Jump to content
stonyat421

crystal mpu4 board with a jpm rom [ update ]

Recommended Posts

22 minutes ago, Wizard said:

I have found this in my stash, is it any use to you?

 

IMG_1302.JPG

that would be great mate 

pm you mate 

thanks 

Edited by stonyat421

Share this post


Link to post
Share on other sites

wrong price of play is what that alpha display is trying to tell you - it could be something as simple as the stake and prize key is not being seen at the switch lines.

i think the stake and prize key sits on the orange switch loom. Probably easy enough to work out what enable/returns its coming from.

 

Share this post


Link to post
Share on other sites

The extra ground lines might well be to do with trying to shut out the interference on the sound that every MPU4 , even the youngest ones out there suffer with!

 

Share this post


Link to post
Share on other sites

 

still no joy getting the elite roms to run but did get slight movement from the 4 main reels and gamble reel [tried the dips on the board same thing ] using the buccaneer settings from the manual

i did get the buccaneer card to run for about 10 mins then i got ER 2.5 i have changed a few dirty plugs in case of connection problems but still got the same thing

did notice the gamble reel was not spinning correct on start up so changed the plugs [still the same ]

Share this post


Link to post
Share on other sites
23 hours ago, MAGIK said:

wrong price of play is what that alpha display is trying to tell you - it could be something as simple as the stake and prize key is not being seen at the switch lines.

i think the stake and prize key sits on the orange switch loom. Probably easy enough to work out what enable/returns its coming from.

 

cheers mate will check that out :D

Share this post


Link to post
Share on other sites
On 2/1/2019 at 11:16 PM, stonyat421 said:

that would be great mate 

pm you mate 

thanks 

betya thats a 7795 sound generator IC on that board.  So the circuit for it will look very similar to the sound section on a scorpion 2.

Share this post


Link to post
Share on other sites
Posted (edited)
2 hours ago, MAGIK said:

betya thats a 7795 sound generator IC on that board.  So the circuit for it will look very similar to the sound section on a scorpion 2.

hi mate

finally got a boot up on the buccaneer crystal card with the buccaneer eproms after sorting out a couple of issues on the reels

machine turns it self of and then restarts again after about 5 mins does this a couple of times then goes dead till i switch it of and restart it again

i have swapped the psus and same thing

i tried the elite roms in 4 cards [ club bullion / club mistique / and the 2 original cards that came with the machine all are dead nothing at all

however when i stick the elite eproms in the buccaneer card i get a bit of movement from all reels for couple of seconds then just the alpha telling me club elite is it possible there is something missing from the the first cards that needs to go on the buccaneer card to boot elite i do no that the 2 ics near the sound eprom also work in the buccaneer card and bot the machine 

Edited by stonyat421

Share this post


Link to post
Share on other sites
9 hours ago, sulzerned said:

Getting  closer Tony. 

yeah mate cant stop now have to complete the mission :D

Share this post


Link to post
Share on other sites

does it have inverted optos? unplug the opto lead and try again.

 

Share this post


Link to post
Share on other sites

OK, from what I worked out there, its a problem with the sound decoding.  If I run it in MFME and then set the sound type away from crystal I get the exact problem you described to me in PM.

So what you could do is first check that the game rom has direct access to the card edge , and that its /OE line is controlled by the card edge and not something on the card. Also, you could try removing ICs 4 and 5 and the sound rom to disable the sound section. Can try removing the GAL22V10 also and see if its to do with sound decoding.

If memory serves I was working on a maygay M1 program card with a similar problem and it was the on board GAL for the sound sample decoding that was dead and causing the databus to be flodded with 1s all the time.

Yours is a slightly different problem , it seems to me like the game is polling the 7795 to initialise it and not getting a response back. That happens with global games on scorpion 1 when they try to play sound.  I did notice in the emulator you do play a small sound sample on game boot so Im wondering if the game is waiting for the sound_playing signal (Busy line ) to clear when it starts the sound and trys to play that sample.

 

Share this post


Link to post
Share on other sites
6 minutes ago, MAGIK said:

OK, from what I worked out there, its a problem with the sound decoding.  If I run it in MFME and then set the sound type away from crystal I get the exact problem you described to me in PM.

So what you could do is first check that the game rom has direct access to the card edge , and that its /OE line is controlled by the card edge and not something on the card. Also, you could try removing ICs 4 and 5 and the sound rom to disable the sound section. Can try removing the GAL22V10 also and see if its to do with sound decoding.

If memory serves I was working on a maygay M1 program card with a similar problem and it was the on board GAL for the sound sample decoding that was dead and causing the databus to be flodded with 1s all the time.

Yours is a slightly different problem , it seems to me like the game is polling the 7795 to initialise it and not getting a response back. That happens with global games on scorpion 1 when they try to play sound.  I did notice in the emulator you do play a small sound sample on game boot so Im wondering if the game is waiting for the sound_playing signal (Busy line ) to clear when it starts the sound and trys to play that sample.

 

ok mate will give this a shot tommorrow thanks for your time il let you no how it goes :D

Share this post


Link to post
Share on other sites
Posted (edited)

well after a few hours on pm last night with magic trying to get elite to run 

we still had no joy as some of you will no i have been trying to get this to run as it was a non starter from the off

after 6 weeks of taking the machine apart and going through it like a fine tooth comb trying to iron out any hiccups  and replacing many bits in the machine 

at this stage the machine is running on club buccaneer/ club bullion bars [ same card ] and club mystique [ different card ]  but sadly not club elite YET

i have done new elite game and sound roms just in case there is a problem with the others i have checked these on the emulator  and run fine

MAGIC IS ALSO LOOKING INTO THIS and if any one has any more imput please fill free to add your comment

 

 

Edited by stonyat421

Share this post


Link to post
Share on other sites

So at the minute we can get club elite to run with the wrong sound if I read what stony is saying properly.  But we cant get it to run with its own sound, no matter what we do. So I think we have a high address line problem and its just hanging when we try and read the address of a particular sound sample that it is trying to play. There has to be a logic problem on the card, either to do with the 74LS latch thats on the upd7759s data lines , or with the ROM connections itself.

I can get a hold of an MPU4 board for a few days to experiment,  if someone has a crystal ROM card mabye I can borrow or buy it from them so I can have a play with some live hardware into getting it running??

Share this post


Link to post
Share on other sites
3 hours ago, Andrew96_ said:

Make sure the links on the card are set right for the correct eprom size

all the eproms are the same mate 64kb for the game and and 512kb for the sound

i did try the eproms in a different card club mystique as it has 1 different link to the bucanner card it is possible that the 2 cards that was with the machine could not be the originals but that i have no idea  

Share this post


Link to post
Share on other sites
3 hours ago, MAGIK said:

So at the minute we can get club elite to run with the wrong sound if I read what stony is saying properly.  But we cant get it to run with its own sound, no matter what we do. So I think we have a high address line problem and its just hanging when we try and read the address of a particular sound sample that it is trying to play. There has to be a logic problem on the card, either to do with the 74LS latch thats on the upd7759s data lines , or with the ROM connections itself.

I can get a hold of an MPU4 board for a few days to experiment,  if someone has a crystal ROM card mabye I can borrow or buy it from them so I can have a play with some live hardware into getting it running??

club elite does not run mate full stop 

at this stage the machine is running on club buccaneer/ club bullion bars [ same card ] and club mystique [ different card ]  but sadly not club elite YET

Share this post


Link to post
Share on other sites
10 minutes ago, stonyat421 said:

all the eproms are the same mate 64kb for the game and and 512kb for the sound

i did try the eproms in a different card club mystique as it has 1 different link to the bucanner card it is possible that the 2 cards that was with the machine could not be the originals but that i have no idea  

Ok, I am looking at this thread on my phone, so bear with me if you have already covered this...

where did the roms come from that don’t work? Did you program them yourselves? And if so do the files fit the roms or are they for bigger capacity roms?

Share this post


Link to post
Share on other sites
29 minutes ago, Andrew96_ said:

Ok, I am looking at this thread on my phone, so bear with me if you have already covered this...

where did the roms come from that don’t work? Did you program them yourselves? And if so do the files fit the roms or are they for bigger capacity roms?

the roms   came with the machine mate

1st set origanals stickers / 2nd set not to sure think are copys with machine

/3rd set from on fmu mecca

i have tried all 3 sets :D and any thing else i can think of 

all 3 sets also run in the emulator   on crystal club bullion 

Edited by stonyat421

Share this post


Link to post
Share on other sites
2 minutes ago, Andrew96_ said:

ah ok, there are many quirks to mpu4!

yeah this is a strange one for sure :D

Share this post


Link to post
Share on other sites

this is what im getting with elite 4/2 origanal game and sound 4/2 and also elite 1.6 roms from the mecca

when i try the other elite rom set i get cash in metre error on the alpha

 

Edited by stonyat421

Share this post


Link to post
Share on other sites

Ok,  I have been having a think about this, first off lets just post this piccy here so we can see what Im gonna be banging on about..

kl.JPG.40e7f0c6e7bab3a900c3a9091eed820b.JPG

and lets go bold and italic so I can see what I am typing ,

Ok , righty thats better lol.  Right,  To understand whats going on here you have to have a good look at the card.  From what I can see , its a very typical JPM design in my opinion.

If you ignore the bottom half of it for the minute and concentrate on the sound section , its design resembles an impact sampled sound card.  We have the sound rom, typically a 27c040 or 27c4001 ,  an NEC7759 speech and audio synthesiser LSI chip, and some handling logic for its data and address buses.  Down towards the left is the sound amplifier, probably a TDA2030 stereo amplifier IC, with two decoupling capacitors on its outputs for left and right channels.  Its capable of 30w output but in a fruity we will get no where near that. Probably in this application we are driving around 5 watts at 8 ohm impedance.

The first thing I see thats of significance here is T1.. They are using that transistor to protect the ROM chip select line, just as they do with the 6264 RAM on an impact CPU. It prevents the select line being driven if there is a low or missing power rail on the card . That helps prevent something called "latch up" where a logic IC is being driven without being supplied power , something that can do quite a bit of damage . I have, and so will many of you, have seen RAM go bad on boards because the 5v supply was missing. Its fatal to older stuff like 4116s etc which have multiple power rails.   The way it typically works is the collector is tied to 5v , the base is tied to the ROM supply rail via a resistor and the emmiter is tied to the MPU4s ROM select line. It functions as a circuit breaker,  if the 5v supply drops down to a level set by the resistor, the transistor doesnt conduct and there is no way to select the ROM.#

Now if that transistor was bad, the ROM might as well be unplugged from the board because its either never gonna transmit its data, or it will transmit its data all the time, niether event is what we want!

Notice also the MAXIM 1232 IC on board.  The voltage monitor.  Thats what will be supplying the base of T1 . We know all about that chip from its use in MPS - and we know the hassle it can cause.

https://datasheets.maximintegrated.com/en/ds/MAX1232.pdf

Its probably being used in two ways here :- first, to reset the sound CPU or put it back to reset if there is a power problem . And second, it likely has control of T1 to switch the ROM select line on and off.  So if that IC is bad, and we know it can go bad,  the sound CPU is never coming out of reset.  And more often or not with this design, it can cause the rest of a game to hang because it is waiting for a response from the sound CPU (7759 is effectively a sound CPU, dont shoot me its easier to say!!)

Next we can focus on the sound CPU and its driving logic.  The circuit is based on a 7474 (used for some of the timing and probably the highest address line ) and a 74HCT373, which sits on the ROM address bus and its used to set the upper address .  Now this is important because the 7759 has a combined address and data bus.

It works like this -  the sound 7759 has a sample select data bus of one byte . This is directly connected to the MPUs data bus coming from the 6809.   The lower address  (the 7759 can address 17 bits ) is then put on A0-A8. The upper address is then send to DA0-DA7 and latched to address bits A9-A16 on the ROM by passing it to the 373 .  Now, heres the trick. Once the address for the ROM is set,  AD0-AD7 are then used to read the data coming back from the ROM and get the sound sample information. DA0-DA7 become inputs for the data, instead of outputting the address.  This is where the 74HC373 can screw things up. Say the chip has a shorted output . No matter what you try and send it , a bit is stuck. That could cause all sorts of funnyness. You could play the wrong sample, or send the ROM to a location where there is no valid sample.  Or what if its inputs were shorted- you could lock up the 7759s data and address buses altogether.   You can tell when you have a bad address select because the sounds have artifacts in them , or samples go missing in the weeds.

Now, the 7759 typically deals with a 27c010. But we are using a 040 or 4001 here.  So there will be extra high address bits that the CPU can set before it asks the sound CPU to play a sound. This is like having a CD multichanger and asking it to play track 4 say, of guns and roses, then hitting disk skip and play track 4 of now thats what I call music.  Its the same track number , but its being played from a different catalouge. The 7759 will play 255 samples, but by controlling the highest address bits yourself,  you can have different sets of 255 samples, so long as each catalouge of samples is designed to work with the chip.  A good example of this is cops and robbers.  Also sets of samples can be combined together to make longer samples.  If you remember my cops and robbers repair I had a situation where a pink panther sound rom worked , but the cops and robbers one didnt. Its not because the sound was designed for different games by different makers. Both techs used the same NEC chip as here to come up with the sound. It was because address line A17 was broken and the CPU couldnt select the highest "page" of samples.  And the game would hang up until it had finished trying to play the sample, leading to really long real spins or crashes on gambles, features etc.

So that 74HC74 sitting near the sound cpu will have a hand in controlling the highest address bits, and probably some of the ICs timing as well.

X1 underneath the chip, and the two small caps provides a clock for the sound CPU - around 640khz. If the oscillator is bad , say X1 is bad or one of the caps is dead, the sound simply doesnt work at all.  It doesnt hang the rest of the game tho because none of the handshake lines telling the MPU the sound is playing are set.

 

So thats the sound section, now for the bottom half.

We have the game ROM. and two PALCE20V8H (one might be a 16v8, cant see clearly).  That lot is easy to work out. IC10 there, closest to the ROM , deals with the data encryption. If you look at a crystal MPU4 ROM in hex editor and compare it to what you see in hex viewer in MFME, it looks like total nonsense.  The reset vectors make no sense, there doesnt appear to be any valid opcodes, and if you try and run it you just crash.  its because the data in the ROM is scrambled.  To use it , it has to be decoded through IC10. The ROM address is set and then the data bits are passed to one side of the PAL. The PAL then outputs the correct data for the MPU4 to read.  So if IC10 is bad or missing, or from the wrong card ( I have looked at a few cards and Im sure the encryption is not the same for every card ) the games not going to run.  It is possible to decrypt the ROM yourself, and then burn an "unprotected " ROM for a standard MPU4 card to run , but would take time to do, and your sound wouldnt work anyway because the MPU4 soundcard has the wrong type of data. I suppose you could bypass out IC10 by directly connecting the data lines from the ROM to the data lines on the card edge.  Youd need to find out the method of the encryption first, proper code breaking excercise!

IC8 , on the other side, will be handling wether the game ROM or sound section gets selected.  Most game cards with onboard sound have this sort of traffic warden to stop the MPU mixing the game and sound ROMs up.  If this IC is bad youll get a crash because the data bus will be transmitting or receiving the wrong information. What you could try tho, is removing this IC out of the board - see , I have had maygay game cards get screwed up because the control PAL for the sound is broken and screwing the databus up. Take it out and the sound doesnt work but you can read the game rom. 

A lot to think about there, but your video gives us some good clues as to whats going on -

The ROM boots -  So the encryption is working as the MPU4 can correctly start from the reset vector and read opcodes.

The sound section seems to work - at least for the Bucaneer card, because you can play that game.

On elite , the game hangs on start up , but AFTER CPU boot - so what does that tell us ? well heres a clue . When you play these games in MFME, they actually play a small sound sample on reset . However, your card gets stuck before it gets that far. And it goes one worse than that, the game execution crashes out and you start getting weird things happen. Now that could be that IC8 is bad, it could be the 74hct373 is bad,  it could also be the MAX1232 is dropping reset after a while and crashing you out. But let me throw in a curve ball. See that terminal stake up top near the sound ROM. Its a ground pin . THAT CARD MUST BE GROUNDED. Crystal games are very sensitive to ground problems for some reason. if you dont have a good ground on that pin , and the MPU4 doesnt have a good ground, you might as well sit there with a piezo firelighter cause you will get all sorts of fun and games with the anti spark lol. Had it with a spectre machine a while back, scorpion 1 but its expansion card needed a PERFECT ground or it crashed!

Im on standby to help. I just sold my elvis machine here and got something as part of the deal....

 

20190316_130350.thumb.jpg.fe31d6e39a2da8f88476ef4249aa1b18.jpg

Share this post


Link to post
Share on other sites

If you fancied sending me one of those crystal cards, Im sure I can have some results for you - or get yourself a multimeter ,and a logic probe (I can send you one if you want ) and lets get some troubleshooting done!

Hope that all helps! let me know what you want to do.

 

Share this post


Link to post
Share on other sites
5 hours ago, MAGIK said:

If you fancied sending me one of those crystal cards, Im sure I can have some results for you - or get yourself a multimeter ,and a logic probe (I can send you one if you want ) and lets get some troubleshooting done!

Hope that all helps! let me know what you want to do.

 

pm mate:D what you are saying above sounds about right as the machine does seem to have a mind of its own :headache:

Share this post


Link to post
Share on other sites

×