I said nothing.

@freddix

Its not a problem to have multiple keyboard functions.

Have a global variable called "programstate" and do:

@Capehill

Giddy asks for a 640x480 16bit surface.

@sofistisoftware

Giddy 3 works perfectly with a recent beta here.

@sofistisoftware

Try this:

* Before running giddy, run "dumpdebugbuffer clear" in a shell
* Run giddy, make it freeze the system
* Reboot, then run "dumpdebugbuffer file=ram:debug.txt"
* Email the debug.txt to pete _at_ petergordon _dot_ org _dot_ uk

@sofistisoftware

Hmmm.. thats weird. It works perfectly here.

Do you have a crash in the debugbuffer when it crashes?

Giddy3 v1.3 just hit OS4Depot. Changes:

* Removed the workaround for a bug in MiniGL. The workaround was actually
causing crashes on the spaceship level. This means that the game should no
longer crash, but the viewscreen on the spaceship won't work properly until
MiniGL 2.1 is released.
* Fixed a bug which meant that Giddy could end up stuck in the ground if you
stood in exactly the right place on one of the bubbles.
* Removed the erroneous bulb effect when starting the game.
* Added credits for MorphOS and Mac builds

Giddy 3 on OS4Depot.net
Giddy 3 on Facebook.

@TSK

You've done some really great tools there

@Menthos

Alright, how about, "when the next beta is ready"

@Menthos

When its done

@Menthos

Well, there is one bug that is in the OS to do with copy/paste, but I think Rigo has a workaround for that.

All the other ones were in the gadget, and are fixed.

@Rigo

Nice site. Look forward to the removal of the Javascript

BTW, my work on richeditor isn't just "initial", i would call it "continuing"

@freddix

Yes. A good tip is to initialise the structure to sensible defaults as much as possible in your constructor. Can save weird bugs and headaches later.

@freddix

Wow! Is it me or has it just got even faster? Its really snappy now. As fast, if not faster than, firefox on my PC. Considering the PC is a much faster machine than my A1, this is very impressive!

@orgin

Quote:

Fixed :)

@freddix

You need to put "break;" at the end of each case part, otherwise the code continues executing through the rest of the cases.

I gave up because it was too much of a stab into the unknown. I have no experience writing AHI drivers, or sound card drivers.

DaveAE gave me an AC'97 based OS4 driver to use as a base, and I was going to put the logic from the Linux driver in where possible.

It just drove me nuts and i gave up, basically.

I still have the code if someone wants to look, but its probably best to just start again.


@DaveAE
The official repulse drivers just plain don't work on OS4. AFAIK, anything that tries to play sound just gets 'stuck'.

@DJBase

I started work on an OS4 native driver using the linux driver as a reference. I gave up, though.

@orgin

Hmm... the creeping slowness could be related to the fact that the editor currently refreshes the whole display, even if it just scrolls. The render function can render subsections of the display, though, so once the functionality of the editor stabilises and the bugs are squashed, the optimisation phase will kick in and i'll fix that.

I was going to suggest that the use of RPTAG_APenColor would cause the slowness on 8bit screens, but then i remembered it falls back to pens anyway

Edit: Oh, and i've already fixed the scrolling oddness.