Automatic Joytick Project

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

I want to build a little experimental device that would sequentially
send electrical impulses through the four(or eight) joystick inputs at
timed lengths and intervals, with of course no more than two circuits
being closed at a time.

This is an experiment involving timing and rendom factors in
videogames, and the idea is to be able to program in the exact sequence
of moves via a joystick and have the pc repeat them exactly as they
were input.(In other words, this would be able to run a Pac-man pattern
without human intervention).

I assume that I'd need to use something like a basic stamp, but I'm
open to any ideas on how to get started on this project at this point).

Thanks a lot.

Darren Harris
Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Darren,

You could use something like a processor or FPGA or even a PIC. That
would require some learning curve. (Well, leanring curve no matter what
you do.)

If you're decent with a soldering iron and/or wire wrap tool, you could
do it in discrete logic. Something like:

555 timer - counter - decoder - glue gates - relays

Email me privately we can talk about it.

- Craig (hyarbr01 .at. harris .dot. com)

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

If you're going to have a PC control this, maybe you can use the
parellel port. Connect four pins to the four directions (somehow) on
the joystick, write a small program/script to send high/low signals to
one or more pins at some time interval.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Searcher7@mail.con2.com wrote:
: I want to build a little experimental device that would sequentially
: send electrical impulses through the four(or eight) joystick inputs at
: timed lengths and intervals, with of course no more than two circuits
: being closed at a time.
:
: This is an experiment involving timing and rendom factors in
: videogames, and the idea is to be able to program in the exact sequence
: of moves via a joystick and have the pc repeat them exactly as they
: were input.(In other words, this would be able to run a Pac-man pattern
: without human intervention).
:
: I assume that I'd need to use something like a basic stamp, but I'm
: open to any ideas on how to get started on this project at this point).

Complete overkill.

--
Mark Spaeth mspaeth@mtl.mit.edu
50 Vassar St., #38.265 mspaeth@mit.edu
Cambridge, MA 02139
(617) 452-2354 http://rgvac.978.org/~mspaeth

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

If it were me.. I would write a quick app that recorded my movements
while playing a game in mame by snooping the joystick port etc. Then
play back the movements via the parallel port.

Then you're talking 0 discret componnets used and the whole thing can
be writn in software.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

ajcrm125 wrote:
> If it were me.. I would write a quick app that recorded my movements
> while playing a game in mame by snooping the joystick port etc. Then
> play back the movements via the parallel port.
>
> Then you're talking 0 discret componnets used and the whole thing can
> be writn in software.

For some odd reason Google totally loses about 1/4 to 1/3 of my posts.
Does anyone else have this problem?

Anyway. I have no idea how to do any of what is mentioned, so it
wouldn't get done anytime soon if I have to study up on this. I'd
probably have to pay someone(or trade arcade parts).

Timing accuracy is *extremely* important, and I guess that I'd have to
run my pattern by playing the game allowing the device to record the
exact time and specific input activation initiated *and* released.

An option to abandon the auto-pattern by just moving the joystick(so I
could take over at any time, and an option(button) to "save" would
allow me to make alterations to the pattern and then run it again. I'd
only have to store about a half dozen patterns. And a 9th key pattern
would just repeat 234 times to get to the last screen in Pac-man.

(Preferrably the device would be something that ran off of AC current).

Anyway, I'd appreciate recommendations for people who do this stuff.

Thanks again.

Darren Harris
Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Darren,

It's still not completely clear what you're trying to accomplish. If
you're just trying to play the game at different levels, or change some
of the game's existing parameters, then it sounds like what you want is
an in-circuit emulator (ICE) if you are running actual hardware. If
you're running MAME then you'd have to edit the source code or ROMs???
I don't know too much about MAME.

- Craig

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Craig Yarbrough wrote:
> Darren,
>
> It's still not completely clear what you're trying to accomplish. If
> you're just trying to play the game at different levels, or change some
> of the game's existing parameters, then it sounds like what you want is
> an in-circuit emulator (ICE) if you are running actual hardware. If
> you're running MAME then you'd have to edit the source code or ROMs???
> I don't know too much about MAME.
>
> - Craig

This will be done on actual hardware.

Look at it this way. How would you get a computer to play Pac-man from
the first to the 256th maze?

Darren Harris
Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

>Look at it this way. How would you get a computer to play Pac-man from
>the first to the 256th maze?

Why?

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Craig Yarbrough wrote:
> >Look at it this way. How would you get a computer to play Pac-man from
> >the first to the 256th maze?
>
> Why?

Using Pac-man was the best way to explain what I want to do, but like I
mentioned, this is an experiment involving timing and random factors in
certain other videogames.

Thanks.

Darren Harris
Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

It's that slow eh?
Wow.. I didn't know that.


Mark C. Spaeth wrote:
> Searcher7@mail.con2.com wrote:
> :
> :
> : ajcrm125 wrote:
> :> If it were me.. I would write a quick app that recorded my movements
> :> while playing a game in mame by snooping the joystick port etc. Then
> :> play back the movements via the parallel port.
> :>
> :> Then you're talking 0 discret componnets used and the whole thing can
> :> be writn in software.
> :
> : For some odd reason Google totally loses about 1/4 to 1/3 of my posts.
> : Does anyone else have this problem?
> :
> : Anyway. I have no idea how to do any of what is mentioned, so it
> : wouldn't get done anytime soon if I have to study up on this. I'd
> : probably have to pay someone(or trade arcade parts).
>
> Now that's funny.
>
> : Timing accuracy is *extremely* important, and I guess that I'd have to
> : run my pattern by playing the game allowing the device to record the
> : exact time and specific input activation initiated *and* released.
>
> ...and thre's certainly no way you're going to get that from a PC parallel port,
> which gives 8kHz data rate at best, with very indeterminate timing.
>
> --
> Mark Spaeth mspaeth@mtl.mit.edu
> 50 Vassar St., #38.265 mspaeth@mit.edu
> Cambridge, MA 02139
> (617) 452-2354 http://rgvac.978.org/~mspaeth

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Searcher7@mail.con2.com wrote:
:
:
: ajcrm125 wrote:
:> If it were me.. I would write a quick app that recorded my movements
:> while playing a game in mame by snooping the joystick port etc. Then
:> play back the movements via the parallel port.
:>
:> Then you're talking 0 discret componnets used and the whole thing can
:> be writn in software.
:
: For some odd reason Google totally loses about 1/4 to 1/3 of my posts.
: Does anyone else have this problem?
:
: Anyway. I have no idea how to do any of what is mentioned, so it
: wouldn't get done anytime soon if I have to study up on this. I'd
: probably have to pay someone(or trade arcade parts).

Now that's funny.

: Timing accuracy is *extremely* important, and I guess that I'd have to
: run my pattern by playing the game allowing the device to record the
: exact time and specific input activation initiated *and* released.

....and thre's certainly no way you're going to get that from a PC parallel port,
which gives 8kHz data rate at best, with very indeterminate timing.

--
Mark Spaeth mspaeth@mtl.mit.edu
50 Vassar St., #38.265 mspaeth@mit.edu
Cambridge, MA 02139
(617) 452-2354 http://rgvac.978.org/~mspaeth

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

ajcrm125 <ajcrm125@gmail.com> wrote:
: It's that slow eh?
: Wow.. I didn't know that.

Everything still limited by interrupt latency. If you want faster transfers
you need to FIFO on a PCI card and use DMA to get the data across.

--
Mark Spaeth mspaeth@mtl.mit.edu
50 Vassar St., #38.265 mspaeth@mit.edu
Cambridge, MA 02139
(617) 452-2354 http://rgvac.978.org/~mspaeth

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Mark C. Spaeth wrote:
> Searcher7@mail.con2.com wrote:
> : Does anyone here know of someone who could undertake this project and
> : the possible cost?
>
> I could... but not for you

No you couldn't. And since we will never do business anyway, what
difference does it make?

Darren Harris
Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Mark C. Spaeth wrote:
> Searcher7@mail.con2.com wrote:
> :
> :
> : Mark C. Spaeth wrote:
> :> Searcher7@mail.con2.com wrote:
> :> : Does anyone here know of someone who could undertake this project and
> :> : the possible cost?
> :>
> :> I could... but not for you
> :
> : No you couldn't. And since we will never do business anyway, what
> : difference does it make?
>
> WEll, i guess you're just SOL then.

And your still a POS.
>
> Maybe you can talk to fred and get another fine, reliable reasonably priced
> product from him

I'd just as well get one of your Universal "board-fryer" adaptors.

Darren Harris
Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Mark C. Spaeth wrote:
> Searcher7@mail.con2.com wrote:
> :
> :
> : Mark C. Spaeth wrote:
> :> Searcher7@mail.con2.com wrote:
> :> :
> :> :
> :> : Mark C. Spaeth wrote:
> :> :> Searcher7@mail.con2.com wrote:
> :> :> : Does anyone here know of someone who could undertake this project and
> :> :> : the possible cost?
> :> :>
> :> :> I could... but not for you
> :> :
> :> : No you couldn't. And since we will never do business anyway, what
> :> : difference does it make?
> :>
> :> WEll, i guess you're just SOL then.
> :
> : And your still a POS.
> :>
> :> Maybe you can talk to fred and get another fine, reliable reasonably priced
> :> product from him
> :
> : I'd just as well get one of your Universal "board-fryer" adaptors.
>
> Lies, lies, lies.

Lying is something you do often.

Nevertheless, everyone can decide for themselves. RGVAC: "Seem a bit
high?"

Darren Harris
Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Seriously Darren, what you're talking about doing would take a
significant amount of design time from an *experienced* engineer; not
only because of the complexity, but because the ambiguity of the
requirements. Chances are VERY high someone would build you something
that did 'A', when you really wanted 'a'. Then re-design, re-work, more
time, etc. There's still too much risk here.

We get this sort of thing all the time in gov't contracting, most times
because the job is top secret and compartmentalized. But that's the
gov't, and they have oodles of tax dollars to spend on us so we really
don't mind the aggravation. I digress.

If you gave someone (not me) a very clear and detailed description of
what you wanted, you might find someone to do it, maybe cheap. But if I
absolutely had to bid right now given the current information it would
be on the order of $30k and 9 months.

No offense meant, but you would probably be better off reading up and
doing this yourself. Even if you fail you still learn something in the
process. Personally that's what I enjoy the most about this hobby!

$0.02

- Craig


Searcher7@mail.con2.com wrote:
> Does anyone here know of someone who could undertake this project and
> the possible cost?
>
> Thanks.
>
> Darren Harris
> Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Craig Yarbrough wrote:
> Seriously Darren, what you're talking about doing would take a
> significant amount of design time from an *experienced* engineer; not
> only because of the complexity, but because the ambiguity of the
> requirements. Chances are VERY high someone would build you something
> that did 'A', when you really wanted 'a'. Then re-design, re-work, more
> time, etc. There's still too much risk here.
>
> We get this sort of thing all the time in gov't contracting, most times
> because the job is top secret and compartmentalized. But that's the
> gov't, and they have oodles of tax dollars to spend on us so we really
> don't mind the aggravation. I digress.
>
> If you gave someone (not me) a very clear and detailed description of
> what you wanted, you might find someone to do it, maybe cheap. But if I
> absolutely had to bid right now given the current information it would
> be on the order of $30k and 9 months.
>
> No offense meant, but you would probably be better off reading up and
> doing this yourself. Even if you fail you still learn something in the
> process. Personally that's what I enjoy the most about this hobby!

$30K and 9 months?!?

The gov't has really spoiled you guys, huh?

But I'm not surprised since the electronics and computer industry
thrive on making even the simpliest things overly complex and
expensive. But that's a discussion about wasted tax dollars.

I could have covered every detail of this project, but the original
post would have been much longer.

The simple version is as follows:

Recording: The ability to record every joystick input's
activation/deactivation in order and at a timing accuracy of 1/60th of
a second(at worst), at the press of a button.

Playback: The ability to repeat those exact movements through the
inputs at a press of a second button. Any manual movement of a joystick
input during playback would instantly *stop* playback and continue
recording.(This would allow manual changes in the recorded sequence
without having to start from the beginning).

Darren Harris
Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

>The gov't has really spoiled you guys, huh?
>
>But I'm not surprised since the electronics and computer industry
>thrive on making even the simpliest things overly complex and
>expensive. But that's a discussion about wasted tax dollars.

ROTFLMAO

I get the impression that your view and experience is probably a little
simplex. But I mean that with the utmost respect. Try 4 years of
engineering college and 9 years in the field, then get back to me. By
the way, $30k @ 9 months engineering at even 1/4 time is CHEAP!

>Recording: The ability to record every joystick input's
>activation/deactivation in order and at a timing accuracy of 1/60th of
>a second(at worst), at the press of a button.

* What kind of joy? 4-way? 8-way?
* What sense are the inputs? Active-high? Active-Low? Open Drain?
* What are the current sink requirements for the switches?
* Does this need to be compatible with multiple platforms? Details on
those?
* How will the data need to be stored/ported?
* Size/weight/power requirements?
* Interface requirements? What kind of plugs? Separate power..???
* ...and so on and so on...

I still think you are the best choice to do this yourself since you
know EXACTLY what you want. Good luck with it.

- Craig

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Mark it pays not to be a student! ;-)

Seriously, for the level of work we do 9 mo @ $30k is cheap. Heck
that's *one* new logic analyzer.

- Craig

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Rob Carroll wrote:
> > This is an experiment involving timing and rendom factors in
> > videogames, and the idea is to be able to program in the exact sequence
> > of moves via a joystick and have the pc repeat them exactly as they
> > were input.(In other words, this would be able to run a Pac-man pattern
> > without human intervention).
>
> Curious here... what's the point of the experiment? How does basically
> scripting joystick "movements" for a game known not to be random have
> anything to do with randomness?

There you go again, failing to understand what has already been
covered.

> Sounds like what you're trying to do is complete overkill... a solution in
> dire need of a dilemna. If I were you, I would consider lower-tech
> alternatives... maybe it won't be as flashy, but it will be a lot cheaper
> and easier... maybe you could even do it yourself, if it didn't require
> soldering (zing).

Why try to explain what you couldn't understand?

Darren Harris
Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Rob Carroll wrote:
> Oh, that's right... how silly of me to forget... "Darren smart. Everybody
> else dumb." I couldn't possibly understand whatever dumbass experiment
> you're trying to put together.
>
> Ok, supergenius, I'll just sit over here, and try not to get in the way of
> your MONSTER intellect... what with your "experiments" and all. *snicker*

Facr the corner and don't forget to keep that cone hat on.

Darren Harris
Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

Rob Carroll wrote:
> Oh, that's right... how silly of me to forget... "Darren smart. Everybody
> else dumb." I couldn't possibly understand whatever dumbass experiment
> you're trying to put together.
>
> Ok, supergenius, I'll just sit over here, and try not to get in the way of
> your MONSTER intellect... what with your "experiments" and all. *snicker*

Face the corner and don't forget to keep that cone hat on.

Darren Harris
Staten Island, New York.

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

In article Mark C. Spaeth says...
>
>ajcrm125 <ajcrm125@gmail.com> wrote:
>: It's that slow eh?
>: Wow.. I didn't know that.
>
>Everything still limited by interrupt latency. If you want faster transfers
>you need to FIFO on a PCI card and use DMA to get the data across.

On a 'modern' PC you can go WAY faster than 8k a sec on a PC Parallel port.

I was bit banging a parallel port under windows at 20k/sec (160kbps)
sending a serial bitstrea, with 1 bit for busy/rdy to control the xfer.
The ST microelectronics eval kit for the STA013 uses the parallel port
to send a mp3 bitstream at upto 192kb/sec (24k/sec) reliably.. that's
just from my experience. No DMA was involved..

Using the port in a more parallel fashion, with the limit of 4 bits of
input for a Standard parallel, you can get near 100k/sec, I was transferring
64k of memory from an old 8-bit machine to a PC in about a second.

Jeez.. I remember the old PDQ leads for DOS which just banged the
parallel port directly, on ISA with no IRQ or DMA and you'd get at
least 28k/sec on a 386-25.

With a bi-directional port I've heard of much faster but not seen that
myself.

- James

 

[Guest]

Archived from groups: rec.games.video.arcade.collecting (More info?)

In article Craig Yarbrough says...
>
>But if I absolutely had to bid right now given the current information it
>would be on the order of $30k and 9 months.

I'll bid $15k and 4.5 months, with the usual software house indemnity of
"hey it's not our fault if it doesn't actually work"

- James

------------------------------
JrokLand http://www.jrok.com
------------------------------