Monday, March 9th 2020

Rumor: Intel to Introduce Big.Little Architecture for Desktop With Alder Lake-S, New LGA 1700 Socket

Hold on to your helmets: a wild rumor that Intel may be looking to introduce the same design considerations as they already did with their Lakefield architecture has appeared. According to momomo via Twitter (a user who has already shared many rumors and details in the PC hardware space) as well as some other sources, Intel is looking to bring a Big.Little-like design (which Intel calls Hybrid architecture) to the desktop platform in the form of Alder Lake-S, to be reportedly built on the 10 nm process. While Intel's Lakefield (especially geared for the mobile market) only sported four Atom (Intel's low power) Tremont cores combined with one high-performance Sunny Cove core, Alder Lake-S could sport as many as an 8+8 configuration, with a TDP currently set up to 80 W (and up to 125 W TDP is also set in the revealing slides with a disclosure regarding investigating performance scaling in up to 150 W TDP).

Should this actual Alder Lake-S product materialize in the 10 nm process, this could be a way for Intel to salvage what it can from the 10 nm process for the desktop platform. As we know from multiple reports on the state of Intel's 10 nm, yields and operating frequencies aren't close to what was expected, and Intel's CFO George Davis even said at last week's Morgan Stanley's Analyst Conference that their 10 nm process wouldn't be as profitable as even 22 nm, which does show that Intel is already looking past this process for their 7 nm deployment. A Big.Little design for a desktop architecture does seem like a more plausible design decision for a struggling process than a full 16-core monolithic die such as those Intel currently employs.
Intel Alder Lake S Lineup Intel CPU Roadmap
The leaked slide also points to a new socket, LGA 1700, which would supersede the LGA 1200 that's being deployed with Comet Lake-S (10th gen) and which could feature support for Intel's Rocket Lake-S family (11th Gen). The leak also plays out a possible PCIe 4.0 support from Alder Lake-S, which could mean this is the first Intel architecture to sport this updated protocol, should it not debut with Rocket Lake-S already. Slightly logic, rational leaps mean that Intel could be looking at leveraging their Golden Cove (high performance) and Gracemont (Atom) CPU cores for this hybrid design.
Sources: via Tom's Hardware, momomo @ Twitter, Ghost Motley @ Twitter, via Videocardz, Anandtech
Add your own comment

68 Comments on Rumor: Intel to Introduce Big.Little Architecture for Desktop With Alder Lake-S, New LGA 1700 Socket

#51
ARF
Vayra86And here is the result of that: at least 20% more idle usage, in this example.
This is the system power consumption which includes mainboard, X570 chipset, SSD, GPU, and so on.
AMD's GPUs can idle at as low as 20-30 MHz. AMD has always been known for not working optimally with the frequencies, they just need time to adjust everything properly.
Example, HD 4890 didn't lower its memory clocks and they always run at 100%.
Posted on Reply
#52
Vayra86
ARFThis is the system power consumption which includes mainboard, X570 chipset, SSD, GPU, and so on.
AMD's GPUs can idle at as low as 20-30 MHz. AMD has always been known for not working optimally with the frequencies, they just need time to adjust everything properly.
Example, HD 4890 didn't lower its memory clocks and they always run at 100%.
We've arrived at our conclusion regardless. The idle consumption of a similar Zen system is higher. It includes the CPU, now, if all else is the same... hmmmmm. I can imagine for Intel the way to gain more profit there is to completely disable a cpu. An idle voltage is still a voltage.
Posted on Reply
#53
ARF
Vayra86We've arrived at our conclusion regardless. The idle consumption of a similar Zen system is higher. It includes the CPU, now, if all else is the same... hmmmmm. I can imagine for Intel the way to gain more profit there is to completely disable a cpu.
Nope, chipset X570 is 12 or 14nm and probably it causes problems.
Posted on Reply
#54
Vayra86
ARFNope, chipset X570 is 12 or 14nm and probably it causes problems.
If that is what you want to believe, who am I to stop you.

Some sources would be good though, instead of just hip shots.
Posted on Reply
#55
ShurikN
Don't really see the point of a big.little uArch on desktop. Mobile/laptop sure, all the way, but desktop...
Posted on Reply
#56
ARF
Vayra86If that is what you want to believe, who am I to stop you.

Some sources would be good though, instead of just hip shots.
You can see that the X570 power consumption starts from a much higher base, idling at 7.35W with no NVMe drive attached. The yellow highlighted bars show max power usage der8auer observed with various combinations of NVMe Gen 3 and SATA drives, and one included a PCI graphics card running the FurMark benchmark too. Finally the bottom two results, in red, show power consumption when there is a Corsair NVMe gen 4 SSD attached, and in/active. Der8auer is surprised by how little difference there is between NVMe Gen 3 and 4 under load in terms of power consumption "because everybody kept telling us (at Computex) that the chipset has much higher power consumption because of PCI Express Gen 4." In summary the OC expert couldn't pin down why the chipset requires so much extra power in actuality. Hopefully things will become clearer as boards are tested by more tech review sites like ourselves, it is still early days.
hexus.net/tech/news/mainboard/132515-der8auer-examines-amd-x570-chipset-power-consumption/
Posted on Reply
#57
Vayra86
ARFhexus.net/tech/news/mainboard/132515-der8auer-examines-amd-x570-chipset-power-consumption/
If you look at B350 vs X570, the difference is 4W in idle on a stock CPU. Still only accounts for half the gap, stock vs stock. OC'd, the AMD board will usually run more efficiently in idle - both X570 and B350.

Looks like its a bit of both, at the very least. Regardless, because we've drifted off topic a little... there is STILL a gain here because being able to shut the CPU off entirely is of course profit anyway, over leaving it at a low voltage. If we discard the board from the total system power equation, there is still a significant chunk of idle usage there from the CPU in both camps. I mean this is 50W idle we're talking about. The board takes 4-8W... on desktop.

Of course we also can't just directly translate these numbers to a laptop situation with ditto CPU, but if Zen is a single chiplet that cannot be shut off versus a BIG little setup where the biggest idle usage can be removed, it needs no explanation what will be preferable for battery life. No matter how low and well it can clock down. Most devices are in idle most of the time...
Posted on Reply
#58
londiste
Vayra86If we discard the board from the total system power equation, there is still a significant chunk of idle usage there from the CPU in both camps. I mean this is 50W idle we're talking about. The board takes 4-8W... on desktop.
I think you have it backwards. Idle CPU should take very little power. My R5-2400G consumes 4W or less at idle. Intel CPUs I have at hand are similarly in a few-watt range. Board, on the other hand, can consume more power with all the controllers and devices in addition to the chipset.
Posted on Reply
#59
Vayra86
londisteI think you have it backwards. Idle CPU should take very little power. My R5-2400G consumes 4W or less at idle. Intel CPUs I have at hand are similarly in a few-watt range. Board, on the other hand, can consume more power with all the controllers and devices in addition to the chipset.
So what then would be the point of a big little setup? I struggle to find one if its not reducing idle usage.
Posted on Reply
#60
londiste
Vayra86So what then would be the point of a big little setup? I struggle to find one if its not reducing idle usage.
Idle usage of an Atom core is definitely in sub-W area and lower. Couple of Atom cores can run at full blast - 2GHz-ish, probably - in 4W. This means idle-ish low performance situations that Atom core can handle will result in much lower power draw.
Posted on Reply
#61
Berfs1
Vya DomusThe impact SMT has is almost universally beneficial, it's so rare that the performance degrades it's not worth eliminating it. I mean that's the whole point, otherwise this feature would have been canned long ago.

The thing is, the 8 core processor with SMT will be less expensive to manufacturer and actually use less power than an 8+8 processor with no SMT. Big.Little makes no sense whatsoever on a desktop.

And if someone is so hung on not having SMT they can just disable it, no need for an entirely different processor.
Big.little is supposed to help efficiency at low loads and at idle, however having a 8c16t cpu will indeed take less power than a 16c16t processor. But yes, 8c16t is cheaper than 16c16t, however not only is 16c16t more secure, it also will have better single threaded performance (and actually will have slightly better multithreaded performance too). That's actually why the i7-9700K can beat the i9-9900K (stock configurations) in certain games, since they have similar clocks but the 9700K has HT disabled. (and yes, you can disable HT on the 9900K too)
Vayra86So what then would be the point of a big little setup? I struggle to find one if its not reducing idle usage.
Perhaps the high performance cores will shut off when the system detects it is idling, so that alone will save power, and the low power cores can be active when the system is idling
Posted on Reply
#62
ratirt
So it will be 8c16t + 8 small cores which performance ain't that great mainly to reduce power while not doing heavy tasks. I think that is the main idea of this. Why Intel shoves 8 small cores into a desktop processor is strange. The only thing that comes to my mind (reasonable explanation) is competing wit AMD with power efficiency? Why would you cripple desktop processor with this? Or maybe Intel can't make 16c processor with regular cores and this is the only way (because of yields? maybe?) Intel can make it if this is going to be a 10nm processor to compete somehow with AMD products.
Posted on Reply
#63
Octopuss
What the hell is big core and small core?
Posted on Reply
#64
trparky
Berfs1Big.little is supposed to help efficiency at low loads and at idle, however having a 8c16t cpu will indeed take less power than a 16c16t processor. But yes, 8c16t is cheaper than 16c16t, however not only is 16c16t more secure, it also will have better single threaded performance (and actually will have slightly better multithreaded performance too). That's actually why the i7-9700K can beat the i9-9900K (stock configurations) in certain games, since they have similar clocks but the 9700K has HT disabled. (and yes, you can disable HT on the 9900K too)
But why would we care about this kind of efficiency in a desktop? Don't get me wrong, in a notebook/mobile setting? Yes, most definitely. But a desktop? Not a chance! Most desktop users expect full power performance from our desktops, it's not like it's limited in power usage or cooling capacity.
Posted on Reply
#65
Assimilator
OctopussWhat the hell is big core and small core?
en.wikipedia.org/wiki/ARM_big.LITTLE
trparkyBut a desktop? Not a chance! Most desktop users expect full power performance from our desktops, it's not like it's limited in power usage or cooling capacity.
By the same "argument", the gas-guzzling cars of the 1960s should never have been superseded by more efficient models. *facepalm*
Posted on Reply
#66
Octopuss
Uh, so the CPU would have twice as many cores, but only half of them can be active at any given time, or?
Posted on Reply
#67
Berfs1
trparkyBut why would we care about this kind of efficiency in a desktop? Don't get me wrong, in a notebook/mobile setting? Yes, most definitely. But a desktop? Not a chance! Most desktop users expect full power performance from our desktops, it's not like it's limited in power usage or cooling capacity.
Because your electricity bill is going to deal with that? And you may not think it is a big deal, but it can also affect your house's power circuitry, if it can't handle multiple computers (which I know some people actually use multiple high performance computers), your whole house's circuit will fault and activate the breakers. That is why we don't want 300W CPUs. Because it is just too much for a mainstream CPU. Sure, for workstation CPU that is fine. But for a common household processor, that is too much. Sorry, but 300W is just too much from a mainstream processor running stock.
Posted on Reply
#68
ToxicTaZ
Intel LGA 1200 socket PCIe 4.0 (DMI 4.0)
With Intel Rocket Lake CPU support

Intel LGA 1700 socket PCIe 5.0 (DMI 5.0) "DDR5 USB4 WiFi-6E
With Intel Meteor Lake CPU support

Meteor Lake is Intel 7nm
Posted on Reply
Add your own comment
Apr 1st, 2025 18:56 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts