home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!sdd.hp.com!swrinde!elroy.jpl.nasa.gov!jato!jdickson
- From: jdickson@jato.jpl.nasa.gov (Jeff Dickson)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: gameport device vs. input device
- Message-ID: <1992Aug28.211858.3001@jato.jpl.nasa.gov>
- Date: 28 Aug 92 21:18:58 GMT
- References: <1992Aug28.174032.4126@galileo.cc.rochester.edu> <1992Aug28.182725.1768@jato.jpl.nasa.gov> <1992Aug28.203427.9520@galileo.cc.rochester.edu>
- Organization: Jet Propulsion Laboratory
- Lines: 50
-
- In article <1992Aug28.203427.9520@galileo.cc.rochester.edu> rjw@merlin.cvs.rochester.edu (Robert Wilson) writes:
- >In article <1992Aug28.182725.1768@jato.jpl.nasa.gov> jdickson@jato.jpl.nasa.gov (Jeff Dickson) writes:
- >>In article <1992Aug28.174032.4126@galileo.cc.rochester.edu> rjw@merlin.cvs.rochester.edu (Robert Wilson) writes:
- >>>I'm trying to use the input device to pick up character input, and the
- >>>gameport device to pick up mouse input from both ports. This is for a
- >>>psych. experiment, so the gameports have to be on equal standing: hence
- >>>my need for using the gameport device.
- >>>
- >>>** My question: is it possible to make the input device unaware of the
- >>>gameports? I want all mouse events to go the gameport device, and only
- >>>to the gameport device.
- >>
- >>The input device task opens the gameport device when it begins and packages
- >>any and all mouse events that occur into IECLASS_RAWMOUSE input events. Key-
- >>board activity too causes IECLASS_RAWKEY input events to be generated. You
- >>can become a handler of the input device chain (See the RKM Libraries and
- >>Devices chapter on the Input Device). As far an equal standing goes, they're
- >>both packaged as input events and you can see them in the order they occurred.
- >
- >Thank you for the prompt reply. I have code already that handles the mouse
- >via the input device, but as far as I can tell (having read the sections you
- >mentioned) the input device only picks up events from one port at a time.
- >This means that the data flows as follows:
- >
- >GAMEPORT 0: mouse->gameport dev 0->input dev->my program
- >GAMEPORT 1: mouse->gameport dev 1->my program
- >
- >Things I'm afraid may slow down my processing an event on gameport 0:
- > 1) the event being queued behind non-mouse events
- > 2) the input device taking a while to process the event
- >
- >I need millisecond accuracy. If anyone can tell me that the input device won't
- >slow things down that much, I'll be happy to go on; but right now it seems that
- >data from gameport 1 could go processed faster than data from gameport 0.
- >
- >If anyone can tell me how to get around the input device, or show that my
- >fears are unfounded, I'd really appreciate it.
- >
- >-Robert
- > rjw@cvs.rochester.edu
-
- I can't imagine that the delays incurred by the input device
- processing mouse and keyboard events would be so great on a more-less
- dry system (nothing going on) to have unruley effects on the accuracy.
- Also, the input device task runs at a fairly high priority. This gives
- it an edge over typical stuff (i.e. Shells). Filesystems too run at a
- high priority, so maintaining this accuracy might be a problem if too
- much DOS activity is drummed up in the interim.
-
- jeff
-