Bug 2378 - geekport button input
: geekport button input
Status: RESOLVED WONTFIX
Product: SB 2/3
Classification: Unclassified
Component: Misc
: unspecified
: Macintosh All
: P2 normal with 7 votes (vote)
: ---
Assigned To: Sean Adams
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2005-10-26 09:07 UTC by Sean Adams
Modified: 2008-12-18 11:38 UTC (History)
1 user (show)

See Also:
Category: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Sean Adams 2005-10-26 09:07:51 UTC
Add a geekport mode for reading a button, or maybe a few buttons using RC timing.
Comment 1 Shawn Rutledge 2005-11-06 08:09:56 UTC
I want to have a control signal for an external amp too.  Maybe a high-current
5V output which could drive a reed relay, or be boosted externally to drive a
bigger relay.
Comment 2 Shawn Rutledge 2005-11-08 11:13:35 UTC
I've read the comments about using the headphone jack for external interface functions.  The trouble with that idea would be the loss of a normal headphone jack, right?  In the bedroom it's good to have both (you need the ability to listen on headphones when others are sleeping, but you could also need amplifier control or some other external stuff).

Maybe for snooze, the box should really have 3 or so buttons built-in on top (a wide one and two smaller ones), but they could be unlabeled and assignable to various functions.  And there should be a geekport as well for the amplifier control etc.

Having the geekport be capable of driving a Dallas 1-wire bus would also be a nice touch; I do happen to have electric curtain rods in my bedroom which can be controlled over 1-wire bus, so if the SqueezeBox could open the curtains in the morning, it would save me the installation of some other device to do this (which I have not gotten around to).  And I could easily design something using a DS2406 to turn the amp on and off, so really the 1-wire bus is good enough for "geek" functions all by itself.  And very extensible; you can put a lot of devices on the same bus.  Likewise external keypads can be implemented with 1-wire chips.  The box could have a standard RJ12 jack in the back with the usual pinout, providing 1-wire data, ground, 5V power and ground.  The more I think about it, the more I like this idea.

The usual pinout is here:
http://www.aagelectronica.com/PDF%20Docs/Connectors.pdf
see the TINI RJ12/DS9490R columns in the chart, those two are compatible with each other and I think should be considered the standard these days

I'd like to work with you on implementing this.
Comment 3 Chris Owens 2006-08-16 14:37:28 UTC
Sean, what shall I do about this bug?
Comment 4 Sean Adams 2006-08-17 13:28:13 UTC
we will not realistically ever get to this.
Comment 5 Shawn Rutledge 2006-08-17 16:59:55 UTC
(In reply to comment #4)
> we will not realistically ever get to this.

Well that's an easy answer, but amplifier control is a really cool thing to have.  I like to use T-amps with my squeezeboxes, and I would really like a way to have the amp turn on along with the squeezebox, when the remote tells it to.  I intend to try to put a T-amp in the same box with a squeezebox.  With the first-generation one, there is that PIO pin which will enable me to control the amp, so I will have to use that; but I will really miss the nice display on the newer ones.  I got one of those first-generation boxes on ebay just so that this would be possible.

(in case you hadn't heard of the t-amp:
http://www.thinkgeek.com/electronics/audio/6cd8/
)

I can't imagine that just having a signal somewhere that goes to high when the squeezebox is "on" is hard.  But most likely there already is a signal like that, you could just tell us where...

Adding buttons shouldn't actually be hard either.  Why do you not want to do this?
Comment 6 Nick Ryman-Tubb 2006-08-18 02:19:42 UTC
Subject: RE:  geekport button input

I have always thought one of the major benefits of the Squeezebox, over the
likes of the big manufactures like Sony, was the open community and open
approach to development.  Given this big interest from the community, it was
a surprise when Slim Devices announced a new unit and failed to add a simple
I/O capability that could have been used for all sorts of creative ideas by
the community.  The earlier versions (see my projects
http://www.moyo.me.uk/squeeze.htm) had some ability for hardware output, but
this was still limited.  I was disappointed that Slim Devices did not
include a simple expansion connector within their designs for their fans!
For a large number of enthusiasts, control for external amplifiers is
important (as well as saving power for the green among us).  Even the big
manufacturers know this.  So why did the new Squeezebox not have an output
on the back for just such control? Oh well.


-----Original Message-----
From: Slim Devices Bugzilla [mailto:bugs@bugs.slimdevices.com] 
Sent: 18 August 2006 01:00
To: nrt@mindsitech.com
Subject: [Bug 2378] geekport button input


https://bugs-archive.lyrion.org/show_bug.cgi?id=2378





------- Comment #5 from shawn.t.rutledge@gmail.com  2006-08-17 16:59 -------
(In reply to comment #4)
> we will not realistically ever get to this.

Well that's an easy answer, but amplifier control is a really cool thing to
have.  I like to use T-amps with my squeezeboxes, and I would really like a
way to have the amp turn on along with the squeezebox, when the remote tells
it to.  I intend to try to put a T-amp in the same box with a squeezebox.
With the first-generation one, there is that PIO pin which will enable me to
control the amp, so I will have to use that; but I will really miss the nice
display on the newer ones.  I got one of those first-generation boxes on
ebay just so that this would be possible.

(in case you hadn't heard of the t-amp:
http://www.thinkgeek.com/electronics/audio/6cd8/
)

I can't imagine that just having a signal somewhere that goes to high when
the squeezebox is "on" is hard.  But most likely there already is a signal
like that, you could just tell us where...

Adding buttons shouldn't actually be hard either.  Why do you not want to do
this?




------- You are receiving this mail because: -------
You are a voter for the bug, or are watching someone who is.

Comment 7 Sean Adams 2006-08-18 11:25:40 UTC
(In reply to comment #5)
> (In reply to comment #4)
> > we will not realistically ever get to this.
> 
> Well that's an easy answer, but amplifier control is a really cool thing to
> have.  I like to use T-amps with my squeezeboxes, and I would really like a way
> to have the amp turn on along with the squeezebox, when the remote tells it to.
>  I intend to try to put a T-amp in the same box with a squeezebox.  With the
> first-generation one, there is that PIO pin which will enable me to control the
> amp, so I will have to use that; but I will really miss the nice display on the
> newer ones.  I got one of those first-generation boxes on ebay just so that
> this would be possible.

Uh... guys, this bug has NOTHING to do with remote amp turn-on.

This was a specific request for a way to read a passive button matrix by scanning for pulse widths determined by an RC time constant for each button.  I doubt that it's _possible_ to do this reliably for more than four buttons or so.  To answer the original request: I would suggest a button pad that sends IR-encoded signals (essentially, a remote) as an alternative.

If you want remote amp turn-on, this is not the right bug.
Comment 8 Shawn Rutledge 2006-08-18 15:53:41 UTC
> Uh... guys, this bug has NOTHING to do with remote amp turn-on.
> 
> This was a specific request for a way to read a passive button matrix by
> scanning for pulse widths determined by an RC time constant for each button.  I

OK, I was attempting to broaden the scope to something that could do both.  A "geekport" reminds me of the BeBox feature of that name, which provided generic PIO.  

Recently I discovered this

http://www.gwendesign.com/slimserver/dev_hard_and_software.htm#io_extension

but it seems to have some limitations when implemented with the Squeezebox hardware.  If it could read buttons, I would think you would agree this is a better way than the RC timing idea.

So it would seem there is an i2c bus already available on some pads.  Maybe just making it easy to use that via the firmware, for both input and output, would be a good start.