Non standard async serial communications
Archiver
Posts: 46,084
Hi All,
Great group here!
I've done a search, but not quite clear on the capabilites of the
Basic Stamp to do what I want, and before I commit to find out that
I may be at a dead end, or need external chips to support my task, I
thought I'd ask here first!
I'm trying to interface to a car radio that has three communication
lines. One comes from the radio, two go to it and all seem to
operate at 1 kb/s. The signalling isn't a standard bus, like RS-232
or I2C. There appear to be state signals as well as data, depending
on the nature of the communication, all on the same lines. For
example, a line may go low to acknowledge data received on another
line, and then a couple of mS later, send 9 to 20 bits of data.
My concern at this point, is if I can use the BS to talk on two (1
bit)ports and listen on the other one, simultaneously. I'll need to
ensure the timing is reasonable, so I'm thinking of having a routine
to both write one port and read the other two, once every
millisecond. Hopefully there's enough processing power left in
between to manage the bits that are going out and coming in, do a
translation to/from RS-232 and both send and receive out the BS 232
port.
Opinions as to if this is something the BS is capable of, are
greatly appreciated and pointers to documents/code that would help
me accomplish this, are even more so!
My biggest concern right now, is if there's the "HP" to do this all
at once.
Thanks for your thoughts! I presently have a Basic Stamp ][noparse][[/noparse] board
with the BS2-IC
Jim
Great group here!
I've done a search, but not quite clear on the capabilites of the
Basic Stamp to do what I want, and before I commit to find out that
I may be at a dead end, or need external chips to support my task, I
thought I'd ask here first!
I'm trying to interface to a car radio that has three communication
lines. One comes from the radio, two go to it and all seem to
operate at 1 kb/s. The signalling isn't a standard bus, like RS-232
or I2C. There appear to be state signals as well as data, depending
on the nature of the communication, all on the same lines. For
example, a line may go low to acknowledge data received on another
line, and then a couple of mS later, send 9 to 20 bits of data.
My concern at this point, is if I can use the BS to talk on two (1
bit)ports and listen on the other one, simultaneously. I'll need to
ensure the timing is reasonable, so I'm thinking of having a routine
to both write one port and read the other two, once every
millisecond. Hopefully there's enough processing power left in
between to manage the bits that are going out and coming in, do a
translation to/from RS-232 and both send and receive out the BS 232
port.
Opinions as to if this is something the BS is capable of, are
greatly appreciated and pointers to documents/code that would help
me accomplish this, are even more so!
My biggest concern right now, is if there's the "HP" to do this all
at once.
Thanks for your thoughts! I presently have a Basic Stamp ][noparse][[/noparse] board
with the BS2-IC
Jim
Comments
>Hi All,
>
>Great group here!
>
>I've done a search, but not quite clear on the capabilites of the
>Basic Stamp to do what I want, and before I commit to find out that
>I may be at a dead end, or need external chips to support my task, I
>thought I'd ask here first!
>
>I'm trying to interface to a car radio that has three communication
>lines. One comes from the radio, two go to it and all seem to
>operate at 1 kb/s. The signalling isn't a standard bus, like RS-232
>or I2C. There appear to be state signals as well as data, depending
>on the nature of the communication, all on the same lines. For
>example, a line may go low to acknowledge data received on another
>line, and then a couple of mS later, send 9 to 20 bits of data.
>
>My concern at this point, is if I can use the BS to talk on two (1
>bit)ports and listen on the other one, simultaneously. I'll need to
>ensure the timing is reasonable, so I'm thinking of having a routine
>to both write one port and read the other two, once every
>millisecond. Hopefully there's enough processing power left in
>between to manage the bits that are going out and coming in, do a
>translation to/from RS-232 and both send and receive out the BS 232
>port.
>
>Opinions as to if this is something the BS is capable of, are
>greatly appreciated and pointers to documents/code that would help
>me accomplish this, are even more so!
>
>My biggest concern right now, is if there's the "HP" to do this all
>at once.
>
>Thanks for your thoughts! I presently have a Basic Stamp ][noparse][[/noparse] board
>with the BS2-IC
>
>Jim
Jim -
If this radio was manufactured by Philips, it probably uses the I2C protocol.
Certain Stamps can work with I2C directly, but all can work with it indirectly
by bit-banging. If it is Philps or you think it may be I2C there is more in the
way of information that can be offered.
Regards,
Bruce Bates
>To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
>from the same email address that you subscribed. Text in the Subject and Body
of the message will be ignored.
>
>Yahoo! Groups Links
>
>
>
>
>
> Jim -
>
> If this radio was manufactured by Philips, it probably uses the
I2C protocol.
> Certain Stamps can work with I2C directly, but all can work with
it indirectly by bit-banging. If it is Philps or you think it may be
I2C there is more in the way of information that can be offered.
>
> Regards,
>
> Bruce Bates
Hi Bruce,
With my limited knowledge of I2C, I don't think it is... However it
would be nice to be wrong!
The Radio is manufactured by Clarion for Nissan and carries the Bose
name. How's that for confusing! The bus I'm trying to talk to is
often called Nbus (Nissan bus) and I wouldn't be surprised if it
shares a portion of Clarion's CeNET. Finding info on either has
been impossible, so I'd at a point where I'm going to try to emulate
it and see if I can talk to it. Sort of like learning a new
language with no books or teacher.
I'll have to read up more on "bit-banging" as I've seen reference to
this already.
What's triggered my concern is reading that the stamp can't talk and
listen at the same time for serial communications.
I'm hopeful that the stamp will be up to the task, as I'll need
quite a few of them, then... If not, I'll just have to shelve it
and get on with using another processor.
can bit-bang code that can talk and listen at the same time. Take a
look at the code for the PSX game controller to see how:
http://www.parallax.com/dl/docs/cols/nv/vol4/col/101.pdf
-- Jon Williams
-- Applications Engineer, Parallax
-- Dallas Office
Original Message
From: ve3mij [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=h68CYWdhfJB4i_ge0OMnzUzQEIz2-hZsElFhvmhidKX_56wlG4wfUZPElCwsau-MrdGQlaDD4L2mMA8]ve3mij@r...[/url
Sent: Friday, February 13, 2004 12:59 PM
To: basicstamps@yahoogroups.com
Subject: [noparse][[/noparse]basicstamps] Re: Non standard async serial communications
--- In basicstamps@yahoogroups.com, Bruce Bates <bvbates@u...> wrote:
>
> Jim -
>
> If this radio was manufactured by Philips, it probably uses the
I2C protocol.
> Certain Stamps can work with I2C directly, but all can work with
it indirectly by bit-banging. If it is Philps or you think it may be
I2C there is more in the way of information that can be offered.
>
> Regards,
>
> Bruce Bates
Hi Bruce,
With my limited knowledge of I2C, I don't think it is... However it
would be nice to be wrong!
The Radio is manufactured by Clarion for Nissan and carries the Bose
name. How's that for confusing! The bus I'm trying to talk to is
often called Nbus (Nissan bus) and I wouldn't be surprised if it
shares a portion of Clarion's CeNET. Finding info on either has
been impossible, so I'd at a point where I'm going to try to emulate
it and see if I can talk to it. Sort of like learning a new
language with no books or teacher.
I'll have to read up more on "bit-banging" as I've seen reference to
this already.
What's triggered my concern is reading that the stamp can't talk and
listen at the same time for serial communications.
I'm hopeful that the stamp will be up to the task, as I'll need
quite a few of them, then... If not, I'll just have to shelve it
and get on with using another processor.
However, there are lots of co-processors
built for it to allow it to act as the 'master'
while the co-processors act as 'slave' devices,
doing things in the 'background' until the
BS2 is ready to get the data and make decisions
on it.
The PIC processors (16F876, 16F877, and the
18Fxxx series) are all very nice for use as
co-processors. Parallax also sells the
Ubicom SX28-class processors, which are
awfully nice -- 50 MIPS in ASM, yet
$8.00 per chip.
--- In basicstamps@yahoogroups.com, "ve3mij" <ve3mij@r...> wrote:
> --- In basicstamps@yahoogroups.com, Bruce Bates <bvbates@u...>
wrote:
> >
> > Jim -
> >
> > If this radio was manufactured by Philips, it probably uses the
> I2C protocol.
> > Certain Stamps can work with I2C directly, but all can work with
> it indirectly by bit-banging. If it is Philps or you think it may
be
> I2C there is more in the way of information that can be offered.
> >
> > Regards,
> >
> > Bruce Bates
>
> Hi Bruce,
>
> With my limited knowledge of I2C, I don't think it is... However
it
> would be nice to be wrong!
>
> The Radio is manufactured by Clarion for Nissan and carries the
Bose
> name. How's that for confusing! The bus I'm trying to talk to is
> often called Nbus (Nissan bus) and I wouldn't be surprised if it
> shares a portion of Clarion's CeNET. Finding info on either has
> been impossible, so I'd at a point where I'm going to try to
emulate
> it and see if I can talk to it. Sort of like learning a new
> language with no books or teacher.
>
> I'll have to read up more on "bit-banging" as I've seen reference
to
> this already.
>
> What's triggered my concern is reading that the stamp can't talk
and
> listen at the same time for serial communications.
>
> I'm hopeful that the stamp will be up to the task, as I'll need
> quite a few of them, then... If not, I'll just have to shelve it
> and get on with using another processor.
Two observations I've made:
1)You can't write a byte to an 8 bit port. This means I can never
have two outputs switch their pins state at the same time. I hope
I've missed a command on my BS2 and that it is possible, but so far,
I only see the ability to control a single pin at a time. This is
necessary in my application.
2)It appears the BS2 is too slow, so I'll have to find some
incredible code efficiency, use a faster version, attach external
components (raises cost, which is already higher than I'd like to
see) or jump to another processor/platform. It appears the BS2 is
dead to the rest of the world while writing out the serial port, as
well, so this creates lot's of dead time when trying to monitor two
lines that can be signalling at 1 ms intervals.
Looks like the BS2 is great, as long as you don't want to go fast...
Any ideas are appreciated!
Jim
--- In basicstamps@yahoogroups.com, "Allan Lane" <allan.lane@h...>
wrote:
> Yes, the BS2 can only do one thing at a time.
> However, there are lots of co-processors
> built for it to allow it to act as the 'master'
> while the co-processors act as 'slave' devices,
> doing things in the 'background' until the
> BS2 is ready to get the data and make decisions
> on it.
>
> The PIC processors (16F876, 16F877, and the
> 18Fxxx series) are all very nice for use as
> co-processors. Parallax also sells the
> Ubicom SX28-class processors, which are
> awfully nice -- 50 MIPS in ASM, yet
> $8.00 per chip.
Dirs = $00FF 'pins 0 ~ 7 outputs, 8 ~ 16 inputs
and then you can write to pins 0 ~ 7 as a group with
OUTL = $FF 'make output pins high
or
OUTL = $F0 'make output pins 0 ~ 3 low, pins 4 ~ 7 high
Pins 8 ~ 16 addresses as OUTH
1)You can't write a byte to an 8 bit port. This means I can never
have two outputs switch their pins state at the same time. I hope
I've missed a command on my BS2 and that it is possible, but so far,
I only see the ability to control a single pin at a time. This is
necessary in my application.
[noparse][[/noparse]Non-text portions of this message have been removed]
silly to not be able to do that...
I just figured it out (duh!) and came back to mention that, to see
your post. So you've confirmed what I just did. Cool....
Thanks for doing it in Hex, I was doing it in binary... e.g. OUTH =
%10101010
I might be able to get this thing to work yet! [noparse]:)[/noparse]
However, I think my hope of converting to RS-232 is pushing the
processing power.
Still, a cool device!
(I knew I was doing something wrong, when I looked at one of the
other pre-programmed stamps I have, and it was changing states on
different pins, at the same time....)
--- In basicstamps@yahoogroups.com, smartdim@a... wrote:
> You CAN write to a group of 8 outputs You can declare pins 0 ~ 7
as outputs by
>
> Dirs = $00FF 'pins 0 ~ 7 outputs, 8 ~ 16 inputs
>
> and then you can write to pins 0 ~ 7 as a group with
>
> OUTL = $FF 'make output pins high
>
> or
>
> OUTL = $F0 'make output pins 0 ~ 3 low, pins 4 ~ 7 high
>
> Pins 8 ~ 16 addresses as OUTH
>
>
> 1)You can't write a byte to an 8 bit port. This means I can never
> have two outputs switch their pins state at the same time. I
hope
> I've missed a command on my BS2 and that it is possible, but so
far,
> I only see the ability to control a single pin at a time. This is
> necessary in my application.
>
>
> [noparse][[/noparse]Non-text portions of this message have been removed]
Parallax, but when it comes to pure speed, it does suffer. However, many have
created awesome projects.
For raw speed, the SX also sold through parallax is awesome. I clock it at 50
MHz which boils down 20 nanoseconds per clock and many instructions take 2 to
4 clock cycles.
One of the drawbacks are assembly language. For something you can do with the
stamp in the time it takes to enter the line of code (15 seconds) can take
easily take an hour or more to program the same function with the SX. But the
stamp cannot come close to the execution speed of an SX.
The SX is inexpensive by comparison......trade offs, trade offs, if you know
what I mean.
ken
Thanks! I thought I must be missing something here, as it seemed
silly to not be able to do that...
I just figured it out (duh!) and came back to mention that, to see
your post. So you've confirmed what I just did. Cool....
Thanks for doing it in Hex, I was doing it in binary... e.g. OUTH =
%10101010
I might be able to get this thing to work yet! [noparse]:)[/noparse]
However, I think my hope of converting to RS-232 is pushing the
processing power.
[noparse][[/noparse]Non-text portions of this message have been removed]
<SNIP> All BASIC Stamps (including BS1) can also transmit RS-232 data through
any of thier I/O pins.
page 295 of Basic Stamp Manual version 2.0
However, I think my hope of converting to RS-232 is pushing the
processing power.
Still, a cool device!
[noparse][[/noparse]Non-text portions of this message have been removed]
I think I may stick with the BS2, just because I have have a couple
and I can do a basic proof of concept with them, even if they won't
do everything I'd dream of! Once I can get the concept right with
an easy to program device, I can work on speed.
Something else I discovered, which really complement's the ease of
the stamp, is the ease of creating schematics, and double sided,
plated through hole, circuit boards, is http://www.expresspcb.com/
I'd like to know if anyone has seen similar services out there.
I've already designed a board for my BS2 for my application, and it
was quite easy.
Between the BS and the ease of making a board, I'm actually quite
excited about how easy it is to make a very professional project!
I'm sure there's more I've yet to discover... [noparse]:)[/noparse]
--- In basicstamps@yahoogroups.com, smartdim@a... wrote:
> The Stamp is an AWESOME device for ease of use, support from this
group and
> Parallax, but when it comes to pure speed, it does suffer.
However, many have
> created awesome projects.
>
> For raw speed, the SX also sold through parallax is awesome. I
clock it at 50
> MHz which boils down 20 nanoseconds per clock and many
instructions take 2 to
> 4 clock cycles.
>
> One of the drawbacks are assembly language. For something you can
do with the
> stamp in the time it takes to enter the line of code (15 seconds)
can take
> easily take an hour or more to program the same function with the
SX. But the
> stamp cannot come close to the execution speed of an SX.
>
> The SX is inexpensive by comparison......trade offs, trade offs,
if you know
> what I mean.
>
> ken
the stamp doing the serial out, and it can't do anything else at the
same time.
I really need it to walk AND chew gum at the same time.
Looks like interleaving port writes and reads will be the only way...
--- In basicstamps@yahoogroups.com, smartdim@a... wrote:
> I dont use the stamp for any RS232 stuff but there is a SEROUT
command
>
> <SNIP> All BASIC Stamps (including BS1) can also transmit RS-232
data through
> any of thier I/O pins.
> page 295 of Basic Stamp Manual version 2.0
>
>
> However, I think my hope of converting to RS-232 is pushing the
> processing power.
>
> Still, a cool device!
>
>
> [noparse][[/noparse]Non-text portions of this message have been removed]
Something else I discovered, which really complement's the ease of
the stamp, is the ease of creating schematics, and double sided,
plated through hole, circuit boards, is http://www.expresspcb.com/
[noparse][[/noparse]Non-text portions of this message have been removed]
a time, and write multiple pins at a time.
The magic is: the pins must be next to
each other.
For input: Once you've said:
DIRA = 0 ' Turn the first 4 pins to input
MyVal = INA ' Read the first 4 pins.
The pins are 'grouped' as:
IN -- All 16 bits
INL, INH -- 8 bits each
INA, INB, INC, IND -- 4 bits each
IN0, IN1, IN2,... -- 1 bit each.
So to WRITE to an 8-bit output port:
DIRH = $FF ' Set pins to output --
' Only need to do this once...
OUTH = MyByte
--- In basicstamps@yahoogroups.com, "ve3mij" <ve3mij@r...> wrote:
> Thanks, for the feedback guys, you've been helpful!
>
> Two observations I've made:
>
> 1)You can't write a byte to an 8 bit port. This means I can never
> have two outputs switch their pins state at the same time. I hope
> I've missed a command on my BS2 and that it is possible, but so
far,
> I only see the ability to control a single pin at a time. This is
> necessary in my application.
>
> 2)It appears the BS2 is too slow, so I'll have to find some
> incredible code efficiency, use a faster version, attach external
> components (raises cost, which is already higher than I'd like to
> see) or jump to another processor/platform. It appears the BS2 is
> dead to the rest of the world while writing out the serial port, as
> well, so this creates lot's of dead time when trying to monitor two
> lines that can be signalling at 1 ms intervals.
>
> Looks like the BS2 is great, as long as you don't want to go fast...
>
> Any ideas are appreciated!
>
> Jim
>
> --- In basicstamps@yahoogroups.com, "Allan Lane" <allan.lane@h...>
> wrote:
> > Yes, the BS2 can only do one thing at a time.
> > However, there are lots of co-processors
> > built for it to allow it to act as the 'master'
> > while the co-processors act as 'slave' devices,
> > doing things in the 'background' until the
> > BS2 is ready to get the data and make decisions
> > on it.
> >
> > The PIC processors (16F876, 16F877, and the
> > 18Fxxx series) are all very nice for use as
> > co-processors. Parallax also sells the
> > Ubicom SX28-class processors, which are
> > awfully nice -- 50 MIPS in ASM, yet
> > $8.00 per chip.
> Thanks, for the feedback guys, you've been helpful!
>
> Two observations I've made:
>
> 1)You can't write a byte to an 8 bit port. This means I can never
> have two outputs switch their pins state at the same time. I hope
> I've missed a command on my BS2 and that it is possible, but so far,
> I only see the ability to control a single pin at a time. This is
> necessary in my application.
I hate to pick nits here, but *everything* that happens in a
processor-driven environment happens iteratively, it's impossible to have
two things that are processor-driven happen at exactly the same time. You
can certainly change the output of two things at nearly the same time by
setting both pins 'high' in succession, but if you want to do precise
timing you may need to utilize some sort of serial->parallel converter.
>
> 2)It appears the BS2 is too slow, so I'll have to find some
> incredible code efficiency, use a faster version, attach external
> components (raises cost, which is already higher than I'd like to
> see) or jump to another processor/platform. It appears the BS2 is
> dead to the rest of the world while writing out the serial port, as
> well, so this creates lot's of dead time when trying to monitor two
> lines that can be signalling at 1 ms intervals.
>
> Looks like the BS2 is great, as long as you don't want to go fast...
>
> Any ideas are appreciated!
>
> Jim
>
> --- In basicstamps@yahoogroups.com, "Allan Lane" <allan.lane@h...>
> wrote:
> > Yes, the BS2 can only do one thing at a time.
> > However, there are lots of co-processors
> > built for it to allow it to act as the 'master'
> > while the co-processors act as 'slave' devices,
> > doing things in the 'background' until the
> > BS2 is ready to get the data and make decisions
> > on it.
> >
> > The PIC processors (16F876, 16F877, and the
> > 18Fxxx series) are all very nice for use as
> > co-processors. Parallax also sells the
> > Ubicom SX28-class processors, which are
> > awfully nice -- 50 MIPS in ASM, yet
> > $8.00 per chip.
>
>
>
>
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and Body
of the message will be ignored.
>
> Yahoo! Groups Links
>
>
>
>
>
>
Sean T. Lamont, CTO / Chief NetNerd, Abstract Software, Inc. (ServNet)
Seattle - Bellingham - Vancouver - Portland - Everett - Tacoma - Bremerton
email: lamont@a... WWW: http://www.serv.net
"Do not fear mistakes, There Are None" - Miles Davis
>
> I hate to pick nits here, but *everything* that happens in a
> processor-driven environment happens iteratively, it's impossible to have
> two things that are processor-driven happen at exactly the same time.
Actually, that turns out not to be the case. Down at the microcode level
each action of a processor affects many bits in parallel, and depending on
the processor each word of microcode fetched from the control store may be
carrying out four, six, or even more concurrent and completely independent
actions.
Iteration is the powerful technique that distinguishes a programmed
architecture from what we loosely call random logic, but it is an advantage,
not a limitation. A processor-driven environment can do anything random
logic can do. For the designer. Afterward, we are limited to the several
hundred instructions the designer chose to create at the level we usually
call "machine code." And at the next level, we are limited to the
instructions the language designer chose to present. But at levels,
processors take actions that affect signals in parallel.
Even the simplest 8-bit microprocessor can affect the state of at least 8
bits in parallel. Hence the name.
The only real question in any particular case for the Basic Stamp is whether
an instruction exists in PBasic that will affect the bits you choose, in
parallel, in exactly the way you need. Changing the state of adjacent bits
is a straightforward action that requires a single PBasic statement, and
(ignoring the housekeeping steps going on) that statement will reduce to a
token that is translated into machine code, one of whose steps will be to
change all eight bits in parallel. In a single clock. This can be checked
easily with an oscilloscope.
Gary
I'm a hardware guy, not a programmer, so for me machine code is the
most efficient, but certainly only in operation, not in
programming! So I've come to expect parallel operations at the bit
capacity of the processor to be a given.
However I have noticed on occasion that parallel operations aren't
always so, on the Stamp. There does seem to be a bit of timing
wander when changing from/to certain bit patterns with the OUT
command. At the most, some bits can drift two cycles. Shouldn't
normally be a problem. I never see any drift when all are set high
or low. (This is observed with a Tektronix TLA715 Logic Analyser at
a 4 nS sample rate.)
Ideally, it would be nice if the Stamp could call a machine code
program (is this possible?) for those routines that need to be very
fast.
As far as my original application, I realised I've made an "oops" on
my Logic Analyser and had the sample rate too low when looking at
the serial bus I'm trying to interface with. So now I see I need to
do serial communications up around 10-12,000 bits/second (Haven't
worked out the exact speed yet.)
The result is, I'll have to make the Stamp go like heck to do what I
want.
The other "First Look" observation I've made is that I've got a 12
bit serial sequence to contend with.. There's a start bit (low) and
and stop bit (high) as well as just before the stop bit an even
parity bit, but the last 9 bits all seem to toggle depending on the
data. If it was 8 bits, I'd be much happier, but this has me
wondering if there's a way to customize the Stamp's serial routine,
or if there's a hardware solution for this. Why it appears to be 9
bits is also strange, as what processor is going to work with 9
bits! What a nuisance. So I must be missing something here...
Anyone have any ideas on this, I would be most grateful!
Thanks for the help so far, I'm certainly going to use the Stamp for
some other applications, even if this one doesn't work out. It's
just so easy to program!
Jim
--- In basicstamps@yahoogroups.com, "Gary W. Sims" <simsgw@c...>
wrote:
> From: "Sean T. Lamont .lost." <lamont@a...>
> >
> > I hate to pick nits here, but *everything* that happens in a
> > processor-driven environment happens iteratively, it's
impossible to have
> > two things that are processor-driven happen at exactly the same
time.
>
> Actually, that turns out not to be the case. Down at the microcode
level
> each action of a processor affects many bits in parallel, and
depending on
> the processor each word of microcode fetched from the control
store may be
> carrying out four, six, or even more concurrent and completely
independent
> actions.
>
> Iteration is the powerful technique that distinguishes a programmed
> architecture from what we loosely call random logic, but it is an
advantage,
> not a limitation. A processor-driven environment can do anything
random
> logic can do. For the designer. Afterward, we are limited to the
several
> hundred instructions the designer chose to create at the level we
usually
> call "machine code." And at the next level, we are limited to the
> instructions the language designer chose to present. But at levels,
> processors take actions that affect signals in parallel.
>
> Even the simplest 8-bit microprocessor can affect the state of at
least 8
> bits in parallel. Hence the name.
>
> The only real question in any particular case for the Basic Stamp
is whether
> an instruction exists in PBasic that will affect the bits you
choose, in
> parallel, in exactly the way you need. Changing the state of
adjacent bits
> is a straightforward action that requires a single PBasic
statement, and
> (ignoring the housekeeping steps going on) that statement will
reduce to a
> token that is translated into machine code, one of whose steps
will be to
> change all eight bits in parallel. In a single clock. This can be
checked
> easily with an oscilloscope.
>
> Gary
bit one stop bit.
Looks like the Stamp allows parity and 7 bits, or no parity and 8
bits, but not parity and 8 bits. (In my perception of the
documentation...)
Anyone know if this can be modified or accomplished either on the
stamp, or with an outboard UART? (Run by the Stamp of course!)
Jim
> The other "First Look" observation I've made is that I've got a 12
> bit serial sequence to contend with.. There's a start bit (low)
and
> and stop bit (high) as well as just before the stop bit an even
> parity bit, but the last 9 bits all seem to toggle depending on
the
> data. If it was 8 bits, I'd be much happier, but this has me
> wondering if there's a way to customize the Stamp's serial
routine,
> or if there's a hardware solution for this. Why it appears to be
9
> bits is also strange, as what processor is going to work with 9
> bits! What a nuisance. So I must be missing something here...
>
> Anyone have any ideas on this, I would be most grateful!