Leaderboard

Many thanks to Eddie C for the supply of the EPROM PCB and ZIF socket! I have started to put it together (my eyes aren't what they were ) I am looking forward to hopefully being able to do some work with EPROMs...

Ordered the components kit, each board needs: 2 x 28pin wide sockets 1 x 4.7UF 25V tantalum bead 5mm 1 x 16 pin socket 1 x 20pin socket (if you want CHR chip) 1 x HD74LS138P IC Wire links Here are the boards assembled and tested with CHR less Super Line Up roms. After understanding the link settings, they all worked perfectly.

In the spirit of sharing, here is a copy of the Gerber files for the SRU program card. Cheers Ed JPM9980_EC.zip

Top man, thank you for sharing

Thanks for sharing Ed. Top man.

Brilliant stuff Ed and thanks for sharing.

In the spirit of sharing, here is the Gerber file for the MPU3 program cards. Cheers Ed MPU3 V2.zip

Visually, when you first met his machine, you think it's a Take Two clone. It is to a point. It's a reglass of the 4quid jackpot, and it plays a lot better. However.. I first saw this in Preston Station about 1989, and I just missed the Lime Street train. I had about 3-4 quid spare The lower features are thick and fast, 30p (why maygay..why 30p??..didn't make you look edgy, in the same way JPM's Nudge Double up MK1 wasn't edgy cutting bells and plums to 30p either). Getting past Number Sprint is a nightmare, as Stop and step is Jackpot if you're good enough. Wins Unlimited, to this day I have never collected, don't think anybody has So the Big Nudge pot..do you risk for a cash repeat or a jackpot, or take a lower win, and hope for the nudge pot repeat?..hint..it's rare, and I think it's broad as it is long. Anyway, down the last pound, gambled and lost wins unlimited, so stop and step. And I hit it blindly, 2 quid note, and it all hangs on up or down. Spawny git guesses right, and 3 on the line. Repeats twice. However, not understanding the machine, I was wondering why a single winline note triggered the repeat. Another Edgy feature, but I'll take the money. I had to collect, and put in a quid, hoping it would repeat a third time, but the Repeater Gods saw my greed, and took my pound as a token of goodwill, So, all in all about a pound up. But I still remember it. Statto

Nice work. Always worth double checking the resistors. I was missing 2 x 27k ohm resistors from the BOM on mine. You will be up and running in no time. Happy new year!

Thanks for sharing Ed. Could be my next project (just one more project - always room for one more). It's like you're a few steps ahead of anticipating what I want to do. Spooky and brilliant 👏

Unbelievably quick! And fantastic job Eddie, you will help keep these brilliant machines working 👌👌

Ordered 5 boards for the prototypes, arrived inside of 7 days.

Here is the layout of the new boards:

Aup.. Long time no speak.. I can't be bothered to try and look through the poor resources section on the forum.. Can anyone tell me if this hardware manual is there.. I dont need it. As i have it. I could up it in time Let me know please Thanks in advance

Hats off to @EddieC - this sort of stuff is painstaking work, thankfully we have the emulator from Chris (RIP man) - which allows for much faster testing of ROM hacks than if the changes had to be tested on real hardware each time I did get a few requests for this gamble block stuff, but I just did the CHR stuff for a bit of a diversion from other projects since it meant fruit machines could live again that otherwise would get broken up for parts... so I couldn't justify the time requirements for this gamble block type work. Like I said above, and I'm sure Eddie can fully attest to, it is absolutely and utterly painstaking, trying to find the relevant routines, then figure out the relevant patches - it's really low level hacking! (plus of course you might need to hack out checksums/chr along the way) - Great work Eddie, really good to see someone has taken this on (takes the pressure off me haha! Very few fruit machine rom hackers around these days)

Here are the changes that I made to the code after the Checksum removal. By using the MFME in debug mode I found that the following ram addresses: M008D = Triggered when Jackpot is achieved M0085 = Gamble pot tens in decimal M0086 = Gamble pot units in decimal M008E = Current gamble pot in HEX This is the area of code that sets the auto collect variables on the roulette gamble. i.e. #$3C = 60 which is £6, #$FA = 250 which is £25, #$32 = 50 which is £5. These are activated at different gamble levels. Simplified the code to raise the limit and then to use the ram address M0085 which is the gamble pot in tens, this will trigger a jackpot and auto collect when the tens address reads below 20 which equates to below £200. This ram address is normally set to FF and then when there is a £10 win its set to F1, £50 is F5 and then £100 is 10. When M0085 is less than 20, then the code triggers the Jump command to a new address #7ED1. I moved some of the code from #6A67 to make room for the JMP command. Next step is to force the limit of the win payout to no more than £100 to limit the payout to the capability of the cash tubes. There is a possibility for the gamble to be £90 x 2 so we need to reset the pot to £100. We also need to find space to insert the new code. Here is a place where there are unused ram addresses. The new code is inserted at #7ED1 where we load the value of 10 in to the accumulator A and the set it back to M0085 to set the tens back to 10. Then we load the value of 00 in to the accumulator A and the set it back to M0086 to set the units back to 00. Then to execute the commands that we moved from the previous ram address of #6A67 I attach a copy of the updated assembly code. In the zip file there are standard and updated £100 dumps. The code has been tested on the emulator and works fine, next step is to burn to eproms and try in a machine. Cabaret 100 Updated Roms.zip d16_No Limit_Modified.rtf

I understand what you mean, this is the complexity of changing the way that a machine was originally designed and it adds additional complications that would never be thought of. You are right in what you are saying, the gamble will allow multiples of the stake and of course the odds are somewhat related to the number of positions on the wheel vs the stake, red, Black, high numbers or low numbers around 50% chance each time. x3 and x4 are much less likely to come in. Overall this is pretty good return when you get to a decent pot value. With the multiples on the gamble you can collect at any time, to get the maximum it will only auto collect if the value of the pot is higher than £100, so could be £90 x2 if you were to win. In my video, I had to get to £210 for the machine to auto collect as the £105 was lower than the auto collect threshold of £110 . This makes total sense now why the limit on the gamble was so low with the early machines with no stake / bank function the gameplay. Would be simpler on later machines with the bank function. We could consider if the odds should be any different based on the size of the pot, however in roulette it's still red or back or house. The odds remain the same it's the pressure of the gamble and risk of loss that increases. Makes me smile on games like Autoroute where I gamble, 20p, 40p,80p, £1.60 then bank a little and then hit the final gamble to get to £3 total. As the programming needs to be completed at and assembler / machine code level and backwardly engineered, changes are extremely difficult. If we had the code in the original high level language these changes would be easily to implement, unfortunately these are unlikely to exist anymore. Would be interesting to understand how the conversion companies developed machines back in the day and if they had customised development kits with companies such as VFS, PCP etc.

Wonder if when it’s all sorted the value/demand of these machines will go up? as most don’t like them because of the tame nature of mpu3 clubbers.

This is closer ed but still not right .the programming needs to.know what can be gambled and what can't with the various odds on offer of x2/x3&x4 .so for instance it has to know the variables of a 20p cherry win upto the max allowed using x2/x3/x4 .I'm pretty certain initially it won't be a straight £100 probably in 90 quid zone or there abouts . For example 20p x4 =80p then 80px4 =£2:40 then £2:40x3 =£9:60 then £9:60x4=£28:80 then finally £28:80p x3=£86:40 If the £86:40 can be banked down to £14:60p and gambled once more x2 then the £100 is achieved. The jp tune should then ring out before payment .the machine will need to know the various options open for total gambling .and as mentioned the % needs to be working properly otherwise the the machine won't play properly. It may well be too generous. Carry on ed as you are closer than most have got with projects like this and hats off to you for unlocking the gamble as this has never been achieved before 👍👍👍👍👍👍

Good stuff Ed and thanks for taking the time to explain it as well. Very enjoyable seeing the results actually working in reel time. (see what I did there). I'll get me coat!