Irrlicht Engine

I use Irrlicht directly for input.

That's why alot of projects use gnu licenses, public domain can seperate the author from his works, while the license implements a type of legal glue that makes sure you are always credited for your original works.

edit: Sorry for continuing an off-topic reply.

I've looked at the shadow nodes code, it uses an IMesh, so limiting them to animated meshes, I believe, wasn't intentional. There is little extra work involved, but IMeshSceneNode should DEFINITELY support shadows.

As for other scene node types, I don't know how wise it would be. Oct tree nodes ...

Have you found a solution to your problem ?? What c++ editor are you using?

*sigh* I found what was going on. I was clearing my structure.

memset((void*)&_ant, 0, sizeof(_ant) );

I don't know why I even had that line in there, but what a lousy mistake to make and to even post about it in the bug reports forum. :oops: shame.

This mistake was even talked about in a ...

It doesn't make any sense why it isn't working. I don't know.. For some reason the virtual function pointer for the allocators are turning up null ??

It keeps coming back to internal_delete in irrallocator with strings..


Do this and see what happens:

1. Create a struct
2. Use struct (ie "_my_struct tempstruct;") and add a stringc var to it (ie "irr::core::stringc Name")

Tell me if you get an error when the struct goes out of scope when ...

My bad, it was 4am and I was walking dead by then.

Hm.. I don't know what to do.. I'm getting into all kinds of trouble here with the allocators.. I added a constructor to my struct which has a stringc in it, to set it to "". But now irr::core::array is complaining, and I think I know why. I must ...

EDIT: I suggest you fullscreen for this post for ease of reading.

I may need a patch, but it is obviously not something I would be concerned with for this issue, as I downloaded and installed vc++ 2005 express three days ago to the day.

Also, you're right, you can't get the proper info from the ...

Here ya go

#ifndef _DEBUG
int APIENTRY _tWinMain(HINSTANCE hInstance,
HINSTANCE hPrevInstance,
LPTSTR lpCmdLine,
int nCmdShow)
#else
int main(int argc[] , char *argv[] )
{
#endif // #ifndef _DEBUG

#define _IRRALLOCATORTEST

#ifdef _IRRALLOCATORTEST
irr::core::stringc IrrString ...

Please devs, look into this.

Your problem is completely unknown to me, the artifacts and so on, I would suggest reporting it as a bug (use a link to this topic) so that the devs will check it.

You should also try different renders, ie. OpenGl if you use Direct3D, in case one acts differently.

//! deallocate memory for an array of objects
void deallocate(T* ptr)
{
internal_delete(ptr);
}

When I init a string as "" (empty), it passes the empty array (ptr=NULL) and crashes at internal_delete(ptr) saying Access violation reading location 0x00000008, which doesn't make any sense ...

:!: Look at Warchiefs reply! He is right!

Code:

// This code will fail. Pointer is not valid
GUI* interface;
interface->createGUI();

// This will work
GUI* interface = new GUI();
interface->createGUI();

// This will work
GUI interface;
interface.createGUI();

Are you creating GUI ...

Ya, I saw that it is used by createDevice.

Didn't know about implicit constructors, now I know. Although, I didn't think it was a constructor issue.

:oops: :!: That must be it! I forgot I had to get an older version of my code (2nd hd crash!!), therefore older versions of the dlls, and probably ...