Apple's advantage is actually that it provides much higher single-thread (really, per-thread) performance than "comparable" x86 processors. A M1 Pro can match a 12900HK in perfectly-threaded high-code-density scenarios like Cinebench, with half the thread count. Real-world IPC is something like 3x (!) that of Intel right now - obviously it also clocks lower but Apple hits very hard in lower-thread-count scenarios.
If your code is bottlenecked on a single thread, or if it doesn't scale well to higher thread counts, Apple is actually great right now. The downside is that you can't get higher core counts, but that's where the Pro and Ultra SKUs come in.
(The real, real downside is that right now you can't get higher core counts on M1 without being tied to a giant GPU you may not even use. What would be really nice is an M1 Ultra-sized chip with 20C or 30C and the same iGPU size as A14 or M1, or a server chip full of e-cores like Denverton or Sierra Forest, but that's very much not Apple's wheelhouse in terms of products unfortunately.)
If your code is bottlenecked on a single thread, or if it doesn't scale well to higher thread counts, Apple is actually great right now. The downside is that you can't get higher core counts, but that's where the Pro and Ultra SKUs come in.
(The real, real downside is that right now you can't get higher core counts on M1 without being tied to a giant GPU you may not even use. What would be really nice is an M1 Ultra-sized chip with 20C or 30C and the same iGPU size as A14 or M1, or a server chip full of e-cores like Denverton or Sierra Forest, but that's very much not Apple's wheelhouse in terms of products unfortunately.)