freeglut

If you are looking for someone who knows joystick programming, I am the
WRONG person to ask.  But since I wrote the words in question, let me try to
clarify what I meant.

When JS (or "freeglut" with the new changes) opens a joystick, it passes it
an "ident" or identifier.  At the moment with "freeglut" this is hard-coded
to be zero.  PLIB has a joystick testing program which checks two joysticks:
one with an "ident" of zero and the other with an "ident" of one.  The
joystick on my computer at home responds to calls with an "ident" of one but
not to calls with an "ident" of zero.

My thinking was that an "active joystick" (as far as "freeglut" is
concerned) is something that responds when polled.  On my system at home,
the "active joystick" would be "1" instead of "0".  Presumably (and I am
speaking far beyond my range of understanding here) I would be able to get a
second joystick, hook it up to some port on the back of my computer, and
have it generate signals for "freeglut" with an "ident" of "0".  In that
case my application would have an "active joystick" with an "ident" of zero.

I would appreciate it if somebody who knows joysticks could straighten me
out.

John F. Fay
john.fay@...



-----Original Message-----
From: Richard Rauch [mailto:sforge@...]
Sent: Thursday, February 05, 2004 3:11 PM
To: freeglut-developer@...
Subject: Re: [Freeglut-developer] Multiple Joysticks


On Thu, Feb 05, 2004 at 01:52:45PM -0600, Fay John F Contr AAC/WMG wrote:
 [...]
> allow the application to specify which joystick the callback applies to.
As
> an additional bit, the function could return a zero if the joystick is
> inactive and a one if it is active.  I realize that all the other callback

For the benefit of those of us who do not know about joystick programming,
what is an "active joystick" as opposed to an "inactive joystick"?

If I plug in a USB joystick, does it become active?  If I unplug an
"active" joystick, what happens?  Can a plugged in joystick be inactive?
An unplugged one active?

If a joystick can change from active to inactive (or vice versa) while
a program is running---e.g., by being plugged in or unplugged---how does
the program find out, having already registered a callback for the
joystick?


-- 
  "I probably don't know what I'm talking about."  http://www.olib.org/~rkr/


-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Freeglut-developer mailing list
Freeglut-developer@...
https://lists.sourceforge.net/lists/listinfo/freeglut-developer

RE: [Freeglut-developer] Multiple Joysticks From: Fay John F Contr AAC/WMG <john.fay@eg...> - 2004-02-06 15:02 Attachments: Message as HTML      Nigel, The comments at your web site sound well-thought-out. (I haven't had a lot of time to think them out myself, but I don't see anything glaring.) The primary problem that I see is a question of coders and testers: we'll need someone with a joystick, with or without a USB port, running Linux, Windows, FreeBSD, and NetBSD at a minimum. (I've invited someone from the PLIB community who has a Macintosh to try "freeglut", but I daresay his biggest problems will be elsewhere in the library.) John F. Fay john.fay@... -----Original Message----- From: freeglut-developer-admin@... [mailto:freeglut-developer-admin@... Behalf Of Nigel Stewart Sent: Thursday, February 05, 2004 7:14 PM To: freeglut-developer@... Subject: Re: [Freeglut-developer] Multiple Joysticks > I propose that we add an argument to "glutJoystickFunc" (and rename it) to > allow the application to specify which joystick the callback applies to. > Suggestions? When I looked into GLUT and FreeGLUT joystick handling, I put a few thoughts here: http://www.nigels.com/glt/news.html Generally, I thought that the API should be revised to support multiple joysticks and connect and disconnect events. The "legacy" callbacks would remain supported for the first detected joystick, I suppose. As the result of some discussion in relation to collaborative interaction or gaming applications - I realised that the same generality would also be useful for keyboard and mouse as well. Again, the "legacy" callbacks would be perfectly supported for less exotic applications. It was the one person, one mouse, one screen interaction model that we were particular interested in experimenting with, or rather, multi-person, multi input, one (big) screen. With the USB hot-plug kind of functionality these days, it makes sense to extend GLUT to generate the connect and disconnect events. Nigel ------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ Freeglut-developer mailing list Freeglut-developer@... https://lists.sourceforge.net/lists/listinfo/freeglut-developer

RE: [Freeglut-developer] Multiple Joysticks From: Fay John F Contr AAC/WMG <john.fay@eg...> - 2004-02-10 15:21 Attachments: Message as HTML      Nigel, I need some details about your proposed joystick interface. You propose the five functions: FGAPI void FGAPIENTRY glutJoystickConnectFunc(void (*joystickConnect)(int)); FGAPI void FGAPIENTRY glutJoystickDisconnectFunc(void (*joystickDisconnect)(int)); FGAPI void FGAPIENTRY glutJoystickButtonFunc(void (*joystickButton)(int,int)); FGAPI void FGAPIENTRY glutJoystickButtonUpFunc(void (*joystickButtonUp)(int,int)); FGAPI void FGAPIENTRY glutJoystickMotionFunc(void (*joystickMotion)(int,int,int,int,int)); (1) Are these window-specific callbacks or general callbacks (like the idle callback)? (2) What do the arguments mean? (3) In view of the potentially increasing number of axes, would it be better for the motion function to have arguments ( int joystick_id, int num_axes, int *axis_values ) ? John F. Fay john.fay@... -----Original Message----- From: freeglut-developer-admin@... [mailto:freeglut-developer-admin@... Behalf Of Nigel Stewart Sent: Thursday, February 05, 2004 7:14 PM To: freeglut-developer@... Subject: Re: [Freeglut-developer] Multiple Joysticks > I propose that we add an argument to "glutJoystickFunc" (and rename it) to > allow the application to specify which joystick the callback applies to. > Suggestions? When I looked into GLUT and FreeGLUT joystick handling, I put a few thoughts here: http://www.nigels.com/glt/news.html Generally, I thought that the API should be revised to support multiple joysticks and connect and disconnect events. The "legacy" callbacks would remain supported for the first detected joystick, I suppose. As the result of some discussion in relation to collaborative interaction or gaming applications - I realised that the same generality would also be useful for keyboard and mouse as well. Again, the "legacy" callbacks would be perfectly supported for less exotic applications. It was the one person, one mouse, one screen interaction model that we were particular interested in experimenting with, or rather, multi-person, multi input, one (big) screen. With the USB hot-plug kind of functionality these days, it makes sense to extend GLUT to generate the connect and disconnect events. Nigel

Re: [Freeglut-developer] Multiple Joysticks From: Nigel Stewart <nigels@ni...> - 2004-02-11 03:25 > FGAPI void FGAPIENTRY glutJoystickConnectFunc(void > (*joystickConnect)(int n)); > FGAPI void FGAPIENTRY glutJoystickDisconnectFunc(void > (*joystickDisconnect)(int n)); > FGAPI void FGAPIENTRY glutJoystickButtonFunc(void > (*joystickButton)(int n,int button)); > FGAPI void FGAPIENTRY glutJoystickButtonUpFunc(void > (*joystickButtonUp)(int n,int button)); > FGAPI void FGAPIENTRY glutJoystickMotionFunc(void > (*joystickMotion)(int n,int x,int y,int u,int v)); > (1) Are these window-specific callbacks or general callbacks (like the idle > callback)? It should follow the keyboard convention, I suppose. A joystick is more like a keyboard than a mouse. > (2) What do the arguments mean? As I recall - n The joystick button Button number (upto 2^32 buttons) x,y,u,v Four analog axes Oh, and while we're thinking about it, converting analog axes to float might also be an option: FGAPI void FGAPIENTRY glutJoystickMotionFunc(void (*joystickMotion)(int n,float x,float y,float u,float v)); Where x,y,u,v are [-1.0,1.0] > (3) In view of the potentially increasing number of axes, would it be better > for the motion function to have arguments ( int joystick_id, int num_axes, > int *axis_values ) ? More general, certainly. Has anyone seen a joystick with more than four axes? I'm averse to introducing pointers into the GLUT API. I am particularly averse to having a 2D array of analog axes for multiple joysticks. We could simply eliminate joystick motion events, it doesn't make sense to me to be polling the joystick state at a different frequency to the frame rate. Perhaps a glutGet interface would have the benefit of being more future-proof. See the discusion at: http://sourceforge.net/tracker/index.php?func=detail&aid=841360&group_id=1032&atid=351032 http://sourceforge.net/tracker/index.php?func=detail&aid=847170&group_id=1032&atid=351032 Overall, I think it is useful to receive button up/down events. Analog axes need to be polled, as well as analog buttons. Perhaps it is reasonable for the joystick state to be updated within FreeGLUT once per frame. glutGet calls should basically be fixed time... Nigel > -----Original Message----- > From: freeglut-developer-admin@... > [mailto:freeglut-developer-admin@... Behalf Of > Nigel Stewart > Sent: Thursday, February 05, 2004 7:14 PM > To: freeglut-developer@... > Subject: Re: [Freeglut-developer] Multiple Joysticks > > >>I propose that we add an argument to "glutJoystickFunc" (and rename it) to >>allow the application to specify which joystick the callback applies to. > > > Suggestions? > > When I looked into GLUT and FreeGLUT joystick handling, I > put a few thoughts here: > > http://www.nigels.com/glt/news.html > > Generally, I thought that the API should be revised to > support multiple joysticks and connect and disconnect events. > The "legacy" callbacks would remain supported for the first > detected joystick, I suppose. > > As the result of some discussion in relation to collaborative > interaction or gaming applications - I realised that the same > generality would also be useful for keyboard and mouse as > well. Again, the "legacy" callbacks would be perfectly supported > for less exotic applications. > > It was the one person, one mouse, one screen interaction model > that we were particular interested in experimenting with, or > rather, multi-person, multi input, one (big) screen. > > With the USB hot-plug kind of functionality these days, it makes > sense to extend GLUT to generate the connect and disconnect > events. > > Nigel

Re: [Freeglut-developer] Multiple Joysticks From: James Jones <jcjones@uf...> - 2004-02-11 12:58 Attachments: Message as HTML      On Tue, 2004-02-10 at 22:23, Nigel Stewart wrote: > More general, certainly. > Has anyone seen a joystick with more than four axes? > I'm averse to introducing pointers into the GLUT API. > I am particularly averse to having a 2D array of analog > axes for multiple joysticks. Two years ago or so I had the experience of playing with a joystick with 5 axes: X, Y, Yaw, Pitch, and Throttle. Unfortunately I neither own one of these nor remember the manufacturer. :( How many applications would make use of more than 4 axes is another story... --=20 James 'J.C.' Jones "No trees were killed in the creation of this message.=20 However, many electrons were terribly inconvenienced." - Anonymous

Re: [Freeglut-developer] Multiple Joysticks From: Steve Baker <sjbaker1@ai...> - 2004-02-11 13:57 James Jones wrote: > On Tue, 2004-02-10 at 22:23, Nigel Stewart wrote: > >>More general, certainly. >>Has anyone seen a joystick with more than four axes? >>I'm averse to introducing pointers into the GLUT API. >>I am particularly averse to having a 2D array of analog >>axes for multiple joysticks. > > > Two years ago or so I had the experience of playing with a joystick with > 5 axes: X, Y, Yaw, Pitch, and Throttle. Unfortunately I neither own one > of these nor remember the manufacturer. :( It's definitely a growing trend to have more axes. Look at a PlayStation-2 controller - many of the buttons are pressure sensitive (ie analog) and it has two analog sticks. In the past, manufacturers were limited by the four analog inputs on the sound card's joystick connector - but now that we have USB sticks, there really isn't a limit. > How many applications would make use of more than 4 axes is another > story... Indeed. Flight simulator games typically need at least 5 axes (Ailerons, Elevator, Rudder, Throttle, Flaps) - and we know of at least one (FlightGear) that uses GLUT...although it uses PLIB for reading joysticks for precisely the reason that GLUT's joystick code sucks. Personally, I don't think joystick code should ever have been added to GLUT because it has nothing to do with the windowing interface. The mouse and keyboard are a different matter because these are traditionally 'focussed' onto a particular window - but joysticks are not used that way and the API's reflect that. ---------------------------- Steve Baker ------------------------- HomeEmail: <sjbaker1@...> WorkEmail: <sjbaker@...> HomePage : http://www.sjbaker.org Projects : http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net http://prettypoly.sf.net -----BEGIN GEEK CODE BLOCK----- GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M- V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++ -----END GEEK CODE BLOCK-----

Re: [Freeglut-developer] Multiple Joysticks From: Richard Rauch <sforge@ol...> - 2004-02-11 14:36 On Wed, Feb 11, 2004 at 07:55:37AM -0600, Steve Baker wrote: [...] > In the past, manufacturers were limited by the four analog inputs on > the sound card's joystick connector - but now that we have USB sticks, > there really isn't a limit. Sounds like a good argument to be extensible. (^& [...] > Personally, I don't think joystick code should ever have been added to > GLUT because it has nothing to do with the windowing interface. The mouse > and keyboard are a different matter because these are traditionally > 'focussed' > onto a particular window - but joysticks are not used that way and the API's > reflect that. I certainly wouldn't have associated a poll-interval with it. If 'twere me, I would have said "Use a GLUT timer." I might or might not have simply barred joystick support, had it been my call. Oh well. It's in there, now. -- "I probably don't know what I'm talking about." http://www.olib.org/~rkr/

Re: [Freeglut-developer] Multiple Joysticks From: Richard Rauch <sforge@ol...> - 2004-02-11 15:04 On Tue, Feb 10, 2004 at 10:23:03PM -0500, Nigel Stewart wrote: [...] > Has anyone seen a joystick with more than four axes? > I'm averse to introducing pointers into the GLUT API. > I am particularly averse to having a 2D array of analog > axes for multiple joysticks. a) No one had seen a joystick at all until someone had made one. We can fairly easily extrapolate to N axes, though, and as Steve points out, there is no longer a real barrier to such being created for USB. b) I would agree re. pointers, in general, but not at the expense of locking ourselves into an API that may become obsolete in 18 months. Fortunately, there are other ways that this can be done... > We could simply eliminate joystick motion events, it doesn't > make sense to me to be polling the joystick state at a different > frequency to the frame rate. Perhaps a glutGet interface would Suppose that in a game, you want a special effect attack that lowers the opponent's display framerate. (A kind of sensory attack or ECM.) What if you use freeglut for joystick and timer operations, but not for graphics? E.g, run a given program (regardless of the window focus or mouse activity) when the user press the "fire" button. (Think of a screen-grabbing type of tool for example.) What if the client is more imaginative than me and thinks of something else that I haven't proposed? It'll probably happen. The current problem with GLUT-style joystick support is that it tries to get these events as an "event". But this is not needed, since we *have* to poll for it anyway, at some level. If we just say that the client polls freeglut for the joystick status, I think that the whole issue melts away very easily. [...] > within FreeGLUT once per frame. glutGet calls should basically > be fixed time... I'm not sure what this last sentence means. -- "I probably don't know what I'm talking about." http://www.olib.org/~rkr/

Re: [Freeglut-developer] Multiple Joysticks From: Steve Baker <sjbaker1@ai...> - 2004-02-11 17:09 Richard Rauch wrote: >>We could simply eliminate joystick motion events, it doesn't >>make sense to me to be polling the joystick state at a different >>frequency to the frame rate. Perhaps a glutGet interface would > > Suppose that in a game, you want a special effect attack that lowers > the opponent's display framerate. (A kind of sensory attack or ECM.) A better example would be when game physics uses the joystick. The physics may well need to be updated at much higher rates than the graphics. Failure to poll peripherals at higher than graphics rates is the kind of thing that causes Quake players to disconnect the swapbuffers from the vertical retrace in order to get higher frame rates. Graphically speaking that's ridiculous. If your CRT is refreshing at 60Hz, there is no earthly reason to recalculate all those expensive graphics at 100Hz because you'll never see it. However, there IS good reason to want your keyboard/mouse/joystick events to be polled at higher rates so that there is less latency between you pushing a button and the results taking effect in the game world. ---------------------------- Steve Baker ------------------------- HomeEmail: <sjbaker1@...> WorkEmail: <sjbaker@...> HomePage : http://www.sjbaker.org Projects : http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net http://prettypoly.sf.net -----BEGIN GEEK CODE BLOCK----- GCS d-- s:+ a+ C++++$ UL+++$ P--- L++++$ E--- W+++ N o+ K? w--- !O M- V-- PS++ PE- Y-- PGP-- t+ 5 X R+++ tv b++ DI++ D G+ e++ h--(-) r+++ y++++ -----END GEEK CODE BLOCK-----

Re: [Freeglut-developer] Multiple Joysticks From: Nigel Stewart <nigels@ni...> - 2004-02-12 03:56 > However, there IS good reason to want your keyboard/mouse/joystick > events to be polled at higher rates so that there is less latency > between you pushing a button and the results taking effect in the > game world. Interesting point. But without multi-threading? Nigel

4 messages have been excluded from this view by a project administrator.