Apple MacBook Pro 16 M4 Max review - The M4 Max is one of the fastest mobile processors

With the release of its new M4 generation, Apple has further extended its performance lead over the competition. Its single-core performance is unrivaled and the high-end version of the M4 Max is one of the fastest mobile processors on the market under multi-core load—plus, it stays super stable (even in battery mode). Its graphics performance, on the other hand, could do with being better and, for the most part, is on par with the competition featuring dedicated GeForce RTX 4070 Laptop GPUs and, in some cases, it even beats lower variants of the 4080 Laptop.

Under simultaneous CPU/GPU load, however, the MacBook Pro 16's cooling clearly limits the performance of the M4 Max and, in fact, the chip's power consumption is even slightly lower than that of the M4 Pro. Power supply is also a limiting factor in this case and the fast Max configuration in particular could benefit from stronger cooling. The 14-inch model is probably even weaker in this aspect if you take the 14-inch model with the M4 Pro as a reference. 

Apart from this, the same advantages and disadvantages of the M4 Pro model apply to this version. These include its high-quality case, extensive features, brilliant mini LED panel with increased brightness and an optional matte finish, excellent sound system and long battery life. Disadvantages include its lack of upgrade/maintenance options in conjunction with the massive surcharges Apple charges for more RAM/SSD capacity.

The MacBook Pro 16 2024 comes either with the full version of the new M4 Pro SoC (as in this detailed review) or two variants of the more powerful M4 Max SoC. With 14 CPU cores, the "small" M4 Max delivers the same CPU performance as the M4 Pro, but has a more powerful GPU with 32 cores for prices starting at $3,499. The full version of the M4 Max with 16 CPU cores and 40 GPU cores starts at $3,999 and it is precisely this variant with 48 GB RAM and a 1-TB SSD that we are testing today. If you choose 128 GB RAM (+$1,000), an 8-TB SSD (+$2,200) and a matte panel (+$150), the price increases up to $7,349.

As there are no differences in its case, connections, input devices, display or speakers compared to the M4 Pro model we have already tested, so we won't cover these sections again here and will instead focus on the performance of the new M4 Max processor.

The 16-inch version of the 2024 MacBook Pro is available with either the M4 Pro SoC or the M4 Max SoC and all versions are very similar in terms of both case and features. This includes Thunderbolt 5 support (up to 120 GBit/s) for all USB-C ports and a cooling unit consisting of two fans; only the heat sinks themselves differ slightly. On the flip side, the M4 Max can control two additional external displays, i.e. a maximum of four external displays at the same time in addition to the internal screen.

Apple has provided an environmental report for the MacBook Pro 16 on its product page. Lots of recycled materials are used for both the device itself and its packaging. A CO₂ footprint of 303 kg is stated for our test configuration.

Inside the MacBook Pro 16, the top-of-the-range version of the new M4 Max can be configured with a maximum of 128 GB RAM, which is a major advantage over almost all comparable devices. Plus, as it is unified RAM, the GPU can also benefit fully from this. However, there will soon be a direct competitor in this respect with the release of the new Strix Halo APUs from AMD, which—according to rumors—will be available with up to 96 GB.

The MacBook Pro 16 features three power modes: Low Charge, Automatic and High Performance. By default, the MacBook works using Automatic mode, which we also used for the following benchmarks and measurements. The High Performance mode didn't always provide an advantage during our tests, but we will take a closer look at this in the individual sections.

Apple has once again changed its CPU strategy with its new M4 chips compared to the old M3 generation. While both M3 Max CPUs were significantly faster than the M3 Pro, the full version of the M4 Pro and the small M4 Max are now on par, as both have 14 CPU cores (10x performance, 4x efficiency) with identical clock rates. Only the top-of-the-range model of the M4 Max is equipped with two additional performance cores, which is of course additionally reflected in its power consumption. While the M4 Pro consumed a maximum of 46 watts under full multi-core load, the M4 Max now consumes up to 57 watts, although this value was only reached very briefly during our test and then leveled off at ~50 watts. In High Performance mode, we noted 57-59 watts and this value was maintained throughout, which is why its multi-core performance in CB 2024, for example, proved to be slightly better.

Its multi-core performance is impressive. In the native Cinebench 2024 and Geekbench tests, its advantage over the old M3 Max was 17-21% and even the full version of the M4 Pro was outperformed by 14-17%. During the Cinebench 2024 test, it was even one of the fastest mobile processors ever—only the overclocked Core i9-14900HX in the XMG Neo 16 with a connected water cooling unit was slightly faster still. The new M4 Max was also only beaten by the Core i9-14900HX and the AMD Ryzen 9 7945HX3D in the Blender CPU test.

Regardless, the new M4 SoCs' single-core performance is beyond any doubt and Apple has been able to further extend its lead over the competition. The old M3 Max was outperformed by a whopping 25-28% and the other competitors simply didn't stand a chance in this scenario. However, its power consumption during the single-core tests was slightly higher than the M4 Pro at 6.2-6.5 watts.

The M4 Max easily passed the Cinebench 2024 Multi stress test (30 minutes) using both energy profiles and its performance remained completely constant. During the old Cinebench R15 Multi loop, its performance fluctuated minimally in Automatic mode, but this was just due to the very short test duration (a few seconds). The brief higher limit of 57 watts and the drop to 50 watts therefore had an effect here. However, we are still only talking about a reduction of ~3-4 %.

In battery mode, its performance remained stable until the battery level reached 4%. Only from this point did we note a significant reduction in performance. Further CPU benchmarks can be found here.

The fast M4 Max SoC is equipped with a 40-core GPU, making it the current top-of-the-range model from Apple. We were able to elicit a maximum of 75 watts from the GPU, but in lots of tests, we noted a maximum of 60 watts. Turbo mode had no direct influence on its power consumption in this case, but we were still able to achieve a better result in the Cinebench 2024 GPU test (+4.7 %). The GPU can access the entire RAM.

Depending on the benchmark, its advantage over the old M3 Max GPU was 14-27% and during cross-system tests, its results were generally slightly better than the GeForce RTX 4070 Laptop inside direct competitors such as the Asus ProArt P16. In the GFX-Bench tests and the CB-2024 GPU test, even slower variants of the RTX 4080 Laptop were beaten. During the Blender test with Metal, the M4 Max SoC took around 18 seconds, making it on par with dedicated Nvidia GPUs (OPTIX or CUDA). The GPU's performance remains stable both in battery mode and under continuous load.

It is still difficult to evaluate its gaming performance, as the number of native games available for macOS is limited. Currently, we can only use Baldur's Gate 3 to compare its performance with Windows laptops. In 1080p and with maximum details, the M4 Max GPU placed between the competitors with the GeForce RTX 4070 Laptop. However, the other games we tested had to be emulated, in turn making the laptop's performance limitations very visible. The game Cyberpunk 2077, which is being released natively for macOS next year, will be interesting to try out.

Its cooling is practically no different from the model we already tested featuring the M4 Pro. When idling and performing simple tasks, the MacBook usually remains completely quiet or you only hear a very low-frequency noise. Still, the M4 Max reaches its maximum volume of ~42.2 dB(A) in Automatic mode much faster as soon as the device is put under a little more load. Using High Performance mode, the fan speed fluctuates significantly more than was the case with the M4 Pro model. In the stress test, we measured a maximum of 54.7 dB(A) and 50.2 dB(A) when running Baldur's Gate 3. We could hear some slight coil whirring during certain GPU benchmarks.

Apple has the MacBook Pro 16's surface temperatures very well under control. Its case practically doesn't heat up at all under low load and we measured a maximum of just under 43 °C even when gaming or during the stress test. This means that the device can still be placed on your thighs without any problems. The keyboard area also heats up noticeably and it can cause warm fingers, but there are no limitations in this respect.

The stress test revealed a slightly different behavior of the M4 Max test device compared to the M4 Pro variant, which surprised us a little. In addition, High Performance mode had no noticeable influence here either. Its CPU/GPU consumption leveled off at around 28 watts in each case, i.e. only around 56 watts in total. The M4 Pro variant measured in at around 60 watts in Automatic mode and even 66 watts in High Performance mode. Under combined load, the MacBook Pro 16's cooling unit cannot utilize the performance potential of the new M4 Max at all.

With the M4 Max, the MacBook Pro 16's power consumption is limited by the 140-watt power supply under very high load—but only for a very short time and no power is tapped from its battery. We measured a maximum of around 144 watts, after which its consumption dropped quickly and was then only around 120 watts. When gaming, we noted around 116 watts, which is around 30 watts more than the M4 Pro configuration.

Its battery life has been improved compared to the M3 Max predecessor model, but the M4 Pro version manages even better runtimes. The direct Windows competition stood no chance against runtimes of 18 hours during the WLAN test and just under 22 hours in the video test (both at 150 cd/m²)—only the Dell XPS 16 9640 and the Asus ProArt P16 could keep up well in the video test.

During the WLAN test at full SDR brightness, its runtime was still 6:45 hours, which is a decent performance considering its bright panel. Under full load, however, you'll pay the price that the laptop's performance isn't throttled. In the worst-case scenario, the test ended after 65 minutes, whereby Low Charge mode was activated after 56 minutes.