6 ms·

Carmageddon (and 2) seemed to have a physics engine qualitatively different to the modern norm, seemingly modelling the time of impact and the impulse response

Carmageddon (and 2) seemed to have a physics engine qualitatively different to the modern norm, seemingly modelling the time of impact and the impulse response to that, rather than correcting everything after intersection has already happened. It also seemed to have different static and sliding friction and modelled air friction and wheel dynamics to the extent that a power-up increasing acceleration and grip could make your car flip up like a hot rod. It made a huge difference to the feel of the physics and made things far less springy and slippery than box3d, havok and almost every modern game that uses a dynamic physics engine. This did mean that you could sometimes do things that made the physics take a noticeable amount of time and cause lag, but it also made things feel far more real than any physics engine I've seen since. Hopefully, someone will be able to reproduce that, now. Edit: I believe it even incidentally reproduced the intermediate axis theorem [1] on cars spinning through the air, presumably as an emergent phenomenon, which I've not seen any other physics engine do. [1] https://en.wikipedia.org/wiki/Tennis_racket_theorem https://en.wikipedia.org/wiki/Tennis_racket_theorem

BrokrnAlgorithm 3y ago
Agree, Carmageddon 2's physics still feel modern to this day. I can't think of any other game except Beam NG perhaps that achieves the same level. But the impacts in Carma 2 seem to suprass even that one.
- butz 3y ago
  "Carmageddon: Reincarnation" had even more advanced physics engine, although game ran a bit slow, so later iteration - "Carmageddon: Max Damage" used less detailed physics engine, suitable for consoles.
  - ghusbands 3y ago
    I remember trying (and funding) Reincarnation and finding it hard to really explore the physics, as it lacked set-pieces and seemed less fun when compared to Carmageddon 2. It's possible that I played after the big physics downgrade of maybe 2015 [1], or on a machine that couldn't handle it. I do wish there was a way to set up physics demonstrations, in all the games, for comparison with other engines. Certainly, the common boxes-made-of-jelly issue [2] is something I'm sure you wouldn't see from Carmageddon, though you'd probably see devastating performance issues with that same test. Carmageddon 2 did notably lack (as far as I could tell) simulations of aerofoils and other surface dynamics and their effects on handling; cars seemed to be modelled as near-blocks as far as drag was concerned. [1] https://steamcommunity.com/app/249380/discussions/0/343788552547448345/ https://steamcommunity.com/app/249380/discussions/0/34378855... [2] https://kripken.github.io/box2d.js/demo/webgl/box2d.html https://kripken.github.io/box2d.js/demo/webgl/box2d.html
  - BrokrnAlgorithm 3y ago
    I believe Carma Reincarnation used a completely new system, I believe they talked about that in the dev blog. While the system may have been similarly powerful in theory, I distinctly remember it also being less fun than the Carma 2's system, somehow less reactive. Carma 2's approach essentially allowed for the car to be crushed into any polygonal form, best illustrated when you drove the car into the underside edge of an angled bridge or similar. It was rad and so much fun, considering and that was 1998 which still blows my mind.
orbital-decay 3y ago
This was my immediate thought when I saw this. Car collision physics seems to be mostly in src/DETHRACE/common/car.c and src/DETHRACE/common/crush.c.
JKCalhoun 3y ago
From what I have heard about Patrick Buckland (the primary author of Carmageddon) he was quite into "banger racing" and so took his joy of the sport very seriously when coding what was I believe originally supposed to be a banger racing sim.
jeff-1amstudios 3y ago
Dethrace author here: I was lucky enough to talk to a couple of the original engineers. They also thought looking back that it was the vehicle physics engine that made the game so fun to play. Once you’d got over the initial excitement of running over pedestrians at least!
skykooler 3y ago
Kerbal Space Program can reproduce the intermediate axis theorem, as demonstrated here: https://www.reddit.com/link/10zd76y/video/gt9kpti7whha1/player https://www.reddit.com/link/10zd76y/video/gt9kpti7whha1/play... Of course, KSP just uses the Unity physics engine which is based on PhysX; this is more of an emergent property of things being represented by several separate physics objects connected by joints, rather than a single object.

BrokrnAlgorithm 3y ago

Agree, Carmageddon 2's physics still feel modern to this day. I can't think of any other game except Beam NG perhaps that achieves the same level. But the impacts in Carma 2 seem to suprass even that one.

butz 3y ago
"Carmageddon: Reincarnation" had even more advanced physics engine, although game ran a bit slow, so later iteration - "Carmageddon: Max Damage" used less detailed physics engine, suitable for consoles.
- ghusbands 3y ago
  I remember trying (and funding) Reincarnation and finding it hard to really explore the physics, as it lacked set-pieces and seemed less fun when compared to Carmageddon 2. It's possible that I played after the big physics downgrade of maybe 2015 [1], or on a machine that couldn't handle it. I do wish there was a way to set up physics demonstrations, in all the games, for comparison with other engines. Certainly, the common boxes-made-of-jelly issue [2] is something I'm sure you wouldn't see from Carmageddon, though you'd probably see devastating performance issues with that same test. Carmageddon 2 did notably lack (as far as I could tell) simulations of aerofoils and other surface dynamics and their effects on handling; cars seemed to be modelled as near-blocks as far as drag was concerned. [1] https://steamcommunity.com/app/249380/discussions/0/343788552547448345/ https://steamcommunity.com/app/249380/discussions/0/34378855... [2] https://kripken.github.io/box2d.js/demo/webgl/box2d.html https://kripken.github.io/box2d.js/demo/webgl/box2d.html
- BrokrnAlgorithm 3y ago
  I believe Carma Reincarnation used a completely new system, I believe they talked about that in the dev blog. While the system may have been similarly powerful in theory, I distinctly remember it also being less fun than the Carma 2's system, somehow less reactive. Carma 2's approach essentially allowed for the car to be crushed into any polygonal form, best illustrated when you drove the car into the underside edge of an angled bridge or similar. It was rad and so much fun, considering and that was 1998 which still blows my mind.

orbital-decay 3y ago

This was my immediate thought when I saw this. Car collision physics seems to be mostly in src/DETHRACE/common/car.c and src/DETHRACE/common/crush.c.

JKCalhoun 3y ago

From what I have heard about Patrick Buckland (the primary author of Carmageddon) he was quite into "banger racing" and so took his joy of the sport very seriously when coding what was I believe originally supposed to be a banger racing sim.

jeff-1amstudios 3y ago

Dethrace author here: I was lucky enough to talk to a couple of the original engineers. They also thought looking back that it was the vehicle physics engine that made the game so fun to play. Once you’d got over the initial excitement of running over pedestrians at least!

skykooler 3y ago

Kerbal Space Program can reproduce the intermediate axis theorem, as demonstrated here: https://www.reddit.com/link/10zd76y/video/gt9kpti7whha1/player https://www.reddit.com/link/10zd76y/video/gt9kpti7whha1/play... Of course, KSP just uses the Unity physics engine which is based on PhysX; this is more of an emergent property of things being represented by several separate physics objects connected by joints, rather than a single object.