| View previous topic - View next topic |
| Author |
Message |
Jon Alma
Monkey-Butler
Joined: 09 May 2005
Posts: 50
Location: The Sunny South of France
|
 Posted: Fri Nov 07, 2008 7:38 am Post subject: Better Waterfalls? |
[quote] |
|
After a lengthy pause I'm back working on the Legends from the Lost Realms game.
One thing I am currently working on is introducing waterfalls. For the moment this makes use of the particle system with three particle sources being used,
- One particle source handling the actual waterfall itself
- A similar, but faster flowing source to add a bit of variety to the downward water flow and add a bit of turberlance just upstream of the waterfall.
- A mist source at the base of the waterfall
The overall result (so far) can be seen here.
The result so far is ok, but there is still plenty of room for improvement. For instance, while the waterfalls look really quite good from a distance, they also need to be equally good when viewed from up close ... most of the waterfalls won't be up to a Niagra Falls scale and are normally going to be small falls on streams with the player able to walk up to and into them.
Unfortunately up close they are still looking a little rough round the edges and I'd like to know if anyone knows of any tricks or tips to use in order to improve the overall look. For information, I would prefer to still with particle waterfalls as the code is already in place and at worst just needs to be tweaked. However, if there are other techniques (using animated textures or models for example) then I'd be interested in hearing about them.
Jon.
_________________
Legends from the Lost Realms
|
|
| Back to top |
|
 |
Captain Vimes
Grumble Teddy
Joined: 12 May 2006
Posts: 225
Location: The City Streets
|
| Posted: Mon Nov 10, 2008 7:11 pm Post subject: |
[quote] |
|
It looks pretty good, if a bit simple. I don't know exactly how a particle engine works since I've never done one, so feel free to ignore any of the below suggestions if they are stupid.
a) Increase the number of slow-moving particles. As of now it looks more like a stream of bubbles than a waterfall, which is still a cool effect.
b) Increase the speed of all particles a little. It looks like the river is in the middle of freezing.
_________________
"Sometimes it is better to light a flamethrower than to curse the darkness."
- Terry Pratchett
|
|
| Back to top |
|
|
Verious
Mage
Joined: 06 Jan 2004
Posts: 409
Location: Online
|
| Posted: Mon Nov 10, 2008 7:27 pm Post subject: |
[quote] |
|
|
The water should fall at approximately 9.81 meters/second (the speed of a falling object in a vacuum due to Earth's gravity).
|
|
| Back to top |
|
|
Rainer Deyke
Demon Hunter
Joined: 05 Jun 2002
Posts: 672
|
| Posted: Mon Nov 10, 2008 7:51 pm Post subject: |
[quote] |
|
| Verious wrote: | | The water should fall at approximately 9.81 meters/second (the speed of a falling object in a vacuum due to Earth's gravity). |
No, it should accelerate for an increase in speed of 9.81 meters/second every second. Probably less actually, since the water is not in a vacuum and air resistance is significant.
|
|
| Back to top |
|
|
BDZ
Slightly Deformed Faerie Princess
Joined: 11 Jan 2007
Posts: 32
Location: Wisconsin
|
| Posted: Tue Nov 11, 2008 3:31 am Post subject: |
[quote] |
|
| Verious wrote: | | The water should fall at approximately 9.81 meters/second (the speed of a falling object in a vacuum due to Earth's gravity). |
I find this ironic considering that the title under your name is "Stephen Hawking."
|
|
| Back to top |
|
|
DeveloperX
202192397
Joined: 04 May 2003
Posts: 1626
Location: Decatur, IL, USA
|
| Posted: Tue Nov 11, 2008 4:06 am Post subject: |
[quote] |
|
| BDZ wrote: | | Verious wrote: | | The water should fall at approximately 9.81 meters/second (the speed of a falling object in a vacuum due to Earth's gravity). |
I find this ironic considering that the title under your name is "Stephen Hawking." |
HAHAHAHAHAHA
_________________
Principal Software Architect
Rambling Indie Games, LLC
See my professional portfolio
|
|
| Back to top |
|
|
Verious
Mage
Joined: 06 Jan 2004
Posts: 409
Location: Online
|
| Posted: Tue Nov 11, 2008 3:28 pm Post subject: |
[quote] |
|
| Rainer Deyke wrote: | | No, it should accelerate for an increase in speed of 9.81 meters/second every second. Probably less actually, since the water is not in a vacuum and air resistance is significant. |
That's true, hence the reason I mentioned "approximately" and "in a vacuum". My point was merely to give a frame of reference for the speed of the falling water, since most 3D engines allow the terminal velocity to be specified for a given object (or system wide).
Keep in mind water flowing towards a waterfall will already have a high velocity since water accelerates towards a void/point of release.
| BDZ wrote: | | I find this ironic considering that the title under your name is "Stephen Hawking." |
For as long as I have been a member of this site, I have never really paid much attention to the titles, but that is a bit ironic.
|
|
| Back to top |
|
|
RedSlash
Mage
Joined: 12 May 2005
Posts: 331
|
| Posted: Tue Nov 11, 2008 6:28 pm Post subject: |
[quote] |
|
I totally agree with Rainer and Verious on this one. By accelerating your water particles at the precise speed of 9.81, you will make your already cool looking waterfall... even cooler. I also recommend multiplying it by the constant PI to 5 decimal places.
Other than that, I the only thing I can tell is the noticeable repeating pattern of the water. I think adding a bit more randomness would solve this problem.
|
|
| Back to top |
|
|
Jon Alma
Monkey-Butler
Joined: 09 May 2005
Posts: 50
Location: The Sunny South of France
|
| Posted: Wed Nov 12, 2008 7:26 am Post subject: |
[quote] |
|
Thanks for all the feedback - much appreciated.
| RedSlash wrote: | | I totally agree with Rainer and Verious on this one. By accelerating your water particles at the precise speed of 9.81, you will make your already cool looking waterfall... even cooler. I also recommend multiplying it by the constant PI to 5 decimal places. |
The particles are already being accelerated by 10m/s - not too noticable on a small waterfall and hopefully not too far from the 9.81m/s real wordl figure ... I really hope that someone won't decide real role playing involves taking a stopwatch and measuring the rate of fall of individual droplets in all the waterfalls in the game world ...
| RedSlash wrote: | | Other than that, I the only thing I can tell is the noticeable repeating pattern of the water. I think adding a bit more randomness would solve this problem. |
The particle sources should be pretty random, but with a high number of particles (as in the waterfalls) patterns do appear (due to fluctuations in the framerate which influences the number of new particles added).
I'm going to try to randomise the vertical position of the particles a bit more as this should break up the patterns a bit more. I might also add another particle source with faster flowing particles to see if this also breaks up the consistency a bit as the water falls downwards.
_________________
Legends from the Lost Realms
|
|
| Back to top |
|
|
Hajo
Demon Hunter
Joined: 30 Sep 2003
Posts: 779
Location: Between chair and keyboard.
|
| Posted: Thu Nov 13, 2008 2:23 pm Post subject: |
[quote] |
|
| Jon Alma wrote: | | I might also add another particle source with faster flowing particles to see if this also breaks up the consistency a bit as the water falls downwards. |
I once experimented with a fountain ( http://www.simugraph.com/smf/index.php?topic=54.msg92#msg92 ) and it's been pretty tricky to get a proper flow of the drops. That's a particle system, too, even if only a still render.
What helped there was to raise the size of the particles during their journey. No idea if that helps with a waterfall, too, but you can try.
|
|
| Back to top |
|
|
Jon Alma
Monkey-Butler
Joined: 09 May 2005
Posts: 50
Location: The Sunny South of France
|
| Posted: Thu Nov 13, 2008 2:37 pm Post subject: |
[quote] |
|
After playing about it really does come down to the flow of the particles. Up til now this has been too regular for my waterfalls, but I've just added the ability to vary the initial speed of the particles and this seems to be really breaking up the regularity of the drops falling.
That and an improved texture for the waterfall means I'm getting closer to what I want. If I can tweak a last few bits and pieces I'll post a new video for feedback and comparison.
_________________
Legends from the Lost Realms
|
|
| Back to top |
|
|
Hajo
Demon Hunter
Joined: 30 Sep 2003
Posts: 779
Location: Between chair and keyboard.
|
| Posted: Thu Nov 13, 2008 2:53 pm Post subject: |
[quote] |
|
| Jon Alma wrote: |
After playing about it really does come down to the flow of the particles. Up til now this has been too regular for my waterfalls, but I've just added the ability to vary the initial speed of the particles and this seems to be really breaking up the regularity of the drops falling. |
In my fountain experiments, I randomized the starting vectors. That is, not only have the drops different speeds (length of the vector) but also slightly different directions. I assume you tried that, too?
Good to hear though that you make progress. Looking forward to a video :)
|
|
| Back to top |
|
|
Jon Alma
Monkey-Butler
Joined: 09 May 2005
Posts: 50
Location: The Sunny South of France
|
| Posted: Fri Nov 14, 2008 8:09 am Post subject: |
[quote] |
|
| Hajo wrote: | In my fountain experiments, I randomized the starting vectors. That is, not only have the drops different speeds (length of the vector) but also slightly different directions. I assume you tried that, too?
Good to hear though that you make progress. Looking forward to a video :) |
Well in response to popular demand (?) a new video can be seen here
I'm actually pretty happy with this and other than a bit of tweaking (mainly to optimise the number of particles so the framerate on slower machines is reasonable).
The thing that made the difference is exactly what you suggest - I randomised the starting vectors so the particles fall at different speeds and fall in slightly different directions. The different vertical speeds break up the pattern of the water (as does reducing the number of particles) while the different horizontal movement gives the impression of water hitting rocks and so on.
_________________
Legends from the Lost Realms
|
|
| Back to top |
|
|
RedSlash
Mage
Joined: 12 May 2005
Posts: 331
|
| Posted: Fri Nov 14, 2008 8:15 am Post subject: |
[quote] |
|
|
I think it is a vast improvement over the last one. The pattern is less noticeable. Now there one minor point.. maybe its not that important, but it looks like the water is jumping over the rocks. But it's great!
|
|
| Back to top |
|
|
Verious
Mage
Joined: 06 Jan 2004
Posts: 409
Location: Online
|
| Posted: Sun Feb 22, 2009 2:30 pm Post subject: |
[quote] |
|
One suggestion, perhaps you could add some particles moving away from the base of the waterfall on the horizontal plane to simulate the whitewater immediately after a waterfall, currently the water looks a bit too calm.
The new video with the three waterfalls looks great.
|
|
| Back to top |
|
|
