Taking a guess here... but WX has a wxOpenGL canvas class afaik, and it could be possible to pass the canvas somehow to irrlicht... Maybe look there.

We have also been having this problem (the main menu is using shadowed). When we showed Frank a screenshot he told us to try a base materialmap for all the textures of 128,128,255 (which I haven't tried yet) and seems a bit redundant to me.

Here are the assumptions in the above post:
1) plugin interface takes less time than rendering engine to update
2) ogre is better than cs for 3d graphics (which seems to be the consensus among developers)

But say crystalspace used the ogre rendering engine... Wouldn't it in the future divert people away from having to spend development time on the graphics engine? Since ogre is suited specifically to 3D graphics I also think it does its job better than the lagging cs renderer does.

2 useful resources for track modelling:
http://www.danielsefton.com/tutorials/blender/index.php
http://netpanzer.berlios.de/tuxkart/index.php/Track Building Tutorial?PHPSESSID=082a21a3cf98200d489c2d2454b908cc

The project is still in early stages so don't expect wonders, but it is progressing well

Have you tried ripping the planeshift camera at all? Could be a solution... although I know they have some code where camera internally changes y rotation, so I would prehaps get rid of that.

what does
printf("%s\n",msg_id);
at the top of your code do?

I wish I knew a piece of bug free software

I am not clear on what you mean. How could I use a collider in this case?

I think I solved the problem... sampling the points and applying a force to 4 points on an object makes the angular momentum go mental when it tilts.
Now I just need a way of setting rotation of object...

yeah, don't forget to query the loop time per last loop and deduct it from your sleep time... also it may be an idea to ask on www.gamedev.net/community/forums about this sort of thing as well (lots of stuff on constant framerates and gameloops there)

http://community.crystalspace3d.org/forum/index.php/topic,121.0.html

heres the one I first used: http://www.amazon.com/exec/obidos/tg/detail/-/0672327112/qid=1132086272/sr=8-2/ref=pd_bbs_2/103-5891960-1048639?v=glance&s=books&n=507846
here is one my sister used:
http://www.amazon.co.uk/exec/obidos/ASIN/0672326817/qid=1132086413/sr=8-1/ref=pd_ka_1/026-7947170-9567653
here is one a friend of mine used to use:
http://www.amazon.com/exec/obidos/tg/detail/-/076450746X/qid=1132086272/sr=8-6/ref=pd_bbs_6/103-5891960-1048639?v=glance&s=books&n=507846

all of these seem to be popular mainstream books everyone uses.

just push through it, didn't stop me making and finishing my rpg when I was 11 (by the time I finished I was fully proficient in C++).

Once you have learnt C++ to some level everything else becomes easier (as its just reading technique and api's)

hi

history:
i want to produce hovercraft flight, so I am using hitbeam using the start point as the position of the ship and extending the beam downwards by a certain amount (my cutoff point when force is so small I ignore it).

I was using inverse square law (which is probably more accurate) but it was explosive when you neared ground so you needed a small multiplier (but it made higher distances more neglible).
So now I'm using
Force upwards on ship = multiplier / height to ground point.

Problem is, applying a force to a single center point on an object is a very bad idea as it will rock about when on uneven ground (unstable), so I came up with instead stabilising the four center points along the base sides of the quad of the ships collision box.

Code:

--------X----------
|                         |
|                         |
X                      X
|                         |
|                         |
 --------X----------

like so (if thats the bottom quad of the ships Mechanics system's collision bounding box)

in the program the box has dimensions  2x0x2 (using cs co-ord system), therefore we need 'stabilisers' at

Code:

0,2--------1,2----------2,2
|                                  |
|                                  |
0,1                             2,1
|                                  |
|                                  |
0,0 --------1,0----------2,0        on XZ plane  (y = 0)

csVector3(1,0,0)
csVector3(1,0,2)
csVector3(0,0,1)
csVector3(2,0,1)

but for some reason, the stabilisers 'fight' and the ship goes mental! eventually flying off into space. here is my stabilising code

Code:

Stabiliser::Stabiliser(csRef<iCelEntity> world,csRef<iCelEntity> aship,float mult,float min, float max, csTicks dur)
: multiplier(mult) , cutoff_height_min(min) , cutoff_height_max(max) , duration(dur)
{
SCF_CONSTRUCT_IBASE(NULL);

ship_mech = CEL_QUERY_PROPCLASS_ENT(aship , iPcMechanicsObject);
world_mesh = CEL_QUERY_PROPCLASS_ENT(world , iPcMesh);
}
Stabiliser::~Stabiliser()
{
SCF_DESTRUCT_IBASE();
}

bool Stabiliser::Perform(iTimerEvent *ev)
{
ForceCheck(csVector3(1,0,0));
ForceCheck(csVector3(1,0,2));
ForceCheck(csVector3(0,0,1));
ForceCheck(csVector3(2,0,1));
return true;
}

void Stabiliser::ForceCheck(csVector3 offset)
{
float distance = Height(offset);

// if distance is extremely small then force upwards will be extremely large
if(distance > cutoff_height_min)
AddForce( multiplier / (distance) , offset );
else
AddForce( multiplier / (cutoff_height_min) , offset );
}
void Stabiliser::AddForce(float force , csVector3 &offset)
{
ship_mech->AddForceDuration(csVector3(0,force,0),false,offset,duration);
}
float Stabiliser::Height(csVector3 &offset)
{
csVector3 start = ship_mech->GetBody()->GetPosition() + offset;
csVector3 end = start + csVector3(0,-cutoff_height_max,0);

csHitBeamResult bres = world_mesh->GetMesh()->HitBeam(start , end);
if(bres.hit) printf("beam hit!\n"); else printf("beam fail!\n");
if(bres.hit)
return cutoff_height_max * bres.r;
else
return 999999999;
}

Thank you