OSDev.org

DavidBG wrote:

Solar wrote:C isn't necessary, I agree. But in 99 cases out of 100, not having a firm grasp of C is a sign of a more severe underlying problem - that the person in question has little or no experience programming close to the machine.

I don't think this is quite correct. Or if it is, it is a slight exaggeration.

Providing a single contrary example, presumably puts you in the 1% to which Solar referred.

David,
I think a firm grasp of C is only surpassed by a firm grasp of assembly. It was not QBasic that gave you experience in programming close to the machine.

I think it's not really related to C, Pascal, QBasic or whatever. The point is that he was doing more than Hello World, which in DOS times almost automatically meant doing low-level stuff - completely independent of any language. There's really not that much difference between C and Pascal (except for syntax, of course). For QBasic I honestly can't say, never did low-level things in it. I think it completely misses pointers, though, so it's probably not the same there...

berkus wrote:

Kevin wrote:I think it completely misses pointers, though, so it's probably not the same there...

It has PEEK and POKE and these work perfectly well instead of pointers. You can even use it as JIT for assembly code

(Yes, I also started with QBASIC)

Peek and Poke, yes I used that a lot!

As I did Call Absolute and Call Interrupt.

Nice to meet another QBASIC programmer!

Hobbes wrote:David,
I think a firm grasp of C is only surpassed by a firm grasp of assembly. It was not QBasic that gave you experience in programming close to the machine.

This might be sort of correct. But I doubt I would have ever learned assembly in the first place if it wasn't for the limitations of QBASIC. If I had C or C++, I most likely would have used someone else's work in a library instead of my own. But because there really wasn't much info on QBASIC, (at that time) I had to resort to my own devices.

All the same, it might be nice to learn C at some point.

gerryg400 wrote:Providing a single contrary example, presumably puts you in the 1% to which Solar referred.

I am still not sure about this. There are probably others like me, but I am not sure if Solar meant this to be an exact survey anyway, but simply a way of making a point.

Still, I disagree. Not knowing C doesn't limit you. What does is not being able to think in computer logic.

David

DavidBG wrote:Still, I disagree. Not knowing C doesn't limit you. What does is not being able to think in computer logic.

If you don't know c you

1. Can't peek at working O/S code (like Linux) to see how others do it.
2. Can't use Bran's tutorial as a guide
3. Can't use Tanenbaum's book and Minux source as a learning tool.
4. Can't use many of the OsDev tutorials
5. Can't read the Mmurtl book (which was my first look at an OS source)
etc.

Not knowing c is very limiting.

DavidBG wrote: 5. Can't read the Mmurtl book (which was my first look at an OS source)

At this point, do you consider it to be the best option as a first extensive OS reading from a book, or would other books be better at this?

The mmurtl book is not a great book on osdev, but there are so few that it should be considered. It is a very practical book, not strong on theory (either os or general programming). The mmurtl os is quite an odd os, IMHO.

I was stranded in Heathrow airport once and found it in a bookstore. Read most of it while I was stuck there and became hooked on osdev. Couldn't wait to get home to my pc and set the PM bit.