Bubble Bobble Bootleg info

[Asure]

Will do tomorrow, need better lighting.. Btw, i already burned the 512's and game plays still, but no treasure rooms, or should there be errors?

[joaoljr]

Sorry, you mean the treasure room still looks the same as in the picture you posted before, i.e., "empty"?

Punji

[Asure]

Sorry, you mean the treasure room still looks the same as in the picture you posted before, i.e., "empty"?

Punji

Yes, still looks empty. (i expected as much)
Better pics tomorrow, i need some sleep now

[joaoljr]

Check whether pin 1 on each of the two sockets is connected to VCC or some other IC. Those are the key points.

You may need a multimeter to confirm it if any top side traces are connected to them, since the corresponding pads are clearly unconnected.

Also check pins 20 and 22.

Punji

[Asure]

I see you previously posted a hires panoramic view of the board top:

https://dl.dropboxusercontent.com/u/677 ... .52.40.jpg"

Please also post one for the bottom.

Punji

Here's a few for the bottom:
https://dl.dropboxusercontent.com/u/677 ... .51.40.jpg"
https://dl.dropboxusercontent.com/u/677 ... .51.52.jpg"
https://dl.dropboxusercontent.com/u/677 ... .51.58.jpg"
https://dl.dropboxusercontent.com/u/677 ... .52.20.jpg"
https://dl.dropboxusercontent.com/u/677 ... .52.33.jpg"
https://dl.dropboxusercontent.com/u/677 ... .52.47.jpg"

I hope they are good enought quality.. lighting is really bad today

[Asure]

I checked the sockets & pins for the 28-pins 27C512

Socket 1, pin 1 & 28 appear to be connected.

Pin 1-goes to pin 1 of the array near it.
Pin 1 & pin 28 are connected together.
Pin 20 goes to G4 10 (LS138)
Pin 22 goes to G4 4 (LS138)
pin 28 goes to VCC 5V rail..

Socket 2,

Pin 1-> ra2 (but via rail on the back.)
Pin 1 & pin 28 are connected together.
Pin 20 goes to g4 10 (LS138)
Pin 22 goes to g4 4 (LS138)
pin 28 goes to VCC 5V rail..

Pin 20 of the other four other sockets / roms are connected to pin 3 on the 74LS138 at position G4.

I also checked where pin 1 of the LS138 goes.. this goes to 6H? (LS374 marked H02) PIN 15. From there i'm lost, i can't visualise in my head what a flip flop does yet..

Comparing to what Simon's board had, my lucky guess would be that my board maybe needs only some cuts, instead of rewiring?

[joaoljr]

Comparing to what Simon's board had, my lucky guess would be that my board maybe needs only some cuts, instead of rewiring?

I think you will need to do some re-wring...

First of all, it seems the mysterious 8700 is a 128Kx8 mask ROM. I could not find a datasheet for it, but the pinout seems to be compatible with Sharp LH531000 / Toshiba TC531000 (see http://www.euro-arcade.de/files/romref.txt").
Also, the 74LS138 that you mentioned is the video ROM address decoder. It is currently set up to 32K page size.

Both 8700 chips have pin 20 (chip enable input) connected to 74LS138 address decoder pin 3 (address C input). That means they will get enabled for the first 128K of video ROM (4 pages). Note that they are not using the 74LS138 address decoder output at all.

Both EPROM sockets have pin 20 (chip selection input) connected to 74LS138 address decoder pin 10 (selection line 5 output). That means they will only get enabled for video ROM page 6. Also, these sockets have pin 1 connected to VCC, currently not allowing more than 32K addressing.

So, if you are to use a pair of 27C512 in those sockets to account for page 5, you need to:

1) Reroute socket pin 1 to allow 64K addressing.
2) Change the address decoder to 64K page size.

More details to come...

Punji

[Asure]

Thank you for your reply! Looking forward to more info, take your time

[Bytestorm]

Hi!
Wow, i sound like to thank you for making this possible time the bootlegs gets some love.
What's the progress of the patch? How close to the original is it?
From what I read here its some slowdown left that ain't in the original?

Best regards

[joaoljr]

rom what I read here its some slowdown left that ain't in the original?

Hi, that is correct. The only difference seems to be a somehow greater slowdown at certain levels (even the original slows down a bit there).
Unfortunately I can't improve things further, since the bootlegs need to use the main Z80 CPU for doing (part of) the work of the PS4 security CPU.

On the plus side, you get the option of adding level skip and game continue features. And if you replace one EPROM with a flash chip, you also get a hiscore save feature.

Punji

[Bytestorm]

Ok, I understand
You did a great job m8!

Wich version is the latest now?

Should I burn it, or du you have anything more for release soon?

[joaoljr]

Wich version is the latest now?

The latest version is pretty stable, I haven't done any changes in a while... you may download from this link:

https://ianc.net/temp/boblbobl/boblbobl ... -03-11.zip"

Please check the readme for more details.

Punji

[Asure]

Both EPROM sockets have pin 20 (chip selection input) connected to 74LS138 address decoder pin 10 (selection line 5 output). That means they will only get enabled for video ROM page 6. Also, these sockets have pin 1 connected to VCC, currently not allowing more than 32K addressing.

So, if you are to use a pair of 27C512 in those sockets to account for page 5, you need to:

1) Reroute socket pin 1 to allow 64K addressing.
2) Change the address decoder to 64K page size.

More details to come...

Punji

I've recently stocked up on IC's to shotgun/repair games with, but my lack of too much technical knowledge stops me from doing anything further with this Bubble Bobble.
I've thought up something, can we not run a LS138 (and maybe some other logic?) in parallel, and have that work when the game should select the 512's bank to display the end boss, and treasure rooms?

[joaoljr]

I've recently stocked up on IC's to shotgun/repair games with, but my lack of too much technical knowledge stops me from doing anything further with this Bubble Bobble.
I've thought up something, can we not run a LS138 (and maybe some other logic?) in parallel, and have that work when the game should select the 512's bank to display the end boss, and treasure rooms?

Sorry, another project of mine got all my attention lately... I will try to provide more details to you over the weekend.

You don't need to add any additional chip. The key thing here is to force the lower address line of the LS138 to VCC, so that it will end up selecting pages twice as large. You also need to remove VCC from pin 1 on both EPROM sockets and connect the correct address line there. I will edit your board pictures to show the exact places you need to cut the traces and add the jumpers.

Punji

[shinobi076]

I've same problem with graphic issue on secret room.
Here is my pcb:
http://benjamin.bouloch.free.fr/bb1.jpg"
http://benjamin.bouloch.free.fr/bb2.jpg"
http://benjamin.bouloch.free.fr/bb3.jpg"
http://benjamin.bouloch.free.fr/bb4.jpg"
Maybe you can help me Punji ???

[Asure]

Both EPROM sockets have pin 20 (chip selection input) connected to 74LS138 address decoder pin 10 (selection line 5 output). That means they will only get enabled for video ROM page 6. Also, these sockets have pin 1 connected to VCC, currently not allowing more than 32K addressing.

So, if you are to use a pair of 27C512 in those sockets to account for page 5, you need to:

1) Reroute socket pin 1 to allow 64K addressing.
2) Change the address decoder to 64K page size.

More details to come...

Punji

I did some deduction, if i am wrong bear with me, i am not an expert

27C512:
As these are socketed, i can just lift pin 1 out of the respective sockets. They are then no longer 'always logic high (1)'.
We will need something to pull them high, as the game needs them high for most of the time, right?
(They are now always high, so most of the gfx must be grabbed from the high part of these roms)

So we connect both pins1 of the two eproms to a pull-up resistor. You used CA15 on the other single board.
I assume this CA15 will go logic low (0) when the game reaches the bonus room? This is arranged by the 138?

74LS138:
No idea how this works but just thinking out loud here


Pin 1/2/3 is select a/b/c
Pin 1 was hardwired low always (assumption, as you connect it to VCC in the other board mod.)
Pin 2/3 are wired up properly, allowing 2x2x0 = 4 configurations (4x8=32?). If we had 2x2x2 we have 8(bit) = 64 right?
If we connect VCC there, it does something with the pullup?

Just thinking out loud here, as i really want to fix this pcb to have the treasure room
I'm guessing when i bend out put 1 from the socket, i can get away with less cutting and use wire wrap for a more clean solution right?

Any pointers are welcome, as i love to learn!

[Asure]

I did the mods as i described to myself, and it (almost) works.

- Gently used a side cutter to cut and then lift 74LS138 pin 1
- Connect this pin with short piece of wire wrap to VCC
- Lift & gently bend the two 27C512 pin 1 from the socket
- Wired up S2 pin 1 to the RA at pin 2 (ca15?)
- Wired up M2 (or N2) pin 1 to pin 1 of Q2 (i guess this is wrong, this is probably always VCC/high right?)

Edit: Wired M2 to N2 pin1 same error

Some pics:

Current treasure room:


Eprom mods

[Asure]

I'll just keep talking to myself here.

- I dumped & verified all my wiring. : OK
- Double checked both 512's are connected to the array pin 2: OK
- Out of ideas for now. PCB back in storage.

[Asure]

Finally, my pcb works with 100% correct treasure room and end boss working perfectly.
What i had to do? Swap around the roms in S2 and M2, then everything is perfect.
Fact: The board seems to run fine with the wrong roms in those sockets, and also with NO roms in those sockets. You can play much of the game perfectly fine with no roms there..

I've same problem with graphic issue on secret room.
Here is my pcb:
http://benjamin.bouloch.free.fr/bb1.jpg"
http://benjamin.bouloch.free.fr/bb2.jpg"
http://benjamin.bouloch.free.fr/bb3.jpg"
http://benjamin.bouloch.free.fr/bb4.jpg"
Maybe you can help me Punji ???

Shinobi076: We need to know if your room looks like mine (A) or like Punji (B)

Problem A:

Cause: If pair of 27C512 installed, these may have pin1 held high and 74LS138 wired wrong/unused.

Problem B:

Cause: 27C256 installed, replace with 27C512 and proper data (need to check w/ Punji still..)

Problem C:

Cause: Wired up correctly, but swap the two 27C512's you burned around

[shinobi076]

It's great that you've found a solution Asure !!
My glitch problem is the first one (Problem A)
I've 2 pcb boot (both the same) with all eproms 27C256. Is it a problem ? Eproms must be 27C512 ?