 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I have an idea for a game that povers can play where one matches wits
against another for coming up with the best algorithm to solve the
following "golf" hole-in-one problem:
1. The course is an isosurface bounded by a sphere {0,600}
2. You start (tee off) at <-450, 5 units above surface, 0>
3. The hole is a cylinder of radius 15 cut into the surface at <450,0,0>
or difference{
[isosurface]
cylinder {<450,-1000,0>,<450,1000,0>,15}
}
4. Specify a vector to shoot your ball at, (a good first guess is
<8,0,0>)
5. The pov program will then animate your ball as it bounces across the
surface, and input back to your algorithm (INC) the closest it came to
the hole. This information is in the form of the vector from the closest
point on trajectory to the hole.
6. Your algorithm will use the information to choose a new trajectory.
7. The loop continues until one player's or the other's INC makes it in
the hole, limited to say 20 or 50 tries...
8. Your ball must fall in by the, say, 250th frame or it is a
stalemate. Your ball must also bounce at least once!
Players could choose a course of nine holes to play, with things ranging
from black holes to mindless putt-putt courses.
An animation could then be made from the tee off, the hole, etc.
The person who "wins" the most holes wins the match!
I've got the mechanics of the game down, as demonstrated in the
following animation. Here I had 200 balls of varying trajectories shot
at the target, and they bounce around on the surface. One finds the
hole! The animation actually starts at the 150th frame.
Now I'm going to work on the feedback loop given to the INC and my first
entry for the algorithm. At that time, I'll set up a web page with the
files.
Would I just be playing this by myself?
Post a reply to this message
Attachments:
Download 'bounce20.mpg' (642 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hmm .. I think, quite a very nice idea... but can't we make a little more
realism ... or are we plaing on Mars ??? with a bit more gravity it would
more realistic ...
Now OK .. the sense of the "game" would be (if I understand you right) to
hit the Target with quite a good algorithm and to have 20-50 tries ...
But would'nt it a possibility, to try (at home) some combinattion for the
parameters, find the perfect ones, and hardcode them in the .inc ... (of
course, dependent on the tourse to be played) ...
Now I think I have to let s.o. other to speek ...
So far, quite nice animation .. Did you do the bouncin' of the Object with
the trace() function ???
--
,', Jan Walzer \V/ http://wa.lzer.net ,',
',',' student of >|< mailto:jan### [at] lzer net ',','
' ComputerScience /A\ +49-177-7403863 '
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Fair criticism, perhaps I need a stronger gravity but also less recoil:
Vbounce / Vinit<1.0.
Perhaps it now looks like marbles bouncing off hard plastic--on Mars (tiny
lol).
Yes, the bounce is with the trace function.
Jan Walzer wrote:
> Hmm .. I think, quite a very nice idea... but can't we make a little more
> realism ... or are we plaing on Mars ??? with a bit more gravity it would
> more realistic ...
>
> Now OK .. the sense of the "game" would be (if I understand you right) to
> hit the Target with quite a good algorithm and to have 20-50 tries ...
>
> But would'nt it a possibility, to try (at home) some combinattion for the
> parameters, find the perfect ones, and hardcode them in the .inc ... (of
> course, dependent on the tourse to be played) ...
>
> Now I think I have to let s.o. other to speek ...
>
> So far, quite nice animation .. Did you do the bouncin' of the Object with
> the trace() function ???
>
> --
> ,', Jan Walzer \V/ http://wa.lzer.net ,',
> ',',' student of >|< mailto:jan### [at] lzernet ',','
> ' ComputerScience /A\ +49-177-7403863 '
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Actually, the ball is a point! I could have represented it with a sphere of
any size.
Serge LAROCQUE wrote:
> Looks like fun! The hole is rather big relative to the ball, but to be
> practical, if the hole was scaled to something of the order of the ball, it
> would be next to impossible to get a hole-in-one in a reasonable number of
> shots...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I don't play golf :-)
Maybe I didn't understand the rules to this idea but I think I'd just make a
translation
from start vector to end-point like so: translate
<-450+(900*Movement),5-(5*Movement),0> where Movement is a linear addition over time
in
a while loop or clock, for example.
Like I said, I didn't read very thoroughly. And I'm just joking around too.
Bob
"Greg M. Johnson" <gre### [at] my-dejanewscom> wrote in message
news:392E6D23.3A1756D3@my-dejanews.com...
| I have an idea for a game that povers can play where one matches wits
| against another for coming up with the best algorithm to solve the
| following "golf" hole-in-one problem:
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
BUT WAIT!
You specify ONLY an INITIAL vector! Kind of like an initial velocity. The ball is then
affected by gravity and bounces & rolls off the surface. Note the rolling hills:
there is
no straight line possible. Then, if /when you miss the hole, you get to specify a new
initial velocity.
In Putt-putt, you would be cheating if you specified the entire trajectory, if you
kept the
club touching the ball the whole time from tee to hole!
I spent a lot of time on this over the weekend. After I implemented Jan's suggestions,
I
found that the algorithm did not correctly handle a ball that had rolled to a stop.
Bouncing worked perfect, but a stationary ball kept falling through the surface.
Rendering
my nth fix as we speak...
Bob Hughes wrote:
> I don't play golf :-)
> Maybe I didn't understand the rules to this idea but I think I'd just make a
translation
> from start vector to end-point like so: translate
> <-450+(900*Movement),5-(5*Movement),0> where Movement is a linear addition over time
in
> a while loop or clock, for example.
> Like I said, I didn't read very thoroughly. And I'm just joking around too.
>
> Bob
>
> "Greg M. Johnson" <gre### [at] my-dejanewscom> wrote in message
> news:392E6D23.3A1756D3@my-dejanews.com...
> | I have an idea for a game that povers can play where one matches wits
> | against another for coming up with the best algorithm to solve the
> | following "golf" hole-in-one problem:
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Greg M. Johnson" <gre### [at] my-dejanewscom> wrote in message
news:3933A3FD.894E51D0@my-dejanews.com...
| BUT WAIT!
| You specify ONLY an INITIAL vector! Kind of like an initial velocity. The ball is
then
| affected by gravity and bounces & rolls off the surface. Note the rolling hills:
there is
| no straight line possible. Then, if /when you miss the hole, you get to specify a
new
| initial velocity.
Ah ha, okay, I see now. Similar to many such software games of golf I've seen, less
the
interface.
| In Putt-putt, you would be cheating if you specified the entire trajectory, if you
kept the
| club touching the ball the whole time from tee to hole!
You've seen me play then. Putt-putt I have done.
| I spent a lot of time on this over the weekend. After I implemented Jan's
suggestions,
I
| found that the algorithm did not correctly handle a ball that had rolled to a stop.
| Bouncing worked perfect, but a stationary ball kept falling through the surface.
Rendering
| my nth fix as we speak...
You also need a gopher popping out of the hole when a ball makes it in.
Bob
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Greg M. Johnson" wrote:
> I have an idea for a game that povers can play where one
> matches wits against another for coming up with the best
> algorithm to solve the following "golf" hole-in-one problem:
<snip>
> Would I just be playing this by myself?
When your game is ready, I would like to try and participate! :-)
Greetings,
Rune
---
Updated April 25: http://rsj.mobilixnet.dk
Containing 3D images, stereograms, tutorials,
The POV Desktop Theme, 350+ raytracing jokes,
miscellaneous other things, and a lot of fun!
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Did you ever finish this? I wann'a go!
"Greg M. Johnson" wrote:
>
> I have an idea for a game that povers can play where one matches wits
> against another for coming up with the best algorithm to solve the
> following "golf" hole-in-one problem:
>
> 1. The course is an isosurface bounded by a sphere {0,600}
> 2. You start (tee off) at <-450, 5 units above surface, 0>
> 3. The hole is a cylinder of radius 15 cut into the surface at <450,0,0>
>
> or difference{
> [isosurface]
> cylinder {<450,-1000,0>,<450,1000,0>,15}
> }
> 4. Specify a vector to shoot your ball at, (a good first guess is
> <8,0,0>)
> 5. The pov program will then animate your ball as it bounces across the
> surface, and input back to your algorithm (INC) the closest it came to
> the hole. This information is in the form of the vector from the closest
> point on trajectory to the hole.
> 6. Your algorithm will use the information to choose a new trajectory.
> 7. The loop continues until one player's or the other's INC makes it in
> the hole, limited to say 20 or 50 tries...
> 8. Your ball must fall in by the, say, 250th frame or it is a
> stalemate. Your ball must also bounce at least once!
>
> Players could choose a course of nine holes to play, with things ranging
> from black holes to mindless putt-putt courses.
> An animation could then be made from the tee off, the hole, etc.
> The person who "wins" the most holes wins the match!
>
> I've got the mechanics of the game down, as demonstrated in the
> following animation. Here I had 200 balls of varying trajectories shot
> at the target, and they bounce around on the surface. One finds the
> hole! The animation actually starts at the 150th frame.
>
> Now I'm going to work on the feedback loop given to the INC and my first
> entry for the algorithm. At that time, I'll set up a web page with the
> files.
>
> Would I just be playing this by myself?
>
> ------------------------------------------------------------------------
> Name: bounce20.mpg
> bounce20.mpg Type: Mplayer2 File (video/mpeg)
> Encoding: base64
--
signature{
"Grey Knight" contact{ email "gre### [at] yahoocom" }
site_of_week{ url "http://digilander.iol.it/jrgpov" }
}
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I had a problem with my bouncing algorithm. Everything looked great (in
tune with physics) on the first several bounces. After the ball lost its
"bounce" and was just rolling along, I found that at below a certain x-z
plane speed, the ball would fall through the surface of the earth!
Still a cool concept but my algo was no good for this.....
"Grey Knight" <s16### [at] namtarqubacuk> wrote in message
news:3C594B1E.55557781@namtar.qub.ac.uk...
> Did you ever finish this? I wann'a go!
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
