I analyze it from 2 aspects:
The first one is based on your own situation, hardware class (FPGA DSP PCB layout, etc.), algorithm class
What is your basic situation? The question is not whether you are disgusted or not, but whether you can persevere or whether the learning process is smooth. Yes, yes, we young people should suffer a little bit, we should do something permanent. But if we can take it easy in this process, why not let ourselves take it easy? Interest is important, but in my opinion, the foundation is also important. It is not only a question of the learning process and time, but also the amount and quality of your future achievements ( graduate students pay special attention to this )
Second, let's talk about the things you are interested in - hardware. In fact, your situation is not pure hardware. It has a bit of software in it, because after all, the PCB we have laid out by ourselves may still need to be done by ourselves. To do some functional debugging of drivers and some boards, you still have to code yourself~
How fun is electronic development, and how interesting is electronic design. (It's as addictive and expensive as TM drug addiction) Seeing the boards you made, the circuits you made, the programs you wrote, and the solutions to some project problems ~ the kind you can touch, how interesting!
Let's put it this way, hardware is easy to get started with, but difficult to upgrade. It takes time, money and manpower, especially experience. It may be more difficult to stand on the upper level of the pyramid~ It is a bit like an accountant, the older you are, the more popular it is.
I can't eat it without years of project experience. Note that my project experience here refers to high-speed signal PCB layout, multi-layer boards, EMC, switching power supplies, FPGAs, etc. There are too many things to consider. It is not simply to write a few microcontroller programs (such engineers are not very troublesome to train)
No major is good, no major is bad, and the one that suits you is the best. (interest dominates)
If you really like to engage in electronics, embedded and hardware development, then go into the pit (I will dig the pit with you and jump together) Anyway, I have no way out~hhh
Attach some layouts that I made before: ( robot control board , which has been mass-produced)
The board is not very complicated. The control system made by STM32 is equipped with motor drives, sensors and other necessary ICs for robots.
Finally: If it is feasible, choose to do 2~ When you are reading, the papers and patents are relatively better, which is definitely good for your graduate career. The domestic environment requires more 2 than 1~ and in recent years, artificial intelligence, Big data has once again driven the wind of computer technology ~ the profession standing on the tuyere must have nothing to say.
In terms of hardware, national policies and industry problems have led to the development of artificial intelligence, deep learning, and big data. This is a fact that we do not want to admit but exist
Getting a bat offer is better than anything else! And the platform is good, the career development speed is fast! Of course, there are some drawbacks behind doing software, and it may also be unrealistic to do research and development all the time! This is all for later.
Only this, personal opinion is for reference only.
continue to add
It should be emphasized here: as a graduate student, achievements, patents, paper, etc. are very important! Be sure to pay attention! For academic research, image recognition and artificial intelligence are easier to produce paper! (It's not easy, just relative to the hardware class), and what paper needs is to do theory, data, and experiments. That is more or less you have to have some innovation! (creation and manufacture difference)
And what about the hardware class? It must be based on some applications, involving actual products. Hardware class is difficult to innovate. It only cares what problems you solve with the existing technology, and the economic value to the society. Of course, I personally admire this very much, because I like it very much myself! (It can be touched, just like a son, born and raised and become a talent!)
But the fact that we need to face is that compared with practical application and theoretical research, the latter must be easier to produce paper!
(Personal opinion, if I am wrong, welcome to correct it)