Recommendations please :-)

Archiver

Hello,

I was hoping to get recommendations on what hardware would satisfy
the needs of a beginner. I'm a complete newbie who knows nothing
about microcontrollers, and was hoping that someone could give me an
idea about where to start.

My immediate project only entails very simple control commands. The
idea is to somehow connect a microcontroller to 4 different leaf
switches, so I can close and open any of the 4 circuits in any of *14
different pre-programmed sequences*. Each sequence will have as many
as 66 total commands, and will be executed within approximately 2
minutes. I want to be able to "press a button" at anytime along a 5
hour time-line to activate any of those 14 command sequences with
timing accuracy between each individual command no worse than 1/60th
of a second(within that sequence).

The microcontroller will in effect be duplicating the exact timing of
the commands in each sequence.(Commands that I will like to manually
program into it via the four leaf switches I mentioned).

Am I asking too much?

Any advice would be greatly appreciated.

Thank you for your time

Thanks

Darren Harris
Staten Island, New York.

Searcher7@m...

Archiver

Hello,

I was hoping to get recommendations on what hardware would satisfy
the needs of a beginner. I'm a complete newbie who knows nothing
about microcontrollers, and was hoping that someone could give me an
idea about where to start.

My immediate project only entails very simple control commands. The
idea is to somehow connect a microcontroller to 4 different leaf
switches, so I can close and open any of the 4 circuits in any of *14
different pre-programmed sequences*. Each sequence will have as many
as 66 total commands, and will be executed within approximately 2
minutes. I want to be able to "press a button" at anytime along a 5
hour time-line to activate any of those 14 command sequences with
timing accuracy between each individual command no worse than 1/60th
of a second(within that sequence).

The microcontroller will in effect be duplicating the exact timing of
the commands in each sequence.(Commands that I will like to manually
program into it via the four leaf switches I mentioned).

Any advice would be greatly appreciated.

Thanks

Darren Harris
Staten Island, New York.

Searcher7@m...

Archiver

> Darren:

Hardware selection depends upon what type of loads you are switching, and
exactly what you mean in the timing you describe. Some timing requirements
may create difficulties with a Basic Stamp, they're not super-fast, and RT
requires additional hardware. If what you described means that it would be
acceptable if the sequence of switch actions occur within 16 mS of each
other in sequence, this can be done. Your 5 hour time line is not clear and
may also be a problem.

What exactly are you trying to build? Without more details or a more
specific question, you might wind up re-posting this 100 times w/o getting
the help you want...

Chris


>
> Hello,
>
> I was hoping to get recommendations on what hardware would satisfy
> the needs of a beginner. I'm a complete newbie who knows nothing
> about microcontrollers, and was hoping that someone could give me an
> idea about where to start.
>
> My immediate project only entails very simple control commands. The
> idea is to somehow connect a microcontroller to 4 different leaf
> switches, so I can close and open any of the 4 circuits in any of *14
> different pre-programmed sequences*. Each sequence will have as many
> as 66 total commands, and will be executed within approximately 2
> minutes. I want to be able to "press a button" at anytime along a 5
> hour time-line to activate any of those 14 command sequences with
> timing accuracy between each individual command no worse than 1/60th
> of a second(within that sequence).
>
> The microcontroller will in effect be duplicating the exact timing of
> the commands in each sequence.(Commands that I will like to manually
> program into it via the four leaf switches I mentioned).
>
> Any advice would be greatly appreciated.
>
> Thanks
>
> Darren Harris
> Staten Island, New York.
>
> Searcher7@m...
>
>
>
>
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>

Archiver

"Hardware selection depends upon what type of loads
you are switching,
and exactly what you mean in the timing you describe."


The loads are not at present an issue. After I figure
out what electronic components(microcontroller) to
get, I will have to figure out a way to interface it
to the leaf switches I mentioned. The only "input"
criteria for the microcontroller would be to accept
any of 8 possible commands it will receive when I
manually close and open any of four circuits. The more
difficult part would be the output of any of those 8
commands, since these circuits(leaf switches) have to
by physically closed in order for the PCB on the other
end to notice, I'd have to come up with a way to send
a strong enough current through the wires to fool
it.(I'm actually going to be using the 4-way joystick
controls of an arcade game).

I don't understand your question about the timing. If
I were to "open", let's say switch "B" exactly 1.2
seconds after I closed it, I want to microcontroller
to output the same(to an accuracy of no worse than
1/60th of a second) when I give it the command to do
so. Actually, the longest of the 14 sequences will be
112 different commands in about a two minute
period.(And only 1 of the 4 circuits can actually be
"closed" at one time).

"Some timing requirements may create difficulties with
a Basic Stamp, they're not super-fast, and RT requires
additional hardware."

That is why I'm trying to find out what I'd need to
accomplish what I want.(And what is "RT"?).

"If what you described means that it would be
acceptable if the sequence of switch actions occur
within 16 mS of each other in sequence, this can be
done. Your 5 hour time line is not clear and may also
be a problem."

Then forget I mentioned the 5 hour time line. :-)

"What exactly are you trying to build? Without more
details or a more
specific question, you might wind up re-posting this
100 times w/o
getting the help you want..."

I honestly don't know what I haven't covered, or how
to explain anything more clear than I already have. It
is a very simple project.

Thanks

Darren Harris
Staten Island, New York.

Searcher7@m...




********************************************************************************\
********************************************************************************\
*****
--- "Chris Loiacono (E-mail)"
<chris01@t...> wrote:
>
> > Darren:
>
> Hardware selection depends upon what type of loads
> you are switching, and
> exactly what you mean in the timing you describe.
> Some timing requirements
> may create difficulties with a Basic Stamp, they're
> not super-fast, and RT
> requires additional hardware. If what you described
> means that it would be
> acceptable if the sequence of switch actions occur
> within 16 mS of each
> other in sequence, this can be done. Your 5 hour
> time line is not clear and
> may also be a problem.
>
> What exactly are you trying to build? Without more
> details or a more
> specific question, you might wind up re-posting this
> 100 times w/o getting
> the help you want...
>
> Chris
>
>
> >
> > Hello,
> >
> > I was hoping to get recommendations on what
> hardware would satisfy
> > the needs of a beginner. I'm a complete newbie who
> knows nothing
> > about microcontrollers, and was hoping that
> someone could give me an
> > idea about where to start.
> >
> > My immediate project only entails very simple
> control commands. The
> > idea is to somehow connect a microcontroller to 4
> different leaf
> > switches, so I can close and open any of the 4
> circuits in any of *14
> > different pre-programmed sequences*. Each sequence
> will have as many
> > as 66 total commands, and will be executed within
> approximately 2
> > minutes. I want to be able to "press a button" at
> anytime along a 5
> > hour time-line to activate any of those 14 command
> sequences with
> > timing accuracy between each individual command no
> worse than 1/60th
> > of a second(within that sequence).
> >
> > The microcontroller will in effect be duplicating
> the exact timing of
> > the commands in each sequence.(Commands that I
> will like to manually
> > program into it via the four leaf switches I
> mentioned).
> >
> > Any advice would be greatly appreciated.
> >
> > Thanks
> >
> > Darren Harris
> > Staten Island, New York.
> >
> > Searcher7@m...
> >
> >
> >
> >
> >
> >
> > Your use of Yahoo! Groups is subject to
> > http://docs.yahoo.com/info/terms/
> >
> >
>
>
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/

Archiver

The Basic Stamp 2 will do this as long as you realize that there are no
interrupts. This implies that you will need to loop through the 4 inputs
continuously and check their state. As long as the program doesn't take more
than 16mS to loop, which is a long time- even for a stamp, you should be OK.
The 'button' instruction will help you 'debounce' your manual switches.
If the toughest loop is 112 commands in 2 minutes or less, you're talking
more than 1 second per, so it also is not too fast. More effort will go into
making it take 2 minutes, if that's what you need. Just realize that again,
while the 112 instructions are executing in their own loop, the only thing
the stamp can do is wait for the loop to execute, or be reset.

If I were you, I'd read the manual and 'stamps in class' downloads at
www.parallaxinc.com


CL

>

---

Original Message

---

> From: Darren Harris [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=anHd1JeNiA_zh2b4Vpqne8Ivd0rDc0D18cPa5IWKHwQO6zEIEGLkfKu0w9_7Q35XtFyZHzdOyCwtUfY]searcher731@y...[/url
> Sent: Thursday, April 12, 2001 11:58 AM
> To: basicstamps@yahoogroups.com
> Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please :-)
>
>
> "Hardware selection depends upon what type of loads
> you are switching,
> and exactly what you mean in the timing you describe."
>
>
> The loads are not at present an issue. After I figure
> out what electronic components(microcontroller) to
> get, I will have to figure out a way to interface it
> to the leaf switches I mentioned. The only "input"
> criteria for the microcontroller would be to accept
> any of 8 possible commands it will receive when I
> manually close and open any of four circuits. The more
> difficult part would be the output of any of those 8
> commands, since these circuits(leaf switches) have to
> by physically closed in order for the PCB on the other
> end to notice, I'd have to come up with a way to send
> a strong enough current through the wires to fool
> it.(I'm actually going to be using the 4-way joystick
> controls of an arcade game).
>
> I don't understand your question about the timing. If
> I were to "open", let's say switch "B" exactly 1.2
> seconds after I closed it, I want to microcontroller
> to output the same(to an accuracy of no worse than
> 1/60th of a second) when I give it the command to do
> so. Actually, the longest of the 14 sequences will be
> 112 different commands in about a two minute
> period.(And only 1 of the 4 circuits can actually be
> "closed" at one time).
>
> "Some timing requirements may create difficulties with
> a Basic Stamp, they're not super-fast, and RT requires
> additional hardware."
>
> That is why I'm trying to find out what I'd need to
> accomplish what I want.(And what is "RT"?).
>
> "If what you described means that it would be
> acceptable if the sequence of switch actions occur
> within 16 mS of each other in sequence, this can be
> done. Your 5 hour time line is not clear and may also
> be a problem."
>
> Then forget I mentioned the 5 hour time line. :-)
>
> "What exactly are you trying to build? Without more
> details or a more
> specific question, you might wind up re-posting this
> 100 times w/o
> getting the help you want..."
>
> I honestly don't know what I haven't covered, or how
> to explain anything more clear than I already have. It
> is a very simple project.
>
> Thanks
>
> Darren Harris
> Staten Island, New York.
>
> Searcher7@m...
>
>
>
>
> **************************************************************
> **************************************************************
> *****************************************
> --- "Chris Loiacono (E-mail)"
> <chris01@t...> wrote:
> >
> > > Darren:
> >
> > Hardware selection depends upon what type of loads
> > you are switching, and
> > exactly what you mean in the timing you describe.
> > Some timing requirements
> > may create difficulties with a Basic Stamp, they're
> > not super-fast, and RT
> > requires additional hardware. If what you described
> > means that it would be
> > acceptable if the sequence of switch actions occur
> > within 16 mS of each
> > other in sequence, this can be done. Your 5 hour
> > time line is not clear and
> > may also be a problem.
> >
> > What exactly are you trying to build? Without more
> > details or a more
> > specific question, you might wind up re-posting this
> > 100 times w/o getting
> > the help you want...
> >
> > Chris
> >
> >
> > >
> > > Hello,
> > >
> > > I was hoping to get recommendations on what
> > hardware would satisfy
> > > the needs of a beginner. I'm a complete newbie who
> > knows nothing
> > > about microcontrollers, and was hoping that
> > someone could give me an
> > > idea about where to start.
> > >
> > > My immediate project only entails very simple
> > control commands. The
> > > idea is to somehow connect a microcontroller to 4
> > different leaf
> > > switches, so I can close and open any of the 4
> > circuits in any of *14
> > > different pre-programmed sequences*. Each sequence
> > will have as many
> > > as 66 total commands, and will be executed within
> > approximately 2
> > > minutes. I want to be able to "press a button" at
> > anytime along a 5
> > > hour time-line to activate any of those 14 command
> > sequences with
> > > timing accuracy between each individual command no
> > worse than 1/60th
> > > of a second(within that sequence).
> > >
> > > The microcontroller will in effect be duplicating
> > the exact timing of
> > > the commands in each sequence.(Commands that I
> > will like to manually
> > > program into it via the four leaf switches I
> > mentioned).
> > >
> > > Any advice would be greatly appreciated.
> > >
> > > Thanks
> > >
> > > Darren Harris
> > > Staten Island, New York.
> > >
> > > Searcher7@m...
> > >
> > >
> > >
> > >
> > >
> > >
> > > Your use of Yahoo! Groups is subject to
> > > http://docs.yahoo.com/info/terms/
> > >
> > >
> >
> >
> >
> >
> > Your use of Yahoo! Groups is subject to
> > http://docs.yahoo.com/info/terms/
> >
> >
>
>
> __________________________________________________
> Do You Yahoo!?
> Get email at your own domain with Yahoo! Mail.
> http://personal.mail.yahoo.com/
>
>
>
> Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/

Archiver

"The Basic Stamp 2 will do this as long as you realize
that there are no
interrupts. This implies that you will need to loop
through the 4 inputs
continuously and check their state."

Providing I understand you correctly, is this a manual
process or will the Basic Stamp do this?

"As long as the program doesn't take more than 16mS to
loop, which is a long time- even for a stamp, you
should be OK."

"The 'button' instruction will help you 'debounce'
your manual switches."

This is all Greek to me.

"If the toughest loop is 112 commands in 2 minutes or
less, you're talking more than 1 second per, so it
also is not too fast. More effort will go into making
it take 2 minutes, if that's what you need."

It is not the length of time. I just said that a
sequence may take about 2 minutes.

"Just realize that again, while the 112 instructions
are executing in their own loop, the only thing the
stamp can do is wait for the loop to execute, or be
reset."

I assume that this means that it will follow the
sequence I initiate to it's end, or I can manually
reset, and start it(or another sequence) over?

"If I were you, I'd read the manual and 'stamps in
class' downloads at www.parallaxinc.com"

I don't understand much of what you said, but I will
check out the manual you mentioned.

Nevertheless, here is just one more clarification of
the basics of what I'm attempting...

***Years ago there was a multiplayer "family" game
advertised on T.V. It entailed a device that would
flash lights in a certain sequence. The player would
then have to remember and duplicate that sequence by
pressing the corresponding lighted buttons. The
farther you got into the game the longer the sequence
of lights you had to duplicate.(The more difficult it
became).

Now if you can imagine the computer in the human's
place, and vice-versa, you will understand bacically
what I'm attempting to accomplish. I will be using a
4-way joystick to "input" 14 separate sequences
containing many "on" and "off" commands to all 4
connections.(As I said, each sequence will last a
maximum of about 2 minutes, and have a total of 112
commands). Now, when I press one of 14 buttons, it is
the microcontroller that will have to duplicate the
corresponding sequence to within 1/60th of a
second(between each and every command in that
sequence). So for example, if I'm using it on a game
like Pacman, the microcontroller will be able to
"repeat" any of my patterns, because I would allow it
"read" and remember my joystick movements. It will
then "output" any of the 14 programmed patterns when I
press the corresponding button.

Thanks

Darren Harris
Staten Island, New York.

Searcher7@m...
********************************************************************************\
********************************************************************************\
*****
--- "Chris Loiacono (E-mail)"
<chris01@t...> wrote:
> The Basic Stamp 2 will do this as long as you
> realize that there are no
> interrupts. This implies that you will need to loop
> through the 4 inputs
> continuously and check their state. As long as the
> program doesn't take more
> than 16mS to loop, which is a long time- even for a
> stamp, you should be OK.
> The 'button' instruction will help you 'debounce'
> your manual switches.
> If the toughest loop is 112 commands in 2 minutes or
> less, you're talking
> more than 1 second per, so it also is not too fast.
> More effort will go into
> making it take 2 minutes, if that's what you need.
> Just realize that again,
> while the 112 instructions are executing in their
> own loop, the only thing
> the stamp can do is wait for the loop to execute, or
> be reset.
>
> If I were you, I'd read the manual and 'stamps in
> class' downloads at
> www.parallaxinc.com
>
>
> CL
>
> >

---

Original Message

---

> > From: Darren Harris [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=f1OWD_l6vjw49Z_5ekrQEx7WjDwH5vH3aKFVccyCY0FZ1YVBgYeFSPcYXSmtjMYFGuFxHL0gu1lDr_qN5A]searcher731@y...[/url
> > Sent: Thursday, April 12, 2001 11:58 AM
> > To: basicstamps@yahoogroups.com
> > Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please
> :-)
> >
> >
> > "Hardware selection depends upon what type of
> loads
> > you are switching,
> > and exactly what you mean in the timing you
> describe."
> >
> >
> > The loads are not at present an issue. After I
> figure
> > out what electronic components(microcontroller) to
> > get, I will have to figure out a way to interface
> it
> > to the leaf switches I mentioned. The only "input"
> > criteria for the microcontroller would be to
> accept
> > any of 8 possible commands it will receive when I
> > manually close and open any of four circuits. The
> more
> > difficult part would be the output of any of those
> 8
> > commands, since these circuits(leaf switches) have
> to
> > by physically closed in order for the PCB on the
> other
> > end to notice, I'd have to come up with a way to
> send
> > a strong enough current through the wires to fool
> > it.(I'm actually going to be using the 4-way
> joystick
> > controls of an arcade game).
> >
> > I don't understand your question about the timing.
> If
> > I were to "open", let's say switch "B" exactly 1.2
> > seconds after I closed it, I want to
> microcontroller
> > to output the same(to an accuracy of no worse than
> > 1/60th of a second) when I give it the command to
> do
> > so. Actually, the longest of the 14 sequences will
> be
> > 112 different commands in about a two minute
> > period.(And only 1 of the 4 circuits can actually
> be
> > "closed" at one time).
> >
> > "Some timing requirements may create difficulties
> with
> > a Basic Stamp, they're not super-fast, and RT
> requires
> > additional hardware."
> >
> > That is why I'm trying to find out what I'd need
> to
> > accomplish what I want.(And what is "RT"?).
> >
> > "If what you described means that it would be
> > acceptable if the sequence of switch actions occur
> > within 16 mS of each other in sequence, this can
> be
> > done. Your 5 hour time line is not clear and may
> also
> > be a problem."
> >
> > Then forget I mentioned the 5 hour time line. :-)
> >
> > "What exactly are you trying to build? Without
> more
> > details or a more
> > specific question, you might wind up re-posting
> this
> > 100 times w/o
> > getting the help you want..."
> >
> > I honestly don't know what I haven't covered, or
> how
> > to explain anything more clear than I already
> have. It
> > is a very simple project.
> >
> > Thanks
> >
> > Darren Harris
> > Staten Island, New York.
> >
> > Searcher7@m...
> >
> >
> >
> >
> >
>
**************************************************************
> >
>
**************************************************************
> > *****************************************
> > --- "Chris Loiacono (E-mail)"
> > <chris01@t...> wrote:
> > >
> > > > Darren:
> > >
> > > Hardware selection depends upon what type of
> loads
> > > you are switching, and
> > > exactly what you mean in the timing you
> describe.
> > > Some timing requirements
> > > may create difficulties with a Basic Stamp,
> they're
> > > not super-fast, and RT
> > > requires additional hardware. If what you
> described
> > > means that it would be
> > > acceptable if the sequence of switch actions
> occur
> > > within 16 mS of each
> > > other in sequence, this can be done. Your 5 hour
> > > time line is not clear and
> > > may also be a problem.
> > >
> > > What exactly are you trying to build? Without
> more
> > > details or a more
> > > specific question, you might wind up re-posting
> this
> > > 100 times w/o getting
> > > the help you want...
> > >
> > > Chris
> > >
> > >
> > > >
> > > > Hello,
> > > >
> > > > I was hoping to get recommendations on what
> > > hardware would satisfy
> > > > the needs of a beginner. I'm a complete newbie
> who
> > > knows nothing
> > > > about microcontrollers, and was hoping that
> > > someone could give me an
> > > > idea about where to start.
> > > >
> > > > My immediate project only entails very simple
> > > control commands. The
> > > > idea is to somehow connect a microcontroller
> to 4
> > > different leaf
> > > > switches, so I can close and open any of the 4
> > > circuits in any of *14
> > > > different pre-programmed sequences*. Each
> sequence
> > > will have as many
> > > > as 66 total commands, and will be executed
> within
> > > approximately 2
> > > > minutes. I want to be able to "press a button"
> at
> > > anytime along a 5
> > > > hour time-line to activate any of those 14
> command
> > > sequences with
> > > > timing accuracy between each individual
> command no
> > > worse than 1/60th
> > > > of a second(within that sequence).
>
=== message truncated ===


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/

Archiver

RE: [noparse][[/noparse]basicstamps] Recommendations please :-)

So basically you want to 'Record' the movement of the joystick for later playback.· In order to do this, you will have to have two 'modes', a record mode and a playback mode. This would have to be switchable via a switch, or maybe a pushbutton that would initiate a 'record' sequence.· Then you will also need a way of choosing which sequence you want to play back, and this could be as simple as a number of pushbutton switches.

You would have two sections to the program that ran on the stamp.

The first would be the recorder and it would be initiated by the record button.· You would then move the joystick around and the stamp would scan the joystick switches on the joystick as quickly as it could and store the state of the switches into some sort of a storage device. This portion contains a potential problem, because each sample will have to be stored for later playback and you could potentially have a large amount of storage required.· If you are sampling at 1/60 of a second, you will have 4 bits of data for every 1/60 of a second which comes out to 30 bytes per second or 1.8K per minute. That means that each 2 minute sequence will require 1.8K of storage.· This will exceed the internal storage ability of the stamp and will require an external storage device.· There may also be problems storing this data fast enough to get it saved and then be ready to scan the joystick switches again within the 1/60 of a second time frame.· The record process could be terminated by pressing one of the buttons that are used for playback. Once this button press was complete the sequence could be assigned to that button for playback.

The playback portion would wait for a button press and when a button was pressed, it would retrieve the data from the external storage device and configure it's pins for outputs.· It would then run through a loop and every 1/60 of a second it would retrieve and 'playback' the stored sample for that time period.

Basically what you are trying to make is a 4 bit digital recorder.·

Someone mentioned 'debouncing'. This refers to the fact that a switch doesn't make a clean on/off transistion.· It arcs and springs and makes dozens of contacts for each press of the switch.· The stamp is fast enough that it can detect these millisecond long contacts.· One switch press could be detected as 10 or 20 or more presses because of this effect.· I wouldn't worry about this though. What you are trying to do is record the actual movement of the switches on the joystick. The video game that you are playing back into expects the signals to be noisy and will be able to handle it.

Your main problems are going to be timing and storage.· Getting the playback to happen at the same speed as the recording will be difficult.· You will have to do some research and find out how long each instruction takes and make sure that your playback and record loops have the EXACT same execution times.

Interupts were also mentioned.· What this means is that there are two methods of letting a microcontroller know that something has happened that needs it's attention.· With the basic stamp the only way that really exists is for the stamp to be in a loop (a section of code that executes over and over) and just keep looking at an input over and over again until it gets pressed and then it knows it needs to do something.· Other Microcontrollers support interupts which are a special input to the microcontroller that will 'interupt' what the processor is doing and FORCE it to execute a section of code when the input happens.· This way the processor doesn't have to worry about whether or not any inputs have been pressed until it is interupted and then has to deal with it.· This is not an option on the stamp.

--Scott

---

Original Message

---

From: Darren Harris [noparse]/noparse][url=mailto:searcher731@yahoo.com]mailto:searcher731@yahoo.com[/url
Sent: Thursday, April 12, 2001 10:18 AM
To: basicstamps@yahoogroups.com
Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please :-)


"The Basic Stamp 2 will do this as long as you realize
that there are no
interrupts. This implies that you will need to loop
through the 4 inputs
continuously and check their state."

Providing I understand you correctly, is this a manual
process or will the Basic Stamp do this?

"As long as the program doesn't take more than 16mS to
loop, which is a long time- even for a stamp, you
should be OK."

"The 'button' instruction will help you 'debounce'
your manual switches."

This is all Greek to me.

"If the toughest loop is 112 commands in 2 minutes or
less, you're talking more than 1 second per, so it
also is not too fast. More effort will go into making
it take 2 minutes, if that's what you need."

It is not the length of time. I just said that a
sequence may take about 2 minutes.

"Just realize that again, while the 112 instructions
are executing in their own loop, the only thing the
stamp can do is wait for the loop to execute, or be
reset."

I assume that this means that it will follow the
sequence I initiate to it's end, or I can manually
reset, and start it(or another sequence) over?

"If I were you, I'd read the manual and 'stamps in
class' downloads at www.parallaxinc.com"

I don't understand much of what you said, but I will
check out the manual you mentioned.

Nevertheless, here is just one more clarification of
the basics of what I'm attempting...

***Years ago there was a multiplayer "family" game
advertised on T.V. It entailed a device that would
flash lights in a certain sequence. The player would
then have to remember and duplicate that sequence by
pressing the corresponding lighted buttons. The
farther you got into the game the longer the sequence
of lights you had to duplicate.(The more difficult it
became).

Now if you can imagine the computer in the human's
place, and vice-versa, you will understand bacically
what I'm attempting to accomplish. I will be using a
4-way joystick to "input" 14 separate sequences
containing many "on" and "off" commands to all 4
connections.(As I said, each sequence will last a
maximum of about 2 minutes, and have a total of 112
commands). Now, when I press one of 14 buttons, it is
the microcontroller that will have to duplicate the
corresponding sequence to within 1/60th of a
second(between each and every command in that
sequence). So for example, if I'm using it on a game
like Pacman, the microcontroller will be able to
"repeat" any of my patterns, because I would allow it
"read" and remember my joystick movements. It will
then "output" any of the 14 programmed patterns when I
press the corresponding button.

Thanks

Darren Harris
Staten Island, New York.

Searcher7@mail.con2.com
*********************************************************************************************************************************************************************

--- "Chris Loiacono (E-mail)"
<chris01@tampabay.rr.com> wrote:
> The Basic Stamp 2 will do this as long as you
> realize that there are no
> interrupts. This implies that you will need to loop
> through the 4 inputs
> continuously and check their state. As long as the
> program doesn't take more
> than 16mS to loop, which is a long time- even for a
> stamp, you should be OK.
> The 'button' instruction will help you 'debounce'
> your manual switches.
> If the toughest loop is 112 commands in 2 minutes or
> less, you're talking
> more than 1 second per, so it also is not too fast.
> More effort will go into
> making it take 2 minutes, if that's what you need.
> Just realize that again,
> while the 112 instructions are executing in their
> own loop, the only thing
> the stamp can do is wait for the loop to execute, or
> be reset.
>
> If I were you, I'd read the manual and 'stamps in
> class' downloads at
> www.parallaxinc.com
>
>
> CL
>
> >

---

Original Message

---

> > From: Darren Harris [noparse]/noparse][url=mailto:searcher731@yahoo.com]mailto:searcher731@yahoo.com[/url
> > Sent: Thursday, April 12, 2001 11:58 AM
> > To: basicstamps@yahoogroups.com
> > Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please
> :-)
> >
> >
> > "Hardware selection depends upon what type of
> loads
> > you are switching,
> > and exactly what you mean in the timing you
> describe."
> >
> >
> > The loads are not at present an issue. After I
> figure
> > out what electronic components(microcontroller) to
> > get, I will have to figure out a way to interface
> it
> > to the leaf switches I mentioned. The only "input"
> > criteria for the microcontroller would be to
> accept
> > any of 8 possible commands it will receive when I
> > manually close and open any of four circuits. The
> more
> > difficult part would be the output of any of those
> 8
> > commands, since these circuits(leaf switches) have
> to
> > by physically closed in order for the PCB on the
> other
> > end to notice, I'd have to come up with a way to
> send
> > a strong enough current through the wires to fool
> > it.(I'm actually going to be using the 4-way
> joystick
> > controls of an arcade game).
> >
> > I don't understand your question about the timing.
> If
> > I were to "open", let's say switch "B" exactly 1.2
> > seconds after I closed it, I want to
> microcontroller
> > to output the same(to an accuracy of no worse than
> > 1/60th of a second) when I give it the command to
> do
> > so. Actually, the longest of the 14 sequences will
> be
> > 112 different commands in about a two minute
> > period.(And only 1 of the 4 circuits can actually
> be
> > "closed" at one time).
> >
> > "Some timing requirements may create difficulties
> with
> > a Basic Stamp, they're not super-fast, and RT
> requires
> > additional hardware."
> >
> > That is why I'm trying to find out what I'd need
> to
> > accomplish what I want.(And what is "RT"?).
> >
> > "If what you described means that it would be
> > acceptable if the sequence of switch actions occur
> > within 16 mS of each other in sequence, this can
> be
> > done. Your 5 hour time line is not clear and may
> also
> > be a problem."
> >
> > Then forget I mentioned the 5 hour time line. :-)
> >
> > "What exactly are you trying to build? Without
> more
> > details or a more
> > specific question, you might wind up re-posting
> this
> > 100 times w/o
> > getting the help you want..."
> >
> > I honestly don't know what I haven't covered, or
> how
> > to explain anything more clear than I already
> have. It
> > is a very simple project.
> >
> > Thanks
> >
> > Darren Harris
> > Staten Island, New York.
> >
> > Searcher7@mail.con2.com
> >
> >
> >
> >
> >
>
**************************************************************
> >
>
**************************************************************
> > *****************************************
> > --- "Chris Loiacono (E-mail)"
> > <chris01@tampabay.rr.com> wrote:
> > >
> > > > Darren:
> > >
> > > Hardware selection depends upon what type of
> loads
> > > you are switching, and
> > > exactly what you mean in the timing you
> describe.
> > > Some timing requirements
> > > may create difficulties with a Basic Stamp,
> they're
> > > not super-fast, and RT
> > > requires additional hardware. If what you
> described
> > > means that it would be
> > > acceptable if the sequence of switch actions
> occur
> > > within 16 mS of each
> > > other in sequence, this can be done. Your 5 hour
> > > time line is not clear and
> > > may also be a problem.
> > >
> > > What exactly are you trying to build? Without
> more
> > > details or a more
> > > specific question, you might wind up re-posting
> this
> > > 100 times w/o getting
> > > the help you want...
> > >
> > > Chris
> > >
> > >
> > > >
> > > > Hello,
> > > >
> > > > I was hoping to get recommendations on what
> > > hardware would satisfy
> > > > the needs of a beginner. I'm a complete newbie
> who
> > > knows nothing
> > > > about microcontrollers, and was hoping that
> > > someone could give me an
> > > > idea about where to start.
> > > >
> > > > My immediate project only entails very simple
> > > control commands. The
> > > > idea is to somehow connect a microcontroller
> to 4
> > > different leaf
> > > > switches, so I can close and open any of the 4
> > > circuits in any of *14
> > > > different pre-programmed sequences*. Each
> sequence
> > > will have as many
> > > > as 66 total commands, and will be executed
> within
> > > approximately 2
> > > > minutes. I want to be able to "press a button"
> at
> > > anytime along a 5
> > > > hour time-line to activate any of those 14
> command
> > > sequences with
> > > > timing accuracy between each individual
> command no
> > > worse than 1/60th
> > > > of a second(within that sequence).
>
=== message truncated ===


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/

·

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Archiver

Darren,
Am I right in thinking that you want to 'record' up to 14 different 'tracks'
each track at most 2 minutes long with a max of 112 moves of the joystick.
([noparse][[/noparse]UP 1.2s] - [noparse][[/noparse]DOWN 0.2] - [noparse][[/noparse]LEFT 1.4s] = 3 moves).

Then when it's done you would like to select which of the 14 tracks you want
to 'play back' and press the 'play button'.
Is this what you like to do?
Is the 'record' joystick the same as the 'playpack' joystick?
If so, you could use relays, transistors, optocouplers etc connected to
the STAMP pins across the switches to simulate 'real' closures of
the contacts.
Is the joystick 5V....?
What will be the longest time that you would hold the stick in any given
direction?

If this is way off please excuse me!

Regards and good luck!
/Henrik Olsson. Sweden.

---

Original Message

---

>
> The loads are not at present an issue. After I figure
> out what electronic components(microcontroller) to
> get, I will have to figure out a way to interface it
> to the leaf switches I mentioned. The only "input"
> criteria for the microcontroller would be to accept
> any of 8 possible commands it will receive when I
> manually close and open any of four circuits. The more
> difficult part would be the output of any of those 8
> commands, since these circuits(leaf switches) have to
> by physically closed in order for the PCB on the other
> end to notice, I'd have to come up with a way to send
> a strong enough current through the wires to fool
> it.(I'm actually going to be using the 4-way joystick
> controls of an arcade game).
>
> I don't understand your question about the timing. If
> I were to "open", let's say switch "B" exactly 1.2
> seconds after I closed it, I want to microcontroller
> to output the same(to an accuracy of no worse than
> 1/60th of a second) when I give it the command to do
> so. Actually, the longest of the 14 sequences will be
> 112 different commands in about a two minute
> period.(And only 1 of the 4 circuits can actually be
> "closed" at one time).
>
> "Some timing requirements may create difficulties with
> a Basic Stamp, they're not super-fast, and RT requires
> additional hardware."
>
> That is why I'm trying to find out what I'd need to
> accomplish what I want.(And what is "RT"?).
>
> "If what you described means that it would be
> acceptable if the sequence of switch actions occur
> within 16 mS of each other in sequence, this can be
> done. Your 5 hour time line is not clear and may also
> be a problem."
>
> Then forget I mentioned the 5 hour time line. :-)
>
> "What exactly are you trying to build? Without more
> details or a more
> specific question, you might wind up re-posting this
> 100 times w/o
> getting the help you want..."
>
> I honestly don't know what I haven't covered, or how
> to explain anything more clear than I already have. It
> is a very simple project.
>
> Thanks
>
> Darren Harris
> Staten Island, New York.
>
> Searcher7@m...

Archiver

It's amazing how apparently simple things turn out to
be so complicated. If I had the tools, I'd actually be
able to accomplish this with a timer based(mechanical)
mechanism. Anyway...

"There may also be problems storing this data fast
enough to get it saved and then be ready to scan the
joystick switches again within the 1/60 of a second
time frame."

Throughout any sequence, two instructions wouldn't
occur within about 1/10th of a second anyway.

"Your main problems are going to be timing and
storage. Getting the playback to happen at the same
speed as the recording will be difficult. You will
have to do some research and find out how long each
instruction takes and make sure that your playback and
record loops have the EXACT same execution times."

Since input would be manual, I don't know how that is
possible. But if the microcontroller can read my
manual inputs of a particular sequence, then why
wouldn't it be able to just duplicate all individual
commands in that sequence as outputs(with that 1/60th
of a second accuracy I mentioned)?

And if storage is an issue, then is there a way to use
my PC(hard drive)?

So I'm assuming that a Basic Stamp 2 is what I would
need for my project? Is there any thing else?

Thanks

Darren Harris
Staten Island, New York.

Searcher7@m...
********************************************************************************\
********************************************************************************\
*****
--- Scott Winn <scott.winn@m...> wrote:
>
> So basically you want to 'Record' the movement of
> the joystick for later
> playback. In order to do this, you will have to
> have two 'modes', a record
> mode and a playback mode. This would have to be
> switchable via a switch, or
> maybe a pushbutton that would initiate a 'record'
> sequence. Then you will
> also need a way of choosing which sequence you want
> to play back, and this
> could be as simple as a number of pushbutton
> switches.
>
> You would have two sections to the program that ran
> on the stamp.
>
> The first would be the recorder and it would be
> initiated by the record
> button. You would then move the joystick around and
> the stamp would scan
> the joystick switches on the joystick as quickly as
> it could and store the
> state of the switches into some sort of a storage
> device. This portion
> contains a potential problem, because each sample
> will have to be stored for
> later playback and you could potentially have a
> large amount of storage
> required. If you are sampling at 1/60 of a second,
> you will have 4 bits of
> data for every 1/60 of a second which comes out to
> 30 bytes per second or
> 1.8K per minute. That means that each 2 minute
> sequence will require 1.8K of
> storage. This will exceed the internal storage
> ability of the stamp and
> will require an external storage device. There may
> also be problems storing
> this data fast enough to get it saved and then be
> ready to scan the joystick
> switches again within the 1/60 of a second time
> frame. The record process
> could be terminated by pressing one of the buttons
> that are used for
> playback. Once this button press was complete the
> sequence could be assigned
> to that button for playback.
>
> The playback portion would wait for a button press
> and when a button was
> pressed, it would retrieve the data from the
> external storage device and
> configure it's pins for outputs. It would then run
> through a loop and every
> 1/60 of a second it would retrieve and 'playback'
> the stored sample for that
> time period.
>
> Basically what you are trying to make is a 4 bit
> digital recorder.
>
> Someone mentioned 'debouncing'. This refers to the
> fact that a switch
> doesn't make a clean on/off transistion. It arcs
> and springs and makes
> dozens of contacts for each press of the switch.
> The stamp is fast enough
> that it can detect these millisecond long contacts.
> One switch press could
> be detected as 10 or 20 or more presses because of
> this effect. I wouldn't
> worry about this though. What you are trying to do
> is record the actual
> movement of the switches on the joystick. The video
> game that you are
> playing back into expects the signals to be noisy
> and will be able to handle
> it.
>
> Your main problems are going to be timing and
> storage. Getting the playback
> to happen at the same speed as the recording will be
> difficult. You will
> have to do some research and find out how long each
> instruction takes and
> make sure that your playback and record loops have
> the EXACT same execution
> times.
>
> Interupts were also mentioned. What this means is
> that there are two
> methods of letting a microcontroller know that
> something has happened that
> needs it's attention. With the basic stamp the only
> way that really exists
> is for the stamp to be in a loop (a section of code
> that executes over and
> over) and just keep looking at an input over and
> over again until it gets
> pressed and then it knows it needs to do something.
> Other Microcontrollers
> support interupts which are a special input to the
> microcontroller that will
> 'interupt' what the processor is doing and FORCE it
> to execute a section of
> code when the input happens. This way the processor
> doesn't have to worry
> about whether or not any inputs have been pressed
> until it is interupted and
> then has to deal with it. This is not an option on
> the stamp.
>
> --Scott
>
>
>

---

Original Message

---

> From: Darren Harris [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=wN1VhxNTPYwCu5t3ZhF5DAGmKjcpaysEQLUpFmYgFtiTuWmIjhYSaFzU3jE7HB19fL5iv6T_0HBMEy9cCwc2zw]searcher731@y...[/url
> Sent: Thursday, April 12, 2001 10:18 AM
> To: basicstamps@yahoogroups.com
> Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please
> :-)
>
>
> "The Basic Stamp 2 will do this as long as you
> realize
> that there are no
> interrupts. This implies that you will need to loop
> through the 4 inputs
> continuously and check their state."
>
> Providing I understand you correctly, is this a
> manual
> process or will the Basic Stamp do this?
>
> "As long as the program doesn't take more than 16mS
> to
> loop, which is a long time- even for a stamp, you
> should be OK."
>
> "The 'button' instruction will help you 'debounce'
> your manual switches."
>
> This is all Greek to me.
>
> "If the toughest loop is 112 commands in 2 minutes
> or
> less, you're talking more than 1 second per, so it
> also is not too fast. More effort will go into
> making
> it take 2 minutes, if that's what you need."
>
> It is not the length of time. I just said that a
> sequence may take about 2 minutes.
>
> "Just realize that again, while the 112 instructions
> are executing in their own loop, the only thing the
> stamp can do is wait for the loop to execute, or be
> reset."
>
> I assume that this means that it will follow the
> sequence I initiate to it's end, or I can manually
> reset, and start it(or another sequence) over?
>
> "If I were you, I'd read the manual and 'stamps in
> class' downloads at www.parallaxinc.com"
>
> I don't understand much of what you said, but I will
> check out the manual you mentioned.
>
> Nevertheless, here is just one more clarification of
> the basics of what I'm attempting...
>
> ***Years ago there was a multiplayer "family" game
> advertised on T.V. It entailed a device that would
> flash lights in a certain sequence. The player would
> then have to remember and duplicate that sequence by
> pressing the corresponding lighted buttons. The
> farther you got into the game the longer the
> sequence
> of lights you had to duplicate.(The more difficult
> it
> became).
>
> Now if you can imagine the computer in the human's
> place, and vice-versa, you will understand bacically
> what I'm attempting to accomplish. I will be using a
> 4-way joystick to "input" 14 separate sequences
> containing many "on" and "off" commands to all 4
> connections.(As I said, each sequence will last a
> maximum of about 2 minutes, and have a total of 112
> commands). Now, when I press one of 14 buttons, it
> is
> the microcontroller that will have to duplicate the
> corresponding sequence to within 1/60th of a
> second(between each and every command in that
> sequence). So for example, if I'm using it on a game
> like Pacman, the microcontroller will be able to
> "repeat" any of my patterns, because I would allow
> it
> "read" and remember my joystick movements. It will
>
=== message truncated ===


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/

Archiver

I'd think about using a BS2P and storing the recordings in the unused
program area. You could compress the storage by "run length encoding". So
you'd say (not real stamp code):

scancount=0
eeaddress=0
laststate=-1
loop:
state=getstate ' 0=up, 1=up/right, 2=up, or whatever... never -1
if state<>laststate then writeit
scancount=scancount+1
goto loop

writeit:
if laststate=-1 then skipwrite
write laststate, scancount (to EEPROM)
eeaddress=eeaddress+1
skipwrite:
scancount=1
laststate=state
goto loop


Of course, you need to know when to stop too. This would wind up writing to
EEPROM a code and a number of "scans" that that code was in force. Playback
would be similar.

Just a thought.

Al Williams
AWC
*Expand your Stamp I/O: http://www.al-williams.com/awce/pak3.htm


>

---

Original Message

---

> From: Darren Harris [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=jd3oELYKBkLyE03pEpfPOnzdb17uPh0E6gtfGbnBq4w_XmoX1IMAbsGm07_69FR5DQiqXFDyb3xnWdM7UQ]searcher731@y...[/url
> Sent: Thursday, April 12, 2001 1:59 PM
> To: basicstamps@yahoogroups.com
> Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please :-)
>
>
> It's amazing how apparently simple things turn out to
> be so complicated. If I had the tools, I'd actually be
> able to accomplish this with a timer based(mechanical)
> mechanism. Anyway...
>
> "There may also be problems storing this data fast
> enough to get it saved and then be ready to scan the
> joystick switches again within the 1/60 of a second
> time frame."
>
> Throughout any sequence, two instructions wouldn't
> occur within about 1/10th of a second anyway.
>
> "Your main problems are going to be timing and
> storage. Getting the playback to happen at the same
> speed as the recording will be difficult. You will
> have to do some research and find out how long each
> instruction takes and make sure that your playback and
> record loops have the EXACT same execution times."
>
> Since input would be manual, I don't know how that is
> possible. But if the microcontroller can read my
> manual inputs of a particular sequence, then why
> wouldn't it be able to just duplicate all individual
> commands in that sequence as outputs(with that 1/60th
> of a second accuracy I mentioned)?
>
> And if storage is an issue, then is there a way to use
> my PC(hard drive)?
>
> So I'm assuming that a Basic Stamp 2 is what I would
> need for my project? Is there any thing else?
>
> Thanks
>
> Darren Harris
> Staten Island, New York.
>
> Searcher7@m...
> ******************************************************************
> ******************************************************************
> *********************************
> --- Scott Winn <scott.winn@m...> wrote:
> >
> > So basically you want to 'Record' the movement of
> > the joystick for later
> > playback. In order to do this, you will have to
> > have two 'modes', a record
> > mode and a playback mode. This would have to be
> > switchable via a switch, or
> > maybe a pushbutton that would initiate a 'record'
> > sequence. Then you will
> > also need a way of choosing which sequence you want
> > to play back, and this
> > could be as simple as a number of pushbutton
> > switches.
> >
> > You would have two sections to the program that ran
> > on the stamp.
> >
> > The first would be the recorder and it would be
> > initiated by the record
> > button. You would then move the joystick around and
> > the stamp would scan
> > the joystick switches on the joystick as quickly as
> > it could and store the
> > state of the switches into some sort of a storage
> > device. This portion
> > contains a potential problem, because each sample
> > will have to be stored for
> > later playback and you could potentially have a
> > large amount of storage
> > required. If you are sampling at 1/60 of a second,
> > you will have 4 bits of
> > data for every 1/60 of a second which comes out to
> > 30 bytes per second or
> > 1.8K per minute. That means that each 2 minute
> > sequence will require 1.8K of
> > storage. This will exceed the internal storage
> > ability of the stamp and
> > will require an external storage device. There may
> > also be problems storing
> > this data fast enough to get it saved and then be
> > ready to scan the joystick
> > switches again within the 1/60 of a second time
> > frame. The record process
> > could be terminated by pressing one of the buttons
> > that are used for
> > playback. Once this button press was complete the
> > sequence could be assigned
> > to that button for playback.
> >
> > The playback portion would wait for a button press
> > and when a button was
> > pressed, it would retrieve the data from the
> > external storage device and
> > configure it's pins for outputs. It would then run
> > through a loop and every
> > 1/60 of a second it would retrieve and 'playback'
> > the stored sample for that
> > time period.
> >
> > Basically what you are trying to make is a 4 bit
> > digital recorder.
> >
> > Someone mentioned 'debouncing'. This refers to the
> > fact that a switch
> > doesn't make a clean on/off transistion. It arcs
> > and springs and makes
> > dozens of contacts for each press of the switch.
> > The stamp is fast enough
> > that it can detect these millisecond long contacts.
> > One switch press could
> > be detected as 10 or 20 or more presses because of
> > this effect. I wouldn't
> > worry about this though. What you are trying to do
> > is record the actual
> > movement of the switches on the joystick. The video
> > game that you are
> > playing back into expects the signals to be noisy
> > and will be able to handle
> > it.
> >
> > Your main problems are going to be timing and
> > storage. Getting the playback
> > to happen at the same speed as the recording will be
> > difficult. You will
> > have to do some research and find out how long each
> > instruction takes and
> > make sure that your playback and record loops have
> > the EXACT same execution
> > times.
> >
> > Interupts were also mentioned. What this means is
> > that there are two
> > methods of letting a microcontroller know that
> > something has happened that
> > needs it's attention. With the basic stamp the only
> > way that really exists
> > is for the stamp to be in a loop (a section of code
> > that executes over and
> > over) and just keep looking at an input over and
> > over again until it gets
> > pressed and then it knows it needs to do something.
> > Other Microcontrollers
> > support interupts which are a special input to the
> > microcontroller that will
> > 'interupt' what the processor is doing and FORCE it
> > to execute a section of
> > code when the input happens. This way the processor
> > doesn't have to worry
> > about whether or not any inputs have been pressed
> > until it is interupted and
> > then has to deal with it. This is not an option on
> > the stamp.
> >
> > --Scott
> >
> >
> >

---

Original Message

---

> > From: Darren Harris [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=jd3oELYKBkLyE03pEpfPOnzdb17uPh0E6gtfGbnBq4w_XmoX1IMAbsGm07_69FR5DQiqXFDyb3xnWdM7UQ]searcher731@y...[/url
> > Sent: Thursday, April 12, 2001 10:18 AM
> > To: basicstamps@yahoogroups.com
> > Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please
> > :-)
> >
> >
> > "The Basic Stamp 2 will do this as long as you
> > realize
> > that there are no
> > interrupts. This implies that you will need to loop
> > through the 4 inputs
> > continuously and check their state."
> >
> > Providing I understand you correctly, is this a
> > manual
> > process or will the Basic Stamp do this?
> >
> > "As long as the program doesn't take more than 16mS
> > to
> > loop, which is a long time- even for a stamp, you
> > should be OK."
> >
> > "The 'button' instruction will help you 'debounce'
> > your manual switches."
> >
> > This is all Greek to me.
> >
> > "If the toughest loop is 112 commands in 2 minutes
> > or
> > less, you're talking more than 1 second per, so it
> > also is not too fast. More effort will go into
> > making
> > it take 2 minutes, if that's what you need."
> >
> > It is not the length of time. I just said that a
> > sequence may take about 2 minutes.
> >
> > "Just realize that again, while the 112 instructions
> > are executing in their own loop, the only thing the
> > stamp can do is wait for the loop to execute, or be
> > reset."
> >
> > I assume that this means that it will follow the
> > sequence I initiate to it's end, or I can manually
> > reset, and start it(or another sequence) over?
> >
> > "If I were you, I'd read the manual and 'stamps in
> > class' downloads at www.parallaxinc.com"
> >
> > I don't understand much of what you said, but I will
> > check out the manual you mentioned.
> >
> > Nevertheless, here is just one more clarification of
> > the basics of what I'm attempting...
> >
> > ***Years ago there was a multiplayer "family" game
> > advertised on T.V. It entailed a device that would
> > flash lights in a certain sequence. The player would
> > then have to remember and duplicate that sequence by
> > pressing the corresponding lighted buttons. The
> > farther you got into the game the longer the
> > sequence
> > of lights you had to duplicate.(The more difficult
> > it
> > became).
> >
> > Now if you can imagine the computer in the human's
> > place, and vice-versa, you will understand bacically
> > what I'm attempting to accomplish. I will be using a
> > 4-way joystick to "input" 14 separate sequences
> > containing many "on" and "off" commands to all 4
> > connections.(As I said, each sequence will last a
> > maximum of about 2 minutes, and have a total of 112
> > commands). Now, when I press one of 14 buttons, it
> > is
> > the microcontroller that will have to duplicate the
> > corresponding sequence to within 1/60th of a
> > second(between each and every command in that
> > sequence). So for example, if I'm using it on a game
> > like Pacman, the microcontroller will be able to
> > "repeat" any of my patterns, because I would allow
> > it
> > "read" and remember my joystick movements. It will
> >
> === message truncated ===
>
>
> __________________________________________________
> Do You Yahoo!?
> Get email at your own domain with Yahoo! Mail.
> http://personal.mail.yahoo.com/
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>

Archiver

At 11:58 AM 4/12/01 -0700, you wrote:

This is just an idea to conserve memory...


Read all 4 inputs at once as a single "NIBBLE"

(forget the button command)

Instead, read the NEW NIBBLE compared to the PREVIOUS NIBBLE

(for debounce, read the NEW NIBBLE a couple of times)

If the NEW NIBBLE is DIFFERENT than the PREVIOUS NIBBLE, then

log the 'time' and NIBBLE information.

(The 'time' variable could be an incremental variable inside the
compare loop between NEW NIBBLE and PREVIOUS NIBBLE it need not
be $TIME as in HH:MM:SS .... Example: Initialize 'time' to
Zero before entering the compare loop... Once inside the
compare loop if NEW NIBBLE equals PREVIOUS NIBBLE then increment
the 'time' variable. Use this value to determine a relative
amount of time the Joystick remained in one position.)

For multiple "tracks" you can segment the memory by a certain

offset, and just index it upon playback.




Beau Schwabe IC Mask Designer
National Semiconductor Network Products Division
500 Pinnacle Court, Suite 525 Mail Stop GA1 Norcross, GA 30071

Archiver

"Am I right in thinking that you want to 'record' up to 14 different
'tracks'
each track at most 2 minutes long with a max of 112 moves of the
joystick.
([noparse][[/noparse]UP 1.2s] - [noparse][[/noparse]DOWN 0.2] - [noparse][[/noparse]LEFT 1.4s] = 3 moves)."

yes. But the return from left, right, up, and down are important also.

"Then when it's done you would like to select which of the 14 tracks
you want
to 'play back' and press the 'play button'.Is this what you like to
do?"

Exactly.

"Is the 'record' joystick the same as the 'playpack' joystick?"

Actually, I have to find a way to interface the microcontroller to
the 4 leaf switches that the joystick touches to open and close those
circuits.

"If so, you could use relays, transistors, optocouplers etc connected
to
the STAMP pins across the switches to simulate 'real'
closures of
the contacts."

Huh? Remember, I'm a newbie. :-)

"Is the joystick 5V....?"

Yes.

"What will be the longest time that you would hold the stick in any
given
direction?"

About 7 seconds.

"If this is way off please excuse me!"

Actually, you seem to have a handle on what I'm attempting. I would
have given even more details as to why I need to do this, but it
would have been somewhat off topic.

Thanks.

Darren Harris
Staten Island, New York.

Searcher7@m...
**********************************************************************
**********************************************************************
**********************************************************************
--- In basicstamps@y..., "Henrik Olsson" <henrik-olsson@t...> wrote:
> Darren,
> Am I right in thinking that you want to 'record' up to 14 different
'tracks'
> each track at most 2 minutes long with a max of 112 moves of the
joystick.
> ([noparse][[/noparse]UP 1.2s] - [noparse][[/noparse]DOWN 0.2] - [noparse][[/noparse]LEFT 1.4s] = 3 moves).
>
> Then when it's done you would like to select which of the 14 tracks
you want
> to 'play back' and press the 'play button'.
> Is this what you like to do?
> Is the 'record' joystick the same as the 'playpack' joystick?
> If so, you could use relays, transistors, optocouplers etc
connected to
> the STAMP pins across the switches to simulate 'real'
closures of
> the contacts.
> Is the joystick 5V....?
> What will be the longest time that you would hold the stick in any
given
> direction?
>
> If this is way off please excuse me!
>
> Regards and good luck!
> /Henrik Olsson. Sweden.
>
>
>

---

Original Message

---

>
> > The loads are not at present an issue. After I figure
> > out what electronic components(microcontroller) to
> > get, I will have to figure out a way to interface it
> > to the leaf switches I mentioned. The only "input"
> > criteria for the microcontroller would be to accept
> > any of 8 possible commands it will receive when I
> > manually close and open any of four circuits. The more
> > difficult part would be the output of any of those 8
> > commands, since these circuits(leaf switches) have to
> > by physically closed in order for the PCB on the other
> > end to notice, I'd have to come up with a way to send
> > a strong enough current through the wires to fool
> > it.(I'm actually going to be using the 4-way joystick
> > controls of an arcade game).
> >
> > I don't understand your question about the timing. If
> > I were to "open", let's say switch "B" exactly 1.2
> > seconds after I closed it, I want to microcontroller
> > to output the same(to an accuracy of no worse than
> > 1/60th of a second) when I give it the command to do
> > so. Actually, the longest of the 14 sequences will be
> > 112 different commands in about a two minute
> > period.(And only 1 of the 4 circuits can actually be
> > "closed" at one time).
> >
> > "Some timing requirements may create difficulties with
> > a Basic Stamp, they're not super-fast, and RT requires
> > additional hardware."
> >
> > That is why I'm trying to find out what I'd need to
> > accomplish what I want.(And what is "RT"?).
> >
> > "If what you described means that it would be
> > acceptable if the sequence of switch actions occur
> > within 16 mS of each other in sequence, this can be
> > done. Your 5 hour time line is not clear and may also
> > be a problem."
> >
> > Then forget I mentioned the 5 hour time line. :-)
> >
> > "What exactly are you trying to build? Without more
> > details or a more
> > specific question, you might wind up re-posting this
> > 100 times w/o
> > getting the help you want..."
> >
> > I honestly don't know what I haven't covered, or how
> > to explain anything more clear than I already have. It
> > is a very simple project.
> >
> > Thanks
> >
> > Darren Harris
> > Staten Island, New York.
> >
> > Searcher7@m...

Archiver

Darren,
I guess I understood you right.
You would need some sort of interface between the joystick and STAMP. If the
joystick's switches is connected to +5V only you could connect the switches
direct to the STAMP pins - for the recordmode, that is. For the playbackmode
you can use transistors conected across the switches in the joystick and
drive the base or gate of the transistor from another STAMP pin.
If it's not 5V on the joystick it gets a little more complicated but it sure
can be done. Is it 5V or 12V or maby 230VAC?
On the software side Scott gave you a couple of good ideas but like he says
the storagespace is the problem. Maby you can use an external EEPROM or RAM
connected to the STAMP.
Or you could just the skip the STAMP and connect the four switches to
LPT-port on your PC and record days and weeks of sequences.

Good luck!
/Henrik Olsson. Sweden.

---

Original Message

---

From: "Darren Harris" <searcher731@y...>
To: <basicstamps@yahoogroups.com>
Sent: Thursday, April 12, 2001 8:58 PM
Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please :-)


> It's amazing how apparently simple things turn out to
> be so complicated. If I had the tools, I'd actually be
> able to accomplish this with a timer based(mechanical)
> mechanism. Anyway...
>
> "There may also be problems storing this data fast
> enough to get it saved and then be ready to scan the
> joystick switches again within the 1/60 of a second
> time frame."
>
> Throughout any sequence, two instructions wouldn't
> occur within about 1/10th of a second anyway.
>
> "Your main problems are going to be timing and
> storage. Getting the playback to happen at the same
> speed as the recording will be difficult. You will
> have to do some research and find out how long each
> instruction takes and make sure that your playback and
> record loops have the EXACT same execution times."
>
> Since input would be manual, I don't know how that is
> possible. But if the microcontroller can read my
> manual inputs of a particular sequence, then why
> wouldn't it be able to just duplicate all individual
> commands in that sequence as outputs(with that 1/60th
> of a second accuracy I mentioned)?
>
> And if storage is an issue, then is there a way to use
> my PC(hard drive)?
>
> So I'm assuming that a Basic Stamp 2 is what I would
> need for my project? Is there any thing else?
>
> Thanks
>
> Darren Harris
> Staten Island, New York.
>
> Searcher7@m...
>

Archiver

I assume that what you said(whatever it all meant) is too complicated
to be covered in a manual that would come along with the BS2P(which I
assume is the Basic Stamp 2?).

It seems as though I'm going to have to do a lot of studying to
figure out how these things work.(But I'll save these messages for
reference).

Thanks

Darren Harris
Staten Island, New York.

Searcher7@m...
**********************************************************************
**********************************************************************
**********************************************************************



--- In basicstamps@y..., "Al Williams" <alw@a...> wrote:
> I'd think about using a BS2P and storing the recordings in the
unused
> program area. You could compress the storage by "run length
encoding". So
> you'd say (not real stamp code):
>
> scancount=0
> eeaddress=0
> laststate=-1
> loop:
> state=getstate ' 0=up, 1=up/right, 2=up, or whatever... never -1
> if state<>laststate then writeit
> scancount=scancount+1
> goto loop
>
> writeit:
> if laststate=-1 then skipwrite
> write laststate, scancount (to EEPROM)
> eeaddress=eeaddress+1
> skipwrite:
> scancount=1
> laststate=state
> goto loop
>
>
> Of course, you need to know when to stop too. This would wind up
writing to
> EEPROM a code and a number of "scans" that that code was in force.
Playback
> would be similar.
>
> Just a thought.
>
> Al Williams
> AWC
> *Expand your Stamp I/O: http://www.al-williams.com/awce/pak3.htm
>
>
> >

---

Original Message

---

> > From: Darren Harris [noparse][[/noparse]mailto:searcher731@y...]
> > Sent: Thursday, April 12, 2001 1:59 PM
> > To: basicstamps@y...
> > Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please :-)
> >
> >
> > It's amazing how apparently simple things turn out to
> > be so complicated. If I had the tools, I'd actually be
> > able to accomplish this with a timer based(mechanical)
> > mechanism. Anyway...
> >
> > "There may also be problems storing this data fast
> > enough to get it saved and then be ready to scan the
> > joystick switches again within the 1/60 of a second
> > time frame."
> >
> > Throughout any sequence, two instructions wouldn't
> > occur within about 1/10th of a second anyway.
> >
> > "Your main problems are going to be timing and
> > storage. Getting the playback to happen at the same
> > speed as the recording will be difficult. You will
> > have to do some research and find out how long each
> > instruction takes and make sure that your playback and
> > record loops have the EXACT same execution times."
> >
> > Since input would be manual, I don't know how that is
> > possible. But if the microcontroller can read my
> > manual inputs of a particular sequence, then why
> > wouldn't it be able to just duplicate all individual
> > commands in that sequence as outputs(with that 1/60th
> > of a second accuracy I mentioned)?
> >
> > And if storage is an issue, then is there a way to use
> > my PC(hard drive)?
> >
> > So I'm assuming that a Basic Stamp 2 is what I would
> > need for my project? Is there any thing else?
> >
> > Thanks
> >
> > Darren Harris
> > Staten Island, New York.
> >
> > Searcher7@m...
> > ******************************************************************
> > ******************************************************************
> > *********************************
> > --- Scott Winn <scott.winn@m...> wrote:
> > >
> > > So basically you want to 'Record' the movement of
> > > the joystick for later
> > > playback. In order to do this, you will have to
> > > have two 'modes', a record
> > > mode and a playback mode. This would have to be
> > > switchable via a switch, or
> > > maybe a pushbutton that would initiate a 'record'
> > > sequence. Then you will
> > > also need a way of choosing which sequence you want
> > > to play back, and this
> > > could be as simple as a number of pushbutton
> > > switches.
> > >
> > > You would have two sections to the program that ran
> > > on the stamp.
> > >
> > > The first would be the recorder and it would be
> > > initiated by the record
> > > button. You would then move the joystick around and
> > > the stamp would scan
> > > the joystick switches on the joystick as quickly as
> > > it could and store the
> > > state of the switches into some sort of a storage
> > > device. This portion
> > > contains a potential problem, because each sample
> > > will have to be stored for
> > > later playback and you could potentially have a
> > > large amount of storage
> > > required. If you are sampling at 1/60 of a second,
> > > you will have 4 bits of
> > > data for every 1/60 of a second which comes out to
> > > 30 bytes per second or
> > > 1.8K per minute. That means that each 2 minute
> > > sequence will require 1.8K of
> > > storage. This will exceed the internal storage
> > > ability of the stamp and
> > > will require an external storage device. There may
> > > also be problems storing
> > > this data fast enough to get it saved and then be
> > > ready to scan the joystick
> > > switches again within the 1/60 of a second time
> > > frame. The record process
> > > could be terminated by pressing one of the buttons
> > > that are used for
> > > playback. Once this button press was complete the
> > > sequence could be assigned
> > > to that button for playback.
> > >
> > > The playback portion would wait for a button press
> > > and when a button was
> > > pressed, it would retrieve the data from the
> > > external storage device and
> > > configure it's pins for outputs. It would then run
> > > through a loop and every
> > > 1/60 of a second it would retrieve and 'playback'
> > > the stored sample for that
> > > time period.
> > >
> > > Basically what you are trying to make is a 4 bit
> > > digital recorder.
> > >
> > > Someone mentioned 'debouncing'. This refers to the
> > > fact that a switch
> > > doesn't make a clean on/off transistion. It arcs
> > > and springs and makes
> > > dozens of contacts for each press of the switch.
> > > The stamp is fast enough
> > > that it can detect these millisecond long contacts.
> > > One switch press could
> > > be detected as 10 or 20 or more presses because of
> > > this effect. I wouldn't
> > > worry about this though. What you are trying to do
> > > is record the actual
> > > movement of the switches on the joystick. The video
> > > game that you are
> > > playing back into expects the signals to be noisy
> > > and will be able to handle
> > > it.
> > >
> > > Your main problems are going to be timing and
> > > storage. Getting the playback
> > > to happen at the same speed as the recording will be
> > > difficult. You will
> > > have to do some research and find out how long each
> > > instruction takes and
> > > make sure that your playback and record loops have
> > > the EXACT same execution
> > > times.
> > >
> > > Interupts were also mentioned. What this means is
> > > that there are two
> > > methods of letting a microcontroller know that
> > > something has happened that
> > > needs it's attention. With the basic stamp the only
> > > way that really exists
> > > is for the stamp to be in a loop (a section of code
> > > that executes over and
> > > over) and just keep looking at an input over and
> > > over again until it gets
> > > pressed and then it knows it needs to do something.
> > > Other Microcontrollers
> > > support interupts which are a special input to the
> > > microcontroller that will
> > > 'interupt' what the processor is doing and FORCE it
> > > to execute a section of
> > > code when the input happens. This way the processor
> > > doesn't have to worry
> > > about whether or not any inputs have been pressed
> > > until it is interupted and
> > > then has to deal with it. This is not an option on
> > > the stamp.
> > >
> > > --Scott
> > >
> > >
> > >

---

Original Message

---

> > > From: Darren Harris [noparse][[/noparse]mailto:searcher731@y...]
> > > Sent: Thursday, April 12, 2001 10:18 AM
> > > To: basicstamps@y...
> > > Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please
> > > :-)
> > >
> > >
> > > "The Basic Stamp 2 will do this as long as you
> > > realize
> > > that there are no
> > > interrupts. This implies that you will need to loop
> > > through the 4 inputs
> > > continuously and check their state."
> > >
> > > Providing I understand you correctly, is this a
> > > manual
> > > process or will the Basic Stamp do this?
> > >
> > > "As long as the program doesn't take more than 16mS
> > > to
> > > loop, which is a long time- even for a stamp, you
> > > should be OK."
> > >
> > > "The 'button' instruction will help you 'debounce'
> > > your manual switches."
> > >
> > > This is all Greek to me.
> > >
> > > "If the toughest loop is 112 commands in 2 minutes
> > > or
> > > less, you're talking more than 1 second per, so it
> > > also is not too fast. More effort will go into
> > > making
> > > it take 2 minutes, if that's what you need."
> > >
> > > It is not the length of time. I just said that a
> > > sequence may take about 2 minutes.
> > >
> > > "Just realize that again, while the 112 instructions
> > > are executing in their own loop, the only thing the
> > > stamp can do is wait for the loop to execute, or be
> > > reset."
> > >
> > > I assume that this means that it will follow the
> > > sequence I initiate to it's end, or I can manually
> > > reset, and start it(or another sequence) over?
> > >
> > > "If I were you, I'd read the manual and 'stamps in
> > > class' downloads at www.parallaxinc.com"
> > >
> > > I don't understand much of what you said, but I will
> > > check out the manual you mentioned.
> > >
> > > Nevertheless, here is just one more clarification of
> > > the basics of what I'm attempting...
> > >
> > > ***Years ago there was a multiplayer "family" game
> > > advertised on T.V. It entailed a device that would
> > > flash lights in a certain sequence. The player would
> > > then have to remember and duplicate that sequence by
> > > pressing the corresponding lighted buttons. The
> > > farther you got into the game the longer the
> > > sequence
> > > of lights you had to duplicate.(The more difficult
> > > it
> > > became).
> > >
> > > Now if you can imagine the computer in the human's
> > > place, and vice-versa, you will understand bacically
> > > what I'm attempting to accomplish. I will be using a
> > > 4-way joystick to "input" 14 separate sequences
> > > containing many "on" and "off" commands to all 4
> > > connections.(As I said, each sequence will last a
> > > maximum of about 2 minutes, and have a total of 112
> > > commands). Now, when I press one of 14 buttons, it
> > > is
> > > the microcontroller that will have to duplicate the
> > > corresponding sequence to within 1/60th of a
> > > second(between each and every command in that
> > > sequence). So for example, if I'm using it on a game
> > > like Pacman, the microcontroller will be able to
> > > "repeat" any of my patterns, because I would allow
> > > it
> > > "read" and remember my joystick movements. It will
> > >
> > === message truncated ===
> >
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Get email at your own domain with Yahoo! Mail.
> > http://personal.mail.yahoo.com/
> >
> >
> >
> > Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/
> >

Archiver

"This is just an idea to conserve memory...
Read all 4 inputs at once as a single "NIBBLE"(forget
the button command)
Instead, read the NEW NIBBLE compared to the PREVIOUS
NIBBLE
(for debounce, read the NEW NIBBLE a couple of times)
If the NEW NIBBLE is DIFFERENT than the PREVIOUS
NIBBLE, then
log the 'time' and NIBBLE information."

1)But the same "NIBBLE" will not occur on consecutive
commands.
2)No two commands will occur within about 1/10th of a
second of each other.
3)It's really 8 commands, since the release is just as
important.

"(The 'time' variable could be an incremental variable
inside the
compare loop between NEW NIBBLE and PREVIOUS NIBBLE
it need not
be $TIME as in HH:MM:SS .... Example:
Initialize 'time' to
Zero before entering the compare loop... Once
inside the
compare loop if NEW NIBBLE equals PREVIOUS NIBBLE
then increment
the 'time' variable. Use this value to determine a
relative
amount of time the Joystick remained in one
position.)"

If I understand you correctly, time would start at
zero as of the "PREVIOUS NIBBLE". If timing accuracy
will not suffer it would be the obvious way to go. BIG
QUESTION: Is it too unreasonable to expect that the
accuracy between the very first and very last
commands(over a span of about 2 minutes) will be
within that 1/60th of a second window?

"For multiple "tracks" you can segment the memory by a
certain
offset, and just index it upon playback."

I assume something simple like 1, 2, 3, 4, 5, 6, 7,
and 8?

If I sound naive, it's because I know nothing about
electronics.(I'm a amateur(mechanical) inventor).

Thanks.

Darren Harris
Staten Island, New York.

Searcher7@m...
********************************************************************************\
********************************************************************************\
*****
--- Beau Schwabe <bschwabe@a...> wrote:
> At 11:58 AM 4/12/01 -0700, you wrote:
>
> This is just an idea to conserve memory...
>
>
> Read all 4 inputs at once as a single "NIBBLE"
>
> (forget the button command)
>
> Instead, read the NEW NIBBLE compared to the
> PREVIOUS NIBBLE
>
> (for debounce, read the NEW NIBBLE a couple of
> times)
>
> If the NEW NIBBLE is DIFFERENT than the PREVIOUS
> NIBBLE, then
>
> log the 'time' and NIBBLE information.
>
> (The 'time' variable could be an incremental
> variable inside the
> compare loop between NEW NIBBLE and PREVIOUS
> NIBBLE it need not
> be $TIME as in HH:MM:SS .... Example:
> Initialize 'time' to
> Zero before entering the compare loop... Once
> inside the
> compare loop if NEW NIBBLE equals PREVIOUS NIBBLE
> then increment
> the 'time' variable. Use this value to determine
> a relative
> amount of time the Joystick remained in one
> position.)
>
> For multiple "tracks" you can segment the memory by
> a certain
>
> offset, and just index it upon playback.
>
>
>
>
> Beau Schwabe IC Mask Designer
> National Semiconductor Network Products
> Division
> 500 Pinnacle Court, Suite 525 Mail Stop GA1
> Norcross, GA 30071
>
>
>
>
>
>
>
>
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/

Archiver

---

Original Message

---

> Huh? Remember, I'm a newbie. :-)

OK!
The 'safest' way to simulate switch-closures is to use an opto-coupler. It
concists of a LED and a transistor in one DIP-package (or SMD for that
matter).

It works like this: When you drive current thru the LED the transistor will
conduct and current can flow from the collector to the emitter - quite
simple! The only connection is by light so there's _allmost_ no way that the
STAMP can be hurt if any thing goes wrong in 'the other end'.

The first thing you need to do is findout if the signals going from the
switches to 'whatever' is at 0V when you don't do anything with the stick
and at 5V when you push up for example.

If that is the case you would connect the COLLECTOR on the optocoupler (pin5
on MOST (not all) 6-pin optocouplers) to the one side of the switch that
allways is at 5V and the EMITTER on the optocoupler (pin6) to the 'other'
side of the switch.

Now, if you ground the cathode of the LED in the optocoupler (often pin 2)
and connect a 330ohm resistor between the anode (pin1) and a STAMP pin you
can 'simulate' a 1 second switch closure by for example:

High 4 'Put 5V on I/O 4 on the STAMP, making the LED in the
optoc. light up.
Pause 1000 'Wait 1000mS
Low 4 'Remove the 5V on I/O 4, making the LED go off.

Look up some optocoupler like TIL111, CNY17 H11A1, etc or the PC847 - four
in one. You'll see what I meen.

There are many other ways to do it, this is just one. Maby you don't really
need the optocoupler at all scince it's a 5V system, but it protects the
stamp.

Did it make it any clearer for you or did I just made more complicated for
you?
If you want I can draw a simple schematic and send to you off-list.
Sorry I can't help you much on the software side :-( I'm going to think
about it though.

Let us know how it goes.

/Henrik Olsson. Sweden.



>
> "Is the joystick 5V....?"
>
> Yes.
>
> "What will be the longest time that you would hold the stick in any
> given
> direction?"
>
> About 7 seconds.
>
> "If this is way off please excuse me!"
>
> Actually, you seem to have a handle on what I'm attempting. I would
> have given even more details as to why I need to do this, but it
> would have been somewhat off topic.
>
> Thanks.
>
> Darren Harris
> Staten Island, New York.

Archiver

"Or you could just the skip the STAMP and connect the
four switches to
LPT-port on your PC and record days and weeks of
sequences."

For real? I heard this mentioned before, but have no
idea how to go about getting it done. Is it
easier/more accurate/cheaper than the Stamp idea(for
this beginner)?

Are there any simular projects out there that I can't
study and adapt for my purposes?

Thanks

Darren Harris
Staten Island, New York.

Searcher7@m...
********************************************************************************\
********************************************************************************\
*****
--- Henrik Olsson <henrik-olsson@t...> wrote:
> Darren,
> I guess I understood you right.
> You would need some sort of interface between the
> joystick and STAMP. If the
> joystick's switches is connected to +5V only you
> could connect the switches
> direct to the STAMP pins - for the recordmode, that
> is. For the playbackmode
> you can use transistors conected across the switches
> in the joystick and
> drive the base or gate of the transistor from
> another STAMP pin.
> If it's not 5V on the joystick it gets a little more
> complicated but it sure
> can be done. Is it 5V or 12V or maby 230VAC?
> On the software side Scott gave you a couple of good
> ideas but like he says
> the storagespace is the problem. Maby you can use an
> external EEPROM or RAM
> connected to the STAMP.
> Or you could just the skip the STAMP and connect the
> four switches to
> LPT-port on your PC and record days and weeks of
> sequences.
>
> Good luck!
> /Henrik Olsson. Sweden.
>
>

---

Original Message

---

> From: "Darren Harris" <searcher731@y...>
> To: <basicstamps@yahoogroups.com>
> Sent: Thursday, April 12, 2001 8:58 PM
> Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please
> :-)
>
>
> > It's amazing how apparently simple things turn out
> to
> > be so complicated. If I had the tools, I'd
> actually be
> > able to accomplish this with a timer
> based(mechanical)
> > mechanism. Anyway...
> >
> > "There may also be problems storing this data fast
> > enough to get it saved and then be ready to scan
> the
> > joystick switches again within the 1/60 of a
> second
> > time frame."
> >
> > Throughout any sequence, two instructions wouldn't
> > occur within about 1/10th of a second anyway.
> >
> > "Your main problems are going to be timing and
> > storage. Getting the playback to happen at the
> same
> > speed as the recording will be difficult. You
> will
> > have to do some research and find out how long
> each
> > instruction takes and make sure that your playback
> and
> > record loops have the EXACT same execution times."
> >
> > Since input would be manual, I don't know how that
> is
> > possible. But if the microcontroller can read my
> > manual inputs of a particular sequence, then why
> > wouldn't it be able to just duplicate all
> individual
> > commands in that sequence as outputs(with that
> 1/60th
> > of a second accuracy I mentioned)?
> >
> > And if storage is an issue, then is there a way to
> use
> > my PC(hard drive)?
> >
> > So I'm assuming that a Basic Stamp 2 is what I
> would
> > need for my project? Is there any thing else?
> >
> > Thanks
> >
> > Darren Harris
> > Staten Island, New York.
> >
> > Searcher7@m...
> >
>
>
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/

Archiver

>"This is just an idea to conserve memory...
>
>1)But the same "NIBBLE" will not occur on consecutive
>commands.

In the time the BS2 or PIC can "read" the four switches,
YES, it could...


>2)No two commands will occur within about 1/10th of a
>second of each other.

SLOW for a PIC... questionable for a BS2

>3)It's really 8 commands, since the release is just as
>important.

YES!, the NIBBLE recording method I mention would ONLY
record a "change" in any of the switch positions. Thus
a switch close or a switch open would be recorded.





Beau Schwabe IC Mask Designer
National Semiconductor Network Products Division
500 Pinnacle Court, Suite 525 Mail Stop GA1 Norcross, GA 30071

Archiver

Well, I guess it could be done i QBASIC or VB but I'm not very good at those
(either). Besides this isn't really the place to disscus non STAMP-things.
The hardware side would be axactly the same for a PC as for a STAMP.
Do you know VB at all?
You could set up a timer that reads the LPT-port once every 1/10th seconds,
put the value in file. And then read the file back one line every 1/10th
second (or whaever) and write it back to the LPT-port. But like I said,
don't ask me axactly how to do it.


> "Or you could just the skip the STAMP and connect the
> four switches to
> LPT-port on your PC and record days and weeks of
> sequences."
>
> For real? I heard this mentioned before, but have no
> idea how to go about getting it done. Is it
> easier/more accurate/cheaper than the Stamp idea(for
> this beginner)?
>
> Are there any simular projects out there that I can't
> study and adapt for my purposes?
>
> Thanks
>
> Darren Harris
> Staten Island, New York.
>
> Searcher7@m...
>
****************************************************************************
****************************************************************************
*************
> --- Henrik Olsson <henrik-olsson@t...> wrote:
> > Darren,
> > I guess I understood you right.
> > You would need some sort of interface between the
> > joystick and STAMP. If the
> > joystick's switches is connected to +5V only you
> > could connect the switches
> > direct to the STAMP pins - for the recordmode, that
> > is. For the playbackmode
> > you can use transistors conected across the switches
> > in the joystick and
> > drive the base or gate of the transistor from
> > another STAMP pin.
> > If it's not 5V on the joystick it gets a little more
> > complicated but it sure
> > can be done. Is it 5V or 12V or maby 230VAC?
> > On the software side Scott gave you a couple of good
> > ideas but like he says
> > the storagespace is the problem. Maby you can use an
> > external EEPROM or RAM
> > connected to the STAMP.
> > Or you could just the skip the STAMP and connect the
> > four switches to
> > LPT-port on your PC and record days and weeks of
> > sequences.
> >
> > Good luck!
> > /Henrik Olsson. Sweden.
> >
> >

---

Original Message

---

> > From: "Darren Harris" <searcher731@y...>
> > To: <basicstamps@yahoogroups.com>
> > Sent: Thursday, April 12, 2001 8:58 PM
> > Subject: RE: [noparse][[/noparse]basicstamps] Recommendations please
> > :-)
> >
> >
> > > It's amazing how apparently simple things turn out
> > to
> > > be so complicated. If I had the tools, I'd
> > actually be
> > > able to accomplish this with a timer
> > based(mechanical)
> > > mechanism. Anyway...
> > >
> > > "There may also be problems storing this data fast
> > > enough to get it saved and then be ready to scan
> > the
> > > joystick switches again within the 1/60 of a
> > second
> > > time frame."
> > >
> > > Throughout any sequence, two instructions wouldn't
> > > occur within about 1/10th of a second anyway.
> > >
> > > "Your main problems are going to be timing and
> > > storage. Getting the playback to happen at the
> > same
> > > speed as the recording will be difficult. You
> > will
> > > have to do some research and find out how long
> > each
> > > instruction takes and make sure that your playback
> > and
> > > record loops have the EXACT same execution times."
> > >
> > > Since input would be manual, I don't know how that
> > is
> > > possible. But if the microcontroller can read my
> > > manual inputs of a particular sequence, then why
> > > wouldn't it be able to just duplicate all
> > individual
> > > commands in that sequence as outputs(with that
> > 1/60th
> > > of a second accuracy I mentioned)?
> > >
> > > And if storage is an issue, then is there a way to
> > use
> > > my PC(hard drive)?
> > >
> > > So I'm assuming that a Basic Stamp 2 is what I
> > would
> > > need for my project? Is there any thing else?
> > >
> > > Thanks
> > >
> > > Darren Harris
> > > Staten Island, New York.
> > >
> > > Searcher7@m...
> > >
> >
> >
> >
> >
> > Your use of Yahoo! Groups is subject to
> > http://docs.yahoo.com/info/terms/
> >
> >
>
>
> __________________________________________________
> Do You Yahoo!?
> Get email at your own domain with Yahoo! Mail.
> http://personal.mail.yahoo.com/
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>

Archiver

Hmmmm. Perhaps I should send the joystick assembly to
someone qualified to do this. :-)

I have a mind like a computer.(Not a good thing). If
every detail is not explained to me perfectly, in a
step-by-step manner, I'd get stuck on the smallest
things.

So I don't know how much good a diagram would do in
view of the fact that those electronic terms you used
are alien to me.

I really need a book to study.(One with project
examples(and pictures)). :-)

Thanks.

Darren Harris
Staten Island, New York.

Searcher7@m...
********************************************************************************\
********************************************************************************\
*****


--- Henrik Olsson <henrik-olsson@t...> wrote:
>
>

---

Original Message

---

> > Huh? Remember, I'm a newbie. :-)
>
> OK!
> The 'safest' way to simulate switch-closures is to
> use an opto-coupler. It
> concists of a LED and a transistor in one
> DIP-package (or SMD for that
> matter).
>
> It works like this: When you drive current thru the
> LED the transistor will
> conduct and current can flow from the collector to
> the emitter - quite
> simple! The only connection is by light so there's
> _allmost_ no way that the
> STAMP can be hurt if any thing goes wrong in 'the
> other end'.
>
> The first thing you need to do is findout if the
> signals going from the
> switches to 'whatever' is at 0V when you don't do
> anything with the stick
> and at 5V when you push up for example.
>
> If that is the case you would connect the COLLECTOR
> on the optocoupler (pin5
> on MOST (not all) 6-pin optocouplers) to the one
> side of the switch that
> allways is at 5V and the EMITTER on the optocoupler
> (pin6) to the 'other'
> side of the switch.
>
> Now, if you ground the cathode of the LED in the
> optocoupler (often pin 2)
> and connect a 330ohm resistor between the anode
> (pin1) and a STAMP pin you
> can 'simulate' a 1 second switch closure by for
> example:
>
> High 4 'Put 5V on I/O 4 on the STAMP,
> making the LED in the
> optoc. light up.
> Pause 1000 'Wait 1000mS
> Low 4 'Remove the 5V on I/O 4, making
> the LED go off.
>
> Look up some optocoupler like TIL111, CNY17 H11A1,
> etc or the PC847 - four
> in one. You'll see what I meen.
>
> There are many other ways to do it, this is just
> one. Maby you don't really
> need the optocoupler at all scince it's a 5V system,
> but it protects the
> stamp.
>
> Did it make it any clearer for you or did I just
> made more complicated for
> you?
> If you want I can draw a simple schematic and send
> to you off-list.
> Sorry I can't help you much on the software side :-(
> I'm going to think
> about it though.
>
> Let us know how it goes.
>
> /Henrik Olsson. Sweden.
>
>
>
> >
> > "Is the joystick 5V....?"
> >
> > Yes.
> >
> > "What will be the longest time that you would hold
> the stick in any
> > given
> > direction?"
> >
> > About 7 seconds.
> >
> > "If this is way off please excuse me!"
> >
> > Actually, you seem to have a handle on what I'm
> attempting. I would
> > have given even more details as to why I need to
> do this, but it
> > would have been somewhat off topic.
> >
> > Thanks.
> >
> > Darren Harris
> > Staten Island, New York.
>
>
>
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/

Archiver

>"This is just an idea to conserve memory...>
>1)But the same "NIBBLE" will not occur on
consecutive>commands.

"In the time the BS2 or PIC can "read" the four
switches,YES, it could..."

Actually, since the input is completely manual, I know
that my joystick movements cannot occur that close
together, and never in the same direction twice in a
row(unless you are somehow talking about "open" and
"Close" on the same pin being considered the same
command).

"> In the time the BS2 or PIC can "read" the four
> switches,
> YES, it could..."

If it takes that long, then the stamp would be useless
to me.

Thanks

Darren Harris
Staten Island, New York.

Searcher7@m...
********************************************************************************\
********************************************************************************\
*****
--- Beau Schwabe <bschwabe@a...> wrote:
>
> >"This is just an idea to conserve memory...
> >
> >1)But the same "NIBBLE" will not occur on
> consecutive
> >commands.
>
> In the time the BS2 or PIC can "read" the four
> switches,
> YES, it could...
>
>
> >2)No two commands will occur within about 1/10th of
> a
> >second of each other.
>
> SLOW for a PIC... questionable for a BS2
>
> >3)It's really 8 commands, since the release is just
> as
> >important.
>
> YES!, the NIBBLE recording method I mention would
> ONLY
> record a "change" in any of the switch positions.
> Thus
> a switch close or a switch open would be recorded.
>
>
>
>
>
> Beau Schwabe IC Mask Designer
> National Semiconductor Network Products
> Division
> 500 Pinnacle Court, Suite 525 Mail Stop GA1
> Norcross, GA 30071
>
>
>
>
>
>
>
>
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/

Archiver

1/10th of a second would not be accurate enough
anyway.

Thanks

Darren Harris
Staten Island, New York.

Searcher7@m...
********************************************************************************\
********************************************************************************\
*****
--- Henrik Olsson <henrik-olsson@t...> wrote:
> Well, I guess it could be done i QBASIC or VB but
> I'm not very good at those
> (either). Besides this isn't really the place to
> disscus non STAMP-things.
> The hardware side would be axactly the same for a PC
> as for a STAMP.
> Do you know VB at all?
> You could set up a timer that reads the LPT-port
> once every 1/10th seconds,
> put the value in file. And then read the file back
> one line every 1/10th
> second (or whaever) and write it back to the
> LPT-port. But like I said,
> don't ask me axactly how to do it.
>
>
> > "Or you could just the skip the STAMP and connect
> the
> > four switches to
> > LPT-port on your PC and record days and weeks of
> > sequences."
> >
> > For real? I heard this mentioned before, but have
> no
> > idea how to go about getting it done. Is it
> > easier/more accurate/cheaper than the Stamp
> idea(for
> > this beginner)?
> >
> > Are there any simular projects out there that I
> can't
> > study and adapt for my purposes?
> >
> > Thanks
> >
> > Darren Harris
> > Staten Island, New York.
> >
> > Searcher7@m...
> >
>
****************************************************************************
>
****************************************************************************
> *************
> > --- Henrik Olsson <henrik-olsson@t...> wrote:
> > > Darren,
> > > I guess I understood you right.
> > > You would need some sort of interface between
> the
> > > joystick and STAMP. If the
> > > joystick's switches is connected to +5V only you
> > > could connect the switches
> > > direct to the STAMP pins - for the recordmode,
> that
> > > is. For the playbackmode
> > > you can use transistors conected across the
> switches
> > > in the joystick and
> > > drive the base or gate of the transistor from
> > > another STAMP pin.
> > > If it's not 5V on the joystick it gets a little
> more
> > > complicated but it sure
> > > can be done. Is it 5V or 12V or maby 230VAC?
> > > On the software side Scott gave you a couple of
> good
> > > ideas but like he says
> > > the storagespace is the problem. Maby you can
> use an
> > > external EEPROM or RAM
> > > connected to the STAMP.
> > > Or you could just the skip the STAMP and connect
> the
> > > four switches to
> > > LPT-port on your PC and record days and weeks of
> > > sequences.
> > >
> > > Good luck!
> > > /Henrik Olsson. Sweden.
> > >
> > >

---

Original Message

---

> > > From: "Darren Harris" <searcher731@y...>
> > > To: <basicstamps@yahoogroups.com>
> > > Sent: Thursday, April 12, 2001 8:58 PM
> > > Subject: RE: [noparse][[/noparse]basicstamps] Recommendations
> please
> > > :-)
> > >
> > >
> > > > It's amazing how apparently simple things turn
> out
> > > to
> > > > be so complicated. If I had the tools, I'd
> > > actually be
> > > > able to accomplish this with a timer
> > > based(mechanical)
> > > > mechanism. Anyway...
> > > >
> > > > "There may also be problems storing this data
> fast
> > > > enough to get it saved and then be ready to
> scan
> > > the
> > > > joystick switches again within the 1/60 of a
> > > second
> > > > time frame."
> > > >
> > > > Throughout any sequence, two instructions
> wouldn't
> > > > occur within about 1/10th of a second anyway.
> > > >
> > > > "Your main problems are going to be timing and
> > > > storage. Getting the playback to happen at
> the
> > > same
> > > > speed as the recording will be difficult. You
> > > will
> > > > have to do some research and find out how long
> > > each
> > > > instruction takes and make sure that your
> playback
> > > and
> > > > record loops have the EXACT same execution
> times."
> > > >
> > > > Since input would be manual, I don't know how
> that
> > > is
> > > > possible. But if the microcontroller can read
> my
> > > > manual inputs of a particular sequence, then
> why
> > > > wouldn't it be able to just duplicate all
> > > individual
> > > > commands in that sequence as outputs(with that
> > > 1/60th
> > > > of a second accuracy I mentioned)?
> > > >
> > > > And if storage is an issue, then is there a
> way to
> > > use
> > > > my PC(hard drive)?
> > > >
> > > > So I'm assuming that a Basic Stamp 2 is what I
> > > would
> > > > need for my project? Is there any thing else?
> > > >
> > > > Thanks
> > > >
> > > > Darren Harris
> > > > Staten Island, New York.
> > > >
> > > > Searcher7@m...
> > > >
> > >
> > >
> > >
> > >
> > > Your use of Yahoo! Groups is subject to
> > > http://docs.yahoo.com/info/terms/
> > >
> > >
> >
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Get email at your own domain with Yahoo! Mail.
> > http://personal.mail.yahoo.com/
> >
> >
> >
> > Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
> >
> >
>
>
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/

Archiver

At 01:27 PM 4/12/01 -0700, you wrote:


One word....


DEBOUNCE


> >"This is just an idea to conserve memory...>
> >1)But the same "NIBBLE" will not occur on
>consecutive>commands.
>
>"In the time the BS2 or PIC can "read" the four
>switches,YES, it could..."
>
>Actually, since the input is completely manual, I know
>that my joystick movements cannot occur that close
>together, and never in the same direction twice in a
>row(unless you are somehow talking about "open" and
>"Close" on the same pin being considered the same
>command).
>
>"> In the time the BS2 or PIC can "read" the four
> > switches,
> > YES, it could..."
>
>If it takes that long, then the stamp would be useless
>to me.




Beau Schwabe IC Mask Designer
National Semiconductor Network Products Division
500 Pinnacle Court, Suite 525 Mail Stop GA1 Norcross, GA 30071

Archiver

I assume that after a command it wouldn't be a problem
to program the microcontroller not to accept the same
command for the next second or so. Or I can just
program it not to accept the same command again until
one of the other 6 out of 8 commands were given
first.(The 7th command I left out would of course be
the release for the initial command).

Thanks.

Darren Harris
Staten Island, New York.
********************************************************************************\
********************************************************************************\
*****
Searcher7@m...
--- Beau Schwabe <bschwabe@a...> wrote:
> At 01:27 PM 4/12/01 -0700, you wrote:
>
>
> One word....
>
>
> DEBOUNCE
>
>
> > >"This is just an idea to conserve memory...>
> > >1)But the same "NIBBLE" will not occur on
> >consecutive>commands.
> >
> >"In the time the BS2 or PIC can "read" the four
> >switches,YES, it could..."
> >
> >Actually, since the input is completely manual, I
> know
> >that my joystick movements cannot occur that close
> >together, and never in the same direction twice in
> a
> >row(unless you are somehow talking about "open" and
> >"Close" on the same pin being considered the same
> >command).
> >
> >"> In the time the BS2 or PIC can "read" the four
> > > switches,
> > > YES, it could..."
> >
> >If it takes that long, then the stamp would be
> useless
> >to me.
>
>
>
>
> Beau Schwabe IC Mask Designer
> National Semiconductor Network Products
> Division
> 500 Pinnacle Court, Suite 525 Mail Stop GA1
> Norcross, GA 30071
>
>
>
>
>
>
>
>
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail.
http://personal.mail.yahoo.com/