Apple MacBook Pro 14 2023 Entry Review: Base model with cut-down M2 Pro, SSD downgrade, and slightly higher power draws

The base 2023 MacBook Pro 14 relies on a nerfed M2 Pro SoC with 10 cores (6x Performance cores + 4x Efficiency cores) while the GPU uses just 16 cores. 16 GB of RAM comes standard with the base model, but this can be optionally upgraded to 32 GB during configuration.

The M2 Pro continues to use the same 5 nm architecture as that of the M1 Pro but with the addition of two additional efficiency cores. The base variant of the M2 Pro offers just six performance cores while the high-end variants of the M2 Pro and M2 Max come with eight performance cores.

The efficiency cores reach a maximum of 2.4 GHz while the performance cores can attain 3.5 GHz in single-core or 3.26 GHz in multi-core. Check out our dedicated M2 Pro SoC page for more information and benchmarks.

In our Cinebench R15 multi-core loop test, the CPU performance remains constant without any performance drops even on battery power (only at 3% battery do we see any performance limitations). Compared to the base M1 Pro, the single-core performance seems to have increased by 23% in the native Cinebench R23 test. This is at the same level as the higher-end M2 Pro and M2 Max chips.

Mutli-core performance of the M2 Pro entry is about 23% faster in Cinebench R23 compared to the M1 Pro entry. The M2 Pro with a 12-core CPU, on the other hand, is 25% faster than the base M2 Pro. Multi-core scores are more or less identical compared to the Intel Raptor Lake Core i7-1360P even with lower consumption in the Apple chip. 

Intel, however, has a single-core advantage here while also requiring higher power.

AMD Zen 3 CPUs such as the Ryzen 7 6800U are at a disadvantage in single-core tests but nearly tie with the M2 Pro entry in multi-core. We will be testing laptops powered by 4 nm Zen 4 chips in the coming weeks.

When it comes to efficiency, the M2 Pro entry trails behind the faster M2 Pro in both single and multi-core measurements, and also behind the regular M2 SoC in the MacBook Air. 

Intel cannot match up to the M2 Pro's efficiency with Raptor Lake, but the AMD Ryzen 7 6800U is just slightly behind the M2 Pro in multi-core efficiency tests.

We look into performance and power efficiency metrics of the M2 Pro and M2 Max SoCs in our comprehensive analysis.

Subjectively, the base 2023 MacBook Pro 14 seems similar to the more advanced MacBook Pro 14 and MacBook Pro 16 models in overall system performance. While native applications indeed ran better, we did not experience any problems with emulated apps during testing.

This is best illustrated by Adobe apps such as Photoshop and Premiere Pro. The PugetBench results shown here were obtained via emulated x86 versions of these apps and not reflect actual performance. Results are excellent in the cross-platform CrossMark test.

The GPU in the base M2 Pro now features 16 instead of 14 cores with a maximum clock of 1.4 GHz. This allows the M2 Pro entry GPU to position itself about 7% ahead of the 16-core GPU of the M1 Pro. The M2 Pro with a 19-core GPU is about 12-17% faster depending on the test. 

In 3DMark WildLife, which runs natively on Apple silicon, even the Nvidia GeForce RTX 3050 Ti Laptop GPU in the Microsoft Surface Laptop Studio and Schenker Vision 14 are beaten by the entry M2 Pro GPU. Integrated GPUs from Intel and AMD are at a clear disadvantage.

The graphics performance of the entry M2 Pro remains completely stable both under prolonged load and on battery power.

More GPU benchmarks and performance comparisons are available here.

It is still quite difficult to assess the gaming performance of MacBooks or compare them to Windows laptops. Native games like Resident Evil Village run without problems at 1080p on the entry M2 Pro, but we do not have the possibility of recording and evaluating the fps numbers.

The following benchmarks, therefore, only show results of emulated titles such as Shadow of the Tomb Raider or Windows games tested with CrossOver. This is the case with games such as GTA V, wherein we can run even our regular Windows benchmark, but you should always keep in mind that these games are not optimized for ARM SoCs.

We could test power efficiency of the entry M2 Pro with The Witcher 3 (run via CrossOver) and compare it with Windows laptops. However, the Ultra setting in the game only runs with SSAO and Nvidia Hairworks disabled on the MacBook Pro (in Windows, we enable both HBAO and Hairworks in Ultra).

Thus, the results are not exactly an apples-to-apples (pun intended) comparison, but they still provide an idea of where the new SoC stands. The entry M2 Pro GPU is ahead of its competition from Nvidia and AMD. The outdated Intel Iris Xe Graphics G7 in Raptor Lake mobile is left even further behind.

The fan behavior in the base 2023 MacBook Pro 14 is slightly different from that of the full configuration M2 Pro. The base model is extremely quiet in daily use and fans are generally not audible for the most part. This is also true in pure GPU loads, wherein the fans kick in after a few minutes but are practically inaudible at a sound pressure level (SPL) of just over 24 dB(A).

The fan noise does increase under full CPU load. However, due to the very conservative fan curve, the fan noise SPL increases briefly up to 47.5 dB(A) to get the chip temperature under control and then levels off around 36-38 dB(A). This applies to scenarios of combined load as well where the CPU is more heavily strained. 

That being said, when the GPU is subjected to higher loads (for example, in our combined load with Cinebench R23 multi-core and 3DMark WildLife Extreme Stress test), we do not see this peak noise and the SPL stays at about 36-38 dB(A) from the start.

We couldn't hear any other random electronic noise.