LPT Switch

[Trimoor]

Has anyone looked at the LPT switch interface?
One of the latest project announcements used it for their control.
(Or did you skip over it just because it was in Portuguese?)

It's an extremely cheap interface that allows you to connect 60 inputs over the parallel port without ghosting or bloacking.  60!

It's too bad the drivers are windoze only.

Here is the circuit:



Edit:
Here is a translated link:
Translated

[Silver]

Are you sure there is no ghosting?

I've had a quick look and it seems to me that say buttons 1,2 and 6 are held down - then there would be ghosting for button 7? (for example) or am I missing something? (quite possible)

[Trimoor]

If you remember the KE series keyboard encoders, some of them use matricies.
Ghosting can be avoided by sequentially scanning the matrix instead of waiting for a button to be pressed--interrupt vs. polling.

From the site: (translated)

it does not have problem of ghosting (keyboard keys ghost), as in hack of keyboard

Here is a translated link:
Translated

[brained]

Hey that's pretty interesting. I've found the original site, that guy made a home made cockpit. This site also contains the documentation for building the interface

http://users.skynet.be/plipke/PF50HC/

BYOCP

[brained]

Ohh I've found out that the drivers are open source, so If some one can port it to DOS that will help a bunch

[Trimoor]

Linux too please

[Binary010]

I started doing a lot of research on this when I read this post.  A DOS driver is available on this very website...at least one you can hack.

http://www.arcadecontrols.com/Mirrors/www.csc.tntech.edu/~jbyork/default.htm

If you modify circuits.ini and sneskey.ini you can create a custom dos controller config.

At least logically this should happen.  I haven't been able to build a working test model in the last 3 hours.  give me a week and I'll let you know. 

[Silver]

If you remember the KE series keyboard encoders, some of them use matricies.
Ghosting can be avoided by sequentially scanning the matrix instead of waiting for a button to be pressed--interrupt vs. polling.

Aha, thanks for that.....

[Trimoor]

Has anyone tried this yet?
I'll build one, if anyone can give me a dos compilation.

This is a major gap in my cabinet construction.  I can't wire up the controls until I decide between the LPT switch or the keywiz.  I would prefer the LPT switch.

I don't want to start building this until I know there will be dos or linux software for it.

[brained]

yeah, it'll be nice to see that someone tried it out

[Grasshopper]

If you remember the KE series keyboard encoders, some of them use matricies.
Ghosting can be avoided by sequentially scanning the matrix instead of waiting for a button to be pressed--interrupt vs. polling.

From the site: (translated)

it does not have problem of ghosting (keyboard keys ghost), as in hack of keyboard

Here is a translated link:
Translated

Unfortunately, I don't believe that's true. The only way to avoid ghosting is to use diodes.

[brained]

Trimoor, I've bought all the materials to build one. I'll give it a try this weekend and let you know how everything did. I hope it works, 'cause it seems to be a very simple device to build, and cheap too.

[Hoagie_one]

what does the red arrow with a bar pointing down mean in teh diagram?

[Thenasty]

what does the red arrow with a bar pointing down mean in teh diagram?

DIODES

[Hoagie_one]

thanks.  what about the little yellow dots after teh diodes?

[Trimoor]

thanks. what about the little yellow dots after teh diodes?

Nothing.  I don't know why they put that in there.
And before you ask, the black lines are the wire connections.

Brained, it's not the materials so much as the software.  I refuse to put windows on my cab, but no one has a dos or Linux version yet.  It would be nice if some programmer would do this, seeing how it's open source...

[Hoagie_one]

thanks. what about the little yellow dots after teh diodes?

Nothing.

[Hoagie_one]

anyone built this yet?

I started on mine tonight, just labeling wires and what not.  I only bought 10 diodes just to check for ghosting.

Anyone have any results with this yet?

[Projekt Produkt]

I went quick and dirty and got results.  Just one button so I don't know about the ghosting issue.  The driver works well and is very complete.  I am definitely going to go ahead and build a more complete set of inputs now though.

[Hoagie_one]

OK then, I'm going to go out and get teh parts to make a 30 button circuit and see where it leads me

[Lilwolf]

Ok everyone... This will be fine for a one player setup...

but take a look at the diagram... and tell me what happens when this happens...

Assume player 1 is buttons 1-10, player 2 is 11 - 20..

[Hoagie_one]

thats a good questions....and I can only say, I'll tell ya when its built, lol

[Trimoor]

I'm guessing it repeatedly scans every switch in the matrix, avoiding this problem.
Besides, button 11 is on pin 4 while button 1 and 2 are on pin 3.  What's the problem?

In any case, we'll find out soon.

[Hoagie_one]

Just ordered all the diodes i need for this, just gotta get a few more terminal blocks.

Question.  Whats teh best way to test for ghosting or interference?  Should i just rig all 60 buttons combinations to one button press and see what teh results are?

[Projekt Produkt]

The driver being used sees the first 30 positions as joy1 and the second 30 positions as joy2.  So the driver might at least reduce some ghosting, if any exists, by guaranteeing that two players will have dedicated lines at the port. 

What would be a good way of testing for ghosting.  If someone could say what condition might show some let us know and I'll hook those up and give em a push.

[Hoagie_one]

When i get this hooked up, im going to use the "joy to key" program and assign all buttons a character value so i can test it in Notepad.....I think

 

[Lilwolf]

better way to test it is in the joystick config.

you can press 10 buttons... and try adding one more.

notepad will not test much... just that it got the initial tick... not that its held... (will test for ghosts but so will the joystick calibration page)

[Trimoor]

Well, I finally tried this, and for the most part it works.

I only have four buttons attatched right now, and they're all on the same matrix row, so I can't test ghosting yet.
I can't even test multiple buttons.  The program crashes as soon as I press any button, but it manages to show a keypress before it dies.

If anyone can provide me with working software (preferably dos), I can test this more thoroughly.

I used the program from brained's link:
http://users.skynet.be/plipke/PF50HC/

[NoOne=NBA=]

This seems like an awful lot of headache to save yourself $20.
Just get a KeyWiz Eco, solder the dang thing up, and KNOW you're not going to get any ghosting.

I don't want any MAY work/SOMEtime/KINDA solutions when it comes to the inputs on my cab.
I want to KNOW that, when I press a button, the CAB knows I pressed a button.

[Trimoor]

But if you need more than 32 inputs, or are a total cheepskake like me, it still seems like the best solution.

[NoOne=NBA=]

I'll skimp on the COMPUTER before I skimp on the controls.

[brained]

Well, I finally tried this, and for the most part it works.

I only have four buttons attatched right now, and they're all on the same matrix row, so I can't test ghosting yet.
I can't even test multiple buttons.  The program crashes as soon as I press any button, but it manages to show a keypress before it dies.

If anyone can provide me with working software (preferably dos), I can test this more thoroughly.

I used the program from brained's link:
http://users.skynet.be/plipke/PF50HC/

trimoor did you try to download the drivers from the brazilian site.

[Trimoor]

It's working perfectly now.  I installed a program called "PPJoy", the one used on the Brazilian site.
Here is a link to the official site:
PPJoy

The program is pretty impressive.  It supports all the directpad pro interfaces, plus a mouse->joystick converter.  It can even derive joystick controls from a dll.

I don't want any MAY work/SOMEtime/KINDA solutions when it comes to the inputs on my cab.

I bet the same thing was said about the keyboard encoders when they were in the beta stage too.
Just you wait. ('enry 'iggens)

[Lord Hiryu]

Hi, guys!
The LPT-switch, works!!
My portugesse friend, told me.
It's the easiest way to achieve a good interface for our controls.
The only "bad" thing about this great interface, it's only for win, not Ms-dos (i Think there's no dos drivers available)
BTW, there's a link (a spanish forum), when me and other guys, working on it.
http://www.forumcommunity.net/?t=1104461&st=15
I made the interface this weekend, and only rest to put the diodes 1N4148 on it.
By default, the directional arrows for each joystick are:
Connection number 1 to 4 are the directions of player 1, the rest of the connections are from 5 to 30, these connections will be buttons (you may remap all the functions, on the mame's options)
From 31 to 34 connections are the directions for player 2, and the rest of connections, 35 to 60, will be the buttons of 2 player.
And yes, the 4700ohm resistor is not needed, only need the 1N4148 diodes to avoid the "gosthkey" effect.
Cheers.
P.D.: Sorry for my english again , I hope that this text will be understandable.

[Trimoor]

From the looks of the driver, the connections can be mapped to any joystick buttons/axis desired.  The only limitation is that 1-30 is for joystick 1 only, and 31-60 are for joystick 2 only.
This means the connections cannot be completely random and still assign buttons however you like.  Not a problem in MAME, but it might be for other games.

I didn't need resistors either.  Don't worry about what part number of diodes to use, any of the small glass switching diodes will work.  Just don't use 60Hz rectifier diodes.

Don't worry about your english Lord Hiryu.  You have successfully managed to convey your message using our atrociously illogically confusing language.

Here is a english translated link to the forum:

forum

If this interface takes off, we really should give our thanks to the author of the PPJoy program.

[oldsage]

One thing i didn't understand, those it support four players??? That wasn't very clear to me...
Also, are there any adaptors that can change it from D-SUB to PS/2, or USB??? It would make it a bit easier if you don't need a gameport to use the interface...

[Trimoor]

It supports 60 individual inputs, which MAME can assign in any way it chooses, so yes.

It doesn't use the game/midi port, it uses the parallel/printer port.  There are devices that convert parallel to USB, but these are made for printers and usually don't support the full parallel port protocol.  I'll try it with my ethernet print server and let you know.

[Hoagie_one]

still waitin on teh 100 pk of diodes i ordered.

[oldsage]

It supports 60 individual inputs, which MAME can assign in any way it chooses, so yes.

It doesn't use the game/midi port, it uses the parallel/printer port.

[Lilwolf]

The first person to get it working.

Do me a favor... Test the hidden buttons..

Force press all of player one buttons...

Then unpress one.  Make sure that it works well.  (ie, player one buttons work well even when all the others are pressed.

Then presss all on player one and all on player two... Then unpress one in player two.  Add one.

I'm worried that it will work in most cases... But some it will drop some presses. 

I already have a MK64... But I keep thinking about building another cabinet.  I was planning on just getting a small encoder... but I might be tempted in trying one myself..