Gp2x Joystick Exposed - The Internals

[Series-8]

I think the best solution would have been modular replaceble joystick "cartridge's".

Or a D-Pad with a small hole in the center for a screw in analog stick.

Best of both worlds.

Right now the only solution I see is for GPH to rotate their switch module around 45°
(which means there will be 3 versions of the GP2X... collectors will be happy but everyone else will not)

Or locate a better part.

Also that microswitch joystick was nearly the first thing I hit when searching, indeed they did not look hard for a better design.

Tsk, Tsk. :angry:

 

[Jarska333]

Or a D-Pad with a small hole in the center for a screw in analog stick.

​

Do you know what analog means?

 

[Paradox]

i hope GPH comes up with a fix for the joystick

 

[RiX0R]

I wish they'd used an analog stick .

Analog sticks are the bomb for all kinds of games, and they can be "digitized" in software to any desired angle/bias/whatever as needed.

 

[TelcoLou]

I wish they'd used an analog stick .

Analog sticks are the bomb for all kinds of games, and they can be "digitized" in software to any desired angle/bias/whatever as needed.

​

I strongly disagree ...

Try playing one of those "arcade classics" compilations on a PS2 using the analog stick for games like Frogger, Pacman, Donkey Kong, etc .... (Though I'll admit, it does work nice for games like Sinistar or Bubbles)

 

[remowilliams]

I strongly disagree ...

Try playing one of those "arcade classics" compilations on a PS2 using the analog stick for games like Frogger, Pacman, Donkey Kong, etc .... (Though I'll admit, it does work nice for games like Sinistar or Bubbles)

​

Definitely agree with that. I'm glad they didn't use an analog stick.

 

[LHC]

Try double tapping a direction on an analogue stick and say it is better than a D-Pad. I dare you.

 

[Radek]

I strongly disagree ...

Try playing one of those "arcade classics" compilations on a PS2 using the analog stick for games like Frogger, Pacman, Donkey Kong, etc .... (Though I'll admit, it does work nice for games like Sinistar or Bubbles)

​

Definitely agree with that. I'm glad they didn't use an analog stick.

​

The problem with analog is that it's very difficult to move only in y or x axis not touching another. So you might want to go up but you will get some of left or right with it. Analogs are best when there are separate sticks for horizontal and vertical movements.

So having only analog stick wouldn't be too hot.

What we have now would be quite decent if not that dumb GPH's mistake not rotating it by 45 degrees.

 

[cakey]

It was dumb making it a stick instead of a d-pad.

 

[telengard]

I wish they'd used an analog stick .

Analog sticks are the bomb for all kinds of games, and they can be "digitized" in software to any desired angle/bias/whatever as needed.

​

I strongly disagree ...

Try playing one of those "arcade classics" compilations on a PS2 using the analog stick for games like Frogger, Pacman, Donkey Kong, etc .... (Though I'll admit, it does work nice for games like Sinistar or Bubbles)

​

Heh, Sinistar is sooo hard to play, not only without an analog joystick, but without that centering "spider" it has. My scores are horrible in MAME.

Bubbles is an awesome game too!

~telengard

 

[Axe_Murderer]

What I suggest for the mod is to solder thin and small sheet metal pads over the X,Y contacts.

​

Perhaps another possible solution for those who might be a bit worried on account of being clumsy using a soldering iron would be to extend the size of the smaller aforementioned orthogonal contacts by applying some kind of durable, electrically conductive paint or lacquer? In this way, the undesirable diagonal bias could similarly be minimized because the orthogonal contacts will have been augmented.

I don't know if such an appropriate substance exists, though. However, I thought that I would mention it in case it helped towards the flow of useful ideas. <_<

 

[thejoeandchip]

is anyone here trying to find a good "2" switch we could use. im googling like never before. and doesn't it seem easy enough to switch this out with a d-pad, a little cutting to the case is all you'd need, right?

 

[DaveC]

is anyone here trying to find a good "2" switch we could use. im googling like never before. and doesn't it seem easy enough to switch this out with a d-pad, a little cutting to the case is all you'd need, right?

​

Pads are only 4 switch this is 8. GPH really fucked us by using an 8 switch stick. If they put in a shitty stick but it was 4 switch we would of had a myriad of options for replacements. It is always something, SNAFU.

 

[nubie]

is anyone here trying to find a good "2" switch we could use. im googling like never before. and doesn't it seem easy enough to switch this out with a d-pad, a little cutting to the case is all you'd need, right?

​

Pads are only 4 switch this is 8. . . .we would of had a myriad of options for replacements.

​

I suppose so, but I wonder if we couldn't make a D-pad anyway.

If we made a 4-way pad and used a system of gates to add a diagonal signal, and then put the switch next to it (it really isn't possible to hit it and hold a direction anyways), we could replace the whole raised surface under the stick with a D-pad.

I just had to, now back to my real Idea.

Could it be possible to switch out the short contacts for long ones in the stick?
I mean disassembling the stick fully to the point of removing those Copper contact spots, then elongating the X and Y pad spaces and putting spare Diagonal copper contacts in there as well.

Ah, forget it all, I would be happy with a cupped cap and a shim mod to reduce the 12° throw (24° from L to R) to maybe half that (I'd honestly be happy with even 1/3 reduction)

I am no stranger to modding a reduced throw, I built my own metal short-shift linkage for my hot rodded VW Rabbit GTI. I have a Audi 2.0 high compression engine and a close-ratio gearbox. I even lightened my own flywheel on a lathe at the local community college, that allowed me to shift gears with no clutch, even though all my syncros are shot.

 

[DaveC]

Could it be possible to switch out the short contacts for long ones in the stick?
I mean disassembling the stick fully to the point of removing those Copper contact spots, then elongating the X and Y pad spaces and putting spare Diagonal copper contacts in there as well.

Ah, forget it all, I would be happy with a cupped cap and a shim mod to reduce the 12° throw (24° from L to R) to maybe half that (I'd honestly be happy with even 1/3 reduction)

​

Those contacts are molded in, really can't be removed easily.

I still think that I could solder in shim pieces over the X,Y contacts. The slightly higher profile would both give more XY bias plus reduce the dead zone/throw. What I would do would be tin a thin strip of brass on one side and sand smooth so there are no solder blobs just a nice thin coat. Tin the contacts with a thin layer of silver solder. Then place in the stick, apply a bit of heat with the iron until the solder joins and then for get about it.

Cupped cap?

You mean like these:

 

[ant512]

It's past one in the morning now, so excuse me if this doesn't make any sense.

All of the posts so far have been concerned with the contacts - how to re-orientate the joystick so that the large contacts are at 0 and 90 degrees; or shortening the large contacts; or lengthening the short contacts.

How about changing the shape of the hat itself, so that it's a squashed circle? Like a rounded diamond shape, achieved by shaving off the sides at 45, 135, 225 and 315 degrees? Or even (and this one's pretty cunning) cutting out sections of the hat, so you end up with a hat that looks something like this:

Black sections represent the bits you leave on the hat; grey sections are the bits to remove.

Not sure if it'd work, and I'm too tired to really consider it properly at the moment. Should give you a proper 8-way stick without dead zones. Probably have to make the diagonal prongs slightly shorter than the vertical/horizontal prongs. Might feel a bit weird, though...

Any thoughts?

 

[Eolair]

It's past one in the morning now, so excuse me if this doesn't make any sense.

All of the posts so far have been concerned with the contacts - how to re-orientate the joystick so that the large contacts are at 0 and 90 degrees; or shortening the large contacts; or lengthening the short contacts.

How about changing the shape of the hat itself, so that it's a squashed circle? Like a rounded diamond shape, achieved by shaving off the sides at 45, 135, 225 and 315 degrees? Or even (and this one's pretty cunning) cutting out sections of the hat, so you end up with a hat that looks something like this:

 \|/
---
 /|\

Not sure if it'd work, and I'm too tired to really consider it properly at the moment. Any thoughts?

​

Heh, I think I'm a bit too tired too. In relation to the problem, what good would that cutting do? :unsure:

 

[ant512]

Heh, I think I'm a bit too tired too. In relation to the problem, what good would that cutting do? :unsure:

​

Ok, the problem is that the hat is circular. Consider this picture:

The red circle represents the contacts. You hold the hat at such an angle that the blue arrow is in contact with the red circle. On the standard hat, a circuit is made with the top-left diagonal, despite the fact that you're really trying to push up. On the modified hat the grey portion of the hat is no longer present to make the circuit - the top prong will still be be making a circuit. If you trim the diagonals down, the effect will be even more pronounced. (Or maybe you can't trim the diagonals down - you'd just end up hitting up + left instead... Or maybe you can. Can't think, going to bed.)

As a secondary benefit, there are no deadzones, because pushing the stick in a direction will *always* hit one of the contacts. With the round hat, you can hit the space between the contacts.

EDIT: Sorry for the miniscule pictures - I'm using my Mac and still haven't installed Photoshop, so I'm having to do all of these in a pixel editor instead.

 

[Radek]

Heh, I think I'm a bit too tired too. In relation to the problem, what good would that cutting do? :unsure:

​

Ok, the problem is that the hat is circular. Consider this picture:

The red circle represents the contacts. You hold the hat at such an angle that the blue arrow is in contact with the red circle. On the standard hat, a circuit is made with the top-left diagonal, despite the fact that you're really trying to push up. On the modified hat the grey portion of the hat is no longer present to make the circuit - the top prong will still be be making a circuit. If you trim the diagonals down, the effect will be even more pronounced. (Or maybe you can't trim the diagonals down - you'd just end up hitting up + left instead... Or maybe you can. Can't think, going to bed.)

As a secondary benefit, there are no deadzones, because pushing the stick in a direction will *always* hit one of the contacts. With the round hat, you can hit the space between the contacts.

EDIT: Sorry for the miniscule pictures - I'm using my Mac and still haven't installed Photoshop, so I'm having to do all of these in a pixel editor instead.

​

Schematic is ok and doesn't have to be done im Photoshop.

Main problem with your solutions is that it isn't adressing major issue of gp2x's stick. The orthogonal diections register at 25 degress each and only "25" is way too less. Because of it up/down/left/right movements are sometimes hard to register. Making the hat more selective you'd prevent some of mistakes with diagonals but it will not make hitting othogonals easier. This is the problem with this stick and fastest fix would be rotating it by 45 degress (as sugested by DaveC).

 

[DaveC]

Main problem with your solutions is that it isn't adressing major issue of gp2x's stick. The orthogonal diections register at 25 degress each and only "25" is way too less. Because of it up/down/left/right movements are sometimes hard to register. Making the hat more selective you'd prevent some of mistakes with diagonals but it will not make hitting othogonals easier. This is the problem with this stick and fastest fix would be rotating it by 45 degress (as sugested by DaveC).

​

Rotating 45 deg then adding a shim under the control hat to reduce dead zone would be the best I think. If we (Craig) can accomplish the almost insurmountable task of getting GPH to actually listen to us the hard part would be done (rotation). Then we would just have to shim it and we would have a useable stick.