New Product: GP-Wiz - 32-input Fast USB Controls Interface

[RandyT]

Got a USB port?

Want to build a standalone panel with a ton of controls?  Want to make it hot-swappable?   Want to be able to add another unit later for even more controls?

Want to add another 32 inputs to your KeyWiz (or other encoder) installation?  Want to do it without it costing a fortune?


Introducing the lightning-fast GP-Wiz USB controls interface from www.GroovyGameGear.com


Features:

-  Super-Fast USB Gamepad functionality.  Reports the status of ALL 32 inputs 100 times a second!!!  Much faster than USB Keyboard technology.
-  No limitation on the number of simultaneous button presses and each input has equal priority.
-  Unique USB Device Names and ID numbers for multiple unit installations on the same computer.
-  Plug and Play HID device.  Tested with 98SE, Win2K/XP.  Will most likely work with other systems that comply with the the HID specification..
-  Completely bus powered.
-  Common GROUND wiring method.  Only one input wire and a shared ground line to each switch.
-  Small footprint.   Fits anywhere.
-  Includes 6' USB A to B cable, and printed instruction sheet.
-  Eco version available for $19.95  (requires soldering of input wires)
-  No-Solder version as shown, only $22.95
-  No-Solder MAX version (upper photo) has screw-terminals for easy installation, only $34.95

*Update*

Cables are here and now shipping with the units.


Thanks for looking,
RandyT

[crashwg]

Does it show up as a single gamepad with 2 axis and 28 buttons?

[RandyT]

Yes.  The 2 Axes are purely digital and you can use rest of the buttons any way you please, even as  stick controls.

Note:  Some gamepads are clunky in the implementation of their buttons and would never work as directional controls. 

The GP-Wiz has no such limitations!!!!   

RandyT

[Lilwolf]

Sounds REALLY cool!  CONGRATES!

Since someone will ask.  Are these faster then the ps2 version?

[Tiger-Heli]

Cool, thanks Randy!!!!

[SirPeale]

Neat!

You should make some variety of JAMMA adaptor to plug into JAMMA cabs.  Maybe even plug right into the socket.

[JoyMonkey]

Neat!

You should make some variety of JAMMA adaptor to plug into JAMMA cabs.  Maybe even plug right into the socket.

That'd be a good idea. All it would be is a JAMMA fingerboard with some screw-terminals for video, sound and power inputs, and an IDE cable for plugging into the Wiz.


RandyT; is there something up with your site, none of the images in the shopping section are displaying for me.

[Hoagie_one]

excellent....hmmmm

[tetsujin]

Hi!

Would you be willing to tell me a few technical details?  I'm curious about what controller the module uses.  Also I'm guessing it (like most USB HID) runs at "low speed".  (Though I don't yet know if there would be any advantage to full-speed for such a device)  Is that correct?

Do you know how the 10ms-or-less polling for USB HID compares to the performance of the PS/2 keyboard port?  You compare the unit's behavior to USB keyboards but not to the legacy types.  (Ah, I guess this was already asked..)  The best I could figure was that it'd take a little less than 1ms for a PS/2 keyboard to send a scancode (button pressed or released) but that it'd have to do one at a time - meaning it could match USB gamepad's performance as long as there are less than 10 state changes every 10ms.  (Not an unreasonable assumption, I think...)

Also, is the difference between the solder and no-solder versions just the presence or absence of the dual-row pin headers?

It looks like a great little unit.  It's very tempting to buy one of these rather than build my own.    Or maybe even if I do build one of my own.

(EDIT):  Hmmm....  Cypress microcontroller, maybe?  It has a 6MHz resonator rather than a crystal, so it must be low-speed...

[Kremmit]

Wow, USB from RandyT?  I thought I'd never see the day! 

Just messing with you, Randy, I'll probably buy a few of these down the road.

[Tiger-Heli]

That'd be a good idea. All it would be is a JAMMA fingerboard with some screw-terminals for video, sound and power inputs, and an IDE cable for plugging into the Wiz.

And depending on the pinout, the same adaptor would work with either the KW Eco or the GP Wiz.

[Tiger-Heli]

I expect (hope) RandyT will provide technical answers later, but I thought I'd say what I can now.

Would you be willing to tell me a few technical details?  I'm curious about what controller the module uses.

<sarcasm>Would you also like him to provide the firmware source code and make a schematic of the PCB etching?</sarcasm>  I don't expect an answer to this one.

Also, is the difference between the solder and no-solder versions just the presence or absence of the dual-row pin headers?

That's it.

[tetsujin]

I expect (hope) RandyT will provide technical answers later, but I thought I'd say what I can now.

Would you be willing to tell me a few technical details?  I'm curious about what controller the module uses.

<sarcasm>Would you also like him to provide the firmware source code and make a schematic of the PCB etching?</sarcasm>  I don't expect an answer to this one.

Oh, come on.  I could find out what controller he's using by buying one and reading the label under the sticker.  Maybe I could even make a good guess based on the placement of the capacitors and power and USB lines.  It's not going to be any great secret.  Nor is the circuit schematic, or the board layout.  Not to people with eyes.  If anyone wants to copy his product, they just need to see the original and have firmware to use - which isn't terribly complicated without ripping off his code.

And there's a big difference between having the designs for a product like this and actually manufacturing it at a reasonable price.  This board does not look like a home-brew one-sided etched board.  Not as far as I know, anyway.  I have no interest in entering his business, I'm just a hobbyist.  When I build my encoder using the PIC18F4550, it'll probably cost me as much to build it as it would to buy his.

It's a simple technical question.  If RandyT would rather not answer, I understand completely.  It's his business what he does with information about his product.  But let's let him make that decision, alright?

[jcroach]

Yes.

[Hoagie_one]

I'm a little weak on the digital axis thing too.  Is this just another keypress?

[Bloinkxp]

Digital = On/Off 

So the answer is yes.

[Tiger-Heli]

I'm a little weak on the digital axis thing too.  Is this just another keypress?

Digital axis would be like the D-pad on a game-pad - It would be identified through Windows as J1 Right, J1 Left, etc., but it is digital so it only sends pressed or released, not how far pressed, and wouldn't work with potentiometers.

AFAIK.

[Tiger-Heli]

Oh, come on.  I could find out what controller he's using by buying one and reading the label under the sticker.

Don't count on it.  (Just trying to save you money if that's your only intent.)

Maybe I could even make a good guess based on the placement of the capacitors and power and USB lines.  It's not going to be any great secret.  Nor is the circuit schematic, or the board layout.  Not to people with eyes.

Fair enough.

And there's a big difference between having the designs for a product like this and actually manufacturing it at a reasonable price.  This board does not look like a home-brew one-sided etched board.

Agreed.

When I build my encoder using the PIC18F4550, it'll probably cost me as much to build it as it would to buy his.

If not considerably more.

It's a simple technical question.  If RandyT would rather not answer, I understand completely.  It's his business what he does with information about his product.  But let's let him make that decision, alright?

Hey, calm down, I just said I didn't think he'd tell you that.  It's a free country, he can post whatever he pleases, regardless of what I say.

[tetsujin]

Oh, come on.  I could find out what controller he's using by buying one and reading the label under the sticker.

Don't count on it.  (Just trying to save you money if that's your only intent.)

That piece of information isn't worth the $20.  The product he's selling, however, is.

Mind your own business if you've nothing useful to contribute.

[Tiger-Heli]

Mind your own business if you've nothing useful to contribute.

I mainly was letting you know that the pin headers were the only difference in the two versions, and explaining what digital axes were.

[RandyT]

Neat!

You should make some variety of JAMMA adaptor to plug into JAMMA cabs. Maybe even plug right into the socket.

That'd be a good idea. All it would be is a JAMMA fingerboard with some screw-terminals for video, sound and power inputs, and an IDE cable for plugging into the Wiz.

That does sound like an interesting idea.  Not sure of the size of the market for that one though.  It would depend on what the final board ended up costing in smaller quantities.  Thanks for the suggestions, I'll do some research on it.

RandyT; is there something up with your site, none of the images in the shopping section are displaying for me.

Gahh!  It seems so.  I'm trying to get things moved to a faster server and it appears as though somthing broke (seemingly by itself) along the way.  I'm on it.  Thanks for the heads-up!

Would you be willing to tell me a few technical details?

Sorry, this is a commercial product, and as such, some of the details are proprietary.

Do you know how the 10ms-or-less polling for USB HID compares to the performance of the PS/2 keyboard port? You compare the unit's behavior to USB keyboards but not to the legacy types. (Ah, I guess this was already asked..)

Just as you suspected, "low-speed" USB devices can't touch PS/2 when it comes to latency, but things start getting a little closer when looking at total throughput.  It's very hard to do a direct comparison between USB and PS/2 keyboards because they can be apples and oranges, depending on the criteria you use to judge them.  I.e. in order to get the max throughput on a USB Keyboard, you would need to use every modifier key (Left and Right Alt, Shift, Control and GUI keys)  in your control layout and deviate from the normal single endpoint traditionally used for such devices.  But even that won't get you past the limit of simultaneous keypresses.

Likewise, the best throughput using PS/2 will be achieved by avoiding the Arrow keys and other "special" keys that use 5 total bytes for an "up/down" transition. instead of the normal 3.  This is why some encoders default to using the numpad keys for direction control instead of the arrows.

This difficulty in comparison extends to USB gamepads as well.  Every 10ms, the status of 32 controls (in the case of the GP-Wiz) is reported to the host.  PS/2 could not hope to send 32 keypresses in that time frame and certainly not that many releases!  However, long before that USB report is sent, the PS/2 port has kicked out the half-dozen or more keys that might have been of the most interest to you, with the rest following in a steady stream of FIFO data, instead of being limited to what the endpont buffers can hold and sending in spurts.   And, maybe you aren't interested in pure Game Controller data, so even though it's fast, it won't do everything you might want it to.

In practice, the GP-Wiz flies and it won't get slower with greater activity or with multiples.   My sixth sense tells me that the pipeline for game controller data at the OS level is slightly more direct than Keyboard data.  Call it a "feeling" after putting the unit through some hardcore testing

Also, is the difference between the solder and no-solder versions just the presence or absence of the dual-row pin headers?

As Tiger-Heli already stated, that is the only difference.


Just a quick note here for those considering "rolling their own":
Just because you have a reference design for one controller, that doesn't mean that a simple mod will allow you to do what you want in the end.  I found out the very hard way that if you expect a design to work cross platform, and you don't want to write your own drivers from scratch,  there is very little help for you out there.  2K/XP will accept things a certain way and 98SE will accept things a certain way.  It up to you to figure out what the common ground is, and it has nothing to do with backward compatibility.  I went through no less than 5 designs that would work on one system wonderfully and failed on the other, one to the point where it wouldn't even enumerate!  But I did learn one important thing:  The only thing standard about USB is the standard itself.  The protocol is so heavily reliant on the host that until all OS manufacturers are on the EXACT same page and there's been a few more years for everything to mesh properly, there will continue to be occasional issues.  But USB isn't alone in this regard as oddball PS/2 implementations still rear their ugly heads on rare occasion.  At least it's not as often as it did in the years past.

RandyT

[tetsujin]

Would you be willing to tell me a few technical details?

Sorry, this is a commercial product, and as such, some of the details are proprietary.

Aw...  Well, thanks anyway!  It looks like a great product.  It's exactly the kind of controller I want to have in my CP, and at a great price.  If I didn't now have an interest in building an encoder for the sake of learning how, I'd certainly use these.

Just a quick note here for those considering "rolling their own":
Just because you have a reference design for one controller, that doesn't mean that a simple mod will allow you to do what you want in the end.  I found out the very hard way that if you expect a design to work cross platform, and you don't want to write your own drivers from scratch,  there is very little help for you out there.

Interesting...

Fortunately, the only one whose needs I need to meet are my own.    But that's an interesting piece of information.  I'd love to experiment when I get mine built and find out more about what you're talking about.

(I didn't mean to trivialize what you've made, of course...)

[RandyT]

Wow, USB from RandyT?  I thought I'd never see the day! 

Just messing with you, Randy, I'll probably buy a few of these down the road.

Hehe.  You know I was waiting for a comment like this one from at least one of the folks who put up with the past threads 

But you did notice that it doesn't think it's a keyboard, didn't you 

RandyT

[Stingray]

Looks like a great product. I'm glad there's finally an USB encoder that we don't have to order from overseas.

Any update on when those translucent balltops are going to be available?

-S

[SirPeale]

Indeed!  This looks like a very nice product.

The one thing I would like to see is a competitior to the J-PAC.  Being able to just plug in a device and be playing in minutes, not hours, is great.

[Howard_Casto]

I love new hardware.... but i've got to be honest I don't know why anyone would choose to use this. 

First off, it's essentially a gamepad. The whole reason eveyone uses keyboard encoders is that not every emulator, pc game or application has gamepad support.  Mind you this isn't as big of a problem as it used to be by any means, but still. 

Secondly, you are billing it as a modular solution. How exactly? It has 32 inputs, but only two axis... so unless you want to use some of the buttons as 2 player axis (which imho is very hacky and you shouldn't have to do), it can't even handle 2 players, despite the enormous amount of button inputs.  It doesn't handle sub-panels well either, because of the cost. 20 bucks seems cheap until you realize that you have 5 panels.  I know some of you are saying.... "Well you can just make a harness and keep the pcb in the cp box, thus only switching out the controls."  Yes, yes you can.... you can also double wire your keywiz/ipac inputs to a custom harness and do the exact same thing, thus eliminating the need for this product.

While we are on the matter of cost...  I'll admit 20 bucks is a decent price for a gamepad with that many inputs, however, I can get a pc pro pad 8 usb joystick for 6 bucks at any local store.  It's easy to hack and although it only has 8 button inputs, I save not only 10 bucks on the initial cost, but also the cost of shipping.  I woldn't mention this, because a non-hacked solution is obvioulsy better for the slight cost increase except:

No analog inputs?  Why?  It's the one thing that nobody has... a non-hacked, cheap, analog interface solution.  As is the usb pcb that you've made doesn't offer any features worth using it over a keyboard encoder.  If anything it's limitations would put some people off from buying it. 

I want you to succeed, it's just that this doesn't seem like a good idea to me.  If it had analog inputs and some additional axis for the same price then I'd be all over it... but as is, it merely limits what you can do with your inputs, as opposed to a keyboard encoder, which potentially can offer just as many inputs, without the hinderance of requiring joystick support or usb support. 

[Howard_Casto]

the server burped 

[Howard_Casto]

*personal note* 


Pointing out that this product is built from a usb kit is just lame.   It doesn't matter.... the tools to properly program the eproms cost about 200 bucks (and I'm being conservative)  so unless you are running a business and already have the tools, this route is unaccessable to the average consumer. 

[Hoagie_one]

1 keywiz eco2 + 1 GPkeywiz = 4 player mame machine at 1/2 the price of an ipac4

Reason enough to me.

[Tiger-Heli]

Pointing out that this product is built from a usb kit is just lame.   It doesn't matter.... the tools to properly program the eproms cost about 200 bucks (and I'm being conservative)  so unless you are running a business and already have the tools, this route is unaccessable to the average consumer. 

Huh?  Who said anything about a USB kit besides HC?  I don't follow you.  More comments coming . . .

[Tiger-Heli]

This is just my take on the product, I did not contribute anything significant to it's development (other than mentioning something similar about a year ago and forgetting about it).

The whole reason eveyone uses keyboard encoders is that not every emulator, pc game or application has gamepad support.  Mind you this isn't as big of a problem as it used to be by any means, but still.

My feeling is that while this is workable for a primary MAME interface, it works better for an add-on panel. 

Secondly, you are billing it as a modular solution. How exactly? It has 32 inputs, but only two axis... so unless you want to use some of the buttons as 2 player axis (which imho is very hacky and you shouldn't have to do), it can't even handle 2 players, despite the enormous amount of button inputs.

Well, most mame games don't need an analog axis (there are exceptions).  Here is how I see this working.

I have a HotRod style controller with a KeyWiz in it now.  So the PS/2 port is used, but this is convenient for all emu, PC Games, MAME, etc.

I eventually would like to add a secondary panel with two trigger sticks in an Assault/Battle Zone configuration.  (This would only be used in MAME and as either a standalone panel, or an add-on to my current panel for the odd 4-player games.)

Theoretically, I could use a USB I-pac for the Assault add-on (but not an I-PAC VE as both it and the KW would lose settings on power down), and come up with a weird key assignment for the I-PAC (Like P1UP = O, etc) and program MAME so P1UP was O or Up (for 2 player games) and P3 Up was O for four-player games, but having the device set up as J1B1 would be a lot easier to deal with.

Even moreso if I want to add a second Assualt Panel for Sarge, Vindicators, because then I use the same wiring and the second Panel becomes P2B1 or P1B1 by DEFAULT depending on when I plug it in.

Actually, this is one advantage of this for using this as a standard interface - translating controls.

No analog inputs?  Why?  It's the one thing that nobody has... a non-hacked, cheap, analog interface solution.

I assume you are familar with Daveb's AKI and it didn't meet your criteria for "cheap", correct?

[tetsujin]

Secondly, you are billing it as a modular solution. How exactly? It has 32 inputs, but only two axis... so unless you want to use some of the buttons as 2 player axis (which imho is very hacky and you shouldn't have to do), it can't even handle 2 players, despite the enormous amount of button inputs.  It doesn't handle sub-panels well either, because of the cost. 20 bucks seems cheap until you realize that you have 5 panels.

No, that's actually still pretty cheap for a quality controller.  The functionality you get is worth it IMO.

I'd be curious to know more about how this is being envisioned as a modular system.  I have some ideas about the kind of approach I want to take but I'm curious about what Randy has in mind, since I'll surely learn something from it.

But basically, one approach goes something like this: since each GP-Wiz has its own unique identifier, the front-end software can identify what controls you've connected to the system, and automatically set up the game's input configuration accordingly.

Another approach would be something more like what Tiger-Heli explained (and Randy seemed to allude to) - using this module for add-on controls.  If the main controls are through the keyboard port, and secondary controls are attached as USB game devices, there are a few advantages.  First, since the add-on controls are not emulating keyboards, their mappings will never conflict with controls already installed in the system.  Second, the add-ons will be able to use USB without using the USB keyboard protocol.  (Which Randy seemed to indicate has some unfavorable technical limitations - I don't know first-hand the extent to which those limitations can be avoided.  Using USB means USB hubs are an option, rather than a strict daisy-chain of cooperating PS/2 devices.)  For me, being able to identify the (modular or otherwise swappable) controls through software is another big plus compared to PS/2.

[size=0](Sorry I was kind of a dick yesterday.)[/size]

[Tiger-Heli]

Hopefully RandyT will post his views as well, but . . .

I'd be curious to know more about how this is being envisioned as a modular system.  I have some ideas about the kind of approach I want to take but I'm curious about what Randy has in mind, since I'll surely learn something from it.

But basically, one approach goes something like this: since each GP-Wiz has its own unique identifier, the front-end software can identify what controls you've connected to the system, and automatically set up the game's input configuration accordingly.

True, but the unique identifier is mainly useful so that panel 2 remains panel 2 if you leave all your controls hooked up and turn the machine off and back on.  For a modular design, where you plug and unplug panels, there are some other interesting effects:

Here's another way to look at it, which I was getting at earlier. 
Let's say you have 6 desktop panels - each using one of these controllers -
The panels are the X-Arcade Solo layout - I.E. a single-player Street fighter layout.
Panels 1 and 2 are standard joystick panels.
Panels 3 and 4 are trigger sticks
Panels 5 and 6 are rotary joysticks (with USB mouse hacks (optical rotary, although thinking about it, there are enough inputs to support mechanical rotaries without Druin's interface).

Now here's where the beauty comes in -
I wire each panel so the Joystick is wired to the directionals.  I wire the buttons to button inputs 1 to 6 (assuming the trigger and thumb buttons share inputs 1 and 2),  Coin and Start are Buttons 7 and 8, and pause and escape are button 9 and 10, I still have 18 buttons left for the 12 inputs for each rotary joystick.

Now -
I can plug in the rotary panel for a quick game of 1Player Ikari - (Either panel 5 or 6 since they are wired identically.)  Then if a friend wants to join in, I can plug the second rotary panel in, and Windows will automatically set it up as Panel 2.

Any panel could be used as Panel 1 on up to Panel 6 (if six-player X-men gets fixed), or anywhere in between with no remapping or software, no adjustment to MAME (other than the initial setup), no complicated wiring changes, etc.  Coin buttons automatically correspond to the order that the panel was connected, etc.

[size=0](Sorry I was kind of a dick yesterday.)[/size]

[size=0]No problem.[/size]

[RandyT]

I love new hardware.... but i've got to be honest I don't know why anyone would choose to use this. 

How about these reasons:

It's the fastest USB solution with the highest number of inputs that can be actuated at the same time. 
They are cheap enough to use one for each  CP you decide to build, for Plug and Play operation.
They are small enough to put inside of a classic gaming controller to convert it to USB or make a box to convert "several" of them.

And I could keep going, so it's possible that you could be lacking somewhat in the imagination department.....

First off, it's essentially a gamepad. The whole reason eveyone uses keyboard encoders is that not every emulator, pc game or application has gamepad support.  Mind you this isn't as big of a problem as it used to be by any means, but still. 

Are you listening to yourself, Howard?  I think you just said, paraphrased, that "games don't use game controllers" 

Virtually any application that uses DirectInput, will support a Gamepad and allow you to change the button definitions to whatever you want.   I did a ton of research and could find only one that was a problem...and that one wouldn't even let you change key assignments.

A number of front-end apps (even your own  ) allow the use of a  game controller to navigate from within it.

Secondly, you are billing it as a modular solution. How exactly? It has 32 inputs, but only two axis... so unless you want to use some of the buttons as 2 player axis (which imho is very hacky and you shouldn't have to do), it can't even handle 2 players, despite the enormous amount of button inputs.

Howard, you aren't listening to yourself again.    You say that it's "hacky" to use some of the buttons as joystick directions, but then talk about  the KeyWiz and IPAC, which use the letters R, F, D and G on the keyboard for the second joystick, while the primary one defaults to the workable, but rather inefficient arrow keys.  Are you going to tell me that the GP-Wiz approach is somehow less elegant?

Your "it can't even handle 2 players" statement couldn't be more wrong.  It can handle as many players as there are inputs to connect joysticks to. 

What is of the utmost importance is that the buttons perform equally to the Axes, and in the case of the GP-Wiz, great care was taken to ensure this.  To be quite honest, I would have made ALL the inputs buttons, were it possible to do so and still have cross-platform compatibility using the HID spec.

  It doesn't handle sub-panels well either, because of the cost. 20 bucks seems cheap until you realize that you have 5 panels.  I know some of you are saying.... "Well you can just make a harness and keep the pcb in the cp box, thus only switching out the controls." 

First of all, it will handle sub-panels just fine, if you choose to use it for that purpose.  I don't know how many people have 5 sub-panels, but if they do, some "fancy dancin' " is going to be in order regardless of the solution chosen.

Yes, yes you can.... you can also double wire your keywiz/ipac inputs to a custom harness and do the exact same thing, thus eliminating the need for this product.

Howard, you used exactly the same "HC Brand" of logic when I introduced the KeyWiz (the other "product without a need" according to your opinion.)  But since then, I have sold at least as many KeyWiz's as Oscar has sold spinners, to very satisfied customers, I am proud to add.

While we are on the matter of cost...  I'll admit 20 bucks is a decent price for a gamepad with that many inputs, however, I can get a pc pro pad 8 usb joystick for 6 bucks at any local store.  It's easy to hack and although it only has 8 button inputs, I save not only 10 bucks on the initial cost, but also the cost of shipping.  I woldn't mention this, because a non-hacked solution is obvioulsy better for the slight cost increase except:

Yes, Howard, and we can all go down to the local Salvation Army store and buy a Keyboard for $2.00 and rip it apart for 8 buttons. But we both know (well, at least I do) that it's not the same as a KeyWiz or an IPAC.

The same applies to gamepads.  I researched a number of these units and they tended to be slow and the buttons do not always function as you would expect them to.  Some are matrixed, or not given the same priority as the Axes, and, and and.....

The GP-Wiz has none of these shortcomings, and to be honest, it's a little naive to think that a little blob of silicon on the circuit board of a $5 game pad will perform equal to a full-fledged microcontroller with custom, performance-enhanced microcode.

No analog inputs?  Why?  It's the one thing that nobody has... a non-hacked, cheap, analog interface solution.

Why?  for the same reason there are no analog inputs on a KeyWiz or Ipac.  It's not designed for that purpose.  A 4 or 8 or even a 49 way joystick and buttons (or dancepad, or whatever) use digital inputs, just like a game pad does.   And yes, I am considering a version for 49-way sticks.

BTW, there are analog joystick solutions out there, but cheap is relative and I don't know what your socio-economic situation is to quantify that statement.

As is the usb pcb that you've made doesn't offer any features worth using it over a keyboard encoder.  If anything it's limitations would put some people off from buying it.

Howard, if "some people" is "you" then I can live with that.  You've never had anything good to say to or about me or any of my products, nor have you purchased any.  I guess I shouldn't be surprised by your post, but I had vainly hoped.....

The first feature is that it is USB and not a keyboard encoder. As a gamepad, it can send twice the data in half the time with no limitations on simultaneous button/switch closures.

I switched out the KeyWiz in my cabinet with a GP-Wiz for testing purposes, and to be quite honest, I can't think of a compelling reason to go through the trouble of putting the KeyWiz back in.  So it's a matter of preference and what an individual's circumstances dictate is the best choice for him/her.  It's not about "limitations."

I want you to succeed, it's just that this doesn't seem like a good idea to me.  If it had analog inputs and some additional axis for the same price then I'd be all over it... but as is, it merely limits what you can do with your inputs, as opposed to a keyboard encoder, which potentially can offer just as many inputs, without the hinderance of requiring joystick support or usb support. 

Your opinion is noted, but please realize that it is your opinion and that you don't speak for everyone.  Your needs are not the needs of everyone, and your limitations are also not the limitations of everyone.  There are a number of very useful applications for the GP-Wiz, not the least of which, and has been pointed out several times, supplemental controls for upgrading the capabilities of a current keyboard encoder installation. 

And if you think daisy-chaining PS/2 devices is comparable, you really need to do some reading on the subject.

*personal note*
Pointing out that this product is built from a usb kit is just lame. It doesn't matter.... the tools to properly program the eproms cost about 200 bucks (and I'm being conservative) so unless you are running a business and already have the tools, this route is unaccessable to the average consumer.

*personal note*

"What the heck is Howard on about and who is this statement directed at?"

But just for fun, try $2000 for dev tools and there are no "USB kits."  You actually have to read books and work hard and stuff 

RandyT

[Hoagie_one]

But since then, I have sold at least as many KeyWiz's as Oscar has sold spinners, to very satisfied customers, I am proud to add.

I'm one

[Tiger-Heli]

They are small enough to put inside of a classic gaming controller to convert it to USB

Well, it's cheaper to buy one already converted from  http://www.sealiecomputing.com/retrozone/ , but the GP-Eco is undoubtedly a higher quality encoder, and the other arquments are valid.

Are you listening to yourself, Howard?  I think you just said, paraphrased, that "games don't use game controllers" 

Well, Howard does have somewhat of a point here.  A lot of the PC simulations that I use (not emulation or arcade related), use keyboard control only, but there's also a caveat - You can use Joy2Key to enable a gamepad (or GP-Eco) with keyboard based software much more easily than you can find software to enable a keyboard to work as a joystick (although there was a thread about that as well - dpadpro if memory serves - someone wanted it for EA virtual tennis).

I did a ton of research and could find only one that was a problem...and that one wouldn't even let you change key assignments.

Just a hunch, but I'm guessing you mean Stella, which I haven't used, but Z26 is an altenative, and CPViewer's author has a version of it which does allow you to change key assignments.  (but I think you'd need Joy2Key for 2 player usage, but maybe not if you used two GP-Eco's).

Your "it can't even handle 2 players" statement couldn't be more wrong.  It can handle as many players as there are inputs to connect joysticks to.

Actually - virtually unlimited b/c you can use multiple GP-Eco's.  I think there's a limit of 128 GP Ecos' per port (or maybe total) but that will be a busy arcade machine. 

I don't know how many people have 5 sub-panels, but if they do, some "fancy dancin' " is going to be in order regardless of the solution chosen.

I'm not sure I agree with this statement.  See my example above - if MAME is used and all panels use their own GP-Eco, I see very little "Fancy dancin'", other than I have to tell MAME that Coin1 is J1Button10, Coin2 is J2Button10, etc.  Am I not considering something???

Howard, you used exactly the same "HC Brand" of logic when I introduced the KeyWiz (the other "product without a need" according to your opinion.)

Yep, I remember that.

No analog inputs?

Why?  for the same reason there are no analog inputs on a KeyWiz or Ipac.  It's not designed for that purpose.  A 4 or 8 or even a 49 way joystick and buttons (or dancepad, or whatever) use digital inputs, just like a game pad does.

True, but MAME games like afterbuner need it, as will Nintendo 64 or Playstation or Dreamcast (I think) emulation.  I wouldn't bash the GP-Eco for not including it, though.  It won't support a spinner or trackball either, but that's not it's function.

And yes, I am considering a version for 49-way sticks.

Cool!!!

"What the heck is Howard on about and who is this statement directed at?"
But just for fun, try $2000 for dev tools and there are no "USB kits."  You actually have to read books and work hard and stuff 

I wondered and asked the same thing myself.  The first mention of USB Kit is Howard's post above???

[NoOne=NBA=]

But since then, I have sold at least as many KeyWiz's as Oscar has sold spinners, to very satisfied customers, I am proud to add.

I'm not one....yet.

The fact that you are finally offering a USB solution may change that though.

I've got a few cabs in the planning stage, and may try to work a few of these into the mix on them.

[tetsujin]

Secondly, you are billing it as a modular solution. How exactly? It has 32 inputs, but only two axis... so unless you want to use some of the buttons as 2 player axis (which imho is very hacky and you shouldn't have to do), it can't even handle 2 players, despite the enormous amount of button inputs.

What is of the utmost importance is that the buttons perform equally to the Axes, and in the case of the GP-Wiz, great care was taken to ensure this.  To be quite honest, I would have made ALL the inputs buttons, were it possible to do so and still have cross-platform compatibility using the HID spec.

You've said that you want to keep technical details to yourself, and given the amount of work that's apparently gone into this I totally respect that: but if you could elaborate a little on that or suggest resources to learn more about the performance issues of buttons vs. axes, I'd be very grateful.  I don't want to be pushy, but as someone who's starting out in USB development this is fascinating stuff.  If not, then I won't bug you for more details.

There are a number of very useful applications for the GP-Wiz, not the least of which, and has been pointed out several times, supplemental controls for upgrading the capabilities of a current keyboard encoder installation. 

Not to mention freaks like me who want USB gamepad controls just as a matter of principle. 

Pointing out that this product is built from a usb kit is just lame. It doesn't matter.... the tools to properly program the eproms cost about 200 bucks (and I'm being conservative) so unless you are running a business and already have the tools, this route is unaccessable to the average consumer.

"What the heck is Howard on about and who is this statement directed at?"

Me, maybe?

I didn't really talk about any kind of "USB Kit" but I said that building a USB Gamepad was a pretty basic USB microcontroller project.  I have a tendency to exaggerate the ease with which my projects come to fruition, especially when the project is an exciting one for me.  But for those who are interested in developing USB devices as a hobby project, you can get a good-quality PIC programmer for around $35 (mine was around $80 when I added the power supply, a ZIF socket, and boxed it), a free assembler or compiler, and build a gamepad circuit for around $15-$20.  As a hobbyist, for years I thought USB was inaccessible.  I saw the (easy-to-interface) legacy ports going away and the new USB ports as this hopelessly complicated thing that a hobby circuit could never hope to interface with.  Maybe that really was the case in 1998 or something, I don't know.  But with programmable USB microcontrollers these days, home-brew USB devices are neither terribly expensive or prohibitively difficult.  Having that capability is exciting.

Of course, it's not a trivial amount of work, either, especially (as I'm learning) if you want a product with high-end capabilities to truly be compatible with the different OSes.  I didn't mean to belittle the work that goes into a real production-quality device, either. 

[RandyT]

Well, it's cheaper to buy one already converted from  http://www.sealiecomputing.com/retrozone/ , but the GP-Eco is undoubtedly a higher quality encoder, and the other arquments are valid.

Cool link.  But I think making a box to support several of them would be a more efficient use for the GP-Wiz.  That was thrown out as something you could do iif you wanted to, but it would have to be one of the larger retro controls like the NES arcade stick (which I already used an Eco with for a friend)

Are you listening to yourself, Howard?  I think you just said, paraphrased, that "games don't use game controllers" 

Well, Howard does have somewhat of a point here.  A lot of the PC simulations that I use (not emulation or arcade related), use keyboard control only, but there's also a caveat - You can use Joy2Key to enable a gamepad (or GP-Eco) with keyboard based software much more easily than you can find software to enable a keyboard to work as a joystick (although there was a thread about that as well - dpadpro if memory serves - someone wanted it for EA virtual tennis).

That's a good point. I will take a look at the Joy2Key software to see how well it performs.  It could be a nifty way to get compatibility with the occasional non-arcade/console title where it might be an issue.

I don't know how many people have 5 sub-panels, but if they do, some "fancy dancin' " is going to be in order regardless of the solution chosen.

I'm not sure I agree with this statement.  See my example above - if MAME is used and all panels use their own GP-Eco, I see very little "Fancy dancin'", other than I have to tell MAME that Coin1 is J1Button10, Coin2 is J2Button10, etc.  Am I not considering something???

Heh.  I only meant that if you have 5 sub-panels, there's a lot of work involved figuring all that out regardless of the interface used.  But once you have it set the way you want it, no problems.

Why?  for the same reason there are no analog inputs on a KeyWiz or Ipac.  It's not designed for that purpose.  A 4 or 8 or even a 49 way joystick and buttons (or dancepad, or whatever) use digital inputs, just like a game pad does.

True, but MAME games like afterbuner need it, as will Nintendo 64 or Playstation or Dreamcast (I think) emulation.  I wouldn't bash the GP-Eco for not including it, though.  It won't support a spinner or trackball either, but that's not it's function.

Understandable, but these will most likely still work with digital controls.  And until someone starts selling an arcade quality analog stick for a price comparable to the digital ones the majority of us use now,  a conventional PC stick with it's own interface might be the best choice.

RandyT

[Bloinkxp]

I am just waiting for the site to go up!

Thanks RandyT.

-David

Proud Owner of a Keywiz STD.