Reactor stress test

[mocaponeuk]

need a bigger bowl man... have a look on turbosquid

[TheVjey1985]

looks cool

[Gamea650]

WTF i thought this was real

[Somethingcrappy]

Lol one of them is running away in the top left corner at 0:04-0:06

[Genomax2007]

Old word. Thanks!

[adamsonamine]

good work

[donniedarkoooooo]

you probably need a good pc for this...right?

[TheAudiopulse]

Great! With this we may avert any future meltdowns! Hamdilullah! :D

[AJleKcaHgep]

It's nice, but why you do this?)

[Genomax2007]

@AJleKcaHgep

Thanks! it's only a test with reactor.

[Genomax2007]

oh, thanks!

Tell me more about your music video... :D

c u!



[Adeptmind]

You have some cool stuff.. Would you be interested in creating a music video to some dance music? Like dubstep, tech house or psy trance?

[Buskie1]

I FUCKING LOVE PEGS.

[honzad1]

i see bug :D 0:04 - 0:07 at left

[BurningHero]

@honzad1 it has rabies :D

[casiel10000]

oye muy bueno enserio podrias hacer un tutorial porfavor es q quiero mejorar en mi 3d

[Genomax2007]

@casiel10000

Gracias, pero no es como para hacer un tutorial. Hay cientos de ellos en la red y ademas este video es muy antiguo. Hoy dia hay otras opciones mas precisas para simulaciones como Physx. He visto algunos de tus videos y veo que usas Krakatoa para las particulas, que tal va este plug?

Seguimos en contacto :D

[casiel10000]

@Genomax2007 pues  si yo he usado krakatoa en las partticulas y es un buen plugin pero quiero aprender a hacer esto con el reactor claro seguimos en contacto

[casiel10000]

@Genomax2007 jaja hace mucho q no veia tu video ese plug es muy util a mi me agrada mucho y tu que has intentao ahora??

[Genomax2007]

hahaha, i dont have PhysX in 2007 :)

[nVidia6000]

Gotta say, reactor is pretty shit. I use PhysX always.

[ggilgil]

the only imperfection on this is the missing of friction

[Flam3swe]

10 years later, we will be able to do this in real-time.

[Genomax2007]

@Flam3swe Hahaha, and with more quality!

[Genomax2007]

@Flam3swe Hahaha, and with more quality!

[naka538]

render time?

[Genomax2007]

@naka538 I don´t remember the render time, about 1hour.

[Genomax2007]

@naka538 I don´t remember the render time, about 1hour.

[Genomax2007]

@naka538 I don´t remember the render time, about 1hour.

[Genomax2007]

@naka538 I don´t remember the render time, about 1hour.



[Mounty2010]

sorry, but why u call this "stress test" ?

350 objects is ridiculous ... try 35.000.

But nevertheless, the look is great :)

[Genomax2007]

@Mounty2010 

[Genomax2007]

@Mounty2010

[JamDP1100]

Did u drop Clothes pins into a bowl?

[TheKaos90]

hummthat took a few days to render :)

[OGMO11]

you cant stress test a pre calculated thing... lol :P

[HavokDelta6]

Wrong, in Havok's case you can, in terms of substeps, i am suprised there was not stuff going through stuff here, so it does count a stress test.

[avion85]

hmm i read the comment you responded to and yours. Can you elaborate a bit about these substeps? this does look prerendered to me..how can this be a stress test?

[HavokDelta6]

substeps per frame still comes into play,

if you have an object 1 meter thick, and an object traveling at 10 m/s towards it, if you have 1 substep per second, chances are it'll go right though it, if you have a substep of 20 per second, each interval is 0.5 meters, this means the collision with the 1 meter thick object will be detected, when you have alot of fast moving stuff. or stuff banging into stuff, banging into other stuff. it becomes a stress test at those sub steps.

[avion85]

oh you mean timesteps, i get it. I've actually done an engine for rigid body dynamics and the problem you are referring to simply does not compute in this case. First, time is not a factor coz it doesnt have to go as fast is it does in realtime, it can te much much longer(and actually does) to precompute physics and collision detection in a prerenderd context. Even in realtime you can detect how much objects are interpenetrating and return 1/2 (or any other fraction) dt recursively to avoid it..

[HavokDelta6]

No i mean substeps. cmon mate do your homework.

[avion85]

we are talking about the same thing here mate, be it timesteps, delta-t(dt), or substeps. The term refers how often you check for collisions and calculate responses. it can also refer to physics and graphics engines, for example my engine updates physics 40times more often then graphics

Again, the only way this can be interpreted as a stress test is if you measure how much time it take to pre-render this, thats all

[HavokDelta6]

FAIL, but whatever you want to call it okay?

also just to prove how right i am,

Echo, Alpha, omega :) i got 3 :P

[avion85]

oh i get it:D

You have no idea what you are talking about do you? you tried to play smart correcting these people in the comments eh?

You are busted pal:)

[HavokDelta6]

Yep totally....noob

Bit of advice everyone.

Don't argue with an idiot, they drag you down to their level, then beat you with experience.

[avion85]

so ironic, this situation:)

btw, you never argued...you kept dancing around, i didnt see any arguments at all...

anyway - im done, good luck in your studies.

ppl can judge for themselves

[koolJ]

haha, I like how in the upper right corner one of them spins away funny.

[RyanBriscall]

upper left corner.

[koolJ]

IDK why I said right.

[isezgz]

jajajaj ke wno "a ver si se le va la pinza" XD

[gielfransen]

How lang did the rendering do?

[RAMIDUScz]

I'm really not sure if that is video or 3d graphics...wow, man

[HKurck]

Haha, nice use for the reactor :D

[phoebus1966]

some real investigation can be done with such tools - a lot of mysteries would be solved ... reports would make more sense ...

[ubermensch10000]

is it a game

so yes where can i download?

[gielfransen]

not a game, for animators and you have to pay 1,000$ for it

[Genomax2007]

Create your objects with a real scale (very important!), and ajust the gravity and colission tolerance.

[jakerismitology]

How do you make it to where they don't twitch and spas out?

[Genomax2007]

Are 350 objects with 5480 pol. Totally 1.918.000.

The power of reactor :D

[Qtittb]

HELL YEA!

Cheers,

Q.

[popoknight1]

????? what does that prove?

[l3xar]

i love the peg in the top left at the end, it just kind of rolls away on one end.

Amazing physics though. I really liked it, because i can appreciate how hard it would be :P

[keltroy]

one was like standing up sorta slanted while sliding

[RandomRobot1]

ROFL