Yep. My brother actually has a Surface Book. It's a pretty cool device but severely let down by the hardware. The separation of the GPU is actually a problem and not something I find worthwhile but they didn't have a choice to enable good performance. But it's actually not relevant today, at least for Apple. They have a perfectly fine chip that they put in their Pro iPad that is good enough for most tasks with quite good performance. No separation needed.
The Surface Book is a 9 years old device at this point. I think it's the Microsoft curse, they have a good idea but launch it half-baked and lose interest. They didn't control the most important piece of the stack anyway: the silicon.
Nowadays with the Snapdragon chips they would have a chance at making something much better, but they failed too early, made a weird compromise with the Surface Laptop Studio (wasn't too bad but too expensive for the compromises) and then gave up.
They focus on the Surface Pro which isn't too bad because unlike Apple they actually allow a full fat OS to run and it's not a very good tablet (at least in the media consumption sense) but it makes for a pretty decent laptop.
Of course, you have to deal with Windows and it's not as efficient as an iPad would be.
For the 13" tablet market the appeal is really easy: it can be a notepad that you draw on, you can annote document directly, write equation/diagrams/whatever with the stylus and at the same time it can be a laptop by just snapping a keyboard base to it. The major problem is the software: ideally you want to be able to use "full-fat computer OS" software when bolted on to the keyboard but at the same time enable touch centric usage when used as a notepad/tablet.
Apple could very much do it: they already enable running iPad apps on macOS, they would just need to figure out a layer on top of macOS that would enable usage of touch for the relevant apps without needing to convert the full OS to be touch ready. Just like they did back in the days with Front Row, to enable media center usage out of a regular Mac.
I'm pretty sure they already know how to solve most of the problems (maybe they even have prototypes) but I believe they won't because it is not in their interest profit wise.