Question Intel Raptor Lake vs AMD Zen 4 vs Apple M2

[Carfax83]

These CPUs are all going to square off against each other at some point this year assuming nothing catastrophic occurs to delay any of the product launches. So going by what we know from official sources and informed rumor mongers (many of which were very accurate before Alder Lake and the M1 launched), which CPU do you think will win out in these categories?

1) Single threaded performance
2) Multithreaded performance
3) Gaming performance
4) Performance per watt
5) Overall performance (who wins the majority of applications)

While I've been keeping a close eye on rumors and leaks for Zen 4 and Raptor Lake, I have not admittedly been doing so for the M2; as I'm unrepentant Apple hater  At least I'm honest about it... That said, this is my ranking based on what I've seen and heard:

I think the single threaded crown will go to Raptor Lake, and I say this based on informed rumors that Raptor Lake will have up to 10% more IPC from microarchitectural updates, cache upgrades and higher clock speeds than Alder Lake. From what I've seen, gauging IPC performance isn't easy as it varies so much based on application, but I'd say Alder Lake already has at least a 15% across the board IPC advantage over Zen 3, so Raptor Lake could conceivably have 25% better IPC than Zen 3, which is similar to what Zen 4 will reportedly possess. But I doubt Zen 4 will match Raptor Lake in clock speeds and memory latency performance, which is why I'm predicting Raptor Lake will take the single threaded performance crown.

For multithreaded performance, Zen 4 should easily take it due to having more big cores than its Intel counterpart and similar IPC.

Gaming performance is more complicated because while some games are inherently more reliant on single core performance (strategy games for instance), more and more 3D engines are becoming increasingly parallel due to the adoption of Vulkan and DX12 in addition to modernized programming methods. Still, very few 3D engines can scale beyond 8 threads and 6 to 8 cores remains the sweet spot for gaming and will be for some time. So overall, I feel more comfortable going with Raptor Lake for the gaming crown. Also if rumors are correct, Raptor Lake will officially support DDR5-5600 off the bat while Zen 4 will reportedly use DDR5-5200. The raw memory speed won't likely be a significant factor, but Intel's memory controller will be right next to the CPU cores while Zen 4's will be in the I/O die which while still on the same package will definitely incur a significant latency penalty; which I'm sure will be offset by a massive L3 cache.

On performance per watt, one would think the M2 should take this category easily......but from the small amount of research that I've collected on it, it seems that there won't be much of a performance increase with the M2, if any. Some rumors are even suggesting there may be a bit of a regression in that aspect. Also since Zen 4 will be on TSMC's 5nm node, it will undoubtedly have excellent performance per watt and I believe it will also easily crush Apple's best in single core. So for performance per watt, I'm going to go with Zen 4.

When it comes to overall performance, I'm leaning towards Zen 4 but it will be close. Raptor Lake will supposedly double the amount of Gracemont efficiency cores which will certainly help in multithreaded performance per watt, but ultimately they won't be a match for Zen 4's 16 big cores with SMT. AMD will have the core count advantage and when that's combined with IPC parity with Raptor Lake, Zen 4 will win the majority of the benchmarks.

 

[mikk]

Minor correction, Meteor Lake will have updated IGP but it's still Xe LP (low power, integrated), not Xe HPG (high performance, discrete). I believe the specific flavor for Meteor Lake is Gen12.722, vs. Gen12.71 for Alchemist (Xe HPG) or Gen12.721 for Ponte Vecchio (Xe HPC).

You are wrong on this, Meteor Lake iGPU is based on Gen12.7 HPG. They might market it as Gen12LPG (because it's a low power version in the end) but the architecture is based upon Gen12.7 which is same as DG2. Xe LP (Gen12.1/Gen12.2) is a different architecture. Meteor Lake iGPU will be using a TSMC process, using the subpar Xe LP wouldn't make sense for them when Gen12.7 is ready roughly 1 year before Meteor Lake is launching.

 

[repoman27]

You are wrong on this, Meteor Lake iGPU is based on Gen12.7 HPG. They might market it as Gen12LPG (because it's a low power version in the end) but the architecture is based upon Gen12.7 which is same as DG2. Xe LP (Gen12.1/Gen12.2) is a different architecture. Meteor Lake iGPU will be using a TSMC process, using the subpar Xe LP wouldn't make sense for them when Gen12.7 is ready roughly 1 year before Meteor Lake is launching.

Pretty sure I'm not wrong. Integrated GPUs are low power by definition, and the generation numbers and what they correspond to have been in Intel drivers for over a year now:

https://www.reddit.com/r/intel/comments/jm3u58

 

[epsilon84]

Zen2 to Zen3 was done on the same process mind you (well obviously a tweaked version). Zen4 will benefit from both the jump to 5nm and the move to the new socket (new electrical specs). The info we have so far is AMD may increase the TDP on their flagship consumer desktop. To summarize the list of probable factors adding towards higher performance:

• Zen4 core will be considerably bigger in terms of transistor count
• AM5 max PPT will likely increase
• 5nm will offer better overall efficiency
• we already know the new core on 5nm can sustain 5Ghz all-core in moderate workloads (gaming), which means dynamic range is excellent (they're bound by power and/or thermals)

The point of all this is Zen4 will likely have the headroom to increase all-core clocks in heavy workloads vs Zen3, unlike RPL which will see stagnation at best. Last time AMD changed nodes and architecture (Zen1>Zen2) they claimed 15% better IPC and 25% better performance at ISO power. Add more power this time around, since I'm pretty sure they'll join the stupid power limit game, and gains of over 25% are very likely.

As you can see, the napkin math says the gains are quite evenly matched (as far as we can extrapolate obviously), in fact it pretty much requires Raptor Cove to be better than Golden.

Thanks for that summary, that makes a lot of sense. I think AMD will most likely play the power game as well to keep clocks high - there is no reason to cede a performance crown with a strict self imposed ~120W TDP when Intel is pushing 240W themselves. I still think 5.0GHz across 16 cores like some posters here are implying is rather optimistic, but time will tell

 

[mikk]

Pretty sure I'm not wrong. Integrated GPUs are low power by definition, and the generation numbers and what they correspond to have been in Intel drivers for over a year now:

https://www.reddit.com/r/intel/comments/jm3u58

The driver and other sources say Gen12.7 and pretty sure the second digit doesn't change the architecture. If it's a different achitecture they wouldn't just change the second digit, you can be sure that all Gen12.7 GPUs from Intel are not Xe LP based. Xe HPG2 is based on Gen12.9, this is a different architecture. You can also see that the driver leak from late 2021 lists DG2, MTL, ARL all as XE_HPG_CORE.

https://twitter.com/x/status/1471560689886052354

 

[Zucker2k]

Some people might like it, but Intel 'pre overclocking' everything kinda takes the fun out of well... overclocking. Maybe I'm just old school.

400W for 40k?

Yeah, Intel went nuts on the overvolting...

https://www.reddit.com/r/intel/comments/srok22

 

[coercitiv]

Yeah, Intel went nuts on the overvolting...

It's not overvolting when you're using the necessary voltage for the frequency. Alder Lake has a clearly defined voltage table and it can also adjust the requested VIDs based on AC/DC loadline information from the motherboard. It's an entire feedback loop aimed at making sure the CPU gets the lowest voltage possible based on the quality of the motherboard VRM.

Intel went nuts on the power limits, that's all. Engineering produced a nice piece of silicon tuned for 125-150W operation, marketing pooped on it for the headlines. If I wanted to write a guide for Alder Lake optimization I would literally write only this:

For 12900/12700(KF) set PL2 to 150W, for 12600K set PL2 to 125W.

 

[repoman27]

The driver and other sources say Gen12.7 and pretty sure the second digit doesn't change the architecture. If it's a different achitecture they wouldn't just change the second digit, you can be sure that all Gen12.7 GPUs from Intel are not Xe LP based. Xe HPG2 is based on Gen12.9, this is a different architecture. You can also see that the driver leak from late 2021 lists DG2, MTL, ARL all as XE_HPG_CORE.

https://twitter.com/x/status/1471560689886052354

All Intel Gen12 graphics are the same architecture, which is called Xe. Under that one architecture are 4 different microarchitectures: Xe LP, Xe HPG, Xe HP, and Xe HPC. Intel added Xe HPG to address the mid-range and enthusiast discrete segment, and has since mothballed Xe HP which was focused on data center / AI accelerators. So now have the following variants of Xe:

Gen12.1 = TGL = Tiger Lake IGP = Xe LP on Intel 10SF
Gen12.1 = DG1 = DG1 and SG1 low power dGPUs = Xe LP on Intel 10SF
Gen12.1 = RKL = Rocket Lake IGP = Xe LP backported to Intel 14+++
Gen12.2 = ADL/RPL = Alder Lake and Raptor Lake IGPs = Xe LP on Intel 7
Gen12.5 = ATS = Arctic Sound server dGPU = Xe HP on Intel 7
Gen12.71 = DG2 = Alchemist client dGPU = Xe HPG on TSMC N6
Gen12.721 = PVC = Ponte Vecchio OAM = Xe HPC on TSMC N5
Gen12.722 = MTL = Meteor Lake IGP = Xe LP on TSMC N3/N4

These are all first generation Xe (Xe1) architecture products, but you have four different microarchitectures, all manufactured on different processes, with Xe LP alone going through 3 revisions across 4 processes. Beyond that, we have:

Gen12.? = ARL = Arrow Lake IGP = Xe LP
Gen12.9 = LNL = Lunar Lake IGP = Xe LP
Gen13 = DG3 = Battlemage client dGPU = Xe2 HPG

edit: DG2 is essentially a slightly modified / client focused Arctic Sound ported to TSMC N6. Jupiter Sound was supposed to be the Gen13 Xe2 HP successor to Arctic Sound on Intel 4. However, Intel never commercialized ATS and seems to have bailed on JPS in favor of focusing on Xe HPG instead of Xe HP. DG3 now appears in drivers as "Elasti", so I'm not sure if Intel scrapped both JPS and the original DG3 in favor of a converged microarchitecture called Elasti, or if this is just DG3 referred to differently to indicate its new position filling multiple roles.

 

[mikk]

Gen12.722 = MTL = Meteor Lake IGP = Xe LP on TSMC N3/N4

These are all first generation Xe (Xe1) architecture products, but you have four different microarchitectures, all manufactured on different processes, with Xe LP alone going through 3 revisions across 4 processes. Beyond that, we have:

Gen12.? = ARL = Arrow Lake IGP = Xe LP
Gen12.9 = LNL = Lunar Lake IGP = Xe LP
Gen13 = DG3 = Battlemage client dGPU = Xe2 HPG

Once again, MTL iGPU is not based on Xe LP, it is based on Xe HPG (marketing might call it Xe LPG for the iGPU). Gen12_7 is based on Xe HP architecture. Gen 12_7 is not Xe LP, don't confuse it. One obvious main difference is that Xe LP has 96EUs per render slice whereas Xe HP has 64 EUs per render slice which is confirmed by Intel. Meteor Lake (per driver) has up to 3 slices with 64 EUs and overall up to 192 EUs (confirmed by Intel as well). Just on this you can see it is not based on Xe LP, it can't with 3 slices and 192 EUs.

This is straight from the test driver:

gen12LP_DG1
gen12LP_TGL
gen12LP_ADLP

Xe_HPG_Core_DG2
Xe_HPG_Core_MTL
Xe_HPG_Core_ARL

Furthermore Battlemage is not based on Gen13, it is based on Gen 12_9. Both Elasti and Lunar Lake will get Gen 12_9 graphics per driver. It's like RDNA1, RDNA2, RDNA3.

 

[Thunder 57]

It's not overvolting when you're using the necessary voltage for the frequency. Alder Lake has a clearly defined voltage table and it can also adjust the requested VIDs based on AC/DC loadline information from the motherboard. It's an entire feedback loop aimed at making sure the CPU gets the lowest voltage possible based on the quality of the motherboard VRM.

Intel went nuts on the power limits, that's all. Engineering produced a nice piece of silicon tuned for 125-150W operation, marketing pooped on it for the headlines. If I wanted to write a guide for Alder Lake optimization I would literally write only this:

I agree with what you are saying, but what dumbass is letting marketing make the decisions? I don't think people really respect marketing people as a whole. Someone more important has to be making the final say. I'm sure marketing gave that person a wonderful presentation with numbers and graphs showing how they could tie/beat Zen 3, and the engineers got overruled. And Intel wonders why they have lost their better engineers. There is another reason, but I won't go there.

 

[repoman27]

Once again, MTL iGPU is not based on Xe LP, it is based on Xe HPG (marketing might call it Xe LPG for the iGPU). Gen12_7 is based on Xe HP architecture. Gen 12_7 is not Xe LP, don't confuse it. One obvious main difference is that Xe LP has 96EUs per render slice whereas Xe HP has 64 EUs per render slice which is confirmed by Intel. Meteor Lake (per driver) has up to 3 slices with 64 EUs and overall up to 192 EUs (confirmed by Intel as well). Just on this you can see it is not based on Xe LP, it can't with 3 slices and 192 EUs.

This is straight from the test driver:

gen12LP_DG1
gen12LP_TGL
gen12LP_ADLP

Xe_HPG_Core_DG2
Xe_HPG_Core_MTL
Xe_HPG_Core_ARL

Furthermore Battlemage is not based on Gen13, it is based on Gen 12_9. Both Elasti and Lunar Lake will get Gen 12_9 graphics per driver. It's like RDNA1, RDNA2, RDNA3.

I'm not actually confident enough about this to argue the point any further without additional information.

Where do you see Elasti / Battlemage indicated as Gen12.9 and not Gen13?

 

[mikk]

I'm not actually confident enough about this to argue the point any further without additional information.

Where do you see Elasti / Battlemage indicated as Gen12.9 and not Gen13?

Disassembled from the test driver (and another source I won't share). You can see it from other people if you prefer.

Page 55 - News - Intel GPUs - Intel launches A580

Page 55 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

 

[Hulk]

Raptor Lake will be outclassed by AMD Zen 4. Unsure why people think that adding more cores will make a huge difference when power limits exist.

I'm not arguing either way but adding more cores does help MT when power limits are involved. Clocks (power) can be lowered while performance is increased.

 

[eek2121]

Thanks for that summary, that makes a lot of sense. I think AMD will most likely play the power game as well to keep clocks high - there is no reason to cede a performance crown with a strict self imposed ~120W TDP when Intel is pushing 240W themselves. I still think 5.0GHz across 16 cores like some posters here are implying is rather optimistic, but time will tell

I don’t think anyone here is claiming AMD will have 16 cores running at 5 ghz all core. AMD’s demo was likely done on an 8 core chip. I do suspect the 16 core version will get close.

The thing is, Zen 4 will be on a new node, they might use a higher power limit and/or tdp, the chip is rumored to be much larger (they wouldn’t throw away silicon for nothing), and we know it also clocks higher.

 

[repoman27]

Disassembled from the test driver (and another source I won't share). You can see it from other people if you prefer.

Page 55 - News - Intel GPUs - Intel launches A580

Page 55 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

Thanks, and from the same source I quoted from earlier, but considerably more recent. So the Xe2 branding does seem to be going to Gen12.9 and Xe3 to Gen13, while the LP nomenclature is being replaced with LPG starting with Meteor Lake. Arrow Lake is LPG Plus and it looks like ELG is a Gen12.9 part that was inserted into the line up either ahead or in lieu of DG3 and JPS. Which gives us:

Gen12.1 = TGL = Tiger Lake IGP = Xe LP on Intel 10SF
Gen12.1 = DG1 = DG1 and SG1 low power dGPUs = Xe LP on Intel 10SF
Gen12.1 = RKL = Rocket Lake IGP = Xe LP backported to Intel 14+++
Gen12.2 = ADL/RPL = Alder Lake and Raptor Lake IGPs = Xe LP on Intel 7
Gen12.5 = ATS = Arctic Sound server dGPU = Xe HP on Intel 7
Gen12.71 = DG2 = Alchemist client dGPU = Xe HPG on TSMC N6
Gen12.721 = PVC = Ponte Vecchio OAM = Xe HPC on TSMC N5
Gen12.722 = MTL = Meteor Lake IGP = Xe LPG on TSMC N3/N4
Gen12.? = ARL = Arrow Lake IGP = Xe LPG Plus
Gen12.9 = LNL = Lunar Lake IGP = Xe2 LPG
Gen12.9 = ELG = Battlemage client dGPU? = Xe2 HPG
Gen13 = JPS = Jupiter Sound server dGPU = canceled?
Gen13 = DG3 = Celestial client dGPU? = Xe3 HPG

But if you're parsing this like a sane person, Xe, Xe2, Xe3 is like RDNA, RDNA2, RDNA3, etc. There are some differences between the HP and HPG microarchitectures, and there may be something aside from just branding differentiating LP and LPG as well, but they're all still the same overall Xe architecture. The driver apparently groups ATS (Xe HP), DG2 (Xe HPG), MTL (Xe LPG), and ARL (Xe LPG Plus) all under the heading "gen12-hp", but these are clearly at least three different microarchitectures manufactured using at least three different processes. They are also targeting totally different applications and power levels.

In the end, MTL is a client platform with a typical integrated GPU that uses a low-power variant of Gen12 Xe with up to 192 EUs and a shared memory interface. I wouldn't expect any miracles here.

 

[epsilon84]

I don’t think anyone here is claiming AMD will have 16 cores running at 5 ghz all core. AMD’s demo was likely done on an 8 core chip. I do suspect the 16 core version will get close.

The thing is, Zen 4 will be on a new node, they might use a higher power limit and/or tdp, the chip is rumored to be much larger (they wouldn’t throw away silicon for nothing), and we know it also clocks higher.

Actually, you implied exactly that when you said that even a straight up Zen 3 port to N5 would yield +25% in MT. The only possible way that happens is with all core clocks at 5GHz since a 5950X hovers around 4GHz for all core loads.

I have no doubt that we will see higher clocks, better power efficiency etc. I'm just hesitant to declare an early winner like you did, since we have little concrete info on actual IPC increases and final clockspeeds, and we most likely won't get this until close to launch with leaked samples etc.

This applies to Raptor Lake as well, since that gets a major boost to cache size and additional voltage rails for the E cores could potentially help keep power levels in check. It's not exactly as simple as tacking on another 8E cluster to a 12900K and calling it a day.

 

[DrMrLordX]

Engineering produced a nice piece of silicon tuned for 125-150W operation, marketing pooped on it for the headlines.

If you look at the Phoronix/OpenBenchMark numbers that have been so widely-discussed on these forums recently, it's easy to see that the 12900k in particular dominates every benchmark where peak power draw never crosses ~90W.

I'm not arguing either way but adding more cores does help MT when power limits are involved. Clocks (power) can be lowered while performance is increased.

That's true. Adding E cores is not necessarily the best strategy, though. Their only real advantage is area efficiency.

 

[coercitiv]

If you look at the Phoronix/OpenBenchMark numbers that have been so widely-discussed on these forums recently, it's easy to see that the 12900k in particular dominates every benchmark where peak power draw never crosses ~90W.

That's no surprise to me, one of the best V/f points for 12700K uses around 75W in Cinebench, add 4 more E cores and offset the curve a bit due to better bin quality and you'll understand why 12900K thrives around 80-90W.

 

[DrMrLordX]

That's no surprise to me, one of the best V/f points for 12700K uses around 75W in Cinebench, add 4 more E cores and offset the curve a bit due to better bin quality and you'll understand why 12900K thrives around 80-90W.

I actually think that most of those benchmarks never touch the E cores. But that's just a working theory. I have no thread count data to prove it.

 

[gruffi]

I don’t think anyone here is claiming AMD will have 16 cores running at 5 ghz all core. AMD’s demo was likely done on an 8 core chip. I do suspect the 16 core version will get close.

I think it was 16 cores. Why else would they emphasize it was done on "all cores"? But of course it was just a game. I don't know how that games scales with more cores. But games usually cannot fully utilize all cores, especially 16 cores. So, it's not really comparable with a more CPU intensive scenario like Cinebench. OTOH we have the numbers from TSMC. N5 improves performance by 20% over N7. AMD is using an improved N5 process. Which means it might be even it little more than 20%. And AMD is increasing TDP by almost 15% (105W -> 120W). So, even if the increased IPC of Zen 4 costs some power per clock it should be enough to reach 5 GHz on all cores (or at least get very close). 5950X can clock ~4 GHz on all cores in CPU intensive apps.

 

[Carfax83]

Again, you're making statements about the core itself, while supporting those statements by using a comparison between a 6+4 core SKU and a 6+0 core SKU. If you want to speak about the strengths of Golden Cove, do so with examples of Zen 3 core vs Golden Cove core.

An apples to apples comparison would be something like the 5600G vs 12400, both 65 W, both having same boost clock/turbo. Fortunately PTS has the comparison that you would want i.e. compilation benchmarks.

Intel Core I5 12400 Linux Benchmarks Performance - OpenBenchmarking.org

openbenchmarking.org

You can see that the performance characteristics depends heavily on the code base that is being compiled. Now you can indeed say that Golden Cove is impressive against Zen 3 in compilation benchmarks.

OK I see your point now. I completely forgot about the 12400 and I didn't even know about the 5600G. They are definitely a better comparison point.

However, I still think the 12600K's efficiency cores aren't a significant factor in its performance in code compilation, so the 12600K vs 5600x is still a good juxtapose (especially as it has a bit more cache than the 12400), but just not as good as the 12400 vs 5600G.

 

[Carfax83]

So, even if the increased IPC of Zen 4 costs some power per clock it should be enough to reach 5 GHz on all cores (or at least get very close). 5950X can clock ~4 GHz on all cores in CPU intensive apps.

One of the things about Golden Cove that I find impressive, is how it just continues to scale with clock speeds. I did a search for Zen 3 frequency scaling and I didn't get much back at all, but from being on these forums over the years, it doesn't seem as though Zen 3 is a noteworthy overclocker. It can hit high clock speeds with proper tuning and cooling, but it seems to scale in performance poorly with increased frequencies so that it may not even be worth the effort.

I could be wrong though. If I am, I'm sure someone will correct me.

Anyway, Golden Cove by comparison is definitely a frequency monster and seems to thrive at high clock speeds because that is how Intel designed it to be. Raptor Cove will follow that same path too I have no doubt. So the question is, Zen 4 may be able to hit higher clock speeds, but will it benefit from doing so in a meaningful way?

Hitting high clock speeds and benefiting greatly from it isn't as trivial as some on this forum might believe. Certain forumites made that argument with the Apple A series CPUs, that if just gave it enough voltage it could hit 4ghz and above and blow any x86-64 CPU out of the water. But under greater scrutiny, these claims quickly fell apart. Even if the A15 CPU could be amped to 4ghz and greater, it doesn't guarantee that it's going to scale in performance with the higher frequencies.

 

[Zucker2k]

Alder Lake has a clearly defined voltage table and it can also adjust the requested VIDs based on AC/DC loadline information from the motherboard.

Could this be the culprit then? It makes no sense to me that even Intel marketing would push for so much power for so little gain after the 150w point. This is the common trend with all the top tier ADL chips. This bodes well for RPL since Intel has a lot of wiggle room to make it run much more efficiently than ADL. They'll need to with Zen 4 in the offing.

 

[coercitiv]

However, I still think the 12600K's efficiency cores aren't a significant factor in its performance in code compilation

Here's the 12600K vs 12400 isolated in the LLVM bench. I doubt that ~12% higher clocks and ~2MB of extra L3 cache can account for more than 25% better performance.



I agree with you that E cores are not a major contributor in this test, since this benchmark clearly cares about other bottlenecks as well, not just pure core count. They still have a contribution though, even if it's 10%.

 

[diediealldie]

Getting back to the thread topic though... I'm curious as to how so many are already declaring a victory for Zen 4 based on IPC and clockspeed gains alone? Whilst RPL is getting an additional 8 E cores that should add 25-30% more MT throughput vs the current 8+8 design. Zen 4 may well bridge the gap in ST and clockspeed, but Intel is also bridging the gap in thread count.

Looking back, Zen 1 -> Zen 2 - > Zen 3 all yielded less than 20% improvement (core for core) per iteration on average. That 20% factors in generational increases in clockspeed as well, if we are talking IPC gains it is probably closer to 10-15% gains per iteration.

Yet I'm hearing people are expecting a 40% improvement from Zen 4 over Zen 3?! Wishful thinking or am I not aware of some magic sauce that AMD is cooking up?

Also, Intel does not cut their cache memory for mobile lineups so the difference between RPL mobile and Zen 4 counterparts will be smaller. Zen 4 might extend AMD leadership for pseudo-HEDT classes (5950X which supports ECC memory) but a hoard of small cores is not really something one can simply ignore in many segments. Also, I expect Intel will retain their smaller power usage on idle even if their process node is inferior, thanks to P-E design and new features like DLVR.

 

[soresu]

Even if the A15 CPU could be amped to 4ghz and greater, it doesn't guarantee that it's going to scale in performance with the higher frequencies.

To say nothing of likely shredding its perf/watt and melting the chip without better thermal management😂