2 seem legit. Here is mine at 5.4

Posted on Reply
#47
ArchStupid
Imsochobo, post: 4065884, member: 66457"
7NM doesn't really help Intel with what they're struggling with, Performance.
It should be worse in fact...
Intel is struggling with performance..?
Posted on Reply
#48
ZoneDymo
Manu_PT, post: 4065991, member: 168799"
Sure, only AMD leaks are accurate. Ive seen the picture now :)

As for the question about AMd misleading benchmarks, ask Steve Burke from GamersNexus.
No Im asking you, you brought it up, you made the claim so the burden of proof lies with you Im afraid.

And no, AMD leaks are pretty much never accurate, thats the reason why this is probably fake as well...
In fact Im kinda convinced AMD leaks are done by anti AMD people to get people hyped up about improbable products so they are disappointed when the final products dont live up to that hype, hence attempting to damage the image.

For example the 5ghz claim, or that Ryzen 3000 would all be one tier difference, like this nonsense: https://eteknix-eteknixltd.netdna-ssl.com/wp-content/uploads/2019/03/AMD-Ryzen-3000-Singapore-Leak.jpg
Posted on Reply
#49
Frick
Fishfaced Nincompoop
Manu_PT, post: 4065937, member: 168799"
This is so funny!!

AMD shows the most biased and misleading benchmarks -> "yes!! Intel is done!! AMD will conquer the world!!"

Intel has leaked benchmarks where 5,2 ghz = the new 3,6ghz, wich is amazimg -> "lol intel. Lol this is fake. Lol only 18% improvement. Lol this is never coming to desktop"

Seriously you guys are annoying. If this IPC improvements are true, oh boy, they gonna rip AMD apart on mobile and they will on desktop once they cam do it. I am for sure gonna be waiting.
The rhethoric is the same from both, ugh, "sides". Entertaining in some ways, but sad when you realize it's just how the world works now.
Posted on Reply
#50
Aquinus
Resident Wat-man
I need to see it to believe it. Let me guess; This is before all of the security vulnerability mitigations compared to a CPU with them enabled? :laugh:
Posted on Reply
Add your own comment