pipmak-users — Content creation for the Pipmak Game Engine

[Pipmak-Users] Re: Blureffect in MYST4

[Christian W. <cwa...@gm...>]

hOSHI wrote:
> looking forward to it ;)
> i'd like to display my pipmak project to other users,
> what about a section on the official site?

In my project application, I promised to SourceForge that I wouldn't 
host other people's Pipmak projects (which could potentially be big) on 
their servers. But I can certainly make a links section. Can you find a 
place to host your project yourself and send me a link?

  -Christian

[Pipmak-Users] Re: Blureffect in MYST4

[hOSHI <ho...@ga...>]

looking forward to it ;)
i'd like to display my pipmak project to other users,
what about a section on the official site?

greetz again...

hSH*

[Pipmak-Users] Re: Pipmak-Users digest, Vol 1 #13 - 2 msgs

[Jeff M. t. G. <gua...@vi...>]

Glad to here the development of pipmak continues. Jeff
-- 
-----------------------------------------------
EYE WREATHED IN PAIN
The eye, it's always watching.
It changes colors, change angles, changes stance.
It is more than an eye wreathed in pain,
More than an eye staring blankly in the dark.
It's a mind, a face, a thought.

It's more than you will ever dream of.

It's beyond your comprehension. And it is frightful.
For it has scared the few who have seen it.
And now they are a blank wall. Yet it burns with longing to see.
To comprehend, to love, for the blank walls are like scars to it.
Failed attempts, living evidence.

                    That all it can do is fail. For the eye has spoken.
The tear has fallen. The end is coming, the flame is dying.
The light is fading. The burnt spot growing.

			Jeff McArthur
			Poems From The Raging Heart
			Full collection at http://jeff.bluegillweb.com
			2002 - 2005

Re: [Pipmak-Users] Blureffect in MYST4

[Christian W. <cwa...@gm...>]

> I just finished MYST4 Revelation. Great Game.
> It used depth of field effects in realtime, i think
> via z-buffer calculation...
> That would be a very cool feature for pipmak, don't you think?

Sure. But I think I'll implement the *essential* features sooner than 
the *cool* ones. :) And besides, I currently don't have any graphics 
hardware capable of fragment programs, so I couldn't test it even if I 
wrote such a thing...

By the way, I'm indeed working on Pipmak, even if there's no sign of it 
from the outside. The goal is a complete internal refactorization that 
will make the code more flexible and easier to maintain - if it works 
out, which I can't tell yet right now.

  -Christian

[Pipmak-Users] Blureffect in MYST4

[hOSHI <ho...@ga...>]

Hi,
I just finished MYST4 Revelation. Great Game.
It used depth of field effects in realtime, i think
via z-buffer calculation...
That would be a very cool feature for pipmak, don't you think?

greetinx

hOSHI*

[Pipmak-Users] Pipmak 0.2.1 released

[Christian W. <cwa...@gm...>]

Pipmak 0.2.1 was released today. The main new feature (which has been in 
the CVS version for a while) is timers for simple animation or other 
time-dependent behavior. The demo project has also been updated to show 
these in action.

Also, in addition to the Mac OS X and Windows versions, Linux binaries 
for PPC and x86 are now available.

Get it from <http://pipmak.sourceforge.net/downloads.php>, or read the 
detailed release notes at 
<http://sourceforge.net/project/shownotes.php?release_id=315788>.

  -Christian

[Pipmak-Users] Re: Hotspot maps

[Will M. <lea...@ho...>]

excellent.  found it.  thanks!!!
i'm new to the Mac, and the find files function didn't locate that file,
but now that i know where to look, 
it's all good.

cheers,
will

Re: [Pipmak-Users] Re: Hotspot maps

[Christian W. <cwa...@gm...>]

Will Meyer wrote:
> unfortunately, while i'm no longer receiving any messages,
> i cannot find the equirect.bmp anywhere on my hard drive.

As I had suspected, it's in Pipmak.app/Contents/MacOS/, next to the 
executable. Use "Show package contents" in the contextual menu on 
Pipmak to get there. I agree that's not a very convenient location, but 
the whole thing is a quick and dirty hack.

  -Christian

[Pipmak-Users] Re: Hotspot maps

[Will M. <lea...@ho...>]

i would agree that changing the keyhandler would be a little more intuitive.
unfortunately, while i'm no longer receiving any messages, 
i cannot find the equirect.bmp anywhere on 
my hard drive.

cheers,
will

[Pipmak-Users] CVS Documentation Snapshot

[Christian W. <cwa...@gm...>]

At <http://pipmak.sourceforge.net/Reference.pdf>, you can now find a 
recent snapshot of the documentation that is more comprehensive than the 
version included with Pipmak 0.2.0. I'm making this available because it 
could be helpful for people who don't want to go through the hassle of 
getting the latest documentation from CVS and TeXing it themselves, 
while I don't have time to make a full Pipmak release right now. It is 
still incomplete, and be aware that it documents some features that 
aren't in 0.2.0, but only in the CVS yet.

  -Christian

Re: [Pipmak-Users] Re: Hotspot maps

[Christian W. <cwa...@gm...>]

Will Meyer wrote:
> ok, a returns 97 and the space bar returns 32.

OK, so we were not on the right track...

> if key == string.byte("a") then
> pipmak.print("ok")
>
> now, when i press a, it prints "ok," and then it says "Unused key 97 
> pressed,"
> and the normal key commands still don't work.

Well, I finally found the problem. I made a mistake when I wrote that 
code off the top of my head without testing it. The onkeydown handler 
needs to return false (or nothing) if it handled the keypress itself, 
and true if it has to be passed on to the global handler, not the other 
way around. Like so:

onkeydown (
   function(key)
     if key == string.byte(" ") then
       pipmak.saveequirect(360, 180) --width, height
       return false --keypress was handled
     else
       return true --keypress needs to be passed along to the global key 
handler
     end
   end
)

My bad! Seeing that it confused myself, I could maybe change that. 
Since it's not documented  yet, I figure it could be changed without 
breaking too many existing projects. What do you think, is it more 
intuitive to have the keydown handler return whether the keydown needs 
to be passed on (as it is now), or whether it has handled the keypress 
(as I was assuming when I wrote the faulty code)?

By the way, I'm not sure (without testing) where pipmak.saveequirect() 
saves its image file in the current version. It might be next to the 
application, or it might be somewhere inside the application (use "Show 
package contents" in the contextual menu in the Finder to find out).

  -Christian

[Pipmak-Users] Re: Hotspot maps

[Will M. <lea...@ho...>]

i am altering node.lua and not main.lua.

[Pipmak-Users] Re: Hotspot maps

[Will M. <lea...@ho...>]

ok, a returns 97 and the space bar returns 32.

alright.  i also tried something else.

i replaced:
if key == string.byte("a") then 
pipmak.saveequirect(360, 180)

with:
if key == string.byte("a") then 
pipmak.print("ok")

now, when i press a, it prints "ok," and then it says "Unused key 97 pressed," 
and the normal key 
commands still don't work.

thanks,
will

Re: [Pipmak-Users] Re: Hotspot maps

[Christian W. <cwa...@gm...>]

> when i do the print Bingo trick, it prints bingo whatever key i press

Yeah, that's what's supposed to happen. Good.

> if key == string.byte("a") then
> and it says "unused key 97 pressed" when i press a.

OK, we're on the right track. Seems to be a bug in Pipmak indeed. 
string.byte("a") is supposed to return 97, but apparently it doesn't. 
Can you find out what it returns instead? For example, replace the 
'pipmak.print("Bingo!")' line with 'pipmak.print("The magic number is 
", string.byte("a"), "!")'.

> when i tried your before and after trick, i got both messages.

OK, fine.

> btw, i do appreciate you taking the time to help with this.

Well, I do appreciate you taking the time to help me squashing bugs (or 
design flaws) in Pipmak... :)

  -Christian

Re: [Pipmak-Users] Re: Hotspot maps

[Christian W. <cwa...@gm...>]

Will Meyer wrote:

> when i use (), it doesn't give me an error line, but i can no longer 
> ESC out of the game

And you are in node.lua, not in main.lua? Because when placed in 
main.lua, an onkeydown handler replaces the default global handler that 
handles ESC and the other keys, but it shouldn't do that when placed in 
node.lua. If it does, that's a bug in Pipmak.

  -Christian

[Pipmak-Users] Re: Hotspot maps

[Will M. <lea...@ho...>]

when i do the print Bingo trick, it prints bingo whatever key i press,
and as far as i can tell it doesn't do anything else.
none of the normal key commands work.  the reload command doesn't
work so i close and reopen the program.
when i press the spacebar, it says bingo, and then says "unused key 32 pressed."
with other keys, it just says bingo.
i tried replacing 
if key == string.byte(" ") then 
with
if key == string.byte("a") then
and it says "unused key 97 pressed" when i press a. 
when i tried your before and after trick, i got both messages.
btw, i do appreciate you taking the time to help with this.

thanks,
will

[Pipmak-Users] Re: Hotspot maps

[Will M. <lea...@ho...>]

> > i'm also getting an error saying:
> > bad argument #1 to 'onkeydown' (function expected, got table)
> 
> That's only when you replace () with {}, right? Because stuff in {}s is 
> a table, and onkeydown {foo} is a shortcut for onkeydown({foo}). 
> "onkeydown" is a function that takes another function (the one to be 
> called on keypresses) as an argument. The ()s are correct here.
> 
>   -Christian

yes, that is only when i used {}.  when i use (),
it doesn't give me an error line, but i can no longer ESC 
out of the game and pressing the spacebar gives me an 
"Unused key 32 pressed" message.

by the way, i am running this on OSX.

thanks,
will

Re: [Pipmak-Users] Hotspot maps

[Christian W. <cwa...@gm...>]

Will Meyer wrote:

> ok. ok.  i'm no longer getting that error after retyping the code.
> cutting and pasting the code
> introduced weird line end symbols (or at least that's my theory).
> unfortunately, with the code in there, pressing the space bar only
> result in a message saying "unused
> key 32 pressed."
> any ideas?

Hmm. You are in the node whose node.lua contains the onkeydown handler 
(press "N" to have Pipmak display the number of the current node), and 
you did reload (or leave and re-enter) the node after saving the 
changes to node.lua?
Is the keydown handler function called at all when you press a key? You 
can check by inserting a print statement, like so:

onkeydown (
   function(key)
     pipmak.print("Bingo!")
     if key == string.byte(" ") then
       ...

The whole node.lua file, although it looks like a data file, is 
actually a Lua script that is executed when the node is loaded. All the 
configuration keywords like "cubic", "hotspotmap", "onkeydown" are 
actually Lua functions that manipulate the internal state of the 
engine. Is the "onkeydown" function that installs the keydown handler 
called on reloading the node, as it should? To check, use print 
statements before and after it:

pipmak.print("Before!")
onkeydown (
   ...
)
pipmak.print("After!")

If these tests don't lead us to a solution, there might be something 
wrong in Pipmak itself, and I'd have to find and fix that.

  -Christian

Re: [Pipmak-Users] Re: Hotspot maps

[Christian W. <cwa...@gm...>]

Will Meyer wrote:

> onkeydown (
> =A0function(key)
> =A0 =A0if key =3D=3D string.byte(" ") then
> =A0 =A0 =A0pipmak.saveequirect(360, 180) --width, height
> =A0 =A0 =A0return true --keypress was handled
> =A0 =A0else
> =A0 =A0 =A0return false --keypress needs to be passed along to the =
global
> key handler from defaults.lua
> =A0 =A0end
> =A0end
> )
>
> to the tail of the node.lua file.  i also tried replacing the
> first and last ( )with {}.  either way, it causes the
> node not to work.  i get an error message telling me
> Error loading lua file ["1/node.lua"]:25:unexpected symbol near'

Sounds strange. Which line is line 25? Did you copy-and-paste that text=20=

from the web browser? There might be some mysterious invisible=20
character somewhere that confuses the Lua interpreter. Try removing=20
line 25 (including the line breaks before and after it) and re-typing=20
it by hand.

> i'm also getting an error saying:
> bad argument #1 to 'onkeydown' (function expected, got table)

That's only when you replace () with {}, right? Because stuff in {}s is=20=

a table, and onkeydown {foo} is a shortcut for onkeydown({foo}).=20
"onkeydown" is a function that takes another function (the one to be=20
called on keypresses) as an argument. The ()s are correct here.

  -Christian=

[Pipmak-Users] Re: Hotspot maps

[Will M. <lea...@ho...>]

i'm also getting an error saying:

bad argument #1 to 'onkeydown' (function expected, got table)

[Pipmak-Users] Re: Hotspot maps

[Will M. <lea...@ho...>]

ok. ok.  i'm no longer getting that error after retyping the code.
cutting and pasting the code 
introduced weird line end symbols (or at least that's my theory).
unfortunately, with the code in there, pressing the space bar only
result in a message saying "unused 
key 32 pressed."
any ideas?

cheers,
will

[Pipmak-Users] Re: Hotspot maps

[Will M. <lea...@ho...>]

ok.  i understand that the hotspot maps to a sphere,
sort of like a rectangular map of the earth.
unfortunately, i can't get your hack to work.
i added

onkeydown (
 function(key)
   if key == string.byte(" ") then
     pipmak.saveequirect(360, 180) --width, height
     return true --keypress was handled
   else
     return false --keypress needs to be passed along to the global
key handler from defaults.lua
   end
 end
)


to the tail of the node.lua file.  i also tried replacing the
first and last ( )with {}.  either way, it causes the 
node not to work.  i get an error message telling me 
Error loading lua file ["1/node.lua"]:25:unexpected symbol near'

i'm reading up on lua scripting to see if i can fix this myself,
but if you know what my problem is off the 
top of your head i'd appreciate it.


cheers,
will

Re: [Pipmak-Users] Hotspot maps

[Christian W. <cwa...@gm...>]

> Right now I can't figure out how the hotspot maps map to a cubic  
> panorama.
> As far as I can tell, the hotspot is a single image file.
> Does this image somehow map to an unwrapped cube?

That's a good question, and the answer unfortunately isn't in the  
documentation shipped with 0.2.0 yet. Here's what I wrote about it in  
the current CVS version of the documentation:

> Hotspot maps are 256-color (indexed) images. Color index 0 means  
> background, color index 1 is the first hotspot, index 2 the second  
> etc. The actual colors don’t matter, so you can use any color palette.  
> A particularly suitable palette with easily distinguishable colors is  
> included in the source distribution of Pipmak in hotspot-palette.gif  
> in the extras folder.
> For slide nodes, the image is stretched to cover the whole slide (it  
> need not have the same size or aspect ratio).
> For panoramic nodes, the hotspot maps are equirectangular, that means  
> that azimuth (0..360°) and elevation (-90°..90°) in the panorama are  
> mapped to the horizontal and vertical axes in the image. The horizon  
> in the panorama is mapped to the vertically centered horizontal line  
> in the image, the point directly above the viewer (zenith) is  
> stretched out along the whole top edge of the image, and  
> correspondingly the point directly below (nadir) along the bottom  
> edge. For example, if your hotspot map is 360 × 180 pixels, then the  
> pixel at (68, 52) is what you see 38° (= 90° − 52°) above the horizon  
> after you turn 68° to the right from the initial orientation. See the  
> pipmak.saveequirect function in section 3.2.8 for a preliminary hack  
> to help you draw equirectangular hotspot maps.

Does that help? (If not, please tell me what points are not clear, so  
that I can improve the documentation.)

Here's how I answered this question before (and maybe other useful  
hints):  
<http://www.mystcommunity.com/board/index.php? 
showtopic=15120&view=findpost&p=241469>

Equirectangular hotspot maps have the advantage of consisting of only  
one file. Cubic hotspot maps, i.e. six images that coincide with the  
face images, are on the to-do list, but aren't implemented yet.

I don't have time for preparing a full release that would include the  
updated documentation right now, but I'll probably post just the  
documentation somewhere on the Pipmak home page in a few days. I'll  
announce that here.

  -Christian

[Pipmak-Users] Hotspot maps

[Will M. <lea...@ho...>]

Hi, I'm trying to use your engine to make a 
walk through of a world's fair exhibit.
I'm figuring things out slowly since I have very little
scripting experience.
Right now I can't figure out 
how the hotspot maps map to a cubic panorama.
As far as I can tell, the hotspot is a single image file.
Does this image somehow map to an unwrapped cube?
Any assistance would be greatly appreciated.

cheers,
will

Re: [Pipmak-Users] Cube panorama picture quality

[Christian W. <cwa...@gm...>]

<gua...@vi...> wrote:

> I was wondering if the cube panoramas always looks like the images 
> have been blown up 50%?

The size of the image pixels on your screen depends on three factors:

o resolution of the face images (512x512 in your case),
o field of view,
o window size.

So, what you can do to make the image sharper (the pixels smaller), is:

o Make the window smaller - probably not an option

o Increase the field of view (zoom out) by pressing X or by calling 
pipmak.setverticalfov() in Lua - probably not an option either since a 
bigger field of view means more perspective distortion (in other words, 
you have to be nearer to the screen for it to look right).

o Increase the face image resolution. You can use 1024x1024 or even 
2048x2048 (if your graphics card supports it), intermediate sizes are 
not possible at the moment. The problems here are that loading bigger 
images takes longer, and that your graphics card might run out of 
texture memory (6*512*512*3 bytes = 4.5 MB, 6*1024*1024*3 bytes = 18 
MB, 6*2048*2048*3 bytes = 72 MB). You can also use different 
resolutions for different faces, e.g. 512 for the top and bottom and 
1024 for the sides.

I'm sorry this stuff isn't in the documentation yet. If you are 
comfortable with CVS and TeX, you can get a slightly more comprehensive 
version of the documentation than the one included with 0.2.0 from CVS. 
I don't have time to make a new release at the moment.

Hope that helps - I know the problem, but there is no real solution. 
You have to find compromises. Full panoramas just do take a lot of 
space.

  -Christian