search

SukkoPera / MegaDrivePlusPlus

Universal Region mod, 50/60 Hz switch and In-Game-Reset (IGR) for Sega Mega Drive (AKA Genesis)
GNU General Public License v3.0
122 stars 29 forks source link

Using an ATtiny88-based Nano clone? #29

Closed rfancella closed 8 months ago

rfancella commented 1 year ago

Hello All! I have a bunch of MH-ET Tiny88 boards and would like to be able to use it in a Japanese VA4 Model 1 Mega Drive. The board does not have a labeled TX and RX port. Comparing the two pinouts, the RX pin on the Nano is also D0/PD0. On the MH-ET board D0/PD0 is on Pin0 and is also labeled LED. I wired up the board as described above using the Pin0 as "Pad Port Pin 9". Everything works but the joypad. The IGR and region switch does nothing. The reset button, language, and video change as they should and the RGB LED changes color respectively.

Has anyone been able to successfully used one of these clone boards before? If so, where does the Pad Port Pin 9 need to be connected on the board?

Here is a pinout of the MH-ET board: MH-ATTiny88

SukkoPera commented 1 year ago

RX is PD0 TX is PD1

The rest of the pins seem to use standard Arduino names, so you should be able to make the correct connections.

If the MCU used on the board is an ATmega328p (or similar), it will work.

DieSkaarj commented 1 year ago

PD0 (Pin 0) is also connected to the onboard activity LED (on the Nano that light is on Pin 13, I believe.) I was getting spooky action when I was playing about. It might be an idea to shift the START/C (pin 9) over to PD1 (Pin 1) for these boards.

SukkoPera commented 1 year ago

The led shouldn't be a problem, but it is surely an odd choice to put it on pin 0 when practically all Arduino boards have it on pin 13.

If it is indeed a problem, I'd just unsolder the led from the board.

DieSkaarj commented 1 year ago

I did end up trying that, but it continued to cause ghosting on hw interrupt, hah. The simpler solution was just to move the connection up one on the port.

Here's a reference to the differentials of the board compared to Arduino: https://github.com/SpenceKonde/ATTinyCore/issues/435

SukkoPera commented 1 year ago

I have just realized that this board uses the ATtiny88, which is a new microcontroller whose existence I wasn't even aware of.

At a first glance it seems compatible enough with the ATmega328p to make this thing work, but be aware that this is NOT a clone Nano, but rather a different and unsupported platform, use at your own risk.

By the way, the Tiny88 does not have an hardware UART, that is why the board has no RX/TX pins. Its pins also have different output driving capabilities, which could explain why the onboard LED was moved to pin 0 (pins 0...3 are "high sink").

DieSkaarj commented 1 year ago

@SukkoPera, I don't have the capacity to check this out but if we change the bitmask on the buttons assignment would that put START/C on pin 1? A definition for Tiny88 could patch this in.

From:

word buttons = (b1 & 0x38) | ((b1 & 0x01) << 2)
                 | (b2 & 0xC0) | ((b2 & 0x08) >> 2) | (b2 & 0x01)
                 | ((b3 & 0xF0) << 4);

To:

#ifdef ATTiny88

word buttons = (b1 & 0x38) | ((b1 & 0x02) << 2)
                 | (b2 & 0xC0) | ((b2 & 0x08) >> 2) | (b2 & 0x02)
                 | ((b3 & 0xF0) << 4);

#endif

This should eliminate the ghosting from the LED.

SukkoPera commented 1 year ago

I think so, but I cannot test it and don't want to support it without studying the microcontroller and board first (which I don't really have time for, ATM). Whoever wants to use this board is on their own.

DieSkaarj commented 1 year ago

@rfancella, without any assurances you can try this layout:

                              +-------+                
               +--------------|  USB  |--------------+ 
               |              +-------+              | 
               | [ ] 15                        14[ ] | 
               | [ ] 16                    SCK 13[ ] | 
               | [ ] A6                   MISO 12[ ] | 
               | [ ] A7                   MOSI 11[*] | [LED Blue]
 Reset Out     | [*] A0                     SS 10[*]~| [LED Green]
 Reset In      | [*] A1                         9[*]~| [LED Red]
 JP3/4 (50/60) | [*] A2                         8[*] | [Single Pin LED]
 JP1/2 (JP/EN) | [*] A3                         7[*] | Pad Up (1)
 LCD SDA       | [*] A4/SDA                     6[*] | Pad Down (2)
 LCD SCL       | [*] A5/SCL                     5[*] | Pad Left (3)
               | [ ] 25                         4[*] | Pad Right (4)
               | [ ] RST                   INT1/3[*] | Pad A/B (6)
(Built in LED) | [ ] 0                         5V[*] | +5V
Pad Start/C (9)| [*] 1      VCC  MOSI GND     GND[*] | GND
Pad Select  (7)| [*] 2/INT0 [ ]  [ ]  [ ]     VIN[ ] | 
               |            [ ]  [ ]  [ ]            | 
               |            MISO SCK  RST            | 
               +--ATtiny88---------------------------+

And, in the file MegaDrivePlusPlus.ino, change lines 709-712 to:

word buttons = (b1 & 0x38) | ((b1 & 0x02) << 2)
                 | (b2 & 0xC0) | ((b2 & 0x08) >> 2) | (b2 & 0x02)
                 | ((b3 & 0xF0) << 4)
                     ;

The MegaDrivePlusPlus should work as intended. However another thing to note is that there is likely a 5 second delay on boot so you may have to upload a new bootloader to the device. (I am looking into this for my own project.)

With respect to SukkoPera and MegaDrivePlusPlus, I am targetting this board with my Xiyoubu mod. It is in alpha, but is why I have ASCII art at hand ^.^

rfancella commented 1 year ago

Thank you so much for the time you have already put into this. :D

I think so, but I cannot test it and don't want to support it without studying the microcontroller and board first (which I don't really have time for, ATM). Whoever wants to use this board is on their own.

rfancella commented 1 year ago

I am on it. I'll give it a go and report back! I dropped in a real Nano just to prove my install is good. It works as expected, but it was my last one. LOL. Please do let me know about that bootloader. I am using this board in a couple other projects and with one, the bootup time is killing me. LOL

@rfancella, without any assurances you can try this layout:

                              +-------+                
               +--------------|  USB  |--------------+ 
               |              +-------+              | 
               | [ ] 15                        14[ ] | 
               | [ ] 16                    SCK 13[ ] | 
               | [ ] A6                   MISO 12[ ] | 
               | [ ] A7                   MOSI 11[*] | [LED Blue]
 Reset Out     | [*] A0                     SS 10[*]~| [LED Green]
 Reset In      | [*] A1                         9[*]~| [LED Red]
 JP3/4 (50/60) | [*] A2                         8[*] | [Single Pin LED]
 JP1/2 (JP/EN) | [*] A3                         7[*] | Pad Up (1)
 LCD SDA       | [*] A4/SDA                     6[*] | Pad Down (2)
 LCD SCL       | [*] A5/SCL                     5[*] | Pad Left (3)
               | [ ] 25                         4[*] | Pad Right (4)
               | [ ] RST                   INT1/3[*] | Pad A/B (6)
(Built in LED) | [ ] 0                         5V[*] | +5V
Pad Start/C (9)| [*] 1      VCC  MOSI GND     GND[*] | GND
Pad Select  (7)| [*] 2/INT0 [ ]  [ ]  [ ]     VIN[ ] | 
               |            [ ]  [ ]  [ ]            | 
               |            MISO SCK  RST            | 
               +--ATtiny88---------------------------+

And, in the file MegaDrivePlusPlus.ino, change lines 709-712 to:

word buttons = (b1 & 0x38) | ((b1 & 0x02) << 2)
               | (b2 & 0xC0) | ((b2 & 0x08) >> 2) | (b2 & 0x02)
               | ((b3 & 0xF0) << 4)
                     ;

The MegaDrivePlusPlus should work as intended. However another thing to note is that there is likely a 5 second delay on boot so you may have to upload a new bootloader to the device. (I am looking into this for my own project.)

With respect to SukkoPera and MegaDrivePlusPlus, I am targetting this board with my Xiyoubu mod. It is in alpha, but is why I have ASCII art at hand ^.^

DieSkaarj commented 1 year ago

Here is a problem-solved that might help with the bootloader: https://forum.arduino.cc/t/burn-bootloader-attiny88-cant-be-programmed-by-usb-solved/1033045

Please, @rfancella, really do let me know how you get on with it, hah!

rfancella commented 1 year ago

@DieSkaarj Ok, I have some feedback to start. I hooked up everything except the pad (pins 1-7) and replaced the code, it works with the reset button. I can change regions and a short press will reset the system. When I connect the wires to the pads as described in your diagram, the games starts up but the Tiny never boots. It seem to be stuck at the bootloader. What can I do to help diagnose this?

@rfancella, without any assurances you can try this layout:

                              +-------+                
               +--------------|  USB  |--------------+ 
               |              +-------+              | 
               | [ ] 15                        14[ ] | 
               | [ ] 16                    SCK 13[ ] | 
               | [ ] A6                   MISO 12[ ] | 
               | [ ] A7                   MOSI 11[*] | [LED Blue]
 Reset Out     | [*] A0                     SS 10[*]~| [LED Green]
 Reset In      | [*] A1                         9[*]~| [LED Red]
 JP3/4 (50/60) | [*] A2                         8[*] | [Single Pin LED]
 JP1/2 (JP/EN) | [*] A3                         7[*] | Pad Up (1)
 LCD SDA       | [*] A4/SDA                     6[*] | Pad Down (2)
 LCD SCL       | [*] A5/SCL                     5[*] | Pad Left (3)
               | [ ] 25                         4[*] | Pad Right (4)
               | [ ] RST                   INT1/3[*] | Pad A/B (6)
(Built in LED) | [ ] 0                         5V[*] | +5V
Pad Start/C (9)| [*] 1      VCC  MOSI GND     GND[*] | GND
Pad Select  (7)| [*] 2/INT0 [ ]  [ ]  [ ]     VIN[ ] | 
               |            [ ]  [ ]  [ ]            | 
               |            MISO SCK  RST            | 
               +--ATtiny88---------------------------+

And, in the file MegaDrivePlusPlus.ino, change lines 709-712 to:

word buttons = (b1 & 0x38) | ((b1 & 0x02) << 2)
               | (b2 & 0xC0) | ((b2 & 0x08) >> 2) | (b2 & 0x02)
               | ((b3 & 0xF0) << 4)
                     ;

The MegaDrivePlusPlus should work as intended. However another thing to note is that there is likely a 5 second delay on boot so you may have to upload a new bootloader to the device. (I am looking into this for my own project.)

With respect to SukkoPera and MegaDrivePlusPlus, I am targetting this board with my Xiyoubu mod. It is in alpha, but is why I have ASCII art at hand ^.^

DieSkaarj commented 1 year ago

@rfancella, if you have a multimeter check for a short. And, if you can, take a picture of the connections and post it here. I'll see if there's anything I can recognise. The wiring in the diagram is a direct port of SukkoPeras Arduino layout, except for the start button, which has been shifted to pin 1.

rfancella commented 1 year ago

@DieSkaarj Ok, no shorts across any pins on the controller port: controller

Top of the board: board_top

Bottom: board_back

Side: board_side_1

The only changes in code are here: code_1

and:

code_2

DieSkaarj commented 1 year ago

Lookin' good to me, too, @rfancella. The next step is to disconnect everything and start introducing things back individually. However, if you're confident the issue is with the controller try just removing those lines first and then reintroduce them, one by one, back to the circuit. I probably should have asked before; did you remove the controller lines and try the device again and then put them back on, and does the device still power up through USB?

Also, are the reset lines attached? The bottom legs need to be disengaged from the motherboard. It could be that sending the system into a loop.

See: https://github.com/SukkoPera/MegaDrivePlusPlus/wiki/images/nextria.jpg nextria

rfancella commented 1 year ago

@DieSkaarj I started with not having the controller lines attached. At that point the reset button worked to change regions/language. Also the RGB LED also changed and saved modes appropriately. Once I soldered in the controller lines, it just flashes the light on the Tiny like it's in bootloader mode. If I disconnect the controller lines, it works again. So, I will start removing the controller lines one at a time and test. ;)

I decided to use the "cut traces" method on mine, the one I am testing with. So the reset button is out of the circuit. Genny

Also, this board and wire setup originally had a Nano on it and it worked 100%.

rfancella commented 1 year ago

@DieSkaarj It is for sure pin 2. I remove pin 2 and it boots just fine.

DieSkaarj commented 1 year ago

Pin 2 was connected to pad pin 7 on the motherboard, @rfancella?

And, did you try turning the machine on with, and without, the controller attached? (not that it should change the behaviour but just wanting to make sure there wasn't anything funny going on there.)

tbh, I'm stumped. Pin 2 should connect to the SELECT signal from the controller and it looks fine from the pictures you've posted. You can test the line, but it's probably fine.

The image for the other bit of code you changed doesn't show up, can you post it again?

Now, we know it won't get the controller to work but try writing the original firmware (with your LED changes) to see if it boots and then again with the hotfix I suggested.

rfancella commented 1 year ago

@DieSkaarj Yes, Pin 2 is connected to pad pin 7. The controller attached or not makes no change in the outcome. :( I re-uploaded the picture of the code into the post above. I'll try the original code and report back. :)

Also, I'll try another ATTiny as well.

Pin 2 was connected to pad pin 7 on the motherboard, @rfancella?

And, did you try turning the machine on with, and without, the controller attached? (not that it should change the behaviour but just wanting to make sure there wasn't anything funny going on there.)

tbh, I'm stumped. Pin 2 should connect to the SELECT signal from the controller and it looks fine from the pictures you've posted. You can test the line, but it's probably fine.

The image for the other bit of code you changed doesn't show up, can you post it again?

Now, we know it won't get the controller to work but try writing the original firmware (with your LED changes) to see if it boots and then again with the hotfix I suggested.

SukkoPera commented 1 year ago

Just my own 2 cents: you can get a "real" Arduino Nano clone for ~4€ shipped from China, which would make things much easier :).

rfancella commented 1 year ago

@DieSkaarj Here is what I have done to the bootloader. I setup an Uno as an ISP Programmer. Burned a new bootloader, well no boot loader that is. And then I programmed "Blink" to it. It boots right up and starts the sketch. Works 100%. But it no longer has a "bootloader" and must be programmed with an ISP every time. I burned the Micronucleus bootloader back to it and there is a 5 second boot up before it starts running the sketch. But you can now program it with a USB cable again. I tested it with "Blink" and one of the projects I am also working on and they are both good without a bootloader installed on the Tiny. I do have to program them with Arduin as an ISP every time, but for my other project, I'm fine with that. As for the MD++, I am going to try the same procedure and see if it works without the controller "Select" wire attached.

Here is a problem-solved that might help with the bootloader: https://forum.arduino.cc/t/burn-bootloader-attiny88-cant-be-programmed-by-usb-solved/1033045

Please, @rfancella, really do let me know how you get on with it, hah!

rfancella commented 1 year ago

@SukkoPera Agreed. This is now an exercise in education for me. ;) And something to do with these POS clone boards. LOL

Just my own 2 cents: you can get a "real" Arduino Nano clone for ~4€ shipped from China, which would make things much easier :).

DieSkaarj commented 1 year ago

They're not so bad tbh. Happy to hear about the performance with a change in bootloader. Thanks for letting me know, @rfancella.

I've got the same controller layout on my mod Xiyoubu to the one I suggested and it works flawlessly.

I'm really baffled as to what's going on with your board , @rfancella. Did you try rewriting the original MD++ firmware with the select wire on?

SukkoPera commented 1 year ago

Note that on this board the USB connection uses PD1 (D+) and PD2 (D-). I don't know the USB protocol well enough but maybe SELECT being connected to PD2 might interfere with the bootloader startup?

You'd better move SELECT to another pin, but you need one with EXTINT capabilities or to switch to PCINTs and make substantial amendments to the code.

DieSkaarj commented 1 year ago

https://github.com/DieSkaarj/MegaDrivePlusPlus - branch mh-attiny88

You'll need to move SELECT to D3, A/B to D2 and START to D1.

If it works I'll send @SukkoPera a pull request.

rfancella commented 1 year ago

@DieSkaarj Let me know when your branch is ready and what changes I need to make and I'll do a pull and test.

DieSkaarj commented 1 year ago

It's good to go, @rfancella. You'll still need to adapt for your LED.

DieSkaarj commented 1 year ago

Hi @rfancella,

Did you have any success with this board? I'm having trouble reading START consistently and I think it might be related to this issue.

UPDATE

The issue stems from the device not being able to read -5V (a bipolar signal) which can lead to all sorts of weirdness. The solutions that I've yet to test are either use an OpAmp or use a DC offset. [The issue here appeared because the pull-up resistors in the device were damaged and not functioning correctly. - Ed]

Or not. I don't know. I don't have the right tools to check. I'm baffled. I think there might be noise on my line.

Yeah, I busted the pull-ups experimenting with it not working, hah.

Thanks, David

rfancella commented 1 year ago

Hi @rfancella,

Did you have any success with this board? I'm having trouble reading START consistently and I think it might be related to this issue.

UPDATE

The issue stems from the device not being able to read -5V (a bipolar signal) which can lead to all sorts of weirdness. The solutions that I've yet to test are either use an OpAmp or use a DC offset.

Thanks, David

David, I have not. I got a handful of consoles in for repair and have not had a chance to look at it again. I am hoping to get mine back out later this week and try it with a new board. :dagger: Thanks, Ron

SukkoPera commented 8 months ago

Closing due to inactivity after 1Y, feel free to reopen if necessary,