Martin Eddy
Regular
Chalnoth said:
People used to say that about 3d Graphics Cards.
Computer generated water
- Thread starter carpediem
- Start date
People used to say that about 3d Graphics Cards.
Well, I'd just like to mention that the other video I found to be vastly more impressive than the shark video. But I don't think that the shark video was as unrealistic as you claim.Basic said:
Bear in mind that there were no bubbles except when the shark broke the surface of the water. Granted, there may still be too many, but that's not an experiment I could perform easily
But yeah, the smaller sharks seemed to be moving through the water a bit more quickly than they should have been...
It's a differen situation, and I don't want to get into it here, but just remember that with graphics cards, if a game supports them at all, provide a tremendous visual quality boost. There's no such automatic boost with physics hardware.madmartyau said:People used to say that about 3d Graphics Cards.
Most bubbles were from breaking the water surface, which is OK. But at the same time, there were a lot of bubbles from parts that didn't break the water surface. That must have come from cavitation, and I doubt the worlds most streamlined animal would generate that amount of cavitation.
But it was the other part that looked worst. Not "Deep Blue See"-bad but still unrealistic.
I have a feeling that the parts I don't like is purely artist driven. The water is modelled with correct physics, but the shark movement is modelled by artists. And then the artist wanted a method to add more bubbles to give a feeling of speed.
But I guess it's best to reiterate one thing:
The water physics looks realy good apart from that.
But it was the other part that looked worst. Not "Deep Blue See"-bad but still unrealistic.
I have a feeling that the parts I don't like is purely artist driven. The water is modelled with correct physics, but the shark movement is modelled by artists. And then the artist wanted a method to add more bubbles to give a feeling of speed.
But I guess it's best to reiterate one thing:
The water physics looks realy good apart from that.
All you need for water is enormous amounts of particles. You can simulate splashes using poligonal pixel shaded surfaces but for realistic splashes you also need small particles that are also created in real life (drops). Water was always the most difficult element to render (i mean moving water, not still pool of water in a small area).
Waterfalls, rivers, water splashing of boats and cars that drive over water... These were always terrible (even techdemos for Ageia PhysiX were ugly and looked like some foam not water).
Waterfalls, rivers, water splashing of boats and cars that drive over water... These were always terrible (even techdemos for Ageia PhysiX were ugly and looked like some foam not water).
Basic said:
yes, i agree, the propulsion was, erm, fishy. but re bubbles, i thought the majority of the bubbles were coming from the gills - i would expect them to produce exactly this effect if the animal just re-submerged after surfacing. so i assume it just did that ; )
Last edited by a moderator:
Kanyamagufa
Regular
Basic said:
Okay, okay, sue them because the sharks don't feel 100% real. It is, after all, a water demo, not an extinct shark demo. (Just bustin' your balls.
)Watch the second reel instead.
I was blown away, personally.
Seriously question: has anyone done a water particle simulator that only bothers with two dimensions? (not unlike the idea of the 2D ray tracer i believe someone here working on)
You could probably do a neat oil/blue water simulation like those toys where you turn it and the oil is forced through a channel, turning a wheel, etc.
You could probably do a neat oil/blue water simulation like those toys where you turn it and the oil is forced through a channel, turning a wheel, etc.
Mate Kovacs
Newcomer
A Bézier-Based Approach to Unstructured Moving Mesheszsouthboy said:
This seems to be applicable to highly viscous fluids, and extensible for general fluid dynamics.
