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

#51
TheGuruStud
So fake that CNN and Fox should be reporting it.
Posted on Reply
#52
Nkd
Imsochobo, post: 4065884, member: 66457"
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...
This. It’s laughable when tech news sites forget that. Actually some people have already called this unrealistic. Otherwise intel would have been bragging left and right about it lol. Like they have been about everything.
Posted on Reply
#53
londiste
Nkd, post: 4066037, member: 42675"
This. It’s laughable when tech news sites forget that. Actually some people have already called this unrealistic. Otherwise intel would have been bragging left and right about it lol. Like they have been about everything.
So, what you are saying is that SPEC 2016, SPEC 2017, SYSmark 2014 SE, WebXPRT and Cinebench R15 are all bugged?
A big slide in their Computex presentation does not count as bragging?

For comparison, AMD had 2 numbers for Zen2 IPC increase for Computex - Cinebench (R20) 1T result was +13% and SPECint result that they eventually decided to use in the slide was +15%.
[MEDIA=twitter]1134098430614093827[/MEDIA]
Posted on Reply
#54
Vya Domus
londiste, post: 4066038, member: 169790"
So, what you are saying is that SPEC 2016, SPEC 2017, SYSmark 2014 SE, WebXPRT and Cinebench R15 are all bugged?
A big slide in their Computex presentation does not count as bragging?

For comparison, AMD had 2 numbers for Zen2 IPC increase for Computex - Cinebench (R20) 1T result was +13% and SPECint result that they eventually decided to use in the slide was +15%.
[MEDIA=twitter]1134098430614093827[/MEDIA]
But the benchmark used here isn't SPEC 2016, or 2017 or any of those. It's CPU-Z and it does not seem right to say the least, as I have said in another thread you can look at the changes in Sunny Cove, nothing there suggests 40% is possible, let alone achievable in real world scenarios. On the other had for Zen 2 13-15% seems at the very least possible and in accordance with what is known to have changed from Zen 1.
Posted on Reply
#55
londiste
Vya Domus, post: 4066056, member: 169281"
But the benchmark used here isn't SPEC 2016, or 2017 or any of those. It's CPU-Z and it does not seem right to say the least, as I have said in another thread you can look at the changes in Sunny Cove, nothing there suggests 40% is possible, let alone achievable in real world scenarios. On the other had for Zen 2 13-15% seems at the very least possible and in accordance with what is known to have changed from Zen 1.
The chinese list is probably bullshit.

40% does come from Intel's slide. Apparently there is one (part of) benchmark that indeed does get 40%. Benchmarks run are listed underneath and there is a bunch of results averaged (well, geomeaned) to get that 18%.

Why exactly does Zen2 13-15% (based on 2 benchmarks) sound more possible than whatever Intel is claiming? Both are first party benchmarks more likely than not designed to show their own CPU in best possible light, both results come from benchmarks. AMD's results seem to be from a subset of the same benchmarks that Intel also ran.
Posted on Reply
#56
Vya Domus
londiste, post: 4066057, member: 169790"
The chinese list is probably bullshit.
Then what's the point of even discussing this. He is right, if 40% IPC uplift was achievable in most scenarios it would have been plastered all over Intel's marketing material.
Posted on Reply
#57
londiste
Vya Domus, post: 4066059, member: 169281"
Then what's the point of even discussing this.
Because someone made a TPU news post out of this. Plus, the first half of the news post and first slide are based on information from Intel themselves.
Posted on Reply
#58
PuiuS
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.

I don't know about that. On the other side 13-15% over 3 years is lauded as a great achievement.
Zen launched in 2017 so that makes it 2 years since then and AMD is comparing Zen2 to Zen+, not 1st gen Zen. The problem with Intel is that they are claiming 18% IPC improvement (which may be true), but they lost 1GHz in clock speed.

In an interview, people from AMD said that they were expecting a regression in clock speeds when moving to 7nm (because of high leakage and other problems associated with new node so small), but they managed to avoid this and actually gain a few 100MHz. From what we've seen so far, Intel has yet to fix their 10nm node and are having trouble scaling the clock speeds and core counts.

The 40% number is most likely from a workload that makes heavy use of AVX512.
Posted on Reply
#59
dirtyferret
stimpy88, post: 4065899, member: 178509"
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.
TheLostSwede, post: 4065903, member: 3382"
I think you mean paltry, not
but I could be wrong...
Improving the size of poultry by 18% is no laughing matter, what has AMD ever done with domesticated fowl? Clearly AMD is trailing Intel in this department!
Posted on Reply
#60
badtaylorx
Why call it "per core performance" and not IPC???

That can NOT be a mistake!
Posted on Reply
#61
efikkan
I was fearing that this piece of "news" was going to be covered.

Now let's first check if it passes the smell test;
- The person making this have access to unreleased products from both AMD and Intel, including an early engineering sample of a 6 core Ice Lake, very unlikely.
- There is no CPU named "Sunny Cove", Sunny Cove is the core design, the first implementation is Ice Lake, so this is obviously wrong.
- Some of these clocks are fairly optimistic.
So most likely a fake.

And even if the benchmarks were accurate, >40% gain in a benchmark does not mean 40% higher "IPC". IPC should never be estimated based on a single workload, but a good selection of representative workloads. Even if Ice Lake truly yields 18% higher IPC, that doesn't mean every score will increase by 18% per clock, some may be only 5% while others are >40%, it all depends on which part of the CPU the benchmark stresses.

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.
~18% would be comparable to Sandy Bridge -> Skylake, which is no small achievement.
Performance loss from security mitigations are negligible for Skylake with later patches.

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.
7nm in large volumes will probably not ship until late 2021 or 2022.
Xeons based on Ice Lake-SP(10nm+) will ship in Q2 2020. HEDT have traditionally been based on the server platforms, and considering HEDT is low volume compared to mainstream, Intel could choose to release HEDT based on 10nm+, but nothing is confirmed so far.

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.<snip>

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.
Skylake have no problem with core speed vs. Zen 2. Zen 2's advantage will be energy efficiency and density allowing for more cores, which is where Intel will be limited by their 14nm node. Intel's 6- and 8-cores will consume more energy than their Zen 2 counterparts, but will have no issues competing in other regards. It's when you go past 8-10 cores that Intel will face problems, which means they will have no direct competitor to 12-/16-core "mainstream" parts.

Even if 10nm was less troubled, they would probably still struggle to achieve higher clock speeds. The long-term expectations is actually lower clock speeds for upcoming nodes, so future progress will depend on IPC gains.

My biggest complaint about Intel's situation is the lack of a proper backup plan. If they had only backported Synny Cove to 14nm, they would have had a good performance gain and lead for the next two years in the mainstream segment.
Posted on Reply
#62
stimpy88
TheLostSwede, post: 4065903, member: 3382"
I think you mean paltry, not
but I could be wrong...
It's fine the way it is... Spade a spade, chicken a chicken and all that...
Posted on Reply
#63
kings
Of course the 40% will be in a very specific workload, just like the 29% more IPC that AMD reported for Zen 2 in a presentation a few months ago.
Posted on Reply
#64
EarthDog
TheLostSwede, post: 4065893, member: 3382"
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?
Didnt Intel simply challenge amd in real world benchmarks? Like a 'hey those are great but...how about real world benchmarks'? I didnt know they specifically mentioned to only use real world benchmarks and this was the (only) way forward.


.....actually... that was "real world gaming". Intel was responding to AMD's 'gaming CPUs' and challenged them on that front.
https://www.techpowerup.com/forums/threads/intel-challenges-amd-to-beat-it-in-real-world-gaming.256393/
Posted on Reply
#65
OSdevr
This may well be genuine. I've suspected for a while that Intel has known all along how to drastically increase their IPC but hasn't done so because once they weren't able to go smaller they'd need some way to improve new CPUs.
Posted on Reply
#66
Vayra86
I'm sure there is an IPC boost in Ice Lake. Too bad it won't clock that well ey :) But 40%? Not in your baddest dreams, Intel.
Posted on Reply
#67
Vya Domus
OSdevr, post: 4066133, member: 170580"
This may well be genuine. I've suspected for a while that Intel has known all along how to drastically increase their IPC but hasn't done so because once they weren't able to go smaller they'd need some way to improve new CPUs.
I am not sure that's the case , new revamped architectures usually have to coincide with new nodes. I doubt Sunny Cove was in the pipeline for a lot of time.
Posted on Reply
#68
Metroid
Typical Intel, "Do not buy Ryzen, we have something better in few months".
Posted on Reply
#69
EarthDog
Metroid, post: 4066159, member: 178915"
Typical Intel, "Do not buy Ryzen, we have something better in few months".
Lol, what from 13 years ago when this last was the case? Intel didnt do this with zen or zen+ as there wasn't a need. Now that IPC is there AMD clockspeeds are up... they do.
Posted on Reply
#71
Mistral
Can we agree the "benchmark" is clearly fudged?
Posted on Reply
#72
TheLostSwede
EarthDog, post: 4066125, member: 79836"
Didnt Intel simply challenge amd in real world benchmarks? Like a 'hey those are great but...how about real world benchmarks'? I didnt know they specifically mentioned to only use real world benchmarks and this was the (only) way forward.


.....actually... that was "real world gaming". Intel was responding to AMD's 'gaming CPUs' and challenged them on that front.
https://www.techpowerup.com/forums/threads/intel-challenges-amd-to-beat-it-in-real-world-gaming.256393/
That was one thing, but before that, Intel did this
https://pcper.com/2019/05/intel-pre-computex-gen11-9900ks/
Posted on Reply
#73
Eric3988
I'll believe it when I see it. I refuse to count chickens before they hatch...
Posted on Reply
#75
OSdevr
Vya Domus, post: 4066155, member: 169281"
I am not sure that's the case , new revamped architectures usually have to coincide with new nodes.
No they don't. Intel's tick-tock scheme alternated between new architecture and new process.
Posted on Reply
Add your own comment