and thats why you don't solder your roms to your damn boards...

I must not have payed attention when I tested it before because I was playing a two player game today and the count sounded wrong, then I tested it again to make sure. The EBD 'bonus bug' is still in the code on my Altek-equipped EBD. Previously earned racks on player 2/4 are still scoring 63K instead of the correct 56K (one extra bonus of 7000.)

Strange, I have an EBD with an Altek board (that I installed a LONG time ago) and do not have this bug. Though I definitely did have the bug with my original Bally board.

For a regular tilt (plumb-bob)? Or a slam tilt? System 3-7 games, when they slam tilt, it does a full reset of the CPU akin to a power cycle. It's harsh but it's intentional.

Mines only about 2 years old. The first time I tested it and believed the bug was gone, I must have forgotten the specifics of the bug and only tested the first full rack scoring at the end of the same ball with which it was completed (when it does score the rack correctly for 56K), not after subsequent ball drains when that rack it is then scored from storage via the '56K' lamp... and scores wrong as 63K.

Yeah, I'm familiar with the bug...and I most definitely used to have it when I was running the original MPU. But I have not seen it since putting in the Altek (probably about 10 years ago).

Can you take a look at your Altek board and see what the version # is on the rom chip? Mine is 9.1. Thanks

If one of you is using the free play setting that could also make a difference

I tried my Altek with the free play turned on and off, no effect on the pesky bonus bug, nothing I tried did. Now that I know the bug is still there I may end up just swapping the Altek into my Lightning, then put the Oliver roms into the Lightning mpu, I already have the Oliver roms from the EBD's previous old kaput mpu. Plus I'd get back the Oliver DELUXE feature back. Or I'll get a new Dash board for the EBD with the Oliver roms and keep the old Lightning board, I could use an extra working one for testing purposes anyway.

I will look for the version this evening...I do have free play enabled.

What is the Oliver DELUXE feature?

Here are some pics of my Altek board inside my EBD-LE.

It shows a version for both the ROM and the CPU (?).

This is the ROM that came with it, and it hasn't been messed with since I installed it.

20210413_214917 (resized).jpg

20210413_214929 (resized).jpg

Yeah that's a much older one than mine for sure, discrete resistors/caps instead of SMDs, some different chips, different layout and still uses a battery pack. Actually I'm kind of surprised 'new' boards would have the EBD bug fix or any other bug fix. I'd think they would need to stick with the licensed code for each game. For buyers too, like "hey I want the original code, not some altered version." Though they did add the free play option which works via one of their own added DIP switches but I don't know if that means they had to actually make changes to each game's code or it functions some other way(?)

... Now I see the Weebly board does have both original and some custom game codes selectable on it including the Oliver EBD version.

What post are you referring to? What is "they" in this case?

Are you saying that they have dual copies of all the machines codes in the same ROM? And that selecting the "freeplay" setting switches between them?
Seems like it would be hard to fit all that in a single 8 meg ROM chip.

If that's the case, then maybe the bug is only fixed in the "free play" half, and not fixed in the "coin play" half.

Get those DARN Duracell batteries off your board!!!!!
They are the worst in batteries.
If it rots like hell, it's Duracell....

I would have thought Altek may have learned from the past Bally boards (or Williams).

The free play DIP on your early Altek appears to be 'on', I guess you could try turning it off and see if the bug comes back. Had no effect on the EBD bug in my newer one either way though.

Also the BH bug where if a ball's first score is other than via one of the upper three rollovers (or the rollover switch hit does not register), it usually goes off track on the next ball and the rest of the game it thinks it's got multiple players and the wrong displays are lighting up etc.

Yeah mines a v2, I didn't even know the bug existed until I had the glass off one day for testing some stuff and noticed when I manually scored things with the ball still sitting in the shooter, it would go bat crazy on the next ball.

Meteor, endless multiplier (well, 255x) countdown if you drain "all rockets lit" with many events happening at the same time (usually the spinner). Fixed by the factory, available on ipdb as "meteora" Can you supply link? I have had it keep scoring endlessly.

Beat the Clock, Eight Ball Champ, and probably the couple other games with same display setup as those games, the restore factory settings in test mode does not work properly. It will leave some audits with bad values that stop the game from taking credits or starting. There is zero error messages and confuses the heck out of people thinking there is a hardware problem.

One of the manuals even says that the the 6803 MPU can detect bad RAM at power up and do an automatic restore factory settings. That does not happen, you have to do it manually.

Would have been nice for Bally to do some RAM checks at power on. These games cannot handle when audits are set to a value out of bounds what is possible. Usually results in attract mode, but no coin up, no game play, even if in free mode. ie if the current credit value is higher than the max credits value then you get no game start, no coin up, and no warnings.

I noticed Future Spa attract mode displays are a bit inconsistent to the pattern. Sometimes there is blank time inserted between showing the high score to date and the last game played score and sometimes no blank time. Occasionally the blank time is a bit longer than usual. Not quite a bug, just something odd that caught my attention. It is running the Oliver 7d capable version which could matter.

I found a good way to wreck the NVRAM chip contents too in at least some games. If you manage the push the audit clear button after the last audit, and before the game reboots itself. Like if you are speeding through the audits clearing them all out and run past the final audit. The software crashes and next boot the audits are wrecked. Not really a bug either, just annoying quirk.

My Playboy did the same thing every time about 30 seconds after power-up.
It is kind of a bug. The display attract sequence is controlled by a zero-crossing timer. The display interrupt on occasions causes the attract sequence to skip a check and it gets temporarily stuck on the previous attract display instead of moving to the next.
It can be fixed.

Yikes!
So this is how it affects game play.
Say you've set score thresholds to award an extra ball and for argument sake a threshold is set at 240,000 points.
You start a new ball with a score of 239,990 points. You plunge the ball and it goes over the very top rollover for 10 points as the first score.
Your score rolls to 240,000 but you don't get awarded your extra ball.
This is what RAM $4C is used for.

Here's another related bug but it's not due to that instruction.
Imagine the previous scenario, but you got an extra ball and are served your new ball with the shoot again lamp lit.
You plunge the ball and it goes over the very top rollover for 10 points as the first score.
Your score rolls to 240,000 but you are then awarded 60,000 points extra because the shoot again lamp was already lit at the start of the ball (and stays lit as it should) but the 60,000 points award shouldn't have happened.
I suspect this one might happen with other Ballys of that era.

Good thing that the byte in $12FC was not $DD, the infamous "halt and catch fire".

what happens in with the free play mod rom?

So what would have been the typical way to handle it when credits are maxed - 1) When Special is lit, award nothing, 2) when Special is supposed to be lit, default it to lighting the 50K instead, or 3) just don't light the Special? I don't even know what the other Specials award at max credits, never tried it on mine.

I kinda think it's not a 'bug' - they did it that way on purpose. I mean, if I was at max credits, and played real hard to get the special lit.. I'd want SOMETHING awarded.

Yeah if it's isolated to that particular Special it's more of a "why did they do it here but not there?". The left lane Special can alternate with 50K too.

OH, Ohohoh.. I missed that. Correct - I'd expect it for all_ specials. So yeah, ignore my comment.

In a Bally 6803 games like Special Force it is possible to get corrupt RAM that breaks audit mode you cannot fix within the game's software.

With the bad corrupt ram, every time I go into book keeping data it locks up on game percent audit and you cannot advance or back up from that point. Keypad just makes ricochet noises but nothing happens. The issue is clear book keeping is past the problem game percent audit and restore factory settings does not seem to clear book keeping. If you have NVRAM or a battery hard soldered to the PCB you are stuck at this point. You have to remove the battery or clear the NVRAM with an eprom burner or different game and hope it writes to that corrupt spot.

If I thought about it, I would have dumped the problem NVRAM chip before wiping it with my eprom burner.

Considering how finicky these 6803 games are with bad adjustments/audits they should have done something like WMS with a RAM checksum.

I've had a slightly similar issue with Bally-35 games, happened to me for the third time just yesterday on a Speakeasy (first two times a few years ago on an Eight Ball Deluxe and a Centaur). The games lock up with all blank displays when reaching the first award score threshold setting in book-keeping when there is some value in non-volatile RAM that's corrupt. Disconnecting and reconnecting the battery fixed it on all three occasions.

I've seen audits with digits skipped in the middle (value higher than 9?) when the battery is off a MPU or the NVRAM has not been formatted. Makes sense the software crash on situations like that if it is a replay value.

should zero it out or something preferably

Sometimes in a Bally -35 game some adjustment that has range 00-03 (e.g. the sound adjustments) does not operate correctly even if the displayed value is OK. To fix, I keep the start button pressed to let the value go from 03 back to 00 twice and then it seems to work.

Usually this happens with an uninitialized CMOS RAM after battery or 5101 change.

Maybe, even if the setting uses only 2 lower bits, it only clears the upper bits when going from 03 to 00, and when testing the value it includes the random upper bits.