PhysX wrapper for FreeBASIC and BASIC4GL.
-
- Posts: 8642
- Joined: May 28, 2005 3:28
- Contact:
PhysX wrapper for FreeBASIC and BASIC4GL.
Removed by me was outof date and senseless :-)
Joshy
Joshy
Last edited by D.J.Peters on Dec 25, 2021 17:31, edited 11 times in total.
-
- Posts: 8642
- Joined: May 28, 2005 3:28
- Contact:
-
- Posts: 8642
- Joined: May 28, 2005 3:28
- Contact:
@TMyke thank you for testing
The NVIDIA PhysX lib and Visual Studio 2010 is totaly new for me
but i found the source code of the PhysX Blitz Basic wrapper.
http://www.blitzbasic.com/Community/pos ... opic=90281
Now it's easy for me to see how things in PhysX works e.g. cloth, water ...
Joshy
The NVIDIA PhysX lib and Visual Studio 2010 is totaly new for me
but i found the source code of the PhysX Blitz Basic wrapper.
http://www.blitzbasic.com/Community/pos ... opic=90281
Now it's easy for me to see how things in PhysX works e.g. cloth, water ...
Joshy
Last edited by D.J.Peters on Jun 13, 2011 16:52, edited 1 time in total.
-
- Posts: 8642
- Joined: May 28, 2005 3:28
- Contact:
@Joshy;
You are a FB machine Joshy - well done.
1) the cubes - only a few (2-4) fell - they stayed stacked - which made me wonder; "how do kick them down" - please add a boot to the scene that I can wave my mouse and kick those cubes that did not fall.
j/k - tho I did wonder how I could get them to drop from a varying axis in lieu of a straight line - I was thinking would help encourage 'instability' in the cube's stacking or to help ensure they do not 'stack' after being dropped.
2) I am curious tho, regarding PhysX runtimes;
I see they are signed by NVidia - and I thought I had what I needed for PhysX to work by installing my driver that supposedly included 'PhysX' drivers.
I take it those are meant to be distributed by devs.
Is this the DX9.03C crap (or what ever) all over again?
Is it just me or should nvidia learn from M$ DX debacle and integrate a backward compatible driver package instead of relying on potentially conflicting dependencies devs may release?
Or Is this just unavoidable?
I mean - isn't physX highly dependent upon the video driver too?
Perhaps it is my misconception that my 'single' card - can do 3d acceleration and physX at the same time.
Any good 'real' info on this for someone like me who is just concerned with being sent to DLL HELL with my graphics driver - the one (of few) thing that can make a PC no longer bootable.
I mean sure right now the demos work - will they when my graphics card is updated?
Its exciting stuff none the less - thus why I bothered running the demos - and for that Joshy - a big stein of your favorite brew is raised in your honor.
edit:
oh - all compiled demos ran - Win XP SP3 - GF GT220 w/ v200.
Tho I seen no frame rate displayed so I can not report that.
You are a FB machine Joshy - well done.
1) the cubes - only a few (2-4) fell - they stayed stacked - which made me wonder; "how do kick them down" - please add a boot to the scene that I can wave my mouse and kick those cubes that did not fall.
j/k - tho I did wonder how I could get them to drop from a varying axis in lieu of a straight line - I was thinking would help encourage 'instability' in the cube's stacking or to help ensure they do not 'stack' after being dropped.
2) I am curious tho, regarding PhysX runtimes;
I see they are signed by NVidia - and I thought I had what I needed for PhysX to work by installing my driver that supposedly included 'PhysX' drivers.
I take it those are meant to be distributed by devs.
Is this the DX9.03C crap (or what ever) all over again?
Is it just me or should nvidia learn from M$ DX debacle and integrate a backward compatible driver package instead of relying on potentially conflicting dependencies devs may release?
Or Is this just unavoidable?
I mean - isn't physX highly dependent upon the video driver too?
Perhaps it is my misconception that my 'single' card - can do 3d acceleration and physX at the same time.
Any good 'real' info on this for someone like me who is just concerned with being sent to DLL HELL with my graphics driver - the one (of few) thing that can make a PC no longer bootable.
I mean sure right now the demos work - will they when my graphics card is updated?
Its exciting stuff none the less - thus why I bothered running the demos - and for that Joshy - a big stein of your favorite brew is raised in your honor.
edit:
oh - all compiled demos ran - Win XP SP3 - GF GT220 w/ v200.
Tho I seen no frame rate displayed so I can not report that.
Line 39 (or there about) in the Cube.bas file, change the 0.5 to 1 or 1.5. The line is:maddogg6 wrote:... I did wonder how I could get them to drop from a varying axis in lieu of a straight line
ActorSetPosition Actors(i),NxVEC(x,y,rnd2*0.5)
Just add this to the main loop in the examplemaddogg6 wrote:Tho I seen no frame rate displayed so I can not report that.
Code: Select all
Dim As Integer fps_count, fps_target = 10
Dim As Double fps_actual, fps_time1, fps_time2 = Timer
Do
If fps_count > fps_target Then
fps_time1 = fps_time2
fps_time2 = Timer
fps_actual = fps_count / (fps_time2 - fps_time1)
fps_count = 1
fps_target = fps_actual * 0.8 ' Update every 0.8 of a second
WindowTitle "FPS: " & fps_actual
Else
fps_count += 1
EndIf
' Rest of loop code here
Loop While Inkey = ""
By the way, if the simulation is running slowly try removing the Sleep command at the end of the main loop.
As for your card being able to run 3d acceleration and physX at the same time, I got this from the PhysX wiki page:
Any CUDA-ready GeForce graphics card (series 8 and newer, with a minimum of 32 cores and 256MB of video memory[14]) can take advantage of PhysX without the need to install a dedicated PhysX card.
-
- Posts: 8642
- Joined: May 28, 2005 3:28
- Contact:
@maddogg6
since PhysX 2.8.4 the PhysX *.dlls must be in the same folder as the *.exe
before PhysX was a part of the gfx driver.
the problem in the past was
if you have 4 games compiled with different versions of PhysX
it was your turn to change from one version to the next.
(4 different gfx card driver on one PC :lol:)
again this problem is gone.
don't worry about the very simple tests
this are only a test not more.
more next week
Joshy
since PhysX 2.8.4 the PhysX *.dlls must be in the same folder as the *.exe
before PhysX was a part of the gfx driver.
the problem in the past was
if you have 4 games compiled with different versions of PhysX
it was your turn to change from one version to the next.
(4 different gfx card driver on one PC :lol:)
again this problem is gone.
don't worry about the very simple tests
this are only a test not more.
more next week
Joshy
I just checked it again. Looks like the same problem, no performance difference between hardware and software mode. Although when I checked it it seemed to be in hardware.
PhysXGetPPUMode() = 1 (which is PPUMODE_HARDWARE)
PhysXGetSDKVersionNumber() = 284
PhysXGetHWVersion() = 1 (seems a bit odd...?)
PhysXGetNbPPUs() = 1
Since I last checked this I've got a new graphic card, a NV550ti, and I'm now running Win7.
PhysXGetPPUMode() = 1 (which is PPUMODE_HARDWARE)
PhysXGetSDKVersionNumber() = 284
PhysXGetHWVersion() = 1 (seems a bit odd...?)
PhysXGetNbPPUs() = 1
Since I last checked this I've got a new graphic card, a NV550ti, and I'm now running Win7.