I'm pretty sure by a comment I read from you earlier that this is a bug.

if you have a library file called say graphics.vh and it refers to a #define that is in the system vc...

#include "graphics.vh"
#define BLAH 1
=fail

#define BLAH 1
#include "graphics.vh"
=sucess

oh that was me, I just got a new computer.
Athlon64 3000+ 1gig DDR400 :D

I ported part of a C++ project into v3, including a switch that would cause verge to state that:
verbal.vc(411): "{" is not a recognized keyword or identifier.
Ironically line 411 contians only a "}".
I had default: cases and each case used {}s
I yanked out some stuff and it would compile, run other stuff fine then crash upon returning a string from within a switch.

This function will duplicate the error:

string Blah(int x)


{


switch(x)


  {


  case 0: return "a";


  case 1: return "b";


  case 2: return "c";


  }


return "err";


}

I rewrote a bit and my code runs perfectly now, so that's cool. It's just odd that it compiles fine.
El

all the functions around songs that deal woth position dont work. position is always 1.

I'm not sure but I believe that rotScale is causing v3 to crash the time. Either that or timer.

I have no evidence of this except before I was using rotscale it it didnt crash.
and
When I do use it, windows will give me the message, 'verge needs to close', but I can still go into verge (if I dont close that error message) and interface with everything. The only difference is that my cursor won't spin.

My cursor uses both rotscale and timer so I dunno.

Looks like SCAN_GPLUS should be re#defined as 78.
Thanks

Here's a bug I found (I posted it in a thread, I didn't know about this thread at the time):

http://www.verge-rpg.com/boards/display_thread.php?id=10964

rage... yeah, for those two reasons its not really possible to put maps in subdirs at this time, but you're absolutely right, that will be fixed.

and you're right about the string expected too. added them both to my list.

Ok, two bugs:

1. The "automax" CFG parameter is being ignored in the latest build of the engine. I run my game in fullscreen with automax set to 0 and yet I can still see the program auto-maximise as it changes resolution.

2. EntityStalk doesn't seem to be working quite as it should be. If you replace the Darin sprite in the Sully demo with a 32x32 sprite with hotspot data like this:

hotspot_x 8
hotspot_y 16
hotspot_w 16
hotspot_h 16

... Then Crystal bumps into the player whenever the party walks sideways, restarting her animation each time which makes it look like she's hovering. Vertical walking is fine. Bug?

This likely has something to do with the distance between stalker and stalkee. Perhaps a few more functions are in order for entity stalking, like being able to set the distance in pixels the stalker stays from the stalkee, like:

SetStalkDistance(int stalker, int stalkdist)

Or alter the EntityStalk function to accept distance input ( EntityStalk(int stalker, int stalkee, int distance) ).

... Oh, and don't forget that sound bug with one of the sound functions interpreting volume as a value between 0 and 666.

I've found another problem, and I most certainly didn't code these in.

I guess you could call them 'Lazy' pixels or artifacts;
However I'm not sure if this is video card related... direct draw related or whatever but here it is (reckoning you're debugging and all)

if you download this and run it you may see what I'm talking about:
hither...
Towards the top there are some random pixels...
Here's a link to a screenshot showing this problem:
Bug.jpg
if you zoom in, you'll see a couple towards the top - left;