YouTuber Luke Miani came up with two 16 “MacBook Pros, one in the basic configuration with a 16-core GPU and 16GB of RAM, while the other was the top of the range with a 32-core GPU and 64GB of RAM. with a MacBook Pro with an Apple M1 core processor, an iMac with a Core i7 and Radeon Pro 5300, and an 18-core Mac Pro with a Radeon Vega 64. Some of the results were really interesting.
–
The power of the new processors was clearly demonstrated by the presence of accelerators for ProRes video. In Final Cut Pro (FCP), previous Macs needed about 650-750 seconds to process (including the 18-core Intel), while the new ones with M1 Pro and Max did the same task in about 105 seconds. But the strange thing was that the double unit for ProRes in the M1 Max brought only 5 seconds better time compared to the M1 Pro. Minor differences from the others were in the rendering in FCP, where the 18-core Intel with Vega 64 was already catching up. The results in DaVinci Resolve were also very good (approx. 35-40 seconds for M1 Pro / Max versus 90-230 seconds for others).
18-core Intel significantly outperformed Apple’s new processors in Blender, for example (its times to about 60% of the new M1 Pro and Max times), while Apple’s new features were the best when rendering via GPU, but the differences were not very significant and even almost never. A surcharge for a better GPU would not make much sense here. The video still contained many benchmarks, where new processors usually showed excellent performance, but we will be more interested in the practical one in real applications.
Of interest was, for example, the game Shadow of the Tomb Raider. Here, Apple’s new processors drove through the Rosetta emulator interface, so you could expect lower performance than when running native. Even so, the M1 Max reached 97 fps, beating the Vega 64 at 88 fps. Recall that the Vega 64 has a performance similar to, for example, the desktop GeForce RTX 2060, so overcoming such GPUs, even more so when running in Rosetta and not natively, is a very impressive result. M1 Pro then had 55 fps and the original M1 only 24 fps. Here, the increase in performance in the GPU part is clearly visible. The high increase in graphics performance was also evident in 3DMark, where it roughly corresponded to the differences in brute force (as in most tests, M1 Pro slightly exceeded the theoretical assumptions, on the contrary, M1 Max did not slightly reach them).
All M1 showed high single-threaded performance in the Cinebench R23 (about + 25% against Intel), in multi-core of course 18-core Intel won, but surprisingly not as much as one would expect due to the number of cores 80% higher. However, he outclassed the new Apply, for example in Novabench or V-Ray, however, even in running through Rosetta, they were still able to compete with the Core i9 in older Macs. You will learn more details in the following video, the conclusion of which is that the performance of Apple’s new CPUs is excellent, especially considering their consumption. GPU performance also went significantly up, which is especially true of the weaker M1 Pro, where it often exceeds theoretical assumptions, so in many GPU tests it is more than 2 times more powerful than M1. On the other hand, the M1 Max was slightly disappointing, as it usually did not achieve twice as high performance against the M1 Pro.
Prices of related products:
–
