Progressive Gaming / Neurizon LED sign

[boffin]

Haven't had too much chance to start hacking this yet, but laid my hands on this interesting LED sign.
Full RGB, and really bright. 

Inside
40A 5V PSU
Main CPU board
LED driver boards (x3)
Interface board between CPU board and the LED driver boards

However, I have no instructions, none of the custom software to program it.  Anyone with a bright idea, or that might have come across one of these please speak up

It would make a great teardown for Dave, but would cost me a small fortune to send to him.

(see attachments for photos)

[viperidae]

With all that ram and flash it's probably running Linux. There might be a serial console port out header somewhere. It would be 115200 8n1 rs232

[SL4P]

There is no way I'd be junking or tearing that apart just yet.
The name Neurizon Pty Ltd says it's an Australian design, and being only ~12 years old - probably has a good chance of being brought back to life.

As indicated as a 'gaming' product, also suggests it could be related to casino style gaming - with lots of blinkenlights.  AU has a well developed gaming technology industry.

The display boards are apparently very predictable, with per-matrix shift-decoder/drivers.
The controller is very 'advanced' for a lowly display controller - so there's a lot to explore!
Keep on hacking!

[SL4P]

Some references
NEURIZON PTY LTD (Suite 4, 23 Overlord Place Acacia Ridge, QLD 4110, AU)
Johnson, Steven Brian (Suite 4, 23 Overlord Place Acacia Ridge, QLD 4110, AU)

 45 Jijaws Street (Unit 4) QLD, 4074, AU

http://patents.justia.com/assignee/neurizon-pty-ltd

An interesting legal tussle - the casino group copied the gaming technology developed by Neurizon, and lost in court, but appealed and it dragged on.  There must be some remnants of the hardware/controller IP floating about.

The unit you have - appears to be one node of a distributed gaming system - so it likely has some anti-tamper technology implemented.

If you can't get an early answers on the whole device - it's probably worth trying to get schematics from the rubble... maybe a port of their proprietary Linux if they'll allow it.  Sorry no gaming code!

[CJay]

Well it's got ethernet and RS232 so I'd probably attach it to a network and look for traffic with Wireshark to see if it's broadcasting anything, ARP, DHCP etc.

[darrellg]

That's a casino progressive jackpot display. I would ask about it at http://forums.delphiforums.com/slottech.

[boffin]

Connected up a serial port tonight -- nothing.  Appears to be normal 9 pin serial, TX on 3  RX=2.   Made a crossover, and even jumped CTS/RTS and DTR/DSR/CD for good measure.  Nothing.  The ethernet port isn't asking for DHCP, I'll have to drag out the Linux laptop and see if I can sniff anything that's coming out of it.

I've left a couple messages here and there,  we'll see what happens.

I did have a look at the patent stuff, looks like both those companies are on crack. The patent implies that progressive gaming invented the progressive concept, something which has been around for ages in Vegas (Megabucks anyone?)

[stj]

it's a "topper" display for clustered slot machines with a progressive jackpot.

Progressive Gaming went bankrupt and the assets got transfered to IGT if that helps.

[Maxlor]

If you can't get that embedded ARM board to talk to you, it should be feasible to just replace it with something else, like a beaglebone. Those matrix displays are probably still sold today (ok, maybe not those exact ones, but ones with the same electric layout,) finding some similar ones online and trying their communication protocol will probably work.

[amyk]

If you can't get that embedded ARM board to talk to you, it should be feasible to just replace it with something else, like a beaglebone. Those matrix displays are probably still sold today (ok, maybe not those exact ones, but ones with the same electric layout,) finding some similar ones online and trying their communication protocol will probably work.

It's not ARM, it's either PowerPC or 68K judging by the Freescale CPU (can't quite read the part number, but it begins with MPC...) Dump the flash and put it in a disassembler, that might give you some clues as to what it does. It could be a standard reference design too.

The LED panels won't be very difficult to RE, they're usually just a bunch of shift registers.

[stj]

it *may* communicate using CCTALK protocol.

[CJay]

I reckon it'll be running a variant of Linux with Busybox like a lot of the hotel IPTV boxes which use the PowerPC chips (Amino springs to mind as I recall the pile of 20 or so in the shed), it's also highly likely there's a console serial port (probably TTL levels though) which may or may not be populated.


My money is on the unpopulated six pin header pads.

[dnwheeler]

It may also be worthwhile to ask around the newlifegames.net forums. There are a lot of slot-machine and progressive system experts there.

--Doug

[@rt]

Hi Guys
Old thread, but I thought it appropriate since I have a display from the same company. Mine is different though.



boffin, I take it you never had any luck finding software then?
It looks like a bunch of addressable chips to control each 8x8 module, so worst case, with a lot of work,
I suppose the controller could be replaced with something else if one wanted to do a new controller PCB and new software.

[boffin]

No such luck finding anything.

[@rt]

No such luck finding anything.

Bummer! I was looking to toss the controller and talk to the display panel.

Mine is only red & green, and the drivers are Texas Instruments TLC5921, for which datasheets are available,
but what looks like the show stopper for me, is even after the controller board, they still have their PGA stuff with it’s own crystal clock
on the LED display panel, which looks to be a multi layer board with no direct interface to the actual shift registers.

 

[@rt]

Ok, just thinking aloud here, and would like to know if I’m not making sense.

AFAIK, These are still just serial shift registers, and the display has to be actively driven through them.
They are sixteen bit each, and control the rows and columns of each 8x8 LED module.
Ideally, for six rows of these 8x8 LED modules, ten modules each row, you’d have to rotate out six channels of 160 bits each,
and do that 64 times (one for each pixel in an 8x8 LED module) to make one video frame.

It occurs to me these chips are on the board to somehow simplify the connection from the LED array to the controller PCB.
It’s still a 50 way IDC connector :O
I’m just hoping if they were gone, that the pins at the end of the IDC connector and the shift register chains
would then all be open so I could connect those together with mod wire, and still have a somewhat neat board.

[stj]

do you even know how to connect the display to a gaming machine yet?
because that's step1 - once you know which interface you can start to get somewhere.

[@rt]

I’m not interested in the gaming machine. I’ve already tossed the main controller board, and the audio board.
I’m interested in the LED array and it’s shift registers, or the way the controller board interfaced to the LED array board.

[stj]

then your a fool, because now you have nothing to analyse.
good luck without the controller.

[@rt]

I’m not normally the grammar police, but please, learn grammar before calling someone a fool.
How about analysing datasheet of the display drivers?
The controller had to talk to the LED array PCB, so it’s possible for another controller to talk to the LED array PCB.
The display drivers are made to ... wait for it.... drive the 8x8 LED modules.

I don’t think the OP even wanted to connect it back to a game machine. Just display something arbitrary with it.
You could quite possibly find out all there is to know about the end product, and still find that it can only display dollar values, which would be useless.
I’ve determined so far that one pin on the 50 way connector is a single clock input for the entire LED array board.

[stj]

as someone who spends all day reverse engineering stuff, i'll explain.

first you find which port is active - just because you see many does not mean the software is using all of them.
a sevice manual for the gaming machine may show you the port in use.

once you know that,
the firmware of the gaming machine can be obtained and reverse engineered or emulated to capture the comm's to the display unit - then you have all you need.........

simple - till you throw things away and start trying to over-complicate things.

also, gaming machines often use standard protocols like MDB and CCTALK that are open and well documented.

[@rt]

You could not, for example, write Tetris for it.... or if you could, it would be taking the difficult road.
This is an electronics forum. I’ve written Tetris for an LED display made from scratch.
It isn’t unreasonable that I see a bunch of LED modules and drivers where you see a product.

[stj]

it may be bitmapped to several pages of ram - or atleast it was.

[@rt]

Anything that is going to drive a display matrix should buffer to RAM first.

I’m confident now the extra chips just buffer, and synchronously distribute the signals that are common to every driver on the board.
These include all of the the serial clock, blanking, and latch pins.
I have sent random bits in serial with a 1kHz clock signal, and they can be seen live, or frozen depending on the state of latch.
There’s an LED on the back of the LED array board that lights whenever it's seeing an input clock signal.

To send 160 bits parallel on all six serial channels x 8 LED rows in each module would take 1280 serial clock cycles to update the entire display once if I’m not mistaken,
and double that if you want to mix red and green to produce orange, or to alternate any colour to produce hues of colour combinations.