Apple MacBook Pro 16 2021 M1 Max Laptop Review: Full Performance without Throttling

The new MacBook Pro 16 2021 with the M1 Max processor has an optional performance mode. However, it only affects the fan speed and therefore the surface temperatures under load, but there is no performance difference to the standard automatic setting. We have performed all the following benchmarks and measurements with the default automatic mode. The optional power saver mode will limit the package power of the M1 Max to 30W, which means the system will stay silent, even under stress.

The CPU part of the Apple M1 Max with 10 cores (1 efficiency cluster with two cores + 2 performance clusters with four cores each) are basically identical to the M1 Pro, so we can keep it short here. We determine a maximum consumption of around 29 Watts (~34W package power) in the multi-core benchmarks, both under sustained workloads and on battery power. These consumption values are slightly higher compared to the M1 Pro, but this is caused by the larger chip with additional GPU cores, larger cache, and more RAM. Our benchmark results are, as expected, almost identical to the M1 Pro with minor fluctuations.

Both the single-core as well as multi-core results are comparable to the rivals, but all of them are equipped with 45W CPUs and the often require much more power in reality. Also, almost every Windows competitor reduces the CPU performance on battery power.

It is not surprising that the high-end SKU of the MacBook Pro 16 2021 is obviously a very powerful and immensely responsive laptop. During everyday tasks or when you only stress the CPU, you will not notice a difference to the entry-level unit with the M1 Pro though. This is also supported by the cross-system benchmark CrossMark, where the two devices are very similar. The BlackMagic RAW Speed Test shows the advantage of the larger cache, because the CPU test shows a small advantage of 7 percent.

We also performed the two PugetBench tests for Adobe Photoshop as well as Adobe Premiere Pro. The M1 Max performs noticeably better than the M1 Pro, but you should be a bit cautious with the results, since the benchmarks are only compatible with the Intel versions of the applications. These are emulated via Rosetta 2, and it works, but the performance is noticeably worse compared to the native versions. These will be much faster in practice, but we cannot determine the exact performance advantage.

The high-end version of the MacBook Pro 16 2021 with the fastest M1 Max is equipped with four GPU clusters and eight cores each. We can determine a maximum consumption of 44W for the GPU during the benchmarks, and the results are much better compared to the M1 Pro with 16 GPU cores with an advantage between 60 up to 95 percent. The Wild Life Extreme Unlimited benchmark, which is running natively on the MacBook, shows comparable performance figures to the GeForce RTX 3070, even though the M1 Max does not even require half of the power. The graphics performance is also completely stable on battery power.

You can play games on the MacBook Pro 16, but the number of available games is the big limitation. There are currently only a few native titles and some can be emulated. Our gaming benchmarks once again show an advantage of around 70-80 percent over the M1 Pro and the performance in Shadow of the Tomb Raider is roughly comparable to fast versions of the GeForce RTX 3060. However, there is once again the problem that it is an emulated title.

Apps like CrossWare add support for Windows titles like Witcher 3 or GTA V and the performance is actually not bad. As an example, we have captured the GTA V benchmark sequence in the video below. You can clearly see that the performance is very good with ultra settings (1920 x 1200 pixels), but there are noticeable stutters. You can still play the game like this, but the MacBook Pro 16, even with the powerful M1 Max, is just not a gaming laptop.

The MacBook Pro 16 with the M1 Max processor is silent while idling and during light workloads, because the fans are turned off and there are no other electronic noises (like coil whine), either. Pure CPU load will result in the lowest fan speed (1600 rpm), which is hardly audible at all. It will only get louder when you start to stress the 32 GPU cores.

The fans of the MacBook Pro 16 M1 Max can reach a higher maximum speed (5348 rpm on the left side and 5776 rpm on the right side) compared to the M1 Pro unit (4280 and 4750 rpm, respectively). However, these maximum speeds are pretty theoretical, because there was only one scenario where we could reach the maximum speed. We had to start the stress test (CB R23 and 3DMark Wildlife Extreme) in the normal mode and switched to the performance mode after a couple of minutes. The MBP 16 will then increase the fans speed to the maximum value for a few minutes, which results in loud 55.6 dB(A). After a while, the fans will level off at 4400 rpm (49.5 dB(A). If you start the stress test with the performance mode, the fan speed will fluctuate between 4000 (47.3 dB(A)) and 4900 rpm (52 dB(A)) with an average of 4400 rpm (49.5 dB(A)). Like we said before, there is no performance advantage, so it does not make sense to use the performance mode.

If you just leave the energy settings in the default automatic mode, the stress test will only result in 3600 rpm, which results in 42.9 dB(A). During our gaming test of the Witcher 3, the fan speed will fluctuate a bit with an average of 3200 rpm (39.5 dB(A)). We can also observe 3600 rpm in the PugetBench benchmark (Premiere Pro), but keep in mind that this is the emulated version of the app. The overall fan noise is a very convenient murmur and you can still work very comfortably without headphones. All these results are identical for mains as well as battery power.

The optional energy-saver mode will limit the package power of the M1 Max to 30W, which results in a completely silent operation, even during the stress test.

The new MacBook Pro 16 does not have any temperature issues, because you will hardly notice any warmer surfaces during everyday tasks and the temperatures are not critical at all, even when you really stress the unit.

Contrary to our standard procedure (Cinebench R15 + FurMark), we use the two native apps Cinebench R23 and 3DMark Wildlife Extreme (stress test for each app) for the stress test on the MacBook Pro. We managed to get the highest consumption figures with this combination. Contrary to powerful Windows laptops, the performance is rock solid over the course of the one-hour test and the surface temperatures are still very low. We can only measure about 42 °C on the top and the bottom of the base unit. The chassis gets a slightly warmer during the Witcher 3 test due to the lower fan speeds, but the difference is hardly worth noticing.

We also performed the stress test with the optional performance mode and looked at the temperatures. As mentioned before, it does not affect the performance, but the surface temperatures are even lower at up to 39.9 °C due to the higher fan speeds. All in all, these results are excellent considering the levels of performance and the fan noise. Even the PSU temperatures are completely okay at up to 43 °C.

We cannot notice a difference in terms of fan speed or surface temperatures when we use an external display.