RealityFactory NGD Alpha1 released!!!
no problem! (but please, say something...)
you have to understand how curious I am about your thoughts... that's odd: probably in those months you were in this situation, now I am
The last video showed some errors due to an erroneus mass displacement. I've played too much with the Offset parameters, then I forgot to reset it... so here's a better video:
Video:
I've integrated the water parameters in the PhysicsSystem. now you can enable/disable the water, define his height (the invisible plane under which the objects are in the fluid, and the liquid density. You can achieve a pudding like density, or make the objects sink...
The only issue is that you can define only one water level that affects all your level. Besides this, the development o this feature is complete.
you have to understand how curious I am about your thoughts... that's odd: probably in those months you were in this situation, now I am
The last video showed some errors due to an erroneus mass displacement. I've played too much with the Offset parameters, then I forgot to reset it... so here's a better video:
Video:
I've integrated the water parameters in the PhysicsSystem. now you can enable/disable the water, define his height (the invisible plane under which the objects are in the fluid, and the liquid density. You can achieve a pudding like density, or make the objects sink...
The only issue is that you can define only one water level that affects all your level. Besides this, the development o this feature is complete.
All the demos working ok for me except the last one. I keep walking through walls and disappearing down below the level. I am running the demo on a low end machine (integrated graphics card) which only runs rf in opengl. As I say they all seem to run fine for me. Great job Federico, again, many many thanks for this release.
thanks bernie. I'm happy to know that physics works also in your low-end system: I have developped on a laptop with intel integrated graphics (one of the worst chipset ever), so I've tried to think also at the low-end machines.
A note about the collision system. The last demo shows physics collision on StaticMesh entities. But the player collision on the staticmesh is the RF per-face collision. The staticmesh-player collision sometimes fails, but this doesn't happen for the staticmesh-rigidbody collision that's what I've developped and I'm showing in the demo. I'm saying this because I was in doubt wheater or not completely integrate in the physics worls the RF entities that usually move, that are the player and the pawns, the spouts, the projectiles. In this way all the collision routines would be taken by Newton, which has a better collision detection. Instead, for semplicity and compatibility sake, I've added a physics collision layer over the existing collision system.
So the system is far from being optimized expecially in the staticmesh collision case, giving an additional calculation burden to the application that raycast the collision detection of the RF/genesis entities over the faces of the staticmesh while computing the collision of the newton rigidbodies over the trimesh collision (that is what your staticmesh is in second place). Interesting to note that the newton collision detection doesn't fail even in this stressful situation, the staticmesh raycast collision instead does.
A note about the collision system. The last demo shows physics collision on StaticMesh entities. But the player collision on the staticmesh is the RF per-face collision. The staticmesh-player collision sometimes fails, but this doesn't happen for the staticmesh-rigidbody collision that's what I've developped and I'm showing in the demo. I'm saying this because I was in doubt wheater or not completely integrate in the physics worls the RF entities that usually move, that are the player and the pawns, the spouts, the projectiles. In this way all the collision routines would be taken by Newton, which has a better collision detection. Instead, for semplicity and compatibility sake, I've added a physics collision layer over the existing collision system.
So the system is far from being optimized expecially in the staticmesh collision case, giving an additional calculation burden to the application that raycast the collision detection of the RF/genesis entities over the faces of the staticmesh while computing the collision of the newton rigidbodies over the trimesh collision (that is what your staticmesh is in second place). Interesting to note that the newton collision detection doesn't fail even in this stressful situation, the staticmesh raycast collision instead does.
Oops. I was shooting crates in the first demo, and two just would not move using a gun. So I switched weapons to a grenade launcher. They jumped when it exploded, but then I killed myself in the explosion. I got stuck in a loop of demos that my player was dead the minute the demo started, and hitting escape just respawned a character who died immediately. I couldn't get back to a menu screen.
I'm going to try them again, but be careful of where I stand.
I just tried again, and it locked up on the actor integration demo. I have no idea why. I killed the two guys, then came around to see if the rag dolls would reacte if I shot one when down. I looked down at the guy in brown, and his legs were 'swinging around' - then it became a slideshow. Then it froze. I had to crt-alt-del to shut it down.
I'm going to try them again, but be careful of where I stand.
I just tried again, and it locked up on the actor integration demo. I have no idea why. I killed the two guys, then came around to see if the rag dolls would reacte if I shot one when down. I looked down at the guy in brown, and his legs were 'swinging around' - then it became a slideshow. Then it froze. I had to crt-alt-del to shut it down.
Steve Dilworth - Resisting change since 1965!
about demo1. probably the GenesisCollide parameter in the PhysicsRigidBody entity is set to false. The die loop doesn't mean anything to me in the physics topic, probably I badly set up the health attribute of the player (keepattribute and such).
about demo6. The slow down is probaby due to the timeslicing paradox I was talking about some post before. The lag in the graphics (due to load the explosion bitmap?) makes the physics clock go out of time. My update loop needs some low/high limitation to avoid this while mantaining the timeslicing. That's what kiko and I were talking about.
It's an alpha release, isn't it? When QoD could take the time to look at my code, surely he can help me improve those issues. But his time is limited right now and there are some other features I want to add that aren't in the Alpha1 like the water buoyancy I added few days ago. Then optimization and stability would be the top priority.
about demo6. The slow down is probaby due to the timeslicing paradox I was talking about some post before. The lag in the graphics (due to load the explosion bitmap?) makes the physics clock go out of time. My update loop needs some low/high limitation to avoid this while mantaining the timeslicing. That's what kiko and I were talking about.
It's an alpha release, isn't it? When QoD could take the time to look at my code, surely he can help me improve those issues. But his time is limited right now and there are some other features I want to add that aren't in the Alpha1 like the water buoyancy I added few days ago. Then optimization and stability would be the top priority.
thanks for your specs steven.
I think I solved the time paradox error, or at least I setup the physics to recover from a time distortion. I feel that the simulation is more fluid. Right now though I don't know if I'm running a simulation with a fixed step or not (that is: I don't know if the simulation is framerate dependent, which is bad)... I'm simply using the time datas coming from the RF clock.
A question for QoD: Is the TimeTicks value passed to the CommonData classes a "sliced time"?
I think I solved the time paradox error, or at least I setup the physics to recover from a time distortion. I feel that the simulation is more fluid. Right now though I don't know if I'm running a simulation with a fixed step or not (that is: I don't know if the simulation is framerate dependent, which is bad)... I'm simply using the time datas coming from the RF clock.
A question for QoD: Is the TimeTicks value passed to the CommonData classes a "sliced time"?
I released a Patch for the Alpha 1 release:
http://rfngd.netsons.org/index.php?section=6
N.B. Includes the water buoyancy feature (PhysicsSystem).
I hope that the time paradox is solved. As always, testing and feedback is appreciated.
Download:Patch for the Alpha1 release.
- Fix errors in the Weapon/projectile system
- New Update method.
Use this values in PhysicsSystem:
LoopEachFrame = 17
IterationTime = 60
http://rfngd.netsons.org/index.php?section=6
N.B. Includes the water buoyancy feature (PhysicsSystem).
I hope that the time paradox is solved. As always, testing and feedback is appreciated.
- fps
- Posts: 504
- Joined: Mon Sep 26, 2005 9:54 pm
- Location: in a magical land devoid of hope, happiness, and sanity.
wow. this holds water.
I cannot wait to get ragdolls working in my game.
they will preform very well in my complex levels w/ moutains.
SSSSSSSSSSOOOOOOOOOOOO EEEEEXXXXXXCCCCCCIITEDDDDDDDDDD!!!!
I cannot wait to get ragdolls working in my game.
they will preform very well in my complex levels w/ moutains.
SSSSSSSSSSOOOOOOOOOOOO EEEEEXXXXXXCCCCCCIITEDDDDDDDDDD!!!!
1 wrote:
for the internet is a cruel and dark place at times, and there's sex and blood everywhere.
2 wrote:
You say that like it's a bad thing.
1 wrote:
You are a bad thing.
for the internet is a cruel and dark place at times, and there's sex and blood everywhere.
2 wrote:
You say that like it's a bad thing.
1 wrote:
You are a bad thing.
I just get a crash when it tries to load after the menu:
Searching for fullscreen driver
Initializing Game Shell...
--------------------------------------
--- Reality Factory 0.76A ---
--- For more Information, visit: ---
--- http://www.realityfactory.info ---
--------------------------------------
Parsed RealityFactory.ini file
Genesis3D Initialized
*INFO* VFS detected (not encrypted)...
Initializing Camera Manager...
Initializing User Input Subsystem...
Initializing Audio Manager Subsystem...
Initializing CD Audio Manager Subsystem...
Initializing Midi Audio Manager Subsystem...
Initializing RF Menu Manager Subsystem...
Initializing Network...
Loading Menu.ini...
Parsing Menu.ini...
Loading Character.ini...
Initializing Menu
Initializing Collision Manager Subsystem...
Initializing Network Manager Subsystem...
Initializing AVIFile Video Subsystem...
Launching Reality Factory Game Shell...
Entering CRFMenu::DoMenu()
Entering CRFMenu::ProcessMenu
Entering Windows Message Loop, Rendering Game Menu
Entering CRFMenu::ProcessMenu
Entering Windows Message Loop, Rendering Game Menu
Entering CRFMenu::ProcessMenu
Entering Windows Message Loop, Rendering Game Menu
Initializing Level: demo1.bsp
Configuring Camera Defaults...
Initializing Entity Registry...
Initializing Terrain Manager...
Initializing Effects Manager...
Initializing Actor Manager...
Parsing Environment Setup Entity...
Initializing Model Manager...
Creating Player Avatar...
Loading Player Avatar...
Initializing HUD...
Initializing Damage Subsystem...
Initializing FixedCamera Manager Subsystem...
Loading Player Configuration...
Loading Attributes and Player Configuration from PlayerSetup.ini
Loading Environmental Audio
Parsing PlayerSetup Entity
Initialize Player Data
Parsing EnvironmentSetup Entity
Initializing Automatic Door Manager Subsystem...
Initializing Moving Platform Manager Subsystem...
Initializing Teleport Manager Subsystem...
Initializing MorphingField Effects Manager Subsystem...
Initializing 3d AudioSource Manager Subsystem...
Initializing Particle Effects Manager Subsystem...
Initializing Static Entity Props Subsystem...
Initializing Static Mesh Subsystem...
Initializing NewtonGameDynamics Physics Manager System...
Steve Dilworth - Resisting change since 1965!
- QuestOfDreams
- Site Admin
- Posts: 1520
- Joined: Sun Jul 03, 2005 11:12 pm
- Location: Austria
- Contact:
Re: RealityFactory NGD Alpha1 released!!!
just tried your demo federico and lovem!
this is AMAZING! it works like a charm....!
I got a few random bugs through, (the drop through the floor as noted above) and also on the demo with the ragdoll humans, sometimes just drops out for some reason ?
it is random but have noticed it does this when i skip from the demo with the joints(doors and lights) too quickly, if i wait and run around in the previous demo it seems to come up.
also sometimes when the crates and object get bounced around the level to quickly it freezes.
this too is random, and doesn't happen all the time
My spec's
Pentium (R) D CPU 3.00GHZ
2GIGs Ram
Nvidia Geforce 6600
this is AMAZING! it works like a charm....!
I got a few random bugs through, (the drop through the floor as noted above) and also on the demo with the ragdoll humans, sometimes just drops out for some reason ?
it is random but have noticed it does this when i skip from the demo with the joints(doors and lights) too quickly, if i wait and run around in the previous demo it seems to come up.
also sometimes when the crates and object get bounced around the level to quickly it freezes.
this too is random, and doesn't happen all the time
My spec's
Pentium (R) D CPU 3.00GHZ
2GIGs Ram
Nvidia Geforce 6600
Re: RealityFactory NGD Alpha1 released!!!
I suppose you downloaded the archive RFNGD_22012008b.zip from this page:
http://sourceforge.net/project/showfile ... _id=570612
so I think we should discuss it here:
http://www.realityfactory.info/forum/vi ... f=9&t=3111
I really appreciate if some admin or moderaor could close this topic just to avoid confusion.
anyway, thanks Spyre. I'm glad you liked it (thanks for the specs either!).
the "fall through the floor" bug isn't expected at all, so I'm quite disappointed. At least I hope that this doesn't happens too frequently.
In the Release 1.0 version (http://www.realityfactory.info/forum/vi ... f=9&t=3111) I disabled any material sound in demo2.bsp to show the real capabilities of the physics engine in a stress test with several primitives (cylinder, spheroid, boxes).
http://sourceforge.net/project/showfile ... _id=570612
so I think we should discuss it here:
http://www.realityfactory.info/forum/vi ... f=9&t=3111
I really appreciate if some admin or moderaor could close this topic just to avoid confusion.
anyway, thanks Spyre. I'm glad you liked it (thanks for the specs either!).
the "fall through the floor" bug isn't expected at all, so I'm quite disappointed. At least I hope that this doesn't happens too frequently.
This is mainly due to the incorrect sound preload. I hope someone could help me to figure out how to correctly preload these resources.also sometimes when the crates and object get bounced around the level to quickly it freezes.
In the Release 1.0 version (http://www.realityfactory.info/forum/vi ... f=9&t=3111) I disabled any material sound in demo2.bsp to show the real capabilities of the physics engine in a stress test with several primitives (cylinder, spheroid, boxes).