Question Alder Lake - Official Thread

Page 40 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Dave3000

Golden Member
Jan 10, 2011
1,343
91
91
I might upgrade from a Ryzen 5800x to a i7-12700k within the next 2 months. Does the BIOS in the Z690 motherboards have an option to disable the E-cores? I read that I need Windows 11 to have the optimized thread scheduler that distinguishes between the P-Cores and the E-Cores of the 12th gen CPU's that have E-cores but I want to stick with Windows 10 for now and that's why I want to know if I can disable the E-cores in the BIOS for the 12700k.
 

nicalandia

Diamond Member
Jan 10, 2019
3,330
5,281
136
I might upgrade from a Ryzen 5800x to a i7-12700k within the next 2 months. Does the BIOS in the Z690 motherboards have an option to disable the E-cores? I read that I need Windows 11 to have the optimized thread scheduler that distinguishes between the P-Cores and the E-Cores of the 12th gen CPU's that have E-cores but I want to stick with Windows 10 for now and that's why I want to know if I can disable the E-cores in the BIOS for the 12700k.
8C/16T Zen3 vs 8C/16T Golden Cove is a side grade at best for gaming if you have a Mid Tier and Up GPU and play at 1440P or better
 

clemsyn

Senior member
Aug 21, 2005
531
197
116
It wasn't really worth it. I've just scored 17,700 for a marginal increase in core clock by 100MHz all core which took temps from 60 to 66 deg C. Total package power at 132 watts.

I guess I'll just stick at stock on the CPU, tweak my RAM timings and go from there. I game at 1440p anyway, so with my 6700XT I'm GPU limited. A fun exercise nonetheless.

FYI my PSU is a RM650x and my cooler is a Corsair hi100 Elite, so I think I'm fairly well covered there.

This is what I basically decided. Left everything in stock speeds then undervolted it.
 

TheELF

Diamond Member
Dec 22, 2012
3,967
720
126
When every retailer was sold out of the 11600k about 45 days ago when I upgraded my son's computer, it was simply because Intel didn't want to sell any? o_O
This could also be the retailers not wanting to stock them anymore to empty their stocks for 12th gen, and or maybe to upsell people.

But yeah, intel does not control the whole world, they have very little influence on worldwide trafficking of goods and mining/creation of raw materials, if that stops intel is just as screwed as any other manufacturer, they may have larger stores of materials bunkered but that only helps you so long.
 

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
Part of the reason for slow uptake is probably also the continued expense and scarcity of high end dgpus, added to the already high platform costs for motherboards and ram.
And as others said, it is late to the party. I am glad to see Intel competitive or even in the lead in some cases, but it is not a compelling upgrade to Zen 3.
But the same thing doesn't stop other CPUs (Intel and AMD both) from selling well.
 

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
If you look at Anandtech's forum users over the last 20+ years, you'll probably notice a lot of "budget" buyers that only maintain one or two systems on their own dollar and that only upgrade occasionally. Remember the QX6700? It was widely discussed, but only a few people bought it (opting for its less-expensive Q6600 sibling, among other chips).

Granted the QX6700 had superficial shortages at launch, leading to sites like NewEgg bidding up prices to absurd levels, so that also explains why only a few of us had one. The 12900k is pretty spendy as a platform, though, so there are only so many people with the cash to buy one. Perhaps there will be more Alder Lake-S adopters once B660 launches.
Or maybe a thread exclusively for elite members of our society?
 

TheELF

Diamond Member
Dec 22, 2012
3,967
720
126
Watercooling doesn't make the heat disappear though 😶😶😶
Using proper setting does though.
If you run both at suggested TDP, meaning 125w for intel and ppt of 142w for ryzen, the 12900k uses less power than the 5900x but more than the 5950x ,it has lower temps than both the 5900x and the 5950x because the big die helps a lot in transferring the heat to the cooler (less heat per surface) ,and in both efficiency as well as performance it is right in-between the 5900x and the 5950x.

Being able to use twice the power to beat the 5950x in a big part of benches WITHOUT EVEN OVERCLOCKING is just the cherry on top.

https://www.hardwareluxx.de/index.p...-desktop-cpus-alder-lake-im-test.html?start=8
grBsmGV.jpg
 

DrMrLordX

Lifer
Apr 27, 2000
21,582
10,785
136
Intel constantly ran short of CPUs in the 14nm era when silicon shortages weren't as bad as they are now.

They still managed to sell boatloads of HCC and XCC dice, among other things. They deprioritized things like dual core Pentiums and Celerons. Those were the chips that were impossible to get. The only high-end desktop products that they never produced en masse were the 9900KS and 8086K.

If you are looking at building a CPU for massively parallel workloads, (Which 16 P cores implies) then using a larger balance of E-cores makes more sense. They are both more Area efficient, and more Power efficient. This is exactly why they exist.

Intel's current heterogenous core implementation is slightly borked. 8+8 should already be driving efficiency, but it really isn't. Again, screwed up voltage planes, cores being pushed outside of the proper part of the v/f curve, etc. Alder Lake-S has the look of a weird, overclocked mobile CPU rather than a desktop CPU for parallel workloads. I'm not entirely certain that just adding more E cores would be the right solution unless Intel could fix the voltage problems and dial back the P core clocks in MT applications. And again, Amdahl's Law rears its ugly head (which would also be an issue for 16c Golden Cove, and is currently an issue for a 5950X).

16c Golden Cove with a dual ring layout and a better memory controller would have been a hell of a thing and probably wouldn't need to push more than 150W to dominate nearly every benchmark.
 
Last edited:
  • Like
Reactions: Space Tyrant

coercitiv

Diamond Member
Jan 24, 2014
6,151
11,686
136
What's the problem with that?! Is that going to use up 200w or is it going to make the cpu run like a celeron?!
You're inventing a reason, no enthusiast will go through the trouble of configuring a 200W+ 12900K build only to use JEDEC compliant memory settings, meaning the same warranty risks apply.
 
  • Haha
Reactions: lobz

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
Yes it's extremely obvious to anybody, both PBO and overclocking voids your warranty on your $500-800 cpus, lifting the power limit on an intel cpu doesn't.
I'm not sure why are you mistaking this thread with a reddit partisanship proving ground, but OK. I promise you I won't ever overclock my 9900K, I never had to. I did, however, _unlift_ the power limit the moment I've figured out why the hell is my living room and mostly around my desk always 5-6 °C warmer than the rest of the flat, even though the CPU temps are always reasonable (hail Noctua). A nicely baked cherry on top indeed 😊
 
  • Haha
Reactions: Thunder 57

insertcarehere

Senior member
Jan 17, 2013
639
607
136
Intel's current heterogenous core implementation is slightly borked. 8+8 should already be driving efficiency, but it really isn't.

It is driving efficiency. As per https://www.computerbase.de/2021-11...hnitt_wie_effizient_ist_die_hybridarchitektur, the 12900k is on average 32% faster with 8+8 vs 8+0 with no power limit, 125w/241w PL2, and 125w PL2 scenarios in the MT suite. For comparison, the uplift between 5800x to 5900x (8 to 12, more L3) is 36%, and the uplift between 5900x to 5950x (12 to 16) is 20%. So in the same power envelope, 8 gracemont ~= 4 GC in MT performance, which is a good tradeoff in terms of die area.

Again, screwed up voltage planes, cores being pushed outside of the proper part of the v/f curve, etc. Alder Lake-S has the look of a weird, overclocked mobile CPU rather than a desktop CPU for parallel workloads.

In some respects, Alder Lake-S is a weird, overclocked mobile CPU, it has the same amount of graphics EUs as Tiger Lake-H, and a very similar die size (190mm^2 vs 210mm^2), difference being that TGL-H isn't threatening 5950Xs no matter how much Watts is pumped into them. (preliminary ADL-P benchmarks look very promising themselves given the constraints, but that's another story)

Agree that ADL not having separate voltage planes is an oversight, but again, I don't see how that's such a fundamental, unfixable issue as to render the entire hybrid model a bad idea, especially since this is essentially Intel's first mainstream implementation of the hybrid model.

I'm not entirely certain that just adding more E cores would be the right solution unless Intel could fix the voltage problems and dial back the P core clocks in MT applications.
Adding more E-cores => lower P-core clocks in MT for a set performance target => lower voltages => better efficiency.
And again, Amdahl's Law rears its ugly head (which would also be an issue for 16c Golden Cove, and is currently an issue for a 5950X).

Amdahl's law is the exact reason a hybrid model does make sense, since speedups become more marginal the more cores/threads there are, it's only logical that the cores/threads responsible for the MT speedup be as area and power efficient as possible.

16c Golden Cove with a dual ring layout and a better memory controller would have been a hell of a thing and probably wouldn't need to push more than 150W to dominate nearly every benchmark.

16c Golden Cove would also be an entirely different die size. Look at ADL-S's floor plan, Alder Lake-S as-is is ~210mm^2, add in another Golden Cove cluster + a ringbus to take care of it and we're probably talking near ~300mm^2, again, die size isn't free.

And let's be very honest here, Intel wouldn't limit a 16P GC part to 150w, so even if such a hypothetical part is released everyone would still be talking about how much power it consumes and how "inefficient" that part is.
 
Last edited:

TheELF

Diamond Member
Dec 22, 2012
3,967
720
126
You're inventing a reason, no enthusiast will go through the trouble of configuring a 200W+ 12900K build only to use JEDEC compliant memory settings, meaning the same warranty risks apply.
An enthusiast like that will not use lifted power limits though, they will do a proper overclock that will use less power and reach higher clocks than just lifting limits.
The point is that the 12900k delivers very good performance at 125w and everything else is at the users discretion, you want to use 400w and o/c everything you can? Go ahead! You want to use lifted power limits and stay under warranty? You can do that as well.
But being able to do something and being forced to do something are not the same thing.
Bottom line, you don't need any water cooling for 125w and the 12900k works very well under 125w.
 
Jul 27, 2020
15,759
9,823
106
Intel Launches New, Israeli-Developed Alder Lake Gaming Processors (nocamels.com)
Meet Performance-Core - Architecture Day 2021 | Intel Technology - YouTube

I find it very interesting that whenever Intel is in a bind against AMD, their Israel R&D center saves them. This happened before too, with the Israeli developed Core architecture.

Intel's new chip design developed in Israel - ISRAEL21c
While Yona was developed in partnership with one of Intel’s California centers, the 65nm microprocessor product is the first to be developed in its entirety, both the architecture and strategy, by Intel engineers at its Israel plants in Haifa and Yakum.
The next-generation design has its roots in Intel’s Pentium M laptop chips, developed in Haifa. That suggests the Israeli chip designers have become the pre-eminent architects – over Intel’s engineers in Santa Clara and Oregon, – Roger Kay, an analyst at Endpoint Technologies, told The San Jose Mercury News.
 
  • Like
Reactions: Arkaign

SAAA

Senior member
May 14, 2014
541
126
116
Not my image, from a friend on Discord: View attachment 52912
First time I see a scaled picture of Palm cove, thank you.
That is basically a Skylake core, but wow Palm is huge compared to a Gracemont core, same IPC on paper!
Lacks what, hyperthreading and 5GHz potential overclock? For twice the area it isn't really worth it... guess adding small cores with raptor lake next is really a great idea, plus some golden cove iteration for P cores till a fresh architecture is ready, maybe the "royal" core line that was leaked.
 

Hulk

Diamond Member
Oct 9, 1999
4,191
1,975
136
Not my image, from a friend on Discord: View attachment 52912

Using the dimensions from this die shot I calculated the CB R23 points/mm^2. In the same amount of die space Gracemont is 28% more efficient than Golden Cove. This is including HT for Golden Cove and running cores at stock speeds. 4.9 for Golden Cove and 3.7 for Gracemont. Gracemont really packs quite a bit of compute into a small amount of die space. As Coercertiv as been telling us.

Furthermore, if all of the die are was used for Gracemont cores, approximately 8.47 Gracemont clusters would fit on the die with a resulting CB R23 score of 32,700.
 

Arkaign

Lifer
Oct 27, 2006
20,736
1,377
126
Using the dimensions from this die shot I calculated the CB R23 points/mm^2. In the same amount of die space Gracemont is 28% more efficient than Golden Cove. This is including HT for Golden Cove and running cores at stock speeds. 4.9 for Golden Cove and 3.7 for Gracemont. Gracemont really packs quite a bit of compute into a small amount of die space. As Coercertiv as been telling us.

Furthermore, if all of the die are was used for Gracemont cores, approximately 8.47 Gracemont clusters would fit on the die with a resulting CB R23 score of 32,700.

I haven't kept up with the layout of Aldy, how does the GPU area relate to all of this?

Also, how would a hypothetical GPUless Alder derivative look with an 8+16 layout (P+E)? Similar size?