GH-122 Making a Stupidly Large MX (or Alps) Keyboard

[jacobolus]

How's this?



I’m not totally satisfied, I’ll fiddle a bit more.

[Hypersphere]

How's this?

Show Image

Show Image

I’m not totally satisfied, I’ll fiddle a bit more.

I like it! The simpler, the better.

[berserkfan]

How's this?

Show Image

Show Image

I’m not totally satisfied, I’ll fiddle a bit more.

I'm happy; elegant and good!

[bazh]

I don't know if it were just me but that "H" is very easily mistaken to "K"

[jacobolus]

I don't know if it were just me but that "H" is very easily mistaken to "K"

Basically putting a whole bunch of stripes across letters makes for all kinds of confusing optical effects, and is definitely not optimized for legibility. I don’t think there’s any obvious way to avoid that without ditching the stripes idea.

[jacobolus]

Some tweaks:



SVG file:
 geekhack-ibm.svg (3.1 kB - downloaded 163 times.)

Did I make the G too wide here?

[jdcarpe]

Some tweaks:

Show Image

Show Image

SVG file:
(Attachment Link)

Did I make the G too wide here?

I think the G is fine. Maybe make the serifs on the H a bit smaller?

[jacobolus]

Maybe make the serifs on the H a bit smaller?

I exactly copied the serifs from the I, B, and M in the original IBM logo (all of which are the same).

[bazh]

maybe the serifs could be longer, I mean like those of the I from IBM logo

[jacobolus]

maybe the serifs could be longer, I mean like those of the I from IBM logo

Are you talking about my design here? Or the ones from the previous page? Mine use literally identical serifs to the original IBM logo. (On the H. The G is rather trickier to make something that still looks good with all the white stripes through it, so it differs a bit more dramatically from the source “City” typeface.)

[jacobolus]

One more slight tweak to the G dimensions:



 geekhack-ibm-2.svg (6.08 kB - downloaded 166 times.)

[HoffmanMyster]

Maybe make the serifs on the H a bit smaller?

I exactly copied the serifs from the I, B, and M in the original IBM logo (all of which are the same).

Show Image

Show Image

I love this.  I was hoping we could get the serifs and lines to exactly match like this.   

[berserkfan]

it is getting better and better. Awesome! 

Shall we consider an official geekhack color? I was hoping for a difference from the standard IBM blue or gray. IE I wanted the logo to be both understated, and also stand out.

I think the logo should not be brown to avoid being mistaken for rust. No black, gray, blue also because these are overly corporate.

I think red, orange,  green will all work. I think forest green is more discreet than red/ orange so I'll vote for that. An alternative is purple; it looks rich and royal, yet is clearly not a corporate color so even ignorant people can guess that it wasn't made by a faceless company.

If our current logo discussion works eventually we should adopt that as the official geekhack badge for all geekhack crowdsourced keyboards and keypads. That way we will eventually be able to get discounts for getting 1000 made; it'll also be a proper badge that can be recognized by people on other forums eg reddit.

Lastly, I've always been opposed to all these variations of the logo: http://geekhack.org/index.php?action=dlattach;topic=42051.0;attach=17899;image

I don't like having the name fully spelled out. For one, geek is still a derogatory word in many circles. For two, hack is also a generally negative word. And geekhack suggests to many outsiders that you are part of some semi-illicit hacker community. It's not good to display the word too blatantly on a keyboard when you are doing business and want to be taken seriously. As an older man I am quite away of these things.

Also consider how IBM handled it. None of the words International Business Machines have any negative connotation, but it's too lengthy and staid. A concise IBM, much like a single pair of golden arches for Macdonalds' or just FedEx for Federal Express, makes for an easily recognizable logo. Remember that most logos have very few letters. They save you that few extra nanoseconds to read. You don't need your eyes to sweep from g e e k h a c k . o r g to recognize it - just GH is enough.

And since GH is also the center of the alpha typing space on a QWERTY keyboard, it's very easy to associate us with being the center of the keyboard world.
 
Consider how elegant and memorable this is

[berserkfan]

it is getting better and better. Awesome! 

Shall we consider an official geekhack color? I was hoping for a difference from the standard IBM blue or gray. IE I wanted the logo to be both understated, and also stand out.

I think the logo should not be brown to avoid being mistaken for rust. No black, gray, blue also because these are overly corporate.

I think red, orange,  green will all work. I think forest green is more discreet than red/ orange so I'll vote for that. An alternative is purple; it looks rich and royal, yet is clearly not a corporate color so even ignorant people can guess that it wasn't made by a faceless company.

If our current logo discussion works eventually we should adopt that as the official geekhack badge for all geekhack crowdsourced keyboards and keypads. That way we will eventually be able to get discounts for getting 1000 made; it'll also be a proper badge that can be recognized by people on other forums eg reddit.

Lastly, I've always been opposed to all these variations of the logo: http://geekhack.org/index.php?action=dlattach;topic=42051.0;attach=17899;image

I don't like having the name fully spelled out. For one, geek is still a derogatory word in many circles. For two, hack is also a generally negative word. And geekhack suggests to many outsiders that you are part of some semi-illicit hacker community. It's not good to display the word too blatantly on a keyboard when you are doing business and want to be taken seriously. As an older man I am quite away of these things.

Also consider how IBM handled it. None of the words International Business Machines have any negative connotation, but it's too lengthy and staid. A concise IBM, much like a single pair of golden arches for Macdonalds' or just FedEx for Federal Express, makes for an easily recognizable logo. Remember that most logos have very few letters. They save you that few extra nanoseconds to read. You don't need your eyes to sweep from g e e k h a c k . o r g to recognize it - just GH is enough.

And since GH is also the center of the alpha typing space on a QWERTY keyboard, it's very easy to associate us with being the center of the keyboard world.
 
Consider how elegant and memorable this is

[fohat.digs]

Since I am a shameless **** I will offer to guide you to my ebay sale for a 122 Space Invaders keyboard.

http://geekhack.org/index.php?topic=62288.new#new

After using an IBM F-122 for a couple of years I have grown to really like the configuration, but I can't see using it for other keyboard switch types. Well, maybe blue or orange Alps ......

The board I am selling is very nice, and, amazingly, plug-and-play PS/2 (on my Gigabyte mobo, anyway).

[dorkvader]

g space on a QWERTY keyboard, it's very easy to associate us with being the center of the keyboard world.
 
Consider how elegant and memorable this is

Right next to eachother on Dvorak as well.

After using an IBM F-122 for a couple of years I have grown to really like the configuration, but I can't see using it for other keyboard switch types. Well, maybe blue or orange Alps ......

This PCB is compatible with alps.

[fohat.digs]

This PCB is compatible with alps.

I know, I know.

Problem is, I need to finish about half a dozen projects that are already waiting for me, and purge my collection of at least that many existing ones.

[pr0ximity]

Gotta say, I've been looking at F-122's, but I think this might have to be my next project. I've never built a keyboard from the ground-up, so I'm not interested in the prototype run, but if there's a production run after I'm in 

The Unicomp cases and GH badge are the icing on the cake.

[fohat.digs]

I've been looking at F-122's

But not typing on them.

Otherwise, you would know.

[Hypersphere]

I am glad to see this project idea, and I hope it moves forward. For my own part, I think I will be content to refurbish and convert to USB the IBM F122s that I already have (along with some M122s, although these will also need to be bolt-modded).

To compete with the noise of the original F122s, I would urge you to consider the Matias Alps-inspired tactile-clicky switch for the GH 122.

[Hak Foo]

While I know it's sort of been a side issue on the discussion, we might want to just give the PCB design a sanity check before launch and see if it would be feasible to use Soarer's Controller firmware.  http://deskthority.net/workshop-f7/soarer-s-keyboard-controller-firmware-t6767.html

It apparently supports the Teensy++ and pretty collosal matrices (supposedly, 256 keys) and it looks like it could be simple to configure (well, as simple as a hand-written config file can be expected to be)  All we'd need to be sure of is that it's wiring to appropriate pins on the Teensy++

While someone might want to write something super-custom later, especially if they want to also bolt a bunch of LEDs on, this could give us the ability to use the board within a day or two of the PCBs coming home.   Such a situation could be especially valuable for testing-- even if you don't want to impose a layout, a map which just issued "Row 1, Column 1" as each key was struck would make it child's play to locate sticky or poorly soldered switches.

[dorkvader]

While I know it's sort of been a side issue on the discussion, we might want to just give the PCB design a sanity check before launch and see if it would be feasible to use Soarer's Controller firmware.  http://deskthority.net/workshop-f7/soarer-s-keyboard-controller-firmware-t6767.html

It apparently supports the Teensy++ and pretty collosal matrices (supposedly, 256 keys) and it looks like it could be simple to configure (well, as simple as a hand-written config file can be expected to be)  All we'd need to be sure of is that it's wiring to appropriate pins on the Teensy++

While someone might want to write something super-custom later, especially if they want to also bolt a bunch of LEDs on, this could give us the ability to use the board within a day or two of the PCBs coming home.   Such a situation could be especially valuable for testing-- even if you don't want to impose a layout, a map which just issued "Row 1, Column 1" as each key was struck would make it child's play to locate sticky or poorly soldered switches.

It's a good idea.

I recommended we use soarer's firmware on a teensy++ from the beginning. Metalliqaz might import the matrix to his firmware opening up other options, but I really like how soarer's works what with the config files and not-having-to-reflash-the-chip.

[Hak Foo]

Cool.  I was just worried that the current design might call for pins that that firmware can't reach.

[samwisekoi]

Cool.  I was just worried that the current design might call for pins that that firmware can't reach.

While I know it's sort of been a side issue on the discussion, we might want to just give the PCB design a sanity check before launch and see if it would be feasible to use Soarer's Controller firmware.  http://deskthority.net/workshop-f7/soarer-s-keyboard-controller-firmware-t6767.html

It apparently supports the Teensy++ and pretty collosal matrices (supposedly, 256 keys) and it looks like it could be simple to configure (well, as simple as a hand-written config file can be expected to be)  All we'd need to be sure of is that it's wiring to appropriate pins on the Teensy++

While someone might want to write something super-custom later, especially if they want to also bolt a bunch of LEDs on, this could give us the ability to use the board within a day or two of the PCBs coming home.   Such a situation could be especially valuable for testing-- even if you don't want to impose a layout, a map which just issued "Row 1, Column 1" as each key was struck would make it child's play to locate sticky or poorly soldered switches.

It's a good idea.

I recommended we use soarer's firmware on a teensy++ from the beginning. Metalliqaz might import the matrix to his firmware opening up other options, but I really like how soarer's works what with the config files and not-having-to-reflash-the-chip.

Please tell me more.  I was just about to use the 'inboard' pins to free up the PWM pins on the outside edges.

I believe metalliqaz' tool already supports the Teensy++, but I also want to be sure we have firmware on Day One.

DV, I'll start sending you the matrix pin-outs as they evolve. Please do let me know if any of them won't work.

Thanks!

 - Ron | samwisekoi

[dorkvader]

Please tell me more.  I was just about to use the 'inboard' pins to free up the PWM pins on the outside edges.

I believe metalliqaz' tool already supports the Teensy++, but I also want to be sure we have firmware on Day One.

DV, I'll start sending you the matrix pin-outs as they evolve. Please do let me know if any of them won't work.

Thanks!

 - Ron | samwisekoi

Ok I can come up with a flashing guide and some default layout files. I have the documentation (sadly split across two files) open in another window as well.

I've never used soarer's code on the teensy++ so I will have to do some reading on pin assignments and the like.

One downside is that soarer's code does not support backlighting very well. So unless you want to map "backlight all LEDs" to scroll-lock (or use a toggle switch: my personal favourite idea), it may be better to go with a different firmware.

[Hak Foo]

Now, although the docs aren't specific, it looks like the compiler should be able to handle specifying the PA series pins (which are the middle ones).

From a physical perspective, this does make it a lot harder to have the Teensy socketed-- you can't just solder in a DIP socket anymore if you're accessing pins mid-PCB.  Yeah, you could solder pins to the mid-PCB openings, and then make up something with SIP connectors... but is this even important?   Probably not.

I do see a useful heads-up in the Controller discussion thread: pin PD6 has a LED sitting on it, which means it can't be used as a sense pin.  The latest image you posted seems to avoid PD6, so that's not a big deal.

[samwisekoi]

Now, although the docs aren't specific, it looks like the compiler should be able to handle specifying the PA series pins (which are the middle ones).

From a physical perspective, this does make it a lot harder to have the Teensy socketed-- you can't just solder in a DIP socket anymore if you're accessing pins mid-PCB.  Yeah, you could solder pins to the mid-PCB openings, and then make up something with SIP connectors... but is this even important?   Probably not.

I do see a useful heads-up in the Controller discussion thread: pin PD6 has a LED sitting on it, which means it can't be used as a sense pin.  The latest image you posted seems to avoid PD6, so that's not a big deal.

The Teensy is the same way.  Using that pin as a sense pin is possible; I did it on the GH36.  According to metalliqaz:

The LED is dim because it is on pin D6, which you used for Col12.  The dimness is caused by the column getting energized every millisecond.

I have a plan for attaching the center pins.  I am going to try it out on a breadboard.

Please tell me more.  I was just about to use the 'inboard' pins to free up the PWM pins on the outside edges.

I believe metalliqaz' tool already supports the Teensy++, but I also want to be sure we have firmware on Day One.

DV, I'll start sending you the matrix pin-outs as they evolve. Please do let me know if any of them won't work.

Thanks!

 - Ron | samwisekoi

Ok I can come up with a flashing guide and some default layout files. I have the documentation (sadly split across two files) open in another window as well.

I've never used soarer's code on the teensy++ so I will have to do some reading on pin assignments and the like.

One downside is that soarer's code does not support backlighting very well. So unless you want to map "backlight all LEDs" to scroll-lock (or use a toggle switch: my personal favourite idea), it may be better to go with a different firmware.

At one point you (I think it was you) said you could hang a transistor off of a keyswitch and energize LEDs that way.  Was that you, and is that true?

 - Ron | samwisekoi

[0100010]

Is there a pro / con chart of the available firmware sources?  Soarers / Metalliqaz / Hasu?  LED support, keymap updates, reflashing, layers, etc.  Maybe as it relates to this project, but also in general for others.

[dorkvader]

At one point you (I think it was you) said you could hang a transistor off of a keyswitch and energize LEDs that way.  Was that you, and is that true?

 - Ron | samwisekoi

That was probably me.

If you look in LED backlit keyboards like the poker2, they have a few transistors. That way they can just send voltage out one pin to activate the transistor and then the transistor will switch power on for the LED matrix. I might have to get a germanium one or something with better specs if we're driving a ton of LEDs here (or get 2 in parallel) but yes I planned to use a transistor.

Actually, I planned to use a panel-mounted toggle switch to turn on the transistor to power the LEDs, but I might use a relay if I can get away with it (they use up power), 'cause toggle switches and relays are really cool, and I think fit with the theme of a big bad industrial 122-key keyboard.

[HoffmanMyster]

At one point you (I think it was you) said you could hang a transistor off of a keyswitch and energize LEDs that way.  Was that you, and is that true?

 - Ron | samwisekoi

That was probably me.

If you look in LED backlit keyboards like the poker2, they have a few transistors. That way they can just send voltage out one pin to activate the transistor and then the transistor will switch power on for the LED matrix. I might have to get a germanium one or something with better specs if we're driving a ton of LEDs here (or get 2 in parallel) but yes I planned to use a transistor.

Actually, I planned to use a panel-mounted toggle switch to turn on the transistor to power the LEDs, but I might use a relay if I can get away with it (they use up power), 'cause toggle switches and relays are really cool, and I think fit with the theme of a big bad industrial 122-key keyboard.

Can you explain how using a transistor hooked to a raw 5V source (USB in this case, I presume) is different than using the teensy itself to switch a pin to +5V using its internal controller?  In both cases you would code it to switch a pin to output a current to accomplish your task, so the coding wouldn't be significantly different, would it?  Or is the difference that the teensy pins can't handle the current output we'd need?

Or am I totally misunderstanding?   

[dorkvader]

At one point you (I think it was you) said you could hang a transistor off of a keyswitch and energize LEDs that way.  Was that you, and is that true?

 - Ron | samwisekoi

That was probably me.

If you look in LED backlit keyboards like the poker2, they have a few transistors. That way they can just send voltage out one pin to activate the transistor and then the transistor will switch power on for the LED matrix. I might have to get a germanium one or something with better specs if we're driving a ton of LEDs here (or get 2 in parallel) but yes I planned to use a transistor.

Actually, I planned to use a panel-mounted toggle switch to turn on the transistor to power the LEDs, but I might use a relay if I can get away with it (they use up power), 'cause toggle switches and relays are really cool, and I think fit with the theme of a big bad industrial 122-key keyboard.

Can you explain how using a transistor hooked to a raw 5V source (USB in this case, I presume) is different than using the teensy itself to switch a pin to +5V using its internal controller?  In both cases you would code it to switch a pin to output a current to accomplish your task, so the coding wouldn't be significantly different, would it?  Or is the difference that the teensy pins can't handle the current output we'd need?

Or am I totally misunderstanding?   

I dont think the teensy has the current capacity to drive all the LEDs at once. We're talking like 200 mA LED matrix, I don't want that coming form one pin on the teensy.

And you can still use the teensy to turn the LED matrix on/off for fancy LED effects. Just have the pin on the teensy connect to the transistor gate.


I'm no EE (still not taken any classes) but I know transistors and LEDs from a nanotechnology class I had once (and from reading into DIY amplifiers for some time). I'm busy today, but I'll write up my ideas tomorrow if no-one else has discussed it

[HoffmanMyster]

At one point you (I think it was you) said you could hang a transistor off of a keyswitch and energize LEDs that way.  Was that you, and is that true?

 - Ron | samwisekoi

That was probably me.

If you look in LED backlit keyboards like the poker2, they have a few transistors. That way they can just send voltage out one pin to activate the transistor and then the transistor will switch power on for the LED matrix. I might have to get a germanium one or something with better specs if we're driving a ton of LEDs here (or get 2 in parallel) but yes I planned to use a transistor.

Actually, I planned to use a panel-mounted toggle switch to turn on the transistor to power the LEDs, but I might use a relay if I can get away with it (they use up power), 'cause toggle switches and relays are really cool, and I think fit with the theme of a big bad industrial 122-key keyboard.

Can you explain how using a transistor hooked to a raw 5V source (USB in this case, I presume) is different than using the teensy itself to switch a pin to +5V using its internal controller?  In both cases you would code it to switch a pin to output a current to accomplish your task, so the coding wouldn't be significantly different, would it?  Or is the difference that the teensy pins can't handle the current output we'd need?

Or am I totally misunderstanding?   

I dont think the teensy has the current capacity to drive all the LEDs at once. We're talking like 200 mA LED matrix, I don't want that coming form one pin on the teensy.

And you can still use the teensy to turn the LED matrix on/off for fancy LED effects. Just have the pin on the teensy connect to the transistor gate.


I'm no EE (still not taken any classes) but I know transistors and LEDs from a nanotechnology class I had once (and from reading into DIY amplifiers for some time). I'm busy today, but I'll write up my ideas tomorrow if no-one else has discussed it

That's what I suspected.  I have no further questions then.  Cause essentially the design in question is using a transistor (teensy) to flip an external transistor to send power.  Which without that design constraint you listed (lack of current) seems silly to me.  But that makes sense.

[samwisekoi]

Teensy pins are rated for 50mA, but can push to 100. However, using the Teensy pin and a simple transistor means you can pump VCC through a string of LEDs.  That is what I am doing on the GH36.

What I don't know is if I can use a closed keyswitch (e.g. the Fn key) to activate a transistor and from there power a string of LEDs.

Also, the transistor could be replaced - not with a relay - but with a vacuum tube!  THAT would be machine-age madness!

 - Ron | samwisekoi

[HoffmanMyster]

Teensy pins are rated for 50mA, but can push to 100. However, using the Teensy pin and a simple transistor means you can pump VCC through a string of LEDs.  That is what I am doing on the GH36.

What I don't know is if I can use a closed keyswitch (e.g. the Fn key) to activate a transistor and from there power a string of LEDs.

Also, the transistor could be replaced - not with a relay - but with a vacuum tube!  THAT would be machine-age madness!

 - Ron | samwisekoi

You should be able to write the firmware to accept the Fn key code and in addition to performing the desired function, also put the "LED pin" high and therefore switch the transistor.

[samwisekoi]

Yes, but my programming language of choice is solder!

[damorgue]

Just some thoughts on lighting up 122keys:
USB is 500mA at 5V
Leds are about 1.6V and 20mA each, connected serially in groups of three to reach 5V.

That results in 813mA if I am not mistaken. It is possible to PWM them to not have them all ON at the same time, or run them at lower voltages or currents, but it is still fairly close to the limit of USB 2.0 at least. Just something to keep in mind unless it has already been mentioned.

Edit: Even if you use transistors to circumvent the limit of a teensy, you might run into issues if you want to power it through one USB port.

[AKmalamute]

Edit: Even if you use transistors to circumvent the limit of a teensy, you might run into issues if you want to power it through one USB port.

Back to PS2 ports? Or have the LEDs draw power from a PS2 cord ... or yes, a 2nd USB port ... Don't the TG3 red LEDs use a 12v car adapter? ... maybe a bluetooth keyboard sucking power from a proper 1A wall-wart instead of a battery...

Just throwing ideas out there! No plans to do any of these things!

I'd step up and ask to be another prototyper but I just wrote down all the projects I want in on, and there's no fewer than 7 keyboards I want ... of course that includes the GH60 when it goes live but there's also JoseyQuinn's 42% and the GH36 and ...

 So I guess I'm relegated to moral support for this one. And spouting gibberish without cue.

[samwisekoi]

Edit: Even if you use transistors to circumvent the limit of a teensy, you might run into issues if you want to power it through one USB port.

Back to PS2 ports? Or have the LEDs draw power from a PS2 cord ... or yes, a 2nd USB port ... Don't the TG3 red LEDs use a 12v car adapter? ... maybe a bluetooth keyboard sucking power from a proper 1A wall-wart instead of a battery...

Just throwing ideas out there! No plans to do any of these things!

I'd step up and ask to be another prototyper but I just wrote down all the projects I want in on, and there's no fewer than 7 keyboards I want ... of course that includes the GH60 when it goes live but there's also JoseyQuinn's 42% and the GH36 and ...

 So I guess I'm relegated to moral support for this one. And spouting gibberish without cue.

Well, we could include a powered USB hub inside the case...

[MOZ]

Do not use a secondary USB, that is very, very bad practice.

Using a USB wall adapter is a good idea, checkout the Arduino Schematics for how they handle switch between a wallport and a USB.

[SL89]

no led's needed for me. Caps, num and scroll lock leds are nice, but i absolutely dont need any LED's.

[dorkvader]

Just some thoughts on lighting up 122keys:
USB is 500mA at 5V
Leds are about 1.6V and 20mA each, connected serially in groups of three to reach 5V.

That results in 813mA if I am not mistaken. It is possible to PWM them to not have them all ON at the same time, or run them at lower voltages or currents, but it is still fairly close to the limit of USB 2.0 at least. Just something to keep in mind unless it has already been mentioned.

Edit: Even if you use transistors to circumvent the limit of a teensy, you might run into issues if you want to power it through one USB port.

You dont need to run the LEDs at their maximum rated brightness. We will likely be running about 1ma through each of them: this is about standard for a KB and they will be plenty bright.

[Melvang]

My vote would be for EL lighting.  Get a mat cut (there are places on line that offer this service) slap it on top of the plate and plug it in.  Though it wouldn't look good if the switch plate is the top of the case.

[0100010]

no led's needed for me. Caps, num and scroll lock leds are nice, but i absolutely dont need any LED's.

Ditto.

This.

+1

[Hak Foo]

In a way, I'm not sure backlit even necessarily fits with the overall vibe of the project-- again, a very retro battleship of a board, from back before backlit was a thing.

[samwisekoi]

I'm not a backlighting user at all. I want a CAPLOCK LED and that is it.  Maybe more indicators on a 40% to indicate state/layer.

However, every time I design a keyboard using a Teensy controller, there are not enough pins for even a few LEDs.  And for the GH122 there are nowhere near enough.  So I moved up to the Teensy++ and suddenly there are PLENTY of pins. And the controller is a daughter card anyhow, so I thought what the heck, let's use those pins and make the *controller* useful for more than this one board.

So the GH122 might not have more than three LEDs, but the controller might be smart enough to control more than that.

 - Ron | samwisekoi

[SL89]

I'm not a backlighting user at all. I want a CAPLOCK LED and that is it.  Maybe more indicators on a 40% to indicate state/layer.

However, every time I design a keyboard using a Teensy controller, there are not enough pins for even a few LEDs.  And for the GH122 there are nowhere near enough.  So I moved up to the Teensy++ and suddenly there are PLENTY of pins. And the controller is a daughter card anyhow, so I thought what the heck, let's use those pins and make the *controller* useful for more than this one board.

So the GH122 might not have more than three LEDs, but the controller might be smart enough to control more than that.

 - Ron | samwisekoi

That sounds good to me

[0100010]

Is the controller board layout finalized?

The most recent image in the thread looks like it has a a couple row traces crossed, and there are traces going to the PA0-7 pins; wasn't sure if that was the final plan or not.

EDIT : Also - are the Ground and VCC pins in the correct positions?

[samwisekoi]

Is the controller board layout finalized?

The most recent image in the thread looks like it has a a couple row traces crossed, and there are traces going to the PA0-7 pins; wasn't sure if that was the final plan or not.

EDIT : Also - are the Ground and VCC pins in the correct positions?

Oh, heck no. You will have to be patient my friend. I am waiting for feedback on the GH36 layout and the Unicomp case.

But we SHALL have our battleship! Yes we shall.

 - Ron | samwisekoi

[SL89]

Is the controller board layout finalized?

The most recent image in the thread looks like it has a a couple row traces crossed, and there are traces going to the PA0-7 pins; wasn't sure if that was the final plan or not.

EDIT : Also - are the Ground and VCC pins in the correct positions?

Oh, heck no. You will have to be patient my friend. I am waiting for feedback on the GH36 layout and the Unicomp case.

But we SHALL have our battleship! Yes we shall.

 - Ron | samwisekoi

For what it's worth, this is like a Halo sort of project for me. I am in NO rush for this bad mofo.

[SL89]

Is the controller board layout finalized?

The most recent image in the thread looks like it has a a couple row traces crossed, and there are traces going to the PA0-7 pins; wasn't sure if that was the final plan or not.

EDIT : Also - are the Ground and VCC pins in the correct positions?

Oh, heck no. You will have to be patient my friend. I am waiting for feedback on the GH36 layout and the Unicomp case.

But we SHALL have our battleship! Yes we shall.

 - Ron | samwisekoi

For what it's worth, this is like a halo level of project for me. I am in NO rush for this bad mofo. I wish i could contribute to it, but i am woefully ignorant of what goes into this stuff.