osdev does my head in sometimes

lukem95 · Post by **lukem95** » Sun Dec 23, 2007 4:28 pm

im so sick of my OS, i have half a mind just to start again, but i cant think of a way of doing it which appeals to me.

i have made no progress in months, and as such, have spent less and less time fiddling with it.

my gets() majorly died on me. i have no VGA/VESA graphics driver that works, and modules suck.

what do people think i should do? i would ideally like to have a bootloader that shoves my OS into a graphics mode from the word go, then i can work from the ground up, making sure all the blocks are in place.

but everyone says that a bootloader is one of the hardest parts an OS has... and i can sorta see their points!

i just needed to vent people, feel free to ignore this. im just having a low with CakeOS here

AndrewAPrice · Post by **AndrewAPrice** » Sun Dec 23, 2007 4:36 pm

I regained motivation and increased the speed of progression by planning a design doc:
http://www.osdev.org/phpBB2/viewtopic.php?t=15362

Even if it is a check list of items labeled "Things to do before 0.1" that you go through 1 by 1, then elaborate and break the items until the list is down to simpilest manageable components you can tick off when done.

crazygray1 · Post by **crazygray1** » Sun Dec 23, 2007 6:35 pm

I think I will do that too. I get headaches all the time when I'm actually coding especially when it comes to filesystems.

Telgin · Post by **Telgin** » Sun Dec 23, 2007 9:10 pm

MessiahAndrw wrote:I regained motivation and increased the speed of progression by planning a design doc:
http://www.osdev.org/phpBB2/viewtopic.php?t=15362

Even if it is a check list of items labeled "Things to do before 0.1" that you go through 1 by 1, then elaborate and break the items until the list is down to simpilest manageable components you can tick off when done.

I agree. Even though I haven't really gotten started coding important stuff yet, I decided that the first thing I would do would be to write design goals and expectations, and plan out certain milestones. That way you sort of feel like you've accomplished something when you get to certain milestones / complete even simple things.

If you think your current design isn't going anywhere (you've written yourself into a corner or you think that the design isn't going to work out), then you might just be better of restarting likeyou said. I generally find that I do much better the second time I try something complicated, but if you can't stand to lose your current work, you might be able to save it. It all depends on your code and what issues exist in your code.

inflater · Post by **inflater** » Mon Dec 24, 2007 5:36 am

lukem_95 wrote:im so sick of my OS

That's normal, you just need a bit rest. Especially from C/C++ code.

That's one of the reasons why I stopped PortixOS.

Regards
inflater

lukem95 · Post by **lukem95** » Tue Dec 25, 2007 2:22 pm

cheers for the replies, im gonna do me a long old list of things to do, and start again.

im using Brans Kernel for a base however, as i found this is a good solid base to build upon.

AndrewAPrice · Post by **AndrewAPrice** » Tue Dec 25, 2007 8:43 pm

inflater wrote:
lukem_95 wrote:im so sick of my OS
That's normal, you just need a bit rest. Especially from C/C++ code.
That's one of the reasons why I stopped PortixOS.

Regards
inflater

That's when I work on my next Dream Build Play entry (it's a Xbox and Windows hybrid game). C# requires a lot less code than C++ (automatic garbage collection, no need for headers, etc), but if we were given the option I'd use C++ (it takes a while to learn to trust the C# garbage collector, and then in the back of your head you're constantly thinking about the overhead for foreach and for loops when they dynamically allocate/unallocate when they iterate through an array of 1000s of items).