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

#1
TheLostSwede
Hang on a second there, didn't Intel say they only wanted to use real world benchmarks from now on?
That means this is against their own policy and clearly irrelevant, no?
Posted on Reply
#2
brian111
FWIW, Where I have seen this posted elsewhere the general consensus seems to be that this is not genuine.
Posted on Reply
#3
Crackong
Assume this graph is true.

Ryzen 3th gen gets 13 points every 100MHz
Intel 9th gen gets 15 points every 100MHz
This "super OMG Ryzen killer" SunnyCove only gets 9 points every 100MHz ?

Numbers don't add up.
Posted on Reply
#4
TheLostSwede
brian111, post: 4065860, member: 101827"
FWIW, Where I have seen this posted elsewhere the general consensus seems to be that this is not genuine.
The Intel slide or the Chinese "benchmark"?
Posted on Reply
#5
brian111
TheLostSwede, post: 4065862, member: 3382"
The Intel slide or the Chinese "benchmark"?
The Chinese part. I believe the Intel slide is from Computex.
Posted on Reply
#6
TheLostSwede
brian111, post: 4065863, member: 101827"
The Chinese part.
Well, they often aren't so...
Posted on Reply
#7
ZoneDymo
even if true, I still say we should let Intel sweat a little, make them work harder and make up for years of slacking.
Posted on Reply
#8
Ferrum Master
That CPU should not have any meltdown/spectre mitigations yet enabled in kernel, that benchmark is very sensitive to it.

Thus the performance uplift is really small.
Posted on Reply
#9
FordGT90Concept
"I go fast!1!11!1!"
18% over four years is pretty terrible, especially if you consider how much performance was lost because of security mitigations.
Posted on Reply
#10
londiste
Ferrum Master, post: 4065868, member: 90058"
That CPU should not have any meltdown/spectre mitigations yet enabled in kernel, that benchmark is very sensitive to it.
Thus the performance uplift is really small.
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.

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 don't know about that. On the other side 13-15% over 3 years is lauded as a great achievement.
Posted on Reply
#11
HwGeek
Heads Up- just posted on the News here in Israel:
"Intel to suppliers: The establishment of the new plant[10nm] is postponed.
As revealed in Calcalist, Intel summoned one of those involved in setting up the factory in Kiryat Gat, one after the other, and announced a delay of at least half a year.
Intel CEO Bob Swan: "We look forward to the new plant"

So whats going on? Intel CEO Bob Swan is currently visiting here, Maybe they have change of plans? going 7nm or other plan?
Posted on Reply
#12
lynx29
TheLostSwede, post: 4065859, member: 3382"
Hang on a second there, didn't Intel say they only wanted to use real world benchmarks from now on?
That means this is against their own policy and clearly irrelevant, no?
What does a random person in China have to do with this? This isn't Intel... it's an engineering sample in the wild.

HwGeek, post: 4065877, member: 185585"
Heads Up- just posted on the News here in Israel:
"Intel to suppliers: The establishment of the new plant[10nm] is postponed.
As revealed in Calcalist, Intel summoned one of those involved in setting up the factory in Kiryat Gat, one after the other, and announced a delay of at least half a year.
Intel CEO Bob Swan: "We look forward to the new plant"

So whats going on? Intel CEO Bob Swan is currently visiting here, Maybe they have change of plans? going 7nm or other plan?
There have been rumors in the past that Intel will skip to 7nm for high end desktops. I hope it is true.
Posted on Reply
#14
Imsochobo
lynx29, post: 4065880, member: 153071"
What does a random person in China have to do with this? This isn't Intel... it's an engineering sample in the wild.



There have been rumors in the past that Intel will skip to 7nm for high end desktops. I hope it is true.
Not reply to above:
Remember when ryzen scored crazy in cpu-z ?... yeah, was a bug in the benchmark :)
I think Ice lake should be Highly competetive in laptops but it's ipc should be marginally better than zen 2.

Reply to above:
7NM doesn't really help Intel with what they're struggling with, Performance.
It should be worse in fact...
Posted on Reply
#16
HwGeek
How hard could it be to make icelake cpus current 14nm++ -Impossible?
Posted on Reply
#17
TheLostSwede
lynx29, post: 4065880, member: 153071"
What does a random person in China have to do with this? This isn't Intel... it's an engineering sample in the wild.
I have no idea, but your reading comprehension clearly needs to improve.
The first image is from an Intel presentation, using only synthetic benchmarks, whereas when AMD used them during their presentation at Computex, Intel went out and said that from now, we should only use real world benchmarks. Yet Intel clearly seems more than happy to use synthetic benchmarks when it suits them. As such, this is irrelevant even by Intel's "new" standards, no?
Posted on Reply
#18
Bwaze
"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."

Did Intel ever mention 10nm desktop in last 6 months? Dell leaks of roadmaps showed 14 nm desktop well into 2020. If there really will be Rocket Lake 14 nm desktop part with Willow Cove architecture in 2020, 10 nm Ice Lake after that would be a regression...
Posted on Reply
#19
HwGeek
If Zen 2.0 already got ~12% IPC advantage over current Intel parts, how do you think the Zen 3.0 gonna perform? another 5% IPC + 10% performance at same TDP[Higher All core boost]? this make very big gap for 2020, so maybe Intel is reconsidering it's timeline?
Posted on Reply
#20
stimpy88
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 year. 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.
Posted on Reply
#21
TheLostSwede
stimpy88, post: 4065899, member: 178509"
so a poultry
I think you mean paltry, not
but I could be wrong...
Posted on Reply
#22
FYFI13
Those numbers are questionable to me. Ryzen 2600X (4250MHz) - 480 points in CPU-Z test, Ryzen 3600X (4500MHz) - 609 points. That's ~28% increase. I wish it was truth though..
Posted on Reply
#23
Xzibit
lynx29, post: 4065880, member: 153071"
There have been rumors in the past that Intel will skip to 7nm for high end desktops. I hope it is true.
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.

Posted on Reply
#24
phanbuey
Xzibit, post: 4065908, member: 105152"
Intel has other plans. At least thats what it told its investors this year.


oof if that's true...
Posted on Reply
#25
londiste
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 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.
stimpy88, post: 4065899, member: 178509"
Especially when you take in to account the complete butchering of Intel CPU performance over the last year. 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.
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.
Xzibit, post: 4065908, member: 105152"
Intel has other plans. At least thats what it told its investors this year.

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.
FYFI13, post: 4065905, member: 105256"
Those numbers are questionable to me. Ryzen 2600X (4250MHz) - 480 points in CPU-Z test, Ryzen 3600X (4500MHz) - 609 points. That's ~28% increase. I wish it was truth though..
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).
Posted on Reply
Add your own comment