The Apple M3 Max (16 Core) is a system on a chip (SoC) from Apple for notebooks that was launched towards the end of 2023. It integrates a new 16-core CPU with 12 performance cores with up to 4.06 GHz and 4 efficiency cores with 2.8 GHz. There is also a slimmed-down 14-core variant with a 30-core GPU.
Thanks to the higher clock rates and architectural improvements, the processor performance is also significantly better than the M2 Max in benchmarks and can keep up with the fastest mobile CPUs (such as a Core i9-13900HX).
The M3 also integrates a new graphics card with dynamic caching, mesh shading and ray tracing acceleration via hardware. In the top model, all 40 cores of the chip are used and support up to 5 displays simultaneously (internal and 4 external).
GPU and CPU can jointly access the shared memory on the package (unified memory). This is available in 48, 64 and 128 GB variants and offers 400 GB/s maximum bandwidth (512 bit bus).
The integrated 16-core Neural Engine has also been revised and now offers 18 TOPS peak performance (compared to 15.8 TOPS in the M2 but 35 TOPS in the new A17 Pro). The video engine now also supports AV1 decoding in hardware. H.264, HEVC and ProRes (RAW) can still be decoded and encoded. Like its predecessor, the Max chip offers two video engines and can therefore encode and decode two streams simultaneously.
Unfortunately, the integrated WLAN only continues to support WiFi 6E (no WiFi 7), unlike the small M3 SoC thunderbolt 4 is also supported (max 40 Gbit/s).
The chip is manufactured in the current 3nm process (N3B) at TSMC and contains 92 billion transistors (+37% vs. Apple M2 Max). Under load, the CPU part consumes up to 56 watts, the chip can use a total of 78 watts.
The Apple M3 Pro (12 Core) is a system on a chip (SoC) from Apple for notebooks that was launched in late 2023. It integrates a new 12-core CPU with 6 performance cores with up to 4.06 GHz and 6 efficiency cores with 2.8 GHz. There is also a slimmed-down 11-core variant with a 14-core GPU.
Compared to the M2 Pro the M3 Pro has been slimmed down somewhat and swaps two performance cores for efficiency cores. This is due to the changed core configuration, as 6 cores are now used per cluster (the M2 Pro and M3 still have 4 cores per cluster). Furthermore, the memory bus has been reduced from 256 bits to 192 bits (150 GB/s vs. 200 GB/s). However, thanks to the new architecture and higher clock rates, the new M3 Pro is still slightly faster.
The M3 Pro also integrates a new graphics card with dynamic caching, mesh shading and ray tracing acceleration via hardware. In the top model, all 18 cores of the chip are used and support up to 3 displays simultaneously (internal and 2 external).
GPU and CPU can jointly access the shared memory on the package (unified memory). This is available in 18 or 36 GB variants and offers 150 GB/s maximum bandwidth (192 bit bus).
The integrated 16-core Neural Engine has also been revised and now offers 18 TOPS peak performance (compared to 15.8 TOPS in the M2 but 35 TOPS in the new A17 Pro). The video engine now also supports AV1 decoding in hardware. H.264, HEVC and ProRes (RAW) can still be decoded and encoded.
Unfortunately, the integrated WLAN only continues to support WiFi 6E (no WiFi 7), unlike the small M3 SoC thunderbolt 4 is also supported (max 40 Gbit/s).
The chip is manufactured in the current 3nm process (N3B) at TSMC and contains 37 billion transistors (-7.5% vs. Apple M2 Pro).
The Intel Core Ultra 7 155H is a high-end laptop processor of the Meteor Lake series that has 16 cores (6 + 8 + 2) and 22 threads at its disposal. This 1st Gen Core Ultra chip has come to replace 13th generation Core chips; its 16 cores are comprised of 6 HT-enabled Performance cores running at up to 4.8 GHz and 10 Efficient cores (8 main cores plus 2 extra ones found in the Low Power Island) running at up to 3.8 GHz.
The 8-core Arc GPU, just out of the oven, serves as the integrated graphics adapter - this runs at up to 2.30 GHz - and there is a bevy of other brand-new technologies on offer as well, including the integrated AI Boost NPU with two Gen 3 engines for hardware AI workload acceleration.
Architecture & Features
Meteor Lake is a case of throwing everything at AMD (including the kitchen sink, yes) and seeing what sticks. With this product family, Intel intends to deliver higher CPU performance, higher GPU performance, higher hardware AI workload acceleration rates and longer battery life than what Raptor Lake chips were capable of. For those counting, that's four birds to kill with a single stone.
This generation of Intel Core processors features Redwood architecture P-cores and Crestwood architecture E-cores. Both come with slight architectural improvements over Raptor Cove and Gracemont respectively for slightly higher performance-per-clock figures; the interesting thing is that of the 10 E-cores, two are actually a separate cluster located on what Intel calls a "Low Power Island". Essentially, the latter is an SoC within an SoC that can stay active while most other parts of the chip are temporarily switched off to save power. The low-power E-cores run at up to 2.5 GHz. Intel hopes this approach will let it deliver unprecedentedly low power consumption figures when under low load, boosting battery life of laptops and tablets powered by Meteor Lake.
To build its Meteor Lake processors, Intel uses the Foveros technology (stacking several chips on top of each other). This is a cost-cutting measure more than anything else, as manufacturing several small dies on several different processes is so much cheaper than making a huge single die and hoping that there are no defects in it that will require disabling some parts of it.
Elsewhere, the Core Ultra 7 155H comes with 24 MB of L3 cache and a very healthy number of PCIe 5 and PCIe 4 lanes for NVMe SSD speeds up to 15.7 GB/s. It supports RAM running at up to 7467 MHz (DDR5-5600, LPDDR5-7467, LPDDR5x-7467, to be specific - which is about as good as what 8040 series Ryzen chips have). Naturally, the chip features built-in Thunderbolt 4 support and Intel CNVi Wi-Fi support; fascinatingly enough, Intel chose to keep native SATA III support that AMD had removed from its Ryzen processors quite a while ago.
The 155H is compatible with 64-bit Windows 10, 64-bit Windows 11 and with many Linux distros.
Performance
If one chooses to trust the official performance data published by Intel, then the 155H is about as fast as the Ryzen 9 6900HX (Zen 3 Plus, 8 cores, 16 threads, up to 4.9 GHz), as far as multi-thread performance is concerned. This isn't a ground-breaking result but let's just wait for our in-house testing results instead of jumping to conclusions.
Either way, real-world performance of the chip may vary significantly depending on how high the CPU power limits are and how competent the cooling solution of the system is.
A proper DX12 Ultimate graphics adapter, the Arc is no stranger to ray tracing and other modern technologies including AI frame generation (XeSS). It will let you connect up to four SUHD 4320p monitors and it will both HW-encode and HW-encode the most widely used video codecs including AVC, HEVC and AV1 in a fast and efficient manner.
Your mileage may vary depending on how high the CPU power limits are, how competent the cooling solution of your system is, how fast the RAM of your system is. The latter is really important; Intel stresses that for the Arc to deliver the best results possible, multi-channel RAM configuration is a must.
Power consumption
This mighty Core Ultra 7 series processor has a Base power consumption of 28 W, while its Turbo power consumption is not supposed to exceed 115 W. Its Base power consumption is supposed to be around 64 W, however, most laptop makers will probably go for a higher value to get higher clock speeds and thus better performance. Either way, a powerful cooling solution will be needed to sort out this chip's hot temper.
The 155H is comprised of five small chips ("tiles") that are connected using Intel's Foveros technology. The tile containing main CPU cores is produced on the fairly modern 7 nm Intel process marketed as Intel 4 while most other tiles (the iGPU, the I/O die, ...) are built with TSMC's N5 and N6 processes. The base tile is built with the old Intel 22FFL process.
Average Benchmarks Apple M3 Max 16-Core → 100%n=15
Average Benchmarks Apple M3 Pro 12-Core → 89%n=15
Average Benchmarks Intel Core Ultra 7 155H → 77%n=15
- Range of benchmark values for this graphics card - Average benchmark values for this graphics card * Smaller numbers mean a higher performance 1 This benchmark is not used for the average calculation
v1.26
log 15. 04:51:20
#0 checking url part for id 15113 +0s ... 0s
#1 checking url part for id 15115 +0s ... 0s
#2 checking url part for id 16906 +0s ... 0s
#3 not redirecting to Ajax server +0s ... 0s
#4 did not recreate cache, as it is less than 5 days old! Created at Mon, 13 May 2024 05:39:12 +0200 +0.001s ... 0.001s
#5 composed specs +0.043s ... 0.044s
#6 did output specs +0s ... 0.044s
#7 getting avg benchmarks for device 15113 +0.003s ... 0.047s
#8 got single benchmarks 15113 +0.01s ... 0.058s
#9 getting avg benchmarks for device 15115 +0.003s ... 0.061s
#10 got single benchmarks 15115 +0.008s ... 0.069s
#11 getting avg benchmarks for device 16906 +0.02s ... 0.089s
#12 got single benchmarks 16906 +0.098s ... 0.187s
#13 got avg benchmarks for devices +0s ... 0.187s
#14 min, max, avg, median took s +0.318s ... 0.505s