MadCatz Dreamcast hack help

[pmc]

OK... so I gave up on the AstroPad due to the multiple grounds. I made the guy at EB-Games dig through a bin of old DC stuff to locate me two MadCatz controllers. One of them even works.

So without even reading a tutorial, I wired it up. And it doesn't work. Button presses seem to result in random input to the game. The start button seems to activate the "turbo" or whatever that button with the light is.

I checked my connectivity with a multimeter and everything seems right. I double-checked my pin matching and it all seems right. I even tied in an extra ground to make sure that every section of the board was well grounded.

Can you see anything obvious from the pic? My wiring colors are random BTW. The upper-right yellow wire is where I chose to pick ground off the board. I later tied that to another ground point (I think the start button ground).

I tried overlaying the original start button and pushing that down, but it doesn't seem to work. Could it just the quality of my soldering (which is admitedly fairly poor) or a short. I'll look for a tutorial, but I'd appreciate any hints or suggestions.

[pmc]

It didn't seem to want to add the pic. I'll try that again.

[Dave_K.]

This is very simple to debug.  First, you didn't mention if these wires were connected to an arcade joystick yet or not...if so, disconnect the joystick (so you are left with your hacked pad and each wire not connect to anything else).

Next, remove the ground connection you made from one part of the pad to the other part...you only need one ground to use as a reference not two...I don't think this is the problem, but lets just take this out of the equation.

Next, make sure your ground is actually touching ground, use a multimeter set to continuity test, and test one end of your ground wire to either that other ground wire, or the ground pin at the end of the DC cable (you will have to look this up on gamesx.com if you don't know which connector pin is ground).

Next, hookup the hacked pad to your dreamcast, and turn it on (with a game loaded).  Start testing with the "start" button wire, and work your way around to all the other wires.  Simple touch the ground wire bare end, to the start wire's bare end.  That should simulate the button press.  If you get nothing, then the start button isn't soldered correctly (assuming the ground was verified in the previous step).  Do this test for each button wire and you are done!

Good luck.

[pmc]

This is very simple to debug.  First, you didn't mention if these wires were connected to an arcade joystick yet or not...if so, disconnect the joystick (so you are left with your hacked pad and each wire not connect to anything else).

Connected to a CP with a j/s and microswitches via an internal SCSI ribbon cable. I can disconnect from the CP and test conitnuity to the pins on the ribbon cable. I've done this part already.

Another note: I skipped the shoulder switches and 5th and 6th play buttons. They are currently connected to nothing. I don't think it matters, but I thought I'd mention it.

Next, remove the ground connection you made from one part of the pad to the other part...you only need one ground to use as a reference not two...I don't think this is the problem, but lets just take this out of the equation.

I'll do that. I was grasping at straws at that point. I noticed that some ground tests gave me big buzzes on my multimeter and others were weaker. So to be safe, I tied the grounds together to be sure I have a good signal everywhere on the board. I'll undo that change.

Next, make sure your ground is actually touching ground, use a multimeter set to continuity test, and test one end of your ground wire to either that other ground wire, or the ground pin at the end of the DC cable (you will have to look this up on gamesx.com if you don't know which connector pin is ground).

I've done that (and it tested good), except that I didn't test all the way to the DC connector. I'll do that.

Next, hookup the hacked pad to your dreamcast, and turn it on (with a game loaded).  Start testing with the "start" button wire, and work your way around to all the other wires.  Simple touch the ground wire bare end, to the start wire's bare end.  That should simulate the button press.  If you get nothing, then the start button isn't soldered correctly (assuming the ground was verified in the previous step).  Do this test for each button wire and you are done!

So this is where things had broken down. I test START solder to be OK. I test it's GND to be OK. But it doesn't work. Hmmmm....

OK. So I'll be more methodical in my approach this time and likely find the culprit. Maybe I've got a button in continuous press due to lousy soldering.

Stay tuned....

OK.... two other things I've just learned:

1) The 5th and 6th buttons (C and Z) are digital versions of the analog shoulder triggers. Cool. I can now wire that easily.

2) The LED button is for switching from 4-button to 6-button mode. What's the default at power up? If it's 4-button, my comment (1) above is a problem. Secondly, the LED button is used for reprograming the pad. How do I reset to default programming? It's possible that in all my playing with the wiring, I've reprogrammed the pad so that I'll never get it working right!  

[TalkingOctopus]

Here is some one else who hacked the mad catz: http://www.darkravenwind.com/cdvision/madcatz.htm.  I'm not sure how helpful his tutorial is, but it might be worth checking out.

[pmc]

Nothing is simple.  

I did this: reduced to a single ground connection. I used the one between X and Y. I checked continuity all the way from the CP to the Dreamcast and it's good.

START still didn't work. So i pushed the original button down onto the circuit board to simlulate a button push. It worked. So I retested my connections. Good. So I jumpered from START signal to my GND. No go. So I jumpered from my START signal to the START GND and BOOYA it works.

OK. So I replace my original chosen ground reference with the one that works for START. START works great, but everything else acts funny. OK. So let's try this....

Y should attack (Soul Calibur). But it blocks. Jumper from my solder gumdrop on the circuit board to ground on the CP and I get a block. Push a button down on top of the Y and I get an attack. OK, so I'll jumper from my Y signal gumdrop to the ground connector that's closest to it. Booya! Attack.

Soooo..... If I use the ground closest to the X and Y, START does not work correctly. If I use the ground closest to the START, X and Y don't work correctly. A similar test indicated the same thing with the directionals.

Do I have one of those early MadCatz that used multiple grounds? Or could it be something else. Further, can I try tying all the grounds together?   Or what else can I try?

[Grasshopper]

Have you tried a continuity test between all the various ground points you've just tried? If there is no resistance then it's probably safe to assume all the buttons share a common ground. If not then you've got a problem....!

I really hope this works out for you as I'll shortly be hacking a couple of Madkatz controllers and I want the experience to be as hassle free as possible!

[Dave_K.]

I did this: reduced to a single ground connection. I used the one between X and Y. I checked continuity all the way from the CP to the Dreamcast and it's good.

START still didn't work. So i pushed the original button down onto the circuit board to simlulate a button push. It worked. So I retested my connections. Good. So I jumpered from START signal to my GND. No go. So I jumpered from my START signal to the START GND and BOOYA it works.

What this should tell you is that your original choosen ground was either not really ground, or that your pad uses multiple ground wires...which is not good...see the Astropad thread for more info.

[pmc]

Regarding ground possibly not being ground: Take a look at the picture above. You can see the common leads in the center between the D-pad switch contacts (on the left). It's obvious which side of each switch is the ground and a continuity test verifies that.

I used the "beep" feature on the multimeter to verify continuity between each of the grounds on the board. You can see that there are several (one in the center of the D-pad, one between X and Y, etc.). I get a beep but the closer I am to my reference, the stronger the beep is. i.e., some of the beeping is a little weaker than others. My astropad reacted the same way and it had multiple grounds.

I haven't tried looking at the resistance between the ground contacts. Just continuity. I'll try that.

I hope it's not a multiple-ground unit. If that's the case, I'll pile it up with the astropad carcasses and go look for another victim. I did read ONE reference to early MadCatz units that were multiple grounds but then I guess they went to a single ground for the vast majority of what's out there. Certainly the behavior is similar to the way the Astropad acted when I used one of the several grounds as a common ground for the whole board.

I have one more MadCatz, but it doesn't seem to work at all. I'm busy for a few days, but then I'll go back to EB-Games and dig for two more controllers.

This is getting old fast. But I'm learning a bunch I guess. I'm a glutton for punishment.  

Thanks for all the help guys.

- p

[gnateye]

i have hacked quite a few of these pads so hopefully i can help:

firstly, there is only one ground on this pad so no worries about mulitple grounds.

next, make sure your solder joints are actually making good contact. your multimeter will com in handy here.

now, this one is very important, that button you left on there is a "program" button used to reprogram the pad buttons to whatever you want them to be (although why anyone would need to do this is beyond me, it screw up stuff way more than helps) so make sure that you haven't accidentally "reprogramed" the inputs, that can really mess things up. to fix this you have to press the button down until the LED lights up, then you push the button to reprogram, the led will blink, then you would hit the button you wanted to assign to that button. what you'll do is hit the program button, hit "A", when it blinks hit "a" again, do this with every input (directions and buttons, hitting the program button then the input twice to reassign it to its original input) its confusing but a lot of times that is what is screwing these pads up. they are great to solder to but have a few short comings

next instead of using a game to check your pad, goto my website and download my "dc controller test" program, burn it to a cd and use that to test your pad, its basically a screen with 2 pads on it, when you make a button press, the corresponding button lights up on the screen, this a realy fast and accurate way to test dc hacks, its at the top of my page ( http://www.cosmicbreaks.com/arcade )

i hope that helps. and please anyone who does dc hacks feel free to dload that little controller test program from my site and use it, it saved my but a few times and is much faster than trying to do the same thing ingame.

[gnateye]

oh yeah one more thing.
these pads are great to hack because they have the holes in them, so stick the wire thru from the back of that pad, then solder on the front, its much cleaner and that is what makes these pads better than others.

[pmc]

oh yeah one more thing.
these pads are great to hack because they have the holes in them, so stick the wire thru from the back of that pad, then solder on the front, its much cleaner and that is what makes these pads better than others.

I always take the non-obvious way. I went through the front and then soldered on the back... between capacitors and everything. Touchy work in some places (for a boob like me anyway).

Thanks for the link to your controller tester app. I'll do that straight away. It sucks to test with a game because you have to remember what normal input and combo key presses look like and you can't get past the title screens if START is busted...

I still think, however, that I have a multiple ground version of the board. But we'll see....  

[unicron]

Hey buddy. You probably have a multiple ground version of the board.

Here is a pic of a 100% working hack I did on a madcatz for 2nd player on my arcade. I know it's messy wiring but I also know it works if you connect those wires to the buttons. It has at least 3 grounds. I found which where which by touching the wires together with the pcb plugged into my dc. if you touch the wrong ones, it'll just give you a blue box (controller disconnected)  just replug the sumbish back in and go back to work.

You too can figure out the grounds this way.
You can try the wires before you even solder them to the board.

 I believe on the madcatz I hacked in that pic, there was 1 ground for kicks, 1 ground for punches, and 1 ground for directions.

here's the ugly pic I snapped, for what it's worth:

http://home.houston.rr.com/lcastles/mcatz.jpg

EDIT:

I use a fighting game (third strike) in training mode to test the wires.

And upon further inspection of your pic, it looks to me like we have the same version of madcatz pcb.

To some extent the wires in my pic are color coded, but you can tell which are grounds because they are daisy chained with multiple quick disconnects, to give you an idea of where to look for the multiple grounds on your pcb.

If you don't use the proper grounds for the proper buttons when you hack this pad, the buttons won't work right, you'll get 2 sets of punches or crazy ---steaming pile of meadow muffin--- like that, on this pad it actually matters.

[Grasshopper]

Hey buddy. You probably have a multiple ground version of the board.

Here is a pic of a 100% working hack I did on a madcatz for 2nd player on my arcade. I know it's messy wiring but I also know it works if you connect those wires to the buttons. It has at least 3 grounds. I found which where which by touching the wires together with the pcb plugged into my dc. if you touch the wrong ones, it'll just give you a blue box (controller disconnected)  just replug the sumbish back in and go back to work.

You too can figure out the grounds this way.
You can try the wires before you even solder them to the board.

 I believe on the madcatz I hacked in that pic, there was 1 ground for kicks, 1 ground for punches, and 1 ground for directions.

here's the ugly pic I snapped, for what it's worth:

http://home.houston.rr.com/lcastles/mcatz.jpg

EDIT:

I use a fighting game (third strike) in training mode to test the wires.

And upon further inspection of your pic, it looks to me like we have the same version of madcatz pcb.

To some extent the wires in my pic are color coded, but you can tell which are grounds because they are daisy chained with multiple quick disconnects, to give you an idea of where to look for the multiple grounds on your pcb.

If you don't use the proper grounds for the proper buttons when you hack this pad, the buttons won't work right, you'll get 2 sets of punches or crazy <auto-censored> like that, on this pad it actually matters.

Can you tell me what your pad looked like externally before you hacked it?

I've bought several Mad Katz pads as they were recommended here as being suitable for hacking. They are in various translucent colours, quite cool looking actually. I'm praying they're not the multiple ground versions!

[unicron]

I threw the shell away like a year ago.

Having multiple grounds is no big deal, it's nothing to 'pray' about, I gaurantee. If you can handle like, driving a car, or feeding yourself, IMO this is hella easy! just connect up and test the wires! you'll find the right ones, stick the wires you are testing in the little holes, just don't solder till you know which is which.

I heard the madcatz I bought wouldn't have multiple grounds either, but then just when testing it in training mode I realized it did, and it was no big deal, just an extra wire or two.

crack it open and test your controller!

edit, it goes like this:

1. get in training mode
2. plug in controller (harvested pcb)
3. stick your ground wire in it's hole on the pcb.
4. touch it, one by one, to the holes for the other buttons.
5. Remember for which buttons that ground produces the right button presses.
6. If you need to use another ground, this is when you'll find out, and it will be no big deal. If you need another ground, just use the same wire, or leave that one in and get a different one.

7. if neccessary, find another ground on the controller and stick your wire in it,
then find out which buttons it works for, etc.

just unplug & solder as you figure the stuff out, or just remember what goes where as you learn, and unplug and solder when you are done!

The HOLES on the madcatz PCB are what really make testing your buttons and finding out what is what a CINCH!

[Dave_K.]

Having multiple grounds is no big deal, it's nothing to 'pray' about, I gaurantee.

As long as you don't plan to use your control panel for anything else but these hacked pads I guess its ok (just a few extra wires to worry about).  But if you plan to use the panel for anything else you will run into problems coming up with a common connection scheme.

[pmc]

Having multiple grounds is no big deal, it's nothing to 'pray' about, I gaurantee.

As long as you don't plan to use your control panel for anything else but these hacked pads I guess its ok (just a few extra wires to worry about).  But if you plan to use the panel for anything else you will run into problems coming up with a common connection scheme.

Apologies for jumping back in after weeks away. It's summer and stuff like fishing and swimming and camping and boating and travel and beaches are way more compelling than arcading right now!  

What you say above is exactly my concern and I agree with you. MY CP already has a common ground and I'm not too interested in rewiring it. You can read the related discussion on matter in the recent Astropad thread which I also started because I ran into multiple grounds when hacking an Astropad. I then ran to the MadCatz and ran into the same problem!  

The good news is that mine does appear to be a multi-ground PCB and it's not a problem with my logic (which was exactly the reason for my original post).

Grasshopper: look at my original picture in this thread. That's the multi-ground board. I have two Astropads that look the same on the outside but have different boards inside so I don't see why the MadCatz would be different. You may be able to tell by inspecting the board and comparing to my picture. You can't easily tell with a multimeter because each ground is seems to have continuity to the next although at various strengths. It's enough that it seems to pass a continuity test, but the controller won't operate correctly when a buttons signal is connected to the wrong ground. That's why I've been through 3 or 4 controllers without success yet. Each seems to be great until game-time when the whole pad seems to operate incorrectly. And jumpering the grounds all together doesn't work!  

I'll let you know how my next attempt goes when I give it a shot sometime over the next couple of months. Meantime, I'm going back to summer fun until labor day! Gotta go throw some ribs on the barbie now...

Good luck!