Thursday, August 19th 2021

Intel's Secret Sauce at Catching Up with AMD Core Count is the Gracemont E-core and its Mind-boggling Perf/Watt

When early benchmarks of the Core i9-12900K "Alder Lake-S" processor showing performance comparable to AMD's top 16-core Ryzen 9 5950X surfaced, we knew something was up. 8 Intel P-cores and 8 E-cores, are able to match 16 "Zen 3" cores that are all performance cores. Apparently Intel is able to turn its P-core deficit around by taking a wacky approach. First, the 8 "Golden Cove" P-cores themselves offer significantly higher IPC than "Zen 3." Second, the 8 "Gracemont" E-cores aren't as "slow" as conventional wisdom would suggest.

Intel in its Architecture Day presentation put out some astounding numbers that help support how 8 big + 8 little cores are able to perform in the league of 16 AMD big cores. Apparently, on "Alder Lake-S," the 8 "Gracemont" E-cores enjoy a lavish power budget, and are able to strike an incredible performance/Watt sweet-spot. Intel claims that the "Gracemont" E-core offers 40% higher performance at ISO power than a "Skylake" core (Intel's workhorse P-core for desktops until as recently as 2020); which means it consumes 40% less power at comparable performance.
A "Gracemont" core hence doesn't end up too far behind "Skylake." The combination of high-IPC P-cores and "fairly fast" E-cores are hence able to attain performance levels comparable to 16 "Zen 3" cores. There are some limitations, though. For starters, "Gracemont" cores don't support HyperThreading, unlike "Skylake," and have a reduced ISA instruction-set compared to the P-cores.
Add your own comment

49 Comments on Intel's Secret Sauce at Catching Up with AMD Core Count is the Gracemont E-core and its Mind-boggling Perf/Watt

#1
Vayra86
If this translates to IRL, we're actually looking at a feasible big.little architecture that CAN extract an advantage from its design. At some point that's what I was hoping for when they did launch this idea in the beginning. Its the only way they can somehow make it tick better than what they have.

If, because how workload dependant is this :) Its interesting nonetheless.
Posted on Reply
#2
BorisDG
Today is probably the day with most Intel news... LOL (7 total)
Posted on Reply
#3
mtcn77
This is great news. For a decade Intel has been stubborn as a goat and as a consumer had me at the end of my patience. If Intel did this before sinking billions into their mobile foray we would still be talking about competition in the mobile sector. I'm sick and tired of their ex-colleagues who go on record saying x86 doesn't have to be uncompetitive - you are the one to make it happen!
Well done and I hope they stop their old habits which die hard. Intel can renovate itself they hired Jim Keller for heaven's sake - they better develop system level changes! Intel with all its might does not need to be the slow adopter!
Posted on Reply
#4
Selaya
Vayra86If this translates to IRL, we're actually looking at a feasible big.little architecture that CAN extract an advantage from its design. At some point that's what I was hoping for when they did launch this idea in the beginning. Its the only way they can somehow make it tick better than what they have.

If, because how workload dependant is this :) Its interesting nonetheless.
Too bad it can't run half the code at all bc it lacks AVX. Oops.
Posted on Reply
#5
Vayra86
SelayaToo bad it can't run half the code at all bc it lacks AVX. Oops.
W... what?
Neither core has AVX? Thats.... special.

Here's what I read...
"Gracemont" is Intel's first low-power core to support AVX, AVX2, and AVX-VNNI instruction sets. "Golden Cove" features a more lavish ISA that includes AVX-512 (select client-relevant instructions)."
Posted on Reply
#6
mtcn77
SelayaToo bad it can't run half the code at all bc it lacks AVX. Oops.
You have to understand those huge MUX'es create huge voltage fluctuations. You cannot have power efficiency if you insist on running wide execution units. Design follows function for a reason. All the researchers in the world wouldn't be able to solve an already 'known issue' unless you want to invest into more expensive motherboards which aren't without their own issues.
PS: Nvidia didn't beat 3dfx by featuring the same supersampling, Intel didn't beat IBM with the same level of performance...
Posted on Reply
#7
Selaya
Vayra86W... what?
Neither core has AVX? Thats.... special.

Here's what I read...
"Gracemont" is Intel's first low-power core to support AVX, AVX2, and AVX-VNNI instruction sets. "Golden Cove" features a more lavish ISA that includes AVX-512 (select client-relevant instructions)."
Huh, interesting.
Posted on Reply
#8
windwhirl
[MEDIA=twitter]1428383565981618180[/MEDIA]
AVX-512 seems to be completely disabled on Alder Lake
Not too much of a loss if anyone asks me.
Posted on Reply
#9
zlobby
With AVX512, intel's biggest 'invention'?
Posted on Reply
#10
windwhirl
zlobbyWith AVX512, intel's biggest 'invention'?
No AVX-512 at all
Posted on Reply
#11
TheoneandonlyMrK
These PR piece's are getting stupid, if Intel are aiming they're next generation at against a part that's about to get a +25%(zen3+cache) performance boost near EOL and not even the competition they will face,(Zen4), I can't see the win they can.

For me it's got too damn many little cores, I mean who needs 16 small better than Skylake core's to run the backend and wants just 8 big cores for in focus gaming?!.

Surely 4/8 small OS cores and 16 Big cores is what I might have bought into not the vice versa version they're peddling.
Posted on Reply
#12
Tsukiyomi91
not including AVX in a new processor lineup is something Intel never done before. Perhaps with Gelsinger at the helm, these are perhaps the changes we're seeing?
Posted on Reply
#13
mtcn77
Tsukiyomi91not including AVX in a new processor lineup is something Intel never done before. Perhaps with Gelsinger at the helm, these are perhaps the changes we're seeing?
Jim Keller is a corporate now - what else were we expecting? The guy is the systems engineer. He will definitely trim Intel's architecture until he can have what he wants: more power!
PS: I have full confidence in him turning the power grab away from the marketing team(Product Assurance HR). Where in the world does stupid HR people know better than the engineers?
Posted on Reply
#14
Kelutrel
Someone should tell'em that I have a 5950X in my chassis now, while Alder Lake is just a tiny timid promise of a possible future...
Posted on Reply
#15
Mistral
Well, if an Intel presentation says so it's probably safe to start getting excited about it...
Posted on Reply
#16
windwhirl
Tsukiyomi91not including AVX in a new processor lineup is something Intel never done before. Perhaps with Gelsinger at the helm, these are perhaps the changes we're seeing?
AVX-512 is not included. I understand AVX and AVX2 are still included in both core types
Posted on Reply
#17
DeathtoGnomes
Take this with a grain of salt, wait for reviews.
Posted on Reply
#18
P4-630
Is a TPM in these CPU's enough for windows 11 hardware requirements or do you need to look for specific TPM motherboards as well?
Posted on Reply
#19
tabascosauz
SelayaToo bad it can't run half the code at all bc it lacks AVX. Oops.
zlobbyWith AVX512, intel's biggest 'invention'?
They removed AVX512 (hardware physically still there for use in Sapphire Rapids) from the P-cores, and gave the E-cores AVX. Although, not sure what they are saying when they claim "Haswell-level AVX support" with "Skylake-level AVX performance". Though to be honest I don't think the thermal/power/clock cost of AVX2 has changed that much since Haswell.

And apparently they migrated the only actually marketed component of 512 to AVX2 (VNNI). Seems like a big win, obvious AVX512 memeing aside due to Intel marketing, don't see how this is a bad thing.

That said, comparing Gracemont to Coffee Lake/Comet Lake seems a bit sus, yes it's going to be a big jump in efficiency for Intel but Skylake was only relevant for so long because it was pushed to 5-5.3GHz. Yeah Gracemont is nice and efficient, but it runs at what, 3.0GHz? Not sure how they're expected to make up the MT perf gap with AMD, push Golden Cove even harder with the power budget freed from Gracemont? :confused:
Posted on Reply
#20
TheoneandonlyMrK
Catching up with core count when?!.

I can throw 32 core's in my age's old pc now and they're offering a revolutionary 24?!, FFS is they're PR team on crack.
Anyone can buy 128 thread CPU today too so yeh go Intel.

If I were AMD I'd just stack 8 arm core's on and work some driver magic and have any arm x86 app run on the most effective core type, have at apple and co while they're at it, those arm core's could run back end tat emulated.
Posted on Reply
#21
TheinsanegamerN
If the gracemont cores are so amazingly efficient and perform this well why on earth does intel even bother with the "big" cores anyway? Just make the gracemont cores the main cores and give them more cache so they can keep up with AMD.
Posted on Reply
#22
Darmok N Jalad
On that middle slide, it's a 4C/4T Gracemont against a 2C/4T Skylake. So 8 Gracemont cores is going to be like a 10-series i3?
Posted on Reply
#23
Dr. Dro
My concern with all of these changes is that Windows is too much of a dinosaur to be able to make use of all this tech. But if this is the x86 processor of the future, I think i'm satisfied. Will be eagerly awaiting reviews!
Posted on Reply
#24
HisDivineOrder
Want competition in the PC space, but I'll believe the claims only after I see the benchmarks.
Posted on Reply
#25
Chrispy_
Tremont/Gracemont architecture really ain't bad at all but the 40% power or performance comparison is BS because HT provides 30-40% improved numbers for Skylake and yet these comparisons are horribly handicapping Skylake to 1C1T

Gracemont doesn't need marketing lies and deception, it's a frickin' amazing bit of ultra-dense computing power that will never beat Skylake in a fair fight and isn't supposed to. It's supposed to run all the background crap at low power consumption and free up the Raptor Cove cores to focus on the threads that need maximum performance. In my experience of Denverton servers, those Tremont cores feel very much like Core2 era processing power - except today they're running 16 cores at 1.5W per core, rather than the Core2 at 65W for just a couple of cores.

Seeing a rack stacked with half-U Supermicro Atom 16C/128GB mini servers each with dual 10GbE connectivity is bonkers. A single 42U rack holds enough compute, RAM, and network bandwidth to host VDI workstations for ~1000 people and yet the whole rack is short-depth and uses barely 2KW. If that doesn't impress you then I guess you're just not that into servers.
Posted on Reply
Add your own comment
Copyright © 2004-2021 www.techpowerup.com. All rights reserved.
All trademarks used are properties of their respective owners.