Geek discretion advised: Apple M1 Max trails behind RTX 3080 Laptop by nearly 30% in GFXBench but is about 8% faster than RTX 3060 Laptop, mileage varies depending on considered benchmarks
Working For Notebookcheck
Are you a techie who knows how to write? Then join our Team! English native speakers welcome!
News Writer (AUS/NZL based) - Details here
There's been considerable excitement around Apple's new MacBook Pro 14 and MacBook Pro 16 models, and why not? The Apple M1 silicon has proven to be a solid performer so far, so the upgrade to an M1 Pro or M1 Max can be equally rewarding if not more.
During the keynote, Apple showed off some performance graphs in comparison to powerful gaming and content creator laptops on the Windows side. In one of the slides, Apple claimed that the M1 Max SoC, which offers up to a 32-core GPU, has performance on par with at least a 105 W RTX 3080 Laptop GPU while also consuming 100 W less power than what a 155 W RTX 3080 Laptop GPU would consume.
A new GFXBench entry looks to offer some perspective. Before we come to any conclusion, do note that one benchmark is nowhere near a complete representation of the chip's performance. Also, the GFXBench results can sway either way depending on which entry and which OS/API you select. Finally, the benchmark does not explicitly mention the exact GPU core count used. With that being said, let's see what's on offer.
First up, comparing to an RTX 3080 Laptop GPU, we see that the M1 Max is about 28.5% slower in Aztec Ruins Normal Tier and High Tier Offscreen tests. The M1 Max's fortunes seem to reverse when compared to an RTX 3060 Laptop GPU, though not by much. Here, we see the M1 Max to be about 8.2% and 2.3% faster than the RTX 3060 Laptop GPU in the aforementioned tests.
So, is the M1 Max's performance more in line with an RTX 3060 than an RTX 3080? That's a difficult question to answer right now. From a purely raw performance perspective, the prospect that an M1 Max could surpass an RTX 3080 seems a bit far-fetched. But then, there are multiple TGP variants of the RTX GPUs, so things could intersect at some point depending on the workload; and those that make good use of Apple-specific features will have an obvious advantage.
For instance, 3DCenter's tabulation of available GFXBench results for the M1 Max, RTX 3070 Laptop GPU, RTX 3080 Laptop GPU, and the RTX 3080 desktop GPU indeed show Apple's latest and greatest chip to have a perceivable advantage overall compared to Nvidia's laptop offerings despite sipping way less power.
Also, the real claim to fame for Apple Silicon is performance-per-watt and not just raw performance alone. Besides, a good deal of Windows-based benchmarks such as 3DMark (save for the Wid Life Extreme test) and popular games cannot run on M1 just yet, so it's difficult to get a holistic perspective of the performance differences.
While there's no doubt with regards to the performance-per-watt advantages with M1 Max in comparison to traditional GPUs, the exact benefits will only become apparent once we get to fully review the new MacBook Pros.