I moved to the new Apple MBP because Intel fan noise was driving me mad, but looking at that power consumption and my recent electric bill I am VERY thankful I moved. Most programs would not notice a 4% bump anyway, and those that do will be the ones drinking electricity.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Benchmarks Confirm Intel's Latest Core i9 Chip Outperforms Apple's M1 Max With Several Caveats
- Thread starter MacRumors
- Start date
- Sort by reaction score
Apple is actually active in the Embree open source project, so they had every opportunity to optimize it. Perhaps the M1 is simply not capable of achieving good performance for this particular workload (for example, its vector instructions only operate on 128-bit vectors while Intel's AVX2 operates on 256-bit ones).
But anyway, in this case it doesn't matter. You can't make a useful argument about power efficiency by using one benchmark to compare power consumption and another to compare performance. It would be more interesting to see the power consumption for both when running Geekbench, where the Intel isn't 50% faster.
We know the caveat: the Intel-based system requires totally ridiculous amounts of power. Hope you have a 500 watt power supply for a desktop system!
Well, right. 14 core CPU vs 10 core CPU. is that fair? Could Apple just add another 4 performance cores? Yes, they could. Would it make sense? Yes, for some workloads, for example developers would love that. However Apple made 10 core CPU that has real TPD 45W. (measured) Intel just starts at 45W. Is Intel faster at 45W limit? No. Case closed. Remember, Intel and AMD are not real competitors for Apple. Qualcomm is. Well, will be, in early 2023. At that point Intel and AMD rather have something really good, because what Intel has today isn't good enough.Why not include the Cinebench R23 numbers when mentioning it and just GB5 that's favorable to M1? It's whopping 50% more multicore R23 full power and still 35% more in 70W balanced mode. Sure, on battery it's still much less efficient, no one expects miracles, but lot of people use laptops plugged in almost all the time, especially when the most power is needed, regardless of consumption and performance per watt.
We should be all glad Intel finally pushes back regardless of preferred platform (and if one is locked in Apple or Windows then the comparison is meaningless anyway
All that for a drop of blood.
Benchmark results have started to surface for MSI's new GE76 Raider, one of the first laptops to be powered by Intel's new 12th-generation Core i9 processor.
Intel previously said that its new high-end Core i9 processor is faster than Apple's M1 Max chip in the 16-inch MacBook Pro and, as noted by Macworld, early Geekbench 5 results do appear to confirm this claim, but there are several caveats as usual.
Geekbench 5 results show that the GE76 Raider with the Core i9-12900HK processor has an average multi-core score of 12,707, while the 16-inch MacBook Pro with the M1 Max chip has an average multi-core score of 12,244. This means the Core i9 processor is around 4% faster than the M1 Max chip in this particular comparison.
One of the biggest caveats is power efficiency. PCWorld measured the new GE76 Raider's power draw from the wall while running a CPU-only Cinebench R23 benchmark and found the Core i9 was consistently in the 100-watts range, and even briefly spiked to 140 watts. By comparison, when running the same Cinebench R23 benchmark on the 16-inch MacBook Pro, AnandTech found the M1 Max chip's power draw from the wall to be around 40 watts.
With the Core i9 consuming much more power, battery life takes a considerable hit, with PCWorld finding the new GE76 Raider achieved nearly six hours of offline video playback. Apple advertises the latest 16-inch MacBook Pro as getting up to 21 hours of battery life for offline video playback. Even with potential differences in display brightness and other factors, the 16-inch MacBook Pro clearly runs longer on battery.
Design is also a factor, with the GE76 Raider being a 17-inch gaming laptop that is just over an inch thick and weighs nearly 6.5 pounds. By comparison, the new 16-inch MacBook Pro is 0.66 inches thick and weighs 4.8 pounds.
All in all, it appears that Intel's claim that its new Core i9 processor is faster than the M1 Max chip holds up, but Apple likely has no regrets with switching to its own power-efficient chips for thin-and-light notebooks like the MacBook Air and MacBook Pro. And with the M2 chip expected later this year, Apple is only just getting started.
Article Link: Benchmarks Confirm Intel's Latest Core i9 Chip Outperforms Apple's M1 Max With Several Caveats
My guess: a beefed-up M1 Pro and Max that runs at a higher clock rate (found on Mac mini and the next iMac models).
It looks like you can mine Bitcoin on the M1 Max using the same power as browsing Macrumours on the Intel chip.
I will give Intel a miss.
I will give Intel a miss.
In other words, it's NOT faster. I'm sure Apple could make the M1 Max faster by using more power too. They wisely choose not to.
Not untrue, but honestly if Maxon wants Cinebench to be taken seriously as a cross-arch benchmark, they might want to consider doing something other than using one CPU vendor-created library and shoehorning it across all archs. Apple has already provided Accelerate (however maybe there's a good reason they're not using that).
I agree. That's also a weakness of using Cinebench. It doesn't fully load an M1 like it does Intel's processors, so that's not an apple-to-apples comparison for total power consumption.
The big jump in power efficiency will come when Intel switches from 10nm to 3nm chips from TSMC.
Why should counting cores be fair? It's a final product, fair is comparing it as it is. The same way we compare the power/consumption, it's fair as it is. And it is a bit artificial anyway, because not many people are that flexible between Mac and Windows, but there's no "case closed", just what's better in specific scenarios - do you need full power with battery? Sure, Mac Pro it is. Are you plugged in almost all the time and want/need the fastest mobile CPU? Alder Lake it is. Do you game? Forget Mac. etc.
Good points. ARMv9 will fix this. We will get official vector extension. What we have now in M1 is NEON, and quite frankly neon is not that great comparing to Intel AVX. Apple has secret AMX extension but can not use it publicly, can not add compiler support for it for use in general public. It is available only via Accelerate.framework. This is real disadvantage for M1 vs Intel. Hope M2 is armv9 based. We will see soon.
MSI GE76 Raider is not a laptop. It is portable desktop replacement. It can not operate on battery and reach that performance at all. Apple Macbook Pro can. When someone publishes performance results on battery it will not be a pleasant view for Intel.
Also, when you are plugged all the time just get desktop with 12900K. This "laptop" is hot and noisy.
It's a portable gaming machine if we want to be specific, and a very good one, any comparison with Mac Pro about this specific intended purpose is just ridiculous. We use it to compare the CPU just because it is one of the first to be available.
And no, being plugged in most of the time doesn't mean you shouldn't want to be mobile, there are more factors at play here. And yes, it costs something like noise and heat in peak performance, but the fact this performance is even available in mobile machine, should be celebrated too.
Personally I'd like to see more comparisons at lower wattage, and I almost got M1 Pro already, but I've realized I want to game more than "almost zero" during my last travel, so that's probably the most important preference when Intel got as fast as this.
Really? Why would they keep that secret? This means no open source project will be able to take full advantage of it.
Intel's Latest Core i9 Chip Outperforms Apple's M1 Max With Several Caveats:
1. Windows2. Sounds like a blow dryer
3. Works as a heater
Are you new to the site? 2019 and prior MacBook Pro way too hot and loud. There were many posts about this.
How can they even say it's faster? If Apple wanted to allow that much power and that much battery drain, the M1 Max would be faster.