 | | | Dir3d-l Digest, Vol 8, Issue 19 | Dir3d-l Digest, Vol 8, Issue 19 2004-01-14 - By dir3d-l-request@(protected)
Back
Send Dir3d-l mailing list submissions to
dir3d-l@(protected)
To subscribe or unsubscribe via the World Wide Web, visit
http://nuttybar.drama.uga.edu/mailman/listinfo/dir3d-l
or, via email, send a message with subject or body 'help' to
dir3d-l-request@(protected)
You can reach the person managing the list at
dir3d-l-owner@(protected)
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Dir3d-l digest..."
Today's Topics:
1. Re: Dir 3D pathfinding (Elia Morling)
2. Re: Dir 3D pathfinding (webmaster@(protected))
3. Windoze and my evil little cursor (Allen Partridge)
4. Re: Dir 3D pathfinding (Neto)
5. Barry goes to Mars? (grimmwerks)
6. Re: Windoze and my evil little cursor (noisecrime)
7. Re: Barry goes to Mars? (Warren Ockrassa)
8. RE: Windoze and my evil little cursor (Allen Partridge)
9. Re: Windoze and my evil little cursor (noisecrime)
10. Re: Dir 3D pathfinding (webmaster@(protected))
11. Re: Windoze and my evil little cursor (Lucas Meijer)
-- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- --
Message: 1
Date: Tue, 13 Jan 2004 18:05:58 +0100
From: "Elia Morling" <elia@(protected)>
Subject: Re: [Dir3d-l] Dir 3D pathfinding
To: <dir3d-l@(protected)>
Message-ID: <008c01c3d9f7$8394c360$680aa8c0@(protected)>
Content-Type: text/plain; charset="iso-8859 (See http://iso-8859.ora-code.com)-1"
> Anyone know of a pathfinding algorithm for Director 3D? I'm sure it has been
> done..perhaps it's time to write my own..but even if I do, it would be nice
to
> see a successful one already scripted
This might help. http://www.tildruin.com/sourcecode.asp
Elia
-- ---- ---- ---- ---- ---- --
Message: 2
Date: Tue, 13 Jan 2004 13:16:18 -0500
From: webmaster@(protected)
Subject: Re: [Dir3d-l] Dir 3D pathfinding
To: "dir3d-l@(protected)" <dir3d-l@(protected)>
Message-ID: <1074017778.400435f2a6d9e@(protected)>
Content-Type: text/plain; charset=ISO-8859 (See http://ISO-8859.ora-code.com)-1
Neto and Elia...thanks. I like Neto's because of the potential to find things
in 3 dimensions. It seems as if Neto's could go up ramps, if needed.
Yeah I think some of us should pitch in a buy you a new mouse or something if
you found that code! Thanks.
\
Quoting Elia Morling <elia@(protected)>:
>
> This might help. http://www.tildruin.com/sourcecode.asp
>
> Elia
-- ---- ---- ---- ---- ---- ---- ---- ---- ------
This mail sent through IMP: http://horde.org/imp/
-- ---- ---- ---- ---- ---- --
Message: 3
Date: Tue, 13 Jan 2004 14:06:39 -0500
From: "Allen Partridge" <partridge@(protected)>
Subject: [Dir3d-l] Windoze and my evil little cursor
To: <dir3d-l@(protected)>
Message-ID: <001201c3da08$5fef55e0$d1fea8c0@(protected)>
Content-Type: text/plain; charset="us-ascii"
So I'm working on this game... There are some 3D things and some
particles and I begin to notice something very peculiar. In a browser -
or in authoring, when the windows cursor (which has been set to
invisible) enters the stage rect, the performance of the particle
systems is devastated, and when the cursor leaves the screen, the
performance is fine.
The game converts mouse movements to a 3D target, which is offset from
the mouse. We've enabled and disabled virtually every feature, and the
only noticeable performance hit remaining in the game -- is the
persistant evil caused by the windows cursor - even though the (*&#^^%#(
isn't even visible.)
So... assuming I haven't overlooked something obvious, is there any way
to terminate windows interest in drawing to the screen? It seems pretty
clear that the OS is slowing things down in this case. In lieu of that
any clever methods for keeping the cursor away from my game?
--al
-- ---- ---- ---- ---- ---- --
Message: 4
Date: Tue, 13 Jan 2004 17:32:15 -0300
From: "Neto" <neto@(protected)>
Subject: Re: [Dir3d-l] Dir 3D pathfinding
To: <dir3d-l@(protected)>
Message-ID: <002b01c3da14$54d1e9f0$1501b2aa@(protected)>
Content-Type: text/plain; charset="iso-8859 (See http://iso-8859.ora-code.com)-1"
http://www.station-zero.com/medion/demos/pathfind.zip
You'll surely need to modify it to use in your projects. There is a .MAX
file included with the scene, there you can see how the path nodes are setup
(check the user properties - where the node sectors are definied).
I'll try to make it more user friendly, when I have some free time.
Low_budget + tight_deadlines + HUGE_game = Total_madness
-- -- Original Message -- --
From: <webmaster@(protected)>
To: <dir3d-l@(protected)>
Sent: Tuesday, January 13, 2004 3:16 PM
Subject: Re: [Dir3d-l] Dir 3D pathfinding
> Neto and Elia...thanks. I like Neto's because of the potential to find
things
> in 3 dimensions. It seems as if Neto's could go up ramps, if needed.
>
> Yeah I think some of us should pitch in a buy you a new mouse or something
if
> you found that code! Thanks.
>
>
> \
> Quoting Elia Morling <elia@(protected)>:
> >
> > This might help. http://www.tildruin.com/sourcecode.asp
> >
> > Elia
>
>
>
> -- ---- ---- ---- ---- ---- ---- ---- ---- ------
> This mail sent through IMP: http://horde.org/imp/
>
> __ ____ ____ ____ ____ ____ ____ ____ ____ ____
> Dir3d-l mailing list
> Dir3d-l@(protected)
> http://nuttybar.drama.uga.edu/mailman/listinfo/dir3d-l
>
-- ---- ---- ---- ---- ---- --
Message: 5
Date: Tue, 13 Jan 2004 17:24:01 -0500
From: grimmwerks <grimm@(protected)>
Subject: [Dir3d-l] Barry goes to Mars?
To: "dir3d-l@(protected)" <dir3d-l@(protected)>
Message-ID: <BC29DA31.140BC%grimm@(protected)>
Content-Type: text/plain; charset="US-ASCII"
Hey Barry -
I was just perusing Richard Hoagland's site after listing to an Art Bell
mp3....
...and I see some of your noodlings??!
-- ---- ---- ---- ---- ---- --
Message: 6
Date: Tue, 13 Jan 2004 23:06:21 -0000
From: "noisecrime" <noise@(protected)>
Subject: Re: [Dir3d-l] Windoze and my evil little cursor
To: <dir3d-l@(protected)>
Message-ID: <006e01c3da29$dbfa15a0$0100a8c0@(protected)>
Content-Type: text/plain; charset="iso-8859 (See http://iso-8859.ora-code.com)-1"
-- -- Original Message -- --
From: "Allen Partridge" <partridge@(protected)>
> So I'm working on this game... There are some 3D things and some
> particles and I begin to notice something very peculiar. In a browser -
> or in authoring, when the windows cursor (which has been set to
> invisible) enters the stage rect, the performance of the particle
> systems is devastated, and when the cursor leaves the screen, the
> performance is fine.
Well first off nothing obvious springs to mind, and without seeing the game
or the code its going to be hard to pin down. However i'm hard pressed to
beleive this is purely an OS issue, if for the simple reason that i've never
heard of a problem like this before.
I have to say I feel the problem is more likely to be somewhere in the code,
and with your experience its most likely to be something simple thats been
overlooked, otherwise you would have fixed it already ;)
You mention converting mouse movements to a 3D target, is this using
modelsunderray or its ilk? If so then I would flag that up as a possibiltiy
as it tallies with the description of when the performanc suffers (i.e. when
you move the cursor into the stage rect). Was this one of the features you
disabled? Are you positive you disabled it completely?
Anyway to dig a little deeper some questions
Does the performance still suffer if the cursor is visible or only if it is
invisible?
Does it still suffer if you display any cursor other than 200 (assuming you
use Director to hide it in the first place)
How do you control the visibility of the cursor - director or an xtra?
What happens if you disable the particles?
Is any other software running, are you networked etc?
Have you tested this on several machines?
There is one piece of information that might be relevant. For a long time I
used to expeirence Director (authoring) trying to quit when pressing just
the 'q' key (no ctrl). I linked it to sw3d early on, as most of the tie its
all I work with. However it wasn't until a couple of months ago that I
worked out how its caused. Basically if you rightclick on a sw3d sprite
whist authoring is running and at some point after press the q key, Director
is given the instruction to quit!
This was on win98, but i think i tested on Xp too.
So perhaps there is an issue with sw3d and and an invisible cursor, but i
think another contributing factor would have to be present, as several other
developers including myself have hidden the cursor with no ill effect.
Noisecrime 2004
-- ---- ---- ---- ---- ---- --
Message: 7
Date: Tue, 13 Jan 2004 18:06:20 -0600
From: Warren Ockrassa <warren@(protected)>
Subject: Re: [Dir3d-l] Barry goes to Mars?
To: dir3d-l@(protected)
Message-ID: <7B17BD26-4625-11D8-A914-00039345DD44@(protected)>
Content-Type: text/plain; charset=US-ASCII; format=flowed
On Jan 13, 2004, at 4:24 PM, grimmwerks wrote:
> Hey Barry -
>
> I was just perusing Richard Hoagland's site after listing to an Art
> Bell
> mp3....
>
> ...and I see some of your noodlings??!
This entire message is just wrong on soooooo many levels.
-- WthmO
-- ---- ---- ---- ---- ---- --
Message: 8
Date: Tue, 13 Jan 2004 20:03:22 -0500
From: "Allen Partridge" <partridge@(protected)>
Subject: RE: [Dir3d-l] Windoze and my evil little cursor
To: <dir3d-l@(protected)>
Message-ID: <001e01c3da3a$350d5390$d1fea8c0@(protected)>
Content-Type: text/plain; charset="us-ascii"
NC,
Thanks for the reply.
>>Well first off nothing obvious springs to mind, and without seeing the
game or the code its going to be hard to pin down. However i'm hard
pressed to beleive this is purely an OS issue, if for the simple reason
that i've never heard of a problem like this before.
Struck me as odd as well. I recall some early discussions about having
to write the rect twice under odd circumstances but it sure didn't seem
tied to the cursor or anything like it. But what would make it so
clearly tied to mouse I thought?
>>I have to say I feel the problem is more likely to be somewhere in the
code, and with your experience its most likely to be something simple
thats been overlooked, otherwise you would have fixed it already ;)
My students say "why the &%#(; did it do that?" I reply, "because you
told it to."
>>You mention converting mouse movements to a 3D target, is this using
modelsunderray or its ilk? If so then I would flag that up as a
possibiltiy
as it tallies with the description of when the performanc suffers (i.e.
when
you move the cursor into the stage rect). Was this one of the features
you
disabled? Are you positive you disabled it completely?
Nope, no raycasting - plenty of collision, many models but very low poly
counts. So say 70-80 models with 3 polygons each then up to 30 with 2
polygons and then usually not more than 5 particles max. Models rotate
and translate on each frame. No lights, one camera (models use emissive
to light.)
>Anyway to dig a little deeper some questions
>Does the performance still suffer if the cursor is visible or only if
it is
invisible?
Yep
>Does it still suffer if you display any cursor other than 200 (assuming
you
use Director to hide it in the first place)
Yep
>How do you control the visibility of the cursor - director or an xtra?
Director cursor(200)
>What happens if you disable the particles?
First level fps with cursor outside director movie (250 fps) inside
director movie (40 fps) -- in most cases the cursor over the movie
causes a dip of somewhere around 30% performance. This opener is the
most extreme example.
SYStems are xp dual pIII 450's with 512 RAM and a geForce 2 64 -- An NT4
sys with a duron 850, 128 MB RAM and a geforce 2 64. Frame rates are
comparable across systems with direct x7, 5, open GL -- in software mode
no drop in frame rate - of course the frame rate is only 14 fps. ;)
>Is any other software running, are you networked etc?
everything else off.
>>Have you tested this on several machines?
Just the two.
Okay, so I tried again, I commented out everything that the movie does
except put the frame rate to the message window. Mouse out 333 fps,
mouse enter, 48 fps.
So... I made an empty w3d member and put in in a movie. Cranked up the
frame rate to 999. Wrote a single behavior to put the fps. Cursor out
500 - 1000 fps, cursor in 166 fps.
:(
Al Grumpy now. Very, very grumpy.
-- ---- ---- ---- ---- ---- --
Message: 9
Date: Wed, 14 Jan 2004 01:58:47 -0000
From: "noisecrime" <noise@(protected)>
Subject: Re: [Dir3d-l] Windoze and my evil little cursor
To: <dir3d-l@(protected)>
Message-ID: <001e01c3da41$f2ef3c00$0100a8c0@(protected)>
Content-Type: text/plain; charset="iso-8859 (See http://iso-8859.ora-code.com)-1"
-- -- Original Message -- --
From: "Allen Partridge" <partridge@(protected)>
> >Does the performance still suffer if the cursor is visible or only if
> it is invisible?
>
> Yep
LOL not quite a yes or no question, you could mean yes to either part of it.
Care to clarrify please ;)
> So... I made an empty w3d member and put in in a movie. Cranked up the
> frame rate to 999. Wrote a single behavior to put the fps. Cursor out
> 500 - 1000 fps, cursor in 166 fps.
Right I wrote a quick test demo. New movie, inserted a new w3d member,
placed it on stage at 640x480 then added my basic FPS counter behaviour. Ran
it at 999 fps, and then again at 300 and there was no difference between
performance regardless of where the mouse was. Tested on both win98 and XP
in D8.5 and DiMX
Perhaps you could email me your test to check on my machines, at least that
might eliminate your version of director as the problem. However at this
point it looks like you've got something screwy on your systems or something
up with Director.
Noisecrime 2004
-- ---- ---- ---- ---- ---- --
Message: 10
Date: Tue, 13 Jan 2004 21:53:20 -0500
From: webmaster@(protected)
Subject: Re: [Dir3d-l] Dir 3D pathfinding
To: "dir3d-l@(protected)" <dir3d-l@(protected)>
Message-ID: <1074048800.4004af202cd98@(protected)>
Content-Type: text/plain; charset=ISO-8859 (See http://ISO-8859.ora-code.com)-1
I appears that it may need to be modified before it works? In the .dir I
cannot get it to even halfway work without putting a go to the frame on the
last frame, and when I do, and try to give it a path, it either stops and
gives error immediately or will go to destination and just before arriving,
gives error.
Always same error: index out of range related to this line:
nNode = prePathFind(pMap.sectors[cSector], cNode, dNode)[2]
in parent script "enemy mouse".
Quoting Neto <neto@(protected)>:
> http://www.station-zero.com/medion/demos/pathfind.zip
>
> You'll surely need to modify it to use in your projects. There is a .MAX
> file included with the scene, there you can see how the path nodes are setup
> (check the user properties - where the node sectors are definied).
>
> I'll try to make it more user friendly, when I have some free time.
>
> Low_budget + tight_deadlines + HUGE_game = Total_madness
>
> -- -- Original Message -- --
> From: <webmaster@(protected)>
> To: <dir3d-l@(protected)>
> Sent: Tuesday, January 13, 2004 3:16 PM
> Subject: Re: [Dir3d-l] Dir 3D pathfinding
>
>
> > Neto and Elia...thanks. I like Neto's because of the potential to find
> things
> > in 3 dimensions. It seems as if Neto's could go up ramps, if needed.
> >
> > Yeah I think some of us should pitch in a buy you a new mouse or something
> if
> > you found that code! Thanks.
> >
> >
> > \
> > Quoting Elia Morling <elia@(protected)>:
> > >
> > > This might help. http://www.tildruin.com/sourcecode.asp
> > >
> > > Elia
> >
> >
> >
> > -- ---- ---- ---- ---- ---- ---- ---- ---- ------
> > This mail sent through IMP: http://horde.org/imp/
> >
> > __ ____ ____ ____ ____ ____ ____ ____ ____ ____
> > Dir3d-l mailing list
> > Dir3d-l@(protected)
> > http://nuttybar.drama.uga.edu/mailman/listinfo/dir3d-l
> >
>
> __ ____ ____ ____ ____ ____ ____ ____ ____ ____
> Dir3d-l mailing list
> Dir3d-l@(protected)
> http://nuttybar.drama.uga.edu/mailman/listinfo/dir3d-l
>
-- ---- ---- ---- ---- ---- ---- ---- ---- ------
This mail sent through IMP: http://horde.org/imp/
-- ---- ---- ---- ---- ---- --
Message: 11
Date: Wed, 14 Jan 2004 10:51:46 +0100
From: Lucas Meijer <lucas@(protected)>
Subject: Re: [Dir3d-l] Windoze and my evil little cursor
To: dir3d-l@(protected)
Message-ID: <40051132.7060708@(protected)>
Content-Type: text/plain; charset=us-ascii; format=flowed
Hey Allen,
The particle system is framebased, not time based (!?@!?@*#& yeah I
know).. So whenever you have a slower fps for whatever reason, you'll
see the effect in your particle system.. It might not 'go as far' as it
used to, in my expirience it's totally unreliable for anything other
than eyecandy..
Your problem sounds like you're game is slowing down, and my first guess
would be that you're doing a modelsUnderLoc() call on the mouse.. that
will slow down the game if the cursor is over the screen, and fail
quickly if not... If it's not that, you're code is probably doing
something intensive with the cursor location, that isn't being done, or
stops quickly if its not over the 3d scene..
Good luck, Lucas
> So I'm working on this game... There are some 3D things and some
> particles and I begin to notice something very peculiar. In a browser -
> or in authoring, when the windows cursor (which has been set to
> invisible) enters the stage rect, the performance of the particle
> systems is devastated, and when the cursor leaves the screen, the
> performance is fine.
>
> The game converts mouse movements to a 3D target, which is offset from
> the mouse. We've enabled and disabled virtually every feature, and the
> only noticeable performance hit remaining in the game -- is the
> persistant evil caused by the windows cursor - even though the (*&#^^%#(
> isn't even visible.)
>
> So... assuming I haven't overlooked something obvious, is there any way
> to terminate windows interest in drawing to the screen? It seems pretty
> clear that the OS is slowing things down in this case. In lieu of that
> any clever methods for keeping the cursor away from my game?
>
> --al
>
> __ ____ ____ ____ ____ ____ ____ ____ ____ ____
> Dir3d-l mailing list
> Dir3d-l@(protected)
> http://nuttybar.drama.uga.edu/mailman/listinfo/dir3d-l
>
>
-- ---- ---- ---- ---- ---- --
__ ____ ____ ____ ____ ____ ____ ____ ____ ____
Dir3d-l mailing list
Dir3d-l@(protected)
http://nuttybar.drama.uga.edu/mailman/listinfo/dir3d-l
End of Dir3d-l Digest, Vol 8, Issue 19
**************************************
|
|
|