What's Wrong With This Picture?

[BigO]

Yes. All 15 conductors are present.

 

However the logic pinouts for 5200, Famicom and NeoGeo are entirely incompatible, so don't go swapping controllers between consoles...

Thanks. Not to worry. I only have a 5200.

And when I'm done abusing the cable it won't be useful as an extension cable anyway. :ponder:

[Lynxpro]

Thanks. Not to worry. I only have a 5200.

And when I'm done abusing the cable it won't be useful as an extension cable anyway. :ponder:

 

BigO, you should get in touch with LedZep. He's looking to add a spinner to the CX-53.

[Andromeda Stardust]

Thanks. Not to worry. I only have a 5200.

And when I'm done abusing the cable it won't be useful as an extension cable anyway. :ponder:

Currently I only have a Famicom that would use these cables. I plan on getting a 5200 someday to complete my Atari collection (as I have zero interest in a Jag), but NeoGeo is beyond my budget for the foreseeable future.

 

They used to only carry the 4 foot variety, but now they have 6-foot cables with the 4-foot are more or less permanently out of stock. Longer is better anyway.

 

I bought a couple of the older 4 foot cables a year or so back with the intention of doing an adapter for P3 and P4 auxillary controllers. Also a DPDT switch to select between Zapper and Vaus modes on the second port. Famicom Vaus/Arkanoid games use different logic pinouts compared to the NES Vaus meaning you cannot play Vaus compatible Famicom games on a NES system with an NES Arkanoid controller. You need the Famicom Vaus pinout for Famicom games and the NES Zapper/Arkanoid pinout for NES games. The adapter will let me use an NES Arkanoid controller for both.

[BigO]

 

BigO, you should get in touch with LedZep. He's looking to add a spinner to the CX-53.

If he's got the mechanical part figured out, the rest is relatively easy. The most complex part will be the interruption of existing circuit traces to inject the new signal (assuming that's his intent). Even that could be a fairly painless operation. I've been in there so much lately that I can just about visualize where he'd need to tap in without even looking.

[BigO]

Currently I only have a Famicom that would use these cables. I plan on getting a 5200 someday to complete my Atari collection (as I have zero interest in a Jag), but NeoGeo is beyond my budget for the foreseeable future.

 

They used to only carry the 4 foot variety, but now they have 6-foot cables with the 4-foot are more or less permanently out of stock. Longer is better anyway.

 

I bought a couple of the older 4 foot cables a year or so back with the intention of doing an adapter for P3 and P4 auxillary controllers. Also a DPDT switch to select between Zapper and Vaus modes on the second port. Famicom Vaus/Arkanoid games use different logic pinouts compared to the NES Vaus meaning you cannot play Vaus compatible Famicom games on a NES system with an NES Arkanoid controller. You need the Famicom Vaus pinout for Famicom games and the NES Zapper/Arkanoid pinout for NES games. The adapter will let me use an NES Arkanoid controller for both.

I wish they still had the cheaper, shorter ones.

I don't need any extra length.

[BigO]

Now what's this clown been up to?

Edited December 3, 2015 by BigO

[BigO]

Hardly practical, but if you're looking for credibility among your retrogaming buddies...

 

Actually, I just wanted to make sure I could drive a mouse with the connections that I had exposed to this point. Check.

[Albert]

That's pretty cool -- I am amused by the sound of the 5200 Trak-Ball as you're mousing around.

[BigO]

That's pretty cool -- I am amused by the sound of the 5200 Trak-Ball as you're mousing around.

Thanks. Amusing is not the word I was thinking at the time. I was looking at what surfaces I could cover or recesses I could fill with sound deadening material.

 

First, I'll invest in bearings. You should have heard the ones that I swapped out so I wouldn't be too embarrassed to even make that video. Considering that all four of my Trak-Balls were just barely prevented from going in the dumpster as unrepairable, they're in fine condition.

[Andromeda Stardust]

 

Now what's this clown been up to?

 

A plugin is required. Not sure why, as Firefox html5 supports mp4 natively. I dug into the source and pulled the direct file if you had trouble viewing it:

http://www.atariage.com/forums/gallery_ips/gallery/album_317/gallery_12370_317_852830.mp4

 

EDIT: I just had an epiphany. No need for MAME encoders. I have an old Apple puck mouse lying about that I have zero use for. It's USB and only does one button. I could just cram the guts inside of my trackball and connect a USB cable. A quick Google search shows that Apple puck mouse is very common and not worth much.

[BigO]

 

EDIT: I just had an epiphany. No need for MAME encoders. I have an old Apple puck mouse lying about that I have zero use for. It's USB and only does one button. I could just cram the guts inside of my trackball and connect a USB cable. A quick Google search shows that Apple puck mouse is very common and not worth much.

Hmm...they seem to be catching onto my schemes.

 

I didn't shove the mouse inside the trackball for my test though. I kept it external and fed it with the conditioned quadrature signals from the comparator output that I already had externally available. (I was testing to see if I should bring the raw optical sensor outputs to the outside to support a future mouse adapter module for MAME, etc.)

 

Just in case you see the same thing: this arrangement freaked out the mouse circuitry at first. It didn't take well to those signals being forced to LO. Both sensors had to be active (not blocked) in the trackball on one axis in order for the other axis to work.

 

I stuck in some diodes on those lines to make a LO be effectively hi-z and the mouse was happy again. Didn't have any small signal diodes laying around so threw some 1n4004's in there and it worked okay.

 

I was just hacking a proof--of-concept to meet a specific objective. You might be able to get away with tapping the unconditioned signals to keep your parts count down. But, those sensors might not drive both sets of circuitry. Or your mouse just might not care.

 

Look forward to seeing the posting of your results.

Edited December 3, 2015 by BigO

[BigO]

That's pretty cool -- I am amused by the sound of the 5200 Trak-Ball as you're mousing around.

As recorded by my cell phone, it sorta sounds like "scratching", which just ain't my cup o' tea.

Pretty sure I couldn't pull off the street name "DJ Trak-Ball" anyway. Or any street name for that matter.

 

Something must be done to remedy this noise in the future.

Edited December 3, 2015 by BigO

[BigO]

Got the 2600 keyboard controller adapter built and tested. Testcart and few games say that all 12 buttons are working.

 

I probably should prototype the joystick emulator adapter just to be sure I can make that work.

 

I wonder if anyone would ever want to play Indy 500 with a trackball (or the Thrust+ D.C). Technically, it can be done now, but the drastically higher resolution of the trackball makes it difficult to play Indy 500.

Edited December 13, 2015 by BigO

[ledzep]

If he's got the mechanical part figured out, the rest is relatively easy. The most complex part will be the interruption of existing circuit traces to inject the new signal (assuming that's his intent). Even that could be a fairly painless operation. I've been in there so much lately that I can just about visualize where he'd need to tap in without even looking.

 

That is my intent (if I understand you correctly). I'm making the assumption that the optical readers in the CX-53 function the same way as the one in Oscar Controls Vortex Spinner (it being a supposed direct replacement for an original Tempest arcade cab spinner) -

 

http://mirrors.arcadecontrols.com/OscarControls/vortex/

 

And it has a very straightforward circuit board (can't find a schematic for it online anywhere) -

 

 

So my naive view/expectation/hope is that the 4 wires coming out could be wired into the CX-53 board with no additional conversion or modification since I (again) assume that there is a point on the CX-53 PCB where the signals from the 2 optical wheels are the same type as the S1 and S2 lines coming out of the spinner. Does the CX-53 use/output 5v? If so then that and the GND are obvious, I figured the S1 and S2 pins could be wired into both the X and Y directions of the CX-53, I was thinking of using a switch to feed them to only one direction at a time (horizontal or vertical) though maybe that's not even necessary? I mean, any game that only wants one trak-ball axis (Tempest, Super Breakout) would never even look at the other, yes? Not sure if there are any vertical-only trak-ball compatible games but if there are then horizontal wouldn't mean anything at that point so the game wouldn't care about that.

 

So, somewhere on this -

 

 

there should be a place to solder/jump in the signals coming out of that Vortex header, right? Looks very confusing to me, though. I had thought of trying to recreate one side of that board to feed the Vortex into and then having that plug into the CX-53's harness but I wouldn't know how to do that.

 

Either way, what I hope to end up with is a 5200 spinner controller that acts/feels like a Tempest arcade spinner (spin it really fast, let it spin for a few seconds, stop it with the side of my hand or my fingers) and looks like the CX-53 but with a spinner knob poking through where the trak-ball would be. Nice an heavy, sitting solidly on a table, the best version of 5200 Tempest ever.

[BigO]

 

That is my intent (if I understand you correctly). I'm making the assumption that the optical readers in the CX-53 function the same way as the one in Oscar Controls Vortex Spinner (it being a supposed direct replacement for an original Tempest arcade cab spinner) -

 

http://mirrors.arcadecontrols.com/OscarControls/vortex/

 

And it has a very straightforward circuit board (can't find a schematic for it online anywhere) -

 

 

So my naive view/expectation/hope is that the 4 wires coming out could be wired into the CX-53 board with no additional conversion or modification since I (again) assume that there is a point on the CX-53 PCB where the signals from the 2 optical wheels are the same type as the S1 and S2 lines coming out of the spinner. Does the CX-53 use/output 5v? If so then that and the GND are obvious, I figured the S1 and S2 pins could be wired into both the X and Y directions of the CX-53, I was thinking of using a switch to feed them to only one direction at a time (horizontal or vertical) though maybe that's not even necessary? I mean, any game that only wants one trak-ball axis (Tempest, Super Breakout) would never even look at the other, yes? Not sure if there are any vertical-only trak-ball compatible games but if there are then horizontal wouldn't mean anything at that point so the game wouldn't care about that.

 

So, somewhere on this -

 

 

there should be a place to solder/jump in the signals coming out of that Vortex header, right? Looks very confusing to me, though. I had thought of trying to recreate one side of that board to feed the Vortex into and then having that plug into the CX-53's harness but I wouldn't know how to do that.

 

Either way, what I hope to end up with is a 5200 spinner controller that acts/feels like a Tempest arcade spinner (spin it really fast, let it spin for a few seconds, stop it with the side of my hand or my fingers) and looks like the CX-53 but with a spinner knob poking through where the trak-ball would be. Nice an heavy, sitting solidly on a table, the best version of 5200 Tempest ever.

 

Seems completely workable to me.

 

Assuming that the S1 & S2 connections are quadrature encoded signals, you could very likely replace the signals from the native sensor(s) with those from the Vortex.If that board you showed is all there is to it, the signals are probably not already buffered when they leave the board, so let's go with that assumption.

 

The trackball does work on 5v.

 

I don't think you'll have good results if you try to pile those new signals on top of the ones coming out of the existing sensors. I think you would need to interrupt the OEM signals and inject the new ones.

 

Myself, I'd consider using a wire wrap socket (with some pins cut) plugged into the original socket with the chip then plugged into the wire wrap socket to be able to interrupt and inject signals at the LM339 without hacking up the circuit board.

 

(Note that once you've plugged a wire wrap socket into the original socket, the original socket may no longer work reliably with a chip plugged in directly)

Edited December 14, 2015 by BigO

[BigO]

You could also just bend out the 4 input pins of the LM339 and solder the S1, S2 signals right onto the pins or desolder the original sensors and wire up S1 and S2 to those pads.

 

With two signals per axis, you have a 50/50 shot at getting them connected in the right order. If you get them backward, the controller will work backward. Not worth much research to assure getting them right on the first pass.

 

Shouldn't be a problem to drive both axes with one set of signals. The high input impedance of the op-amps in the LM339 should keep from loading the signals down too much.

[Andromeda Stardust]

Question about modding trackballs. If I use a USB trackball encoder such as this one:

http://groovygamegear.com/webstore/index.php?main_page=product_info&cPath=76_85&products_id=260

 

Can I just wire the PCB directly to the quadrature outputs of the photo diode, or do I need to run the signals through a logic buffer first? I'm thinking about converting my Sega sports trackball into a USB mouse since it appears to be mostly useless outside of the 3 SMS sorts titles that use it. Can I just remove the stock PCB and route the quadrature signal directly onto the encoder PCB, or do I need a quad comparator or equivalent to buffer the signal?

[BigO]

I can't find any real technical information about the Opti-Wiz.

 

But, based on the hardware they claim to support, and a little bit of searching for that hardware, such as the Taito Arkanoid Spinner, it looks to me like it would accept the unbuffered photodiode signals.

 

Probably best to email the manufacturer for confirmation. If it were mine, I'd hook it up directly and give it a try. I wouldn't expect to cause any damage doing that, but proceed at your own risk.

 

I did read something that said it accepts active low inputs. Depending on how the photosensors are built and what kind of signal the Opti-Wiz really expects, you may need pullup or pulldown resistors to supply the proper level when the diode is non-conducting.

 

As for the "active low" part from a logic standpoint or needing to do an inverting buffer or anything like that: if you find that it works backward it's just a matter of reversing the wiring on the affected axis. Quadrature encoding is, essentially, a 50% duty cycle signal and the initial level doesn't mean anything. Inverting the levels just changes the phase relationship of the signals with respect to decoding. Flipping them around makes the decoder logic recognize the opposite direction.

Edited December 14, 2015 by BigO

[Lynxpro]

Question about modding trackballs. If I use a USB trackball encoder such as this one:

http://groovygamegear.com/webstore/index.php?main_page=product_info&cPath=76_85&products_id=260

 

Can I just wire the PCB directly to the quadrature outputs of the photo diode, or do I need to run the signals through a logic buffer first? I'm thinking about converting my Sega sports trackball into a USB mouse since it appears to be mostly useless outside of the 3 SMS sorts titles that use it. Can I just remove the stock PCB and route the quadrature signal directly onto the encoder PCB, or do I need a quad comparator or equivalent to buffer the signal?

 

Have you checked to see whether the Sega homebrew hackers haven't modded any other SMS or Genesis titles to use the Sports Pad?

[ledzep]

You could also just bend out the 4 input pins of the LM339 and solder the S1, S2 signals right onto the pins or desolder the original sensors and wire up S1 and S2 to those pads.

 

With two signals per axis, you have a 50/50 shot at getting them connected in the right order. If you get them backward, the controller will work backward. Not worth much research to assure getting them right on the first pass.

 

Shouldn't be a problem to drive both axes with one set of signals. The high input impedance of the op-amps in the LM339 should keep from loading the signals down too much.

 

Thank you, I will have to look into that at some point. I was just wondering about those resistors on the Vortex PCB, if they were doing something that was duplicated on the CX-53 and therefore the feeds from the Vortex harness should be soldered onto the CX-53 PCB at some point downstream of the similar resistors (if they're on the CX-53 for the same reason).

 

I know this is in the 2600 category but I was wondering how feasible it would be to adapt the CX-53 PCB logic to the Vectrex. The Vectrex, like the 5200, uses an analog joystick. And it doesn't have a native or official trak-ball option. I would think that it might be possible to either create a PCB similar to the CX-53 board to fit a CX-22 or CX-80 and plug that into the Vectrex (and the Vectrex would read it like analog joystick inputs much like the Atari 5200 reads the CX-53, if I understand that process correctly) or to make an adapter for the CX-53 to plug it into the Vectrex (and possibly rewire the fire buttons so there are 4 unique ones instead 2 sets of the stock 2 buttons). That way there probably wouldn't even need to be new or reprogramming of existing games that take advantage of the analog joystick. And then, maybe, a Centipede type game for the Vectrex! Or at least a Missile Command that feels like the arcade version. And Quantum!

[BigO]

 

Thank you, I will have to look into that at some point. I was just wondering about those resistors on the Vortex PCB, if they were doing something that was duplicated on the CX-53 and therefore the feeds from the Vortex harness should be soldered onto the CX-53 PCB at some point downstream of the similar resistors (if they're on the CX-53 for the same reason).

 

I know this is in the 2600 category but I was wondering how feasible it would be to adapt the CX-53 PCB logic to the Vectrex. The Vectrex, like the 5200, uses an analog joystick. And it doesn't have a native or official trak-ball option. I would think that it might be possible to either create a PCB similar to the CX-53 board to fit a CX-22 or CX-80 and plug that into the Vectrex (and the Vectrex would read it like analog joystick inputs much like the Atari 5200 reads the CX-53, if I understand that process correctly) or to make an adapter for the CX-53 to plug it into the Vectrex (and possibly rewire the fire buttons so there are 4 unique ones instead 2 sets of the stock 2 buttons). That way there probably wouldn't even need to be new or reprogramming of existing games that take advantage of the analog joystick. And then, maybe, a Centipede type game for the Vectrex! Or at least a Missile Command that feels like the arcade version. And Quantum!

I assumed those resistors were acting as current limiters to drive the LEDs in the opto-interrupters.

 

Not sure about using the CX-53 on the Vectrex. The Vectrex controllers use a voltage divider to provide a specific voltage level representing the position of the controller. If the circuitry reading the controller is doing A/D conversion on that voltage, I don't think the Trak-Ball will give it a signal it can work with. The 5200 relies on the controller to be a current limiting device to control the rate at which a capacitor charges. As best I've determined, the CX-53 essentially creates a variable resistance by switching some resistors in and out of the circuit quickly according to pulses resulting from the spin speed and direction.

 

In the back of my mind, I recall the Vectrex voltage divider going from -5v to +5v. If that's the case, the CX-53 won't come close to working.

 

Anything I can think of with an unmodified CX-53 seems like it would have a slim possibility for success. At the very least, some sort of adapter would probably be needed.

Edited December 17, 2015 by BigO

[BigO]

As for making a Trak-Ball work with the Vectrex with whatever level of modification is required, I'm sure that could be done if you're willing to spend the time and money to do it. Conceptually, you could probably pick off existing signals from the circuitry and run them through an integrator or something like that to convert to a proportional voltage level that the Vectrex would understand, assuming that full analog control is the goal.

 

Splitting up the 4 buttons would require breaking traces on one of the circuit boards and tapping in some new connections. No big deal. A serious drawback here is the placement of the buttons on opposite sides of the ball. I think you'd have a hard time using all four of them without switching hands.

 

I don't know if there's enough consistency across games for this to make sense, but, rather than splitting the fire buttons a couple of the keypad buttons on the CX-53 could be tied in and read from the Vectrex for less frequent/critical use. The keypad buttons could be used for selecting game variation and other less taxing tasks. This scheme is problematic though. Mapping of the hardcore fire buttons would have to be variable as I assume there's no standard button usage across games.

Edited December 17, 2015 by BigO

[BigO]

When wiring, handmaking harnesses is truly for the tedium tolerant.

[Andromeda Stardust]

When wiring, handmaking harnesses is truly for the tedium tolerant.

This. I still haven't finished my "ultimate" 16-bit arcade joystick. Has three PCBs (clone Genny, clone SNES, and a Cthulhu) and supports 2600-SMS-Genesis, NES-SNES, Turbo16, and Mame-USB. It's already got a rats nest inside and I'm not even close to done yet...

[BigO]

Indeed, rat's nest avoidance is a significant part of the tedium.