(Original post by choco_latte)
Just posting to thank you all for your replies. They were useful.
Sometimes you get tunnel visioned while pursuing some goal, and don't see when events show it isn't the right one anymore. I spent a lot of time and money on the masters thing and it seemed a waste to abandon academia. Your posts helped me realize that I should go get a job.
As nonswimmer says, you can only predict future success based on past performance. I did well at my job before studying, but badly at studying. Doing what I'm good at - working, is much more likely to yield good results - happiness, money, etc. That isn't to say that returning to study was valueless - I did gain new ways of looking at things, and a broader understanding of computer science than you get studying electronics. They are 2 different subjects, far more than I'd previously appreciated.
I got a job at a good company within 2 weeks of starting looking. The future looks pretty bright.
Good luck to you all.
RE 'do you need a PhD':
If a person wants to earn a lot of money, I would suggest that they work in finance. The pay in finance, versus the kind of work they do / skill they need, is much higher than elsewhere. For me, finance work would be unfulfilling, but for friends it has been very profitable.
For the super elite - those who are in a position to study for a quantitative PhD at a top university, the higher degree may be unnecessary. If e.g. you already have a 1st in math from oxford, or made a contribution to your field at undergrad, you can probably get that quant job without a PhD.
For the more normal, unless you have some reason to believe you are exceptional, you should note that the average salary for a worker with a PhD is 40k lifetime mean, and assume that will be you too. There is nothing wrong with aspiration, but delusion is rarely a path to happiness.
I have had the pleasure of working with board members in a number of engineering companies, and although these people are more likely than average to have a PhD, the majority do not. I suspect that PhD graduates and board members are drawn from the same elite section of a distribution of ability. They are high earners because they are naturally able, due to good parenting and early education. That they have a PhD is caused by the same early factors, but did not cause the other good things in their lives. A rich man is likely to have a rolex and a sports car, but nice watches do not cause sports cars.
However, with these elite people, it is important that their success is a narrative. They are X today because they achieved Y and Z previously. Doing academic research is a good way to put yourself in a position achieve Y and Z. So is starting a business, whether it succeeds or fails. Anticipating which technologies will be game changing, and making sure you work on them as they mature also works.
But as I say, most readers probably will not be exceptional. I believe they should choose the things that will lead to them being happy, solvent, and fulfilled. In engineering and science, a research degree is a conservative choice for achieving these things - for 3 years, it returns a high quality of life in return for low wages. After that, it probably won't do more than get you an interview you wouldn't otherwise have got. Relative to working those 3 years, I wonder if it gives you better demonstrations of skill, and experiences. Likely more than unskilled work, and less than a rotational graduate program. The rest is on you.
I agree with sj27 that in some fields, a PhD is often a requirement. In the sciences, there is a structure where PhD grads direct research, MSc grads do the research grunt work, and BSc graduates largely do structured work like quality control. This isn't so much true in my field - engineering companies tend to consider both MSc and PhD grads for research. I'm sure if you had a PhD in that specific topic, you'd get the job, but it's not weighted so highly for the transferable skills that come with research training, because in engineering the emphasis is on producing a product (demonstrator).
But er, this is assertion on subjective experience, so believe what you will.
[edit to add para agreeing with sj]