Forums

Hi Folks,

I've been going around to various forums hoping to connect with other Electronics Hobbyists working on similar projects to exchange research.

I just recently completed building a Flashcart for the Standard Gameboy and put together a five part instructional video, aimed for the novice electronics hobbyist.
http://www.youtube.com/user/NLEproGUY#p/u/4/08sBZSciXWk

I'm now working on one for the SEGA GameGear and hope some of you Guru's can shed some light on the subject and help me get this puppy running.

The real issue I find is verification of the actual pinout connection for the Stock issued ROM chips that are used in the actual Sega GG carts.
There are allot of different pinout diagrams around on the net, but it's hard to confirm the actual connections without any legit Sega schematics
or Datasheets available for the GG system. If you know of any please direct me to them.

Shown below is one of the Pinout Diagrams I found online:
[Sega Game Gear - Cartridge Pinouts.jpg]

Below is the Cart scan of a standard Sega GAMEGEAR cart.
[SEGA_GG_FLASH_CART_ 002b.jpg]

Many of us Backtrack the PCB trace lines using a standard GG cartridge along with these diagrams and then devise a possible
ROM Pinout scheme for the ROM contacts on the PCB. Yet, these connections are all based on the "Hobbyist" provided Cart Edge diagrams
which are questionable as to their accuracy.

I still have some issues with my GameGear? Flash cart but I'm not giving up.

Here is my Current Mess of Wires and Solder! UUGGgghhhhhrrrr!!
[SEGA_GG_FLASH_CART_ 032.jpg]

In the photo of my current cart (seen above) you can see the Red wires connected to pins 41 & 42 of the cart edge. These I have connected
to a jumper on the outside of the cartridge enclosure. From what I've read, when these two contacts are bridged or unbridged determines if
the GameGear is in a SMS or GameGear mode. There are a few Cartridge adapters for the GG that allow you to use SMS cartridges.
So I'm guessing one could also use this flash cart to play SMS Roms. All the more motivation to get this puppy running :)

******Now this design is still in the testing phase.*******
Here is my latest updated "GameGear ROM Pinout & Cart Edge Connections Diagram" that I've come up with and this is what I'm working on and
still trying to get working. So by all means if you can see any errors I've made or misshaps.... I'd love to trade research with others working on similar
Projects.

[SEGA_GG_FLASH_CART_ 002C.jpg]

I look forward to your replies.

Regards,
-Gerry

[Admin: huge photos attached, uninlined]

Sega Game Gear - Cartridge Pinouts.jpg (129.08 KB)

SEGA_GG_FLASH_CART_ 2b.jpg (187.65 KB)

SEGA_GG_FLASH_CART_ 032.jpg (168.03 KB)

SEGA_GG_FLASH_CART_ 002C.jpg (307.08 KB)

Here is a quick overlay of the wiring setup I used with an ATMEL AT49F040 Flash chip.....Forgive the crudity of this Graphic.
I would have provided a schematic but I'm not in my Lab. :(

sega_gg_flash_cart_142.jpg (105.25 KB)

I've been looking at the /OE and /CE of the AT49F040 Flash chip and thinking that the /OE is the proper connection for "MREQ" cartridge edge tab #24.

In my diagram I labeled the MREQ with /CE so depending there's still some changes to be made. I've been trying a bunch of different connections.
But still no luck getting it to work.

I've been thinking of trying a different Chip altogether.
Can anyone recommend a better Flash or EEPROM chip for the Job?

Still lots to toy with.
-Gerry

Pinouts:

http://www.smspower.org/Development/GGCartridge

/GG should be GND (GG mode) or +5V (SMS mode), not left floating.

The cart pinouts match the pin arrangement on typical mask ROMs of the era.

Note that to support games larger than 48KB you will need to add some mapper hardware.

Maxim wrote

Pinouts:

http://www.smspower.org/Development/GGCartridge

/GG should be GND (GG mode) or +5V (SMS mode), not left floating.

The cart pinouts match the pin arrangement on typical mask ROMs of the era.

Note that to support games larger than 48KB you will need to add some mapper hardware.

I came across that CartEdge diagram you sent last year. I'm curious as to where it come from? As it doesn't look like it came from SEGA. I'm wondering how it was derived? Was it backtraced from the Z80 Processor onboard the GG? Curious.....Hmmmmm

Also the 48KB max file size you mentioned.....you're referring to the SMS games and not the GG correct? Or does this apply to the GG games as well?

Thanks for all your help,
-Gerry

Gerry_MAN wrote

Also the 48KB max file size you mentioned.....you're referring to the SMS games and not the GG correct? Or does this apply to the GG games as well?

Count the address lines.

I counted 16, A0-A15

paul_ wrote

I counted 16, A0-A15

Yes, that's right!

The data lines have two possible states (binary system) so:
2^16 = 65536 = 64KB

Theoretically, the largest game (without extra hardware) can be 64KB. There are no games this big because the system RAM also shares the same bus, and it would be pretty restrictive to create a game without using any RAM. The RAM is 8KB, mirrored once, which makes it take up 16KB of address space.

Anyway, my point was that a quick look at the data available could answer the question. Are there any GG games of less than 48KB anyway?

There are 3 32KB GG games.

All other games will require some mapper support. If you exclude Codemasters games, you can use the mapper chip found in some SMS games, but it may be large for a GG cart. Custom mapper hardware is trickier.

Pinouts are based on the source of the document you first posted (which is from an unknown source, but seems Sega-internal due to info on the TV Tuner mode connections), and on other people's work on programmable GG carts (verifying it).

Maxim wrote

There are 3 32KB GG games.

All other games will require some mapper support. If you exclude Codemasters games, you can use the mapper chip found in some SMS games, but it may be large for a GG cart. Custom mapper hardware is trickier.

Pinouts are based on the source of the document you first posted (which is from an unknown source, but seems Sega-internal due to info on the TV Tuner mode connections), and on other people's work on programmable GG carts (verifying it).

Interesting....OK Now, in my archive I have several GG game dumps that vary in size from 64KB, 128KB, 256KB and 512KB. 1 Game "Garfield" is 1024KB.

Now the Cartridge PCB that I used in my diagrams above was SONIC and this is a 256KB Game file from what I've seen.
There was no other MAPPER IC's on the PCB except the ROM chip itself.
SEGA -14459A - J11 32PIN TSOP

Did I get a bad Dump file of the Sonic game here?

Well I verified all my hardware and the connections are all working so far.
For a bit I feared something trivial like a short but I interfaced the Original SEGA ROM chip onto my ZIF And it is working. So at least that's something positive for today. :(

-Gerry

The Mapper is inside of what you call the ROM. For space and cost saving it is combined into a single custom chip. More then common on Sega systems. I know Ni****do allmost allways stayed with separated paging chips, so maybe that confused you.

And, /OE should be /RD, not MREQ.

buggus wrote

The Mapper is inside of what you call the ROM. For space and cost saving it is combined into a single custom chip. More then common on Sega systems. I know Ni****do allmost allways stayed with separated paging chips, so maybe that confused you.

And, /OE should be /RD, not MREQ.

Interesting how certain Carts will use the /RD connection (On my Diagram Cart Edge TAB# 26) Like the SONIC cart PCB I used above, the Original ROM chip does not interface to it at all. However the GARFIELD Cart PCB Seen in the Picture I've attached here below, does use the /RD TAB. Although it connects to the Mapper IC.

As you can see in the Photo there is a 315-5912 TQFP44 Mapper chip present. Would the ROM on this particular cart still use additional Mapper integration? or do you think that this is possibly a normal MASK ROM?

Perhaps I should try this Cart PCB instead.

Any thoughts?

-Gerry

sega_gg_garfield_442.jpg (170.03 KB)

GARFIELD CART PCB

Discrete mapper hardware on GG carts has not been investigated very deeply. The image you show certainly looks a lot like a discrete mapper (the 315- chip) and an ordinary mask ROM (the MPR- chip). You'd need to analyse the connections (hopefully they mostly go connector <-> mapper <-> ROM) and figure out the meaning of any extra lines. Then you need to figure out the capabilities - I guess it must support 1MB ROMs, but maybe only slot 2 mapping, and no save RAM.

If you have the capabilities, you can try desoldering the ROM and dumping it by trying common pinouts. That'd get you most of the way to your goal.

Gerry_MAN wrote

As you can see in the Photo there is a 315-5912 TQFP44 Mapper chip present. Would the ROM on this particular cart still use additional Mapper integration? or do you think that this is possibly a normal MASK ROM?

Could you please scan the bottom side if this cart aswell?

Maxim wrote

Discrete mapper hardware on GG carts has not been investigated very deeply. The image you show certainly looks a lot like a discrete mapper (the 315- chip) and an ordinary mask ROM (the MPR- chip). You'd need to analyse the connections (hopefully they mostly go connector <-> mapper <-> ROM) and figure out the meaning of any extra lines. Then you need to figure out the capabilities - I guess it must support 1MB ROMs, but maybe only slot 2 mapping, and no save RAM.

If you have the capabilities, you can try desoldering the ROM and dumping it by trying common pinouts. That'd get you most of the way to your goal.

Actually I did this with the Original SONIC Mask ROM (Photo Below) I wanted to make sure I didn't have any shorts present in my wiring from the ZIF socket. Using the Original ROM and connecting it to the ZIF would verify this. Which worked and the Sonic Game played fine on the GG, so no shorts in my soldering work to worry about. But of course the Game image dump I performed with the Willem didn't work when I tried it with MEKA EMU. I didn't think it would as there were too many uncertainties.

I tried setting the Willem to an ATMEL AT49F020 a 256KB size Flash chip. I knew the Game Sonic was of this size so it seemed somewhat doable. Anyone know of a better setting to use with the Willem to perform a chip read?

Now if the reason the Dump failed is due to the Embedded Mapper in the Sonic IC or most likely my Pin connections I'm not sure. I'll try it with the Garfield ROM and see if there is a Difference.

I'm thinking of rigging the original ROM chips to the Logic analyzer now that I'll have them connected to the DIP 32 Adapters and have access to the Pins while in operation. Comparing the Waveforms between the MASK ROMs and my Target Flash chip of Choice. I'll monitor the ATMEL Flash chip while running a Flashed game on the Gameboy. This might show something we could not see otherwise. Although still not sure as they are two different Platforms. But I'm mainly interested in looking at the I/O pins to make sure I have those right. The Address & Data pins will no doubt be very different from one and other.

Regardless, I see it as the more info I can collect the better.
Any thoughts anyone?

-Gerry

mpr_sonic_002_669.jpg (75.79 KB)

Sonic TSOP to DIP

gg_flash_cart_003_478.jpg (124.4 KB)

My GG Flash Cart

mpr_sonic_004_651.jpg (134.79 KB)

Willem - Sonic ROM Dump

I just de-soldered the MASK ROM but from looking at the Back side, it looks like I will have to remove the Mapper IC as well. There are a few through holes on the underside of the Mapper chip.

I think I'll do that later tonight.

-Gerry

gg_garf_cart_001_151.jpg (102.73 KB)

gg_garf_cart_002_836.jpg (86.25 KB)

The ROM+mapper only exposes 16 address lines. You need to write to address $ffff to select the 16KB chunk to expose in the $8000-$bfff region. Your ROM dumper may not be able to do that...

The Garfield ROM is more likely to work.

Maxim wrote

The ROM+mapper only exposes 16 address lines. You need to write to address $ffff to select the 16KB chunk to expose in the $8000-$bfff region. Your ROM dumper may not be able to do that...

The Garfield ROM is more likely to work.

Ahh Ok, That is motivating.

I'll try that out first.

Now in my Diagram I had " NC" displayed for the ROM Pins that were literally Unconnected and had not trace lines connected to them at all.

But on this particular Cartridge they are all being used. They are probably the additional address pins right?

Any idea what they would be?

-Gerry

I checked, there is NO standard mask rom with 64mbit / 1mbyte in a 32pin package. Judgeing on the needed address lines and the pcb layout, i suspect this is again a custom chip, which looks like it obeys mostly the pinout of the rom with mapper chip. They seem to use the NC pins AND some of the pins previously used by the internal mapper as address lines.

I would begin tracing the lines from the edge connector to the rom chip, ruling out as much pins as possible. whats then left unknown AND is ONLY connected to the suspected mapper chip can only be address lines and an /OE line. Find out the /OE pin (i reckon pin 22), and try to read the Chip. From the shifting in data blocks in the dump you can get the layout of the left address lines, compared to the known good dump.

This makes sense because the paging window of SMS/GG is 16k, i see 7 lines going from mapper to rom, would count as 14 lines directly from the sms plus 6 from the mapper (7th is /OE), being 20 -> 1024kByte / 1Mbyte.

But then, the chip marking starts with "MPR", maybe giving the impression its more then a rom...

All Sega ROMs (with and without integrated mappers) have MPR numbers, unique across all systems/arcade/etc from the beginning of time for a particular ROM image.

Here are the Scans of the Garfield Cartridge PCB with all the chips removed.

I'll be connecting the ROM to the Logic Analyzer tomorrow if all goes well.

And I should probably put together a Schematic of this PCB also.

One step at a time. :)

-Gerry

gg_garf_cart_no_ic_180.jpg (156.75 KB)

gg_garf_cart_002_172.jpg (86.25 KB)

I suggest you a pinout you may want to try with a flash chip:

22 - /OE
29 - A14
30 - A15
31 - A17

01 - A19
02 - A18
03 - A16

There are now 5 pins left which go to the cut-out short circuits. Wonder what the mapper may be capable of...

Hey buggus,

That's a great suggestion. After I mapped out all the PCB traces from this Garfield cart PCB, and along with your pin suggestions
I came up with this Pinout diagram for the SEGA 315-5912 (TQFP44) Mapper Chip.

Now certain Cartridge edge TAB's are connected by traces to both the Mapper and the ROM contacts, and some are not.
There are some traces that will go from the Mapper chip to only the Cart edge TAB's or only to the ROM/Flash chip Pins.
So I have labeled those accordingly in the Diagram. I labeled certain contacts with "Z80" to designate the connections that go
only to the Cartridge edge TAB connections.

In any case, It looks quite promising.

I'll be rigging this up tonight with my Flash chip and hopefully confirm this diagram.

I'll see how it goes.
-Gerry

SEGA 315-5912 - Mapper.jpg (183.76 KB)

Still non functional. Dang namit!!

Tested all my wiring connections again and tried a new Flash chip.
But still Zip!!

I was waiting on getting my Logic Analyzer POD back from a friend so I just picked it up this morning and I'm
going to be running some scans on the original ROM chips and hopefully I'll be able to see some logical solution.

No Pun intended. :)

I'll keep you all posted.

Later Folks,
-Gerry

Hi Everyone,

I just got another 1 MByte GG Cartridge in from a friend.
The cart is entitled "Virtua Fighter Animation". It is exactly the same PCB as the Garfield 1MByte cart. PCB 171-7000A and it also has the 315-5912 Mapper chip.

I'm getting another 1MB cartridge "Sonic Blast" in next week. I'm assuming this will be the same PCB as well.

In any case, I just wanted to share that info.
I'll post my progress as I go.

Cheers Folks,
-Gerry

sega_vf_315_5912_004_133.jpg (90.66 KB)

sega_vf_315_5912_003_208.jpg (157.96 KB)

sega_vf_315_5912_002_138.jpg (96.03 KB)

Hey Folks,

I got my Logic Analyzer up and running this Morning. I had some problems with the POD interface wires being too short so I built a 60 pin extension with some old ribbon cable and an old Genesis Cartridge to hold the connector PCB. I etched the PCB for the 60 pin connectors yesterday. It has worked out nicely so far.

I'll be interfacing to various ROM chips today to check for pattern similarities in hopes to see something that may shed some light on how to get this sucker to work with a Flash chip.

The photos Below are some shots of the Logic Analyzer and the connecting wires interfaced to one of the ROMs. I used a 32 Pin DIP socket and a pin array to give the POD interface connectors something to grip onto, and also to protect the delicate ROM TSOP pins from damage.

Anyhow, I'll post my Scan results hopefully over the next couple of days.

Later folks,
-Gerry

logic_analyzer_004_198.jpg (85.05 KB)

logic_analyzer_003_214.jpg (86.48 KB)

logic_analyzer_006_691.jpg (105.96 KB)

gg_rom_2_la_001_530.jpg (53.86 KB)

gg_rom_2_la_008_199.jpg (185.07 KB)

gg_rom_2_la_013_201.jpg (170.92 KB)

I just got the "Sonic Blast" GG Cartridge in Today and as I suspected, the cartridge is identical to the other 1MB cartridges.

You can see the Images below.

-Gerry

sonic_b_gg_cart_006_971.jpg (84.31 KB)

sonic_b_gg_cart_004_529.jpg (108.41 KB)

sonic_b_gg_cart_007_717.jpg (100.21 KB)

Hey Folks,

So I got my Logic Analyzer software configured and captured the first 4096 pulse samples off the Original ROM chip from the "Garfield" 1MByte Cartridge.

I put all the Trace Data into separate fields. Such as "Address", "Data" and also had the designated /OE and /CE lines displayed. But I also separated all the Pins and show their Logic Waveforms individually.

Now remember this is only the first 4096 Pulses off all the pins on the ROM chip so that is essentially a fraction of a second of activity.

So now the task is to try and analyze the individual Waveform patterns and see if they make sense as the pin Labels they have been given. /OE, /CE and such specifically.

So now the Pictures below are the 4096 Samples in sections and I took Screen captures as I scrolled across the Waveforms. Observe the Scroll bar indicator at the bottom of the Captured images to see the location of the samples in view.

The First image I have zoomed in closely so you can see the detail in the Pulses and I also show the Address and DATA Bus's in HEX.
The other waveform images are more of a wider view so they detail is less but still able to view.

In any case I'm still playing with the Analyzer unit and will be taking more captures and also performing some Pattern Recognition scans.

Take a look at the Images and let me know your thoughts.

Regards,
-Gerry

**********************************************************************************
Analyzer Capture Images Re-posted further down with corrections and additional Pin information.
**********************************************************************************

There's so much information missing here that it's hard to tell what the goal is.

Are the signal numbers ROM pins 1 through 32? Shouldn't the signal names list what Z80 or mapper chip pin/signal they connect to as well? I see the ROM, but where is the mapper chip? If it isn't present what did you connect the unknown ROM pins to?

There's no way to tell what portions of the dump correspond to cartridge access versus Z80 RAM or I/O access. The signal that is ROM CE from the cartridge port should be labeled to indicate that.

Is /RD the same as Z80 /RD? Is /OE also /RD, if so why is it included?

Charles MacDonald wrote

There's so much information missing here that it's hard to tell what the goal is.

Are the signal numbers ROM pins 1 through 32? Shouldn't the signal names list what Z80 or mapper chip pin/signal they connect to as well? I see the ROM, but where is the mapper chip? If it isn't present what did you connect the unknown ROM pins to?

There's no way to tell what portions of the dump correspond to cartridge access versus Z80 RAM or I/O access. The signal that is ROM CE from the cartridge port should be labeled to indicate that.

Is /RD the same as Z80 /RD? Is /OE also /RD, if so why is it included?

Hi Charles,

I guess I jumped the Gun here. I just got the Logic analyzer working and that's where I was at. I'll add that Pin info for you and perform a new scan.

All the Data here is off the ROM chip connections from the 1MByte Cartridge "Garfield".

My goal is to confirm the Designated Pins for the ROM contacts
on the GG cart. As you know Datasheets cease to exist to confirm the pins.

I have used the Provided GG Cartridge Pinout diagrams from SMSPower and backtraced to the ROM and derived a ROM Pinout like most of us do. I posted the diagram previously.
Now regardless of my attempts to interface a Flash or EEPROM in place of the ROM, it still does not function. As seen in the previous posts some people are speculating that the ROM chips
on GG carts are a ROM and Mapper chip integrated together into one.

I was hoping to try and determine if this was the case with the 1MByte cartridges such as "Garfield" which use the 315-5912 Mapper chip.
I had hopes that it was a mapper and ROM completely separate from each other. But still not sure.

And yes, Pins 1-32 are the ROM pins individually. But I also displayed them together as a BUS labeled as ADDRBUS, DATABUS (these are displayed in HEX.) I also show /OE, /CE of the ROM.

My Goal is a GG Flash cart of course, and hoping to use an official GG cartridge and Modify it by simply interfacing a Flash chip.
As opposed to etching my own PCB and using other Mapper chips. But I may end up having to do this if all GG carts are in fact all designed around using a ROM+Mapper integrated Chip.

So Charles, Thanks so much for the input here. Sorry for the lack of info though. I'll make sure I add that before the next post.

I guess I'll get back to the Lab.
Cheers!
-Gerry

Quote

1MByte cartridges such as "Garfield" which use the 315-5912 Mapper chip. I had hopes that it was a mapper and ROM completely separate from each other.

It seems highly likely to me that where the 315-5912 is present, the other chip is just a normal ROM.

Quote

My goal is to confirm the Designated Pins for the ROM contacts on the GG cart.

OK, let's assume the other chip is indeed a normal ROM. In "Garfield", it will be a 32-pin 1MByte ROM chip. This means that there will be 20 address lines and 8 data lines, leaving just 4 unknown pins.

It should be straightforward to trace the address and data lines to derive a mostly-complete ROM pinout. I would do it, but it seems you already have:

Quote

I have [...] derived a ROM Pinout like most of us do. I posted the diagram previously.

Maybe I've missed something, but I can't see where you've posted a proposed ROM pinout - only one for the 315-5912.

I took the time to trace the lines on your PCB images. You'll be pleased to hear that I agree with your proposed pinout for the 315-5912. However, I would say that we have no way of knowing which is which among the connections which do not go to the cartridge connector - those in red in your diagram.

Do you agree that the ROM pinout implied by the 315-5912 pinout is as follows? The 7 pins marked "???" are A14-A19 and /OE (in some order):

1: ??? 2: ??? 3: ??? 4: A12
5: A7 6: A6 7: A5 8: A4
9: A3 10: A2 11: A1 12: A0
13: D0 14: D1 15: D2 16: GND
17: D3 18: D4 19: D5 20: D6
21: D7 22: ??? 23: A10 24: /RD
25: A11 26: A9 27: A8 28: A13
29: ??? 30: ??? 31: ??? 32: Vcc

I assume you have created a flash cartridge using this pinout, with A14-A19 and /OE as in your diagram? And this flash cartridge is not working? If so, the cause may well be that the "???" connections are in the wrong order.

Here's what I would do to try and figure out the correct order:

The first step is to find out which ROM pin is /OE. To do this, try taking the Garfield ROM chip and grounding A0-A13 and /RD. Attach the "???" pins to Vcc. In other words, connect up the ROM as follows:

GND: Pins 4-12 (inclusive), 16, 23-28
Vcc: Pins 1-3, 22, 29-32

Then, watch what happens to the data lines (13-15 and 17-21) as you try grounding pins 1, 2, 3, 22, 29, 30, 31, one at a time. One of these combinations should produce a valid byte on the data lines. Which combination is it? What is the output byte?

Maxim wrote

Then you need to figure out the capabilities - I guess it must support 1MB ROMs, but maybe only slot 2 mapping, and no save RAM.

Using the MEKA debugger, we can see that Garfield uses mapping in slots 1 and 2, so the 315-5912 presumably supports at least those slots. I guess once Gerry_MAN has his flash cart working, we might be able to find out about slot 0 mapping using PFR Detect.

As for save RAM, clearly it is not supported on this PCB, but the 315-5912 may still support it - perhaps this is the purpose of (some of) the unconnected pins 21 to 25?

Paul Baker wrote

I took the time to trace the lines on your PCB images. You'll be pleased to hear that I agree with your proposed pinout for the 315-5912. However, I would say that we have no way of knowing which is which among the connections which do not go to the cartridge connector - those in red in your diagram.

Do you agree that the ROM pinout implied by the 315-5912 pinout is as follows? The 7 pins marked "???" are A14-A19 and /OE (in some order):

1: ??? 2: ??? 3: ??? 4: A12
5: A7 6: A6 7: A5 8: A4
9: A3 10: A2 11: A1 12: A0
13: D0 14: D1 15: D2 16: GND
17: D3 18: D4 19: D5 20: D6
21: D7 22: ??? 23: A10 24: /RD
25: A11 26: A9 27: A8 28: A13
29: ??? 30: ??? 31: ??? 32: Vcc

I assume you have created a flash cartridge using this pinout, with A14-A19 and /OE as in your diagram? And this flash cartridge is not working? If so, the cause may well be that the "???" connections are in the wrong order.

Here's what I would do to try and figure out the correct order:

The first step is to find out which ROM pin is /OE. To do this, try taking the Garfield ROM chip and grounding A0-A13 and /RD. Attach the "???" pins to Vcc. In other words, connect up the ROM as follows:

GND: Pins 4-12 (inclusive), 16, 23-28
Vcc: Pins 1-3, 22, 29-32

Then, watch what happens to the data lines (13-15 and 17-21) as you try grounding pins 1, 2, 3, 22, 29, 30, 31, one at a time. One of these combinations should produce a valid byte on the data lines. Which combination is it? What is the output byte?

Maxim wrote

Then you need to figure out the capabilities - I guess it must support 1MB ROMs, but maybe only slot 2 mapping, and no save RAM.

Using the MEKA debugger, we can see that Garfield uses mapping in slots 1 and 2, so the 315-5912 presumably supports at least those slots. I guess once Gerry_MAN has his flash cart working, we might be able to find out about slot 0 mapping using PFR Detect.

As for save RAM, clearly it is not supported on this PCB, but the 315-5912 may still support it - perhaps this is the purpose of (some of) the unconnected pins 21 to 25?

Hi Paul,

The Pinouts you provided for the Garfield ROM chip are exactly as I had laid out.
Here is my Pinout Diagram for the GG Garfield ROM chip (MPR-18201-S) that I neglected to post. Sorry about that.
I've been posting in many forums so I got mixed up and assumed I had posted it earlier.

http://www.digital-circuitry.com/IMAGES/My%20LAB/Sega/GameGear/1200%20X%20960/GG_TSOP_ROM_001.jpg

Quote

The 7 pins marked "???" are A14-A19 and /OE (in some order):

Yes I agree. And I have been trying various combinations with these pins but have yet to find any that work.
I will try to determine the ??? pins for the ROM chip by the procedures you gave.Thanks for that suggestion. :)
I also just updated the 315-5912 Mapper Pinout diagram.
Here is the link below:

http://www.digital-circuitry.com/IMAGES/My%20LAB/Sega/GameGear/1200%20X%20960/SEGA%20315-5912%20-%20Mapper_003.jpg

I'll hopefully have some time in the Lab later tonight. I'll try and a few things and also update my Logic analyzer Scan images with more detailed Info.

Again I really appreciate the help and advice here from everyone.
And if you see any inconsistencies with my diagrams let me know.

Talk to you all later,
-Gerry

*************************************************************************
Pinout Diagram for the GG Garfield ROM chip (MPR-18201-S)
HAS BEEN UPDATED.
*************************************************************************

SEGA 315-5912 - Mapper_003.png (65.83 KB)

GG_TSOP_ROM_001.jpg (119.4 KB)

Gerry_MAN wrote

The Pinouts you provided for the Garfield ROM chip are exactly as I had laid out.

Good. This means that the pinouts we have so far (for both ICs) are probably correct, since two of us have come to the same results independently. Now, to find out what the ??? pins are...

Quote

I will try to determine the ??? pins for the ROM chip by the procedures you gave.Thanks for that suggestion. :)

I hope it works. That is, I hope you get a good data byte from one of the combinations. That will tell us which ROM pin is /OE. From there, I can put together a similar procedure to find out the exact order of address lines.

Quote

And if you see any inconsistencies with my diagrams let me know.

I spotted a few small issues with GG_TSOP_ROM_001.jpg:

Updated the GG_TSOP_ROM_001.jpg: image link.

Thanks Paul.
-Gerry

GG_TSOP_ROM_001.jpg (119.4 KB)

Hey Folks,

So I have performed a new scan with the Logic Analyzer and I added some more information regarding the PIN connections.
Figured you all might want to have a look at my progress just for kicks.

This scan is once again taken from the 1Mbyte Garfield ROM. Pins 1-32.
All pins are scanned using positive logic. I also have the Address Bus and Data Bus displayed as its own field in HEX format.

Also, the scans I take are only 4096 samples and is roughly around 800uS of activity and I only show a small portion of this capture in the photos.

Now I have the pins labeled as such: 5_R_A7*

- 5 being the PIN #
- R indicating it as a ROM pin
- A7 being the PIN function (Address BUS 7)
- the ( * ) indicates the pin connects to both the Cartridge edge (Z80)
and the 315-5912 Mapper IC.
- The ( ** ) indicates the PIN connects only to the 315-5912 Mapper IC only.
These pins are of course the ?? unknown pins as we have no reference
to identify them.

I still have to go through the steps Paul suggested to identify the ?? pins but figured I would post the new updated scans first for you all to have a look.

The First 3 capture images are zoomed in quite close for added detail. The last image is a zoomed out view, to give you a wider perspective of the ROM activity.

I remarked how the ?? PIN 31 is in a constant logic HIGH state. This could possibly be a /WR function could it not?
Although ROMs of course do not have this feature. It just caught my eye and it made me think.

I'll be posting Scans of the ROM activity during the Test that Paul suggested and get some feedback from everyone.
Hopefully I'll have some time to play in the Lab tomorrow.

Talk to you later,
-Gerry

GG_ROM_1MB_002A.jpg (355.58 KB)

GG_ROM_1MB_002B.jpg (355.55 KB)

GG_ROM_1MB_002C.jpg (354.67 KB)

GG_ROM_1MB_002D.jpg (433.56 KB)

If you have a device programmer I would recommend connecting the ROM to a socket with the known address signals, CS, and OE hooked up, and then tie the unknown pins to sequential high-order address lines, each through a series resistor (say 4.7K) if you want to be careful.

If then you then dump the ROM as whatever type has that many address lines (27C080), you'll exercise all possible states of the unknown pins. That way you (or any of us) could examine the dump and figure out which pin corresponds to what.

Charles MacDonald wrote

If you have a device programmer I would recommend connecting the ROM to a socket with the known address signals, CS, and OE hooked up, and then tie the unknown pins to sequential high-order address lines, each through a series resistor (say 4.7K) if you want to be careful.

If then you then dump the ROM as whatever type has that many address lines (27C080), you'll exercise all possible states of the unknown pins. That way you (or any of us) could examine the dump and figure out which pin corresponds to what.

Hi Charles,

I appreciate the help. You guys know allot more than I do when it comes to the "code" so I'll let you all take a shot at examining the DUMP data.

I'm in the midst of trying a few things but as soon as I'm done, I'll hook the chip up to the Universal Programmer and download the data to a .rom file for your review.

Hopefully soon we will have this 1MByte Programmable cartridge up and running.

One step at a time. :)

Regards,
-Gerry

Charles MacDonald wrote

If you have a device programmer I would recommend connecting the ROM to a socket with the known address signals, CS, and OE hooked up, and then tie the unknown pins to sequential high-order address lines, each through a series resistor (say 4.7K) if you want to be careful.

If then you then dump the ROM as whatever type has that many address lines (27C080), you'll exercise all possible states of the unknown pins. That way you (or any of us) could examine the dump and figure out which pin corresponds to what.

In one of the previous posts "buggus" had made this same suggestion.

Quote

I checked, there is NO standard mask rom with 64mbit / 1mbyte in a 32pin package. Judgeing on the needed address lines and the pcb layout, i suspect this is again a custom chip, which looks like it obeys mostly the pinout of the rom with mapper chip. They seem to use the NC pins AND some of the pins previously used by the internal mapper as address lines.

I would begin tracing the lines from the edge connector to the rom chip, ruling out as much pins as possible. whats then left unknown AND is ONLY connected to the suspected mapper chip can only be address lines and an /OE line. Find out the /OE pin (i reckon pin 22), and try to read the Chip. From the shifting in data blocks in the dump you can get the layout of the left address lines, compared to the known good dump.

This makes sense because the paging window of SMS/GG is 16k, i see 7 lines going from mapper to rom, would count as 14 lines directly from the sms plus 6 from the mapper (7th is /OE), being 20 -> 1024kByte / 1Mbyte.

But then, the chip marking starts with "MPR", maybe giving the impression its more then a rom...

I got so wrapped up with the Logic analyzer that I forgot to try this. :)

I'm thinking I may build a 512KB Flash cart first while continuing to work on this 1MByte version. And I'll go with using the SMS mappers like Maxim suggested in one of the first posts on this topic.

Now by using a Mapper such as the 315-5235 would exclude some ROMs from functioning correct ?...... but having 512Kb of memory to play with is still pretty good. Still lots of fun if your playing around with your own code.

But regardless, a 1Mbyte version is much more attractive so my money is on this one. I'm really curious if the Garfield cart is using a ROM+Mapper integrated chip and also the 315-5912 Mapper all in unison to achieve the 1MByte memory size?

Guess I need to spend some more time in the Lab.

Talk to you all later,
-Gerry

Gerry_MAN wrote

Guess I need to spend some more time in the Lab.

Nothing new?

buggus wrote

Gerry_MAN wrote

Guess I need to spend some more time in the Lab.

Nothing new?

I've been going insane fixing up my Mom's house as she is trying to sell. And also going nuts lining things up for the coming College semester. I hate dealing with the Red tape.

Needless to say I've been busy with other things.

However, I have been working on something and I will be placing a post in the coming weeks before Xmas with all the details. :)

Nothing too crazy but still SMSPower worthy.

I'll keep you posted.

-Gerry

Gerry_MAN wrote

buggus wrote

Gerry_MAN wrote

Guess I need to spend some more time in the Lab.

Nothing new?

I've been going insane fixing up my Mom's house as she is trying to sell. And also going nuts lining things up for the coming College semester. I hate dealing with the Red tape.

Needless to say I've been busy with other things.

However, I have been working on something and I will be placing a post in the coming weeks before Xmas with all the details. :)

Nothing too crazy but still SMSPower worthy.

I'll keep you posted.

-Gerry

Well, as I mentioned earlier..... I would be making a Posting before the holidays so here it is.

A 512Kb Flash cartridge for the SEGA Game Gear.

Here is the New Topic thread if you haven't seen it yet:

http://www.smspower.org/forums/viewtopic.php?t=12734

I'll continue to work on this 1 MegaByte version over the holidays.
But this should keep us busy for the time being. :)

Cheers!
-Gerry

Just adding another Rare GG PCB scan for reference. This PCB has RAM interconnected to the board, along with a Battery backup.

I'm now able to officially confirm certain connections on the ROM by having access to available datasheets for the RAM chip that is interconnected.

See the pictures below:

-Gerry

GG_ROM+RAM_ 007.jpg (217.9 KB)

GG_ROM+RAM_ 003.jpg (175.41 KB)

GG_ROM+RAM_ 005.jpg (191.29 KB)

docs.google.com2010.09.20.wip.zip updated:

315-5912 8M mapper added
Round testpad block of 2 and 3 added
Slot drill and pad sizes reduced by 30%
small drill footprint added for DIL42-6.07

Victory at last.

A 1 Megabyte DIY SEGA Game Gear Flash cart is now possible !! :)

Check out the details on the new Topic (Link Posted below:)

http://www.smspower.org/forums/viewtopic.php?t=12931

Kind Regards,
Gerry O'Brien

a bit late , but i want you to show my reengeneering method just using Paint.

This three Pictures are just examples, can´t find a finished Sonic Blast PCB
at the moment maybe i´ve done that but i think i hoped for having a better scan.

As you can see, it would be helpful if the scans are hrizontally and vertically
good and the lightning should be optimized before taking a photo.
personally I never got good results with a scanner, too many light.

first step is to save the picture with 256 colors, than try it with 16
colors. If you have good contrast with 16 colors you can use the fill option
and are finish within minutes. If not, it´s helpfull to redraw the wires
with different colors, ending with 2 colors.

I prefer to use eagle from cadsoft to do my pcbs, it´s the most used hobbyist
ecad software used here in germany, do their free version.

eagle has a bitmap import funtion so you open a new board, make a new layer
and give it a dark color e.g. dark blue for the bottom layer.

than you can add the parts and the edge connector to the schematic just use the "ic-packages"
make the nets and route them with your imported bitmap as background.

Sonic Blast GG Top-paint.PNG (161.59 KB)

Sonic Blast GG Bottom-paint.PNG (129.36 KB)

gg_romram_003_297-256colors-paint.PNG (291.6 KB)

Nicely Done!

I agree, the Cadsoft Eagle software is my current PCB design software of choice.
Although, I am starting to use the National Instruments PCB layout software.

The main advantage for this software is having a seamless transition from the Schematic design entry and Simulation, using the NI "Multisim" simulator; and then going directly to the PCB layout all in one software package.

This will save you allot of time overall and is simply a much more efficient way of doing things in my opinion. So I'm in the midst of switching over to the NI software.

Anyhow, If you get a finished PCB layout for this cart, please post it here.
I'd love to see it.

Cheers!
-Gerry

I think you need to desolder everything so you can be sure where the tracks go??

I don't like Eagle because it will let you put tracks too close together, violating the clearances you have set and you won't know about it until afterwards when you run DRC.

I prefer KiCAD as DRC runs in realtime, if you try to lay a track too close to the other it will stop you doing it - Also it has an outline display to show where the clearance is so you know exactly what's going on.

Author	Message
Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	SEGA GameGear Flash Cart - Custom Electronics Posted: Thu Sep 23, 2010 11:18 am Last edited by Gerry_MAN on Sun Oct 10, 2010 12:14 am; edited 1 time in total
	Hi Folks, I've been going around to various forums hoping to connect with other Electronics Hobbyists working on similar projects to exchange research. I just recently completed building a Flashcart for the Standard Gameboy and put together a five part instructional video, aimed for the novice electronics hobbyist. http://www.youtube.com/user/NLEproGUY#p/u/4/08sBZSciXWk I'm now working on one for the SEGA GameGear and hope some of you Guru's can shed some light on the subject and help me get this puppy running. The real issue I find is verification of the actual pinout connection for the Stock issued ROM chips that are used in the actual Sega GG carts. There are allot of different pinout diagrams around on the net, but it's hard to confirm the actual connections without any legit Sega schematics or Datasheets available for the GG system. If you know of any please direct me to them. Shown below is one of the Pinout Diagrams I found online: [Sega Game Gear - Cartridge Pinouts.jpg] Below is the Cart scan of a standard Sega GAMEGEAR cart. [SEGA_GG_FLASH_CART_ 002b.jpg] Many of us Backtrack the PCB trace lines using a standard GG cartridge along with these diagrams and then devise a possible ROM Pinout scheme for the ROM contacts on the PCB. Yet, these connections are all based on the "Hobbyist" provided Cart Edge diagrams which are questionable as to their accuracy. I still have some issues with my GameGear? Flash cart but I'm not giving up. Here is my Current Mess of Wires and Solder! UUGGgghhhhhrrrr!! [SEGA_GG_FLASH_CART_ 032.jpg] In the photo of my current cart (seen above) you can see the Red wires connected to pins 41 & 42 of the cart edge. These I have connected to a jumper on the outside of the cartridge enclosure. From what I've read, when these two contacts are bridged or unbridged determines if the GameGear is in a SMS or GameGear mode. There are a few Cartridge adapters for the GG that allow you to use SMS cartridges. So I'm guessing one could also use this flash cart to play SMS Roms. All the more motivation to get this puppy running :) ****Now this design is still in the testing phase.***** Here is my latest updated "GameGear ROM Pinout & Cart Edge Connections Diagram" that I've come up with and this is what I'm working on and still trying to get working. So by all means if you can see any errors I've made or misshaps.... I'd love to trade research with others working on similar Projects. [SEGA_GG_FLASH_CART_ 002C.jpg] I look forward to your replies. Regards, -Gerry [Admin: huge photos attached, uninlined] Sega Game Gear - Cartridge Pinouts.jpg (129.08 KB) SEGA_GG_FLASH_CART_ 2b.jpg (187.65 KB) SEGA_GG_FLASH_CART_ 032.jpg (168.03 KB) SEGA_GG_FLASH_CART_ 002C.jpg (307.08 KB)

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Thu Sep 23, 2010 1:11 pm
	Here is a quick overlay of the wiring setup I used with an ATMEL AT49F040 Flash chip.....Forgive the crudity of this Graphic. I would have provided a schematic but I'm not in my Lab. :( sega_gg_flash_cart_142.jpg (105.25 KB)

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Thu Sep 23, 2010 1:26 pm
	I've been looking at the /OE and /CE of the AT49F040 Flash chip and thinking that the /OE is the proper connection for "MREQ" cartridge edge tab #24. In my diagram I labeled the MREQ with /CE so depending there's still some changes to be made. I've been trying a bunch of different connections. But still no luck getting it to work. I've been thinking of trying a different Chip altogether. Can anyone recommend a better Flash or EEPROM chip for the Job? Still lots to toy with. -Gerry

Maxim Site Admin Joined: 19 Oct 1999 Posts: 14798 Location: London	Posted: Thu Sep 23, 2010 2:57 pm
	Pinouts: http://www.smspower.org/Development/GGCartridge /GG should be GND (GG mode) or +5V (SMS mode), not left floating. The cart pinouts match the pin arrangement on typical mask ROMs of the era. Note that to support games larger than 48KB you will need to add some mapper hardware.

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Thu Sep 23, 2010 7:23 pm
	Maxim wrote Pinouts: http://www.smspower.org/Development/GGCartridge /GG should be GND (GG mode) or +5V (SMS mode), not left floating. The cart pinouts match the pin arrangement on typical mask ROMs of the era. Note that to support games larger than 48KB you will need to add some mapper hardware. I came across that CartEdge diagram you sent last year. I'm curious as to where it come from? As it doesn't look like it came from SEGA. I'm wondering how it was derived? Was it backtraced from the Z80 Processor onboard the GG? Curious.....Hmmmmm Also the 48KB max file size you mentioned.....you're referring to the SMS games and not the GG correct? Or does this apply to the GG games as well? Thanks for all your help, -Gerry

viletim! Guest	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Fri Sep 24, 2010 12:56 am
viletim! Guest	Gerry_MAN wrote Also the 48KB max file size you mentioned.....you're referring to the SMS games and not the GG correct? Or does this apply to the GG games as well? Count the address lines.

paul_ Joined: 20 Feb 2007 Posts: 56 Location: Wellington, New Zealand.	Posted: Fri Sep 24, 2010 5:12 am
	I counted 16, A0-A15

viletim Joined: 24 Sep 2006 Posts: 191 Location: Sydney, Australia	Posted: Fri Sep 24, 2010 8:31 am
	paul_ wrote I counted 16, A0-A15 Yes, that's right! The data lines have two possible states (binary system) so: 2^16 = 65536 = 64KB Theoretically, the largest game (without extra hardware) can be 64KB. There are no games this big because the system RAM also shares the same bus, and it would be pretty restrictive to create a game without using any RAM. The RAM is 8KB, mirrored once, which makes it take up 16KB of address space. Anyway, my point was that a quick look at the data available could answer the question. Are there any GG games of less than 48KB anyway?

Maxim Site Admin Joined: 19 Oct 1999 Posts: 14798 Location: London	Posted: Fri Sep 24, 2010 8:50 am
	There are 3 32KB GG games. All other games will require some mapper support. If you exclude Codemasters games, you can use the mapper chip found in some SMS games, but it may be large for a GG cart. Custom mapper hardware is trickier. Pinouts are based on the source of the document you first posted (which is from an unknown source, but seems Sega-internal due to info on the TV Tuner mode connections), and on other people's work on programmable GG carts (verifying it).

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Fri Sep 24, 2010 11:28 am
	Maxim wrote There are 3 32KB GG games. All other games will require some mapper support. If you exclude Codemasters games, you can use the mapper chip found in some SMS games, but it may be large for a GG cart. Custom mapper hardware is trickier. Pinouts are based on the source of the document you first posted (which is from an unknown source, but seems Sega-internal due to info on the TV Tuner mode connections), and on other people's work on programmable GG carts (verifying it). Interesting....OK Now, in my archive I have several GG game dumps that vary in size from 64KB, 128KB, 256KB and 512KB. 1 Game "Garfield" is 1024KB. Now the Cartridge PCB that I used in my diagrams above was SONIC and this is a 256KB Game file from what I've seen. There was no other MAPPER IC's on the PCB except the ROM chip itself. SEGA -14459A - J11 32PIN TSOP Did I get a bad Dump file of the Sonic game here? Well I verified all my hardware and the connections are all working so far. For a bit I feared something trivial like a short but I interfaced the Original SEGA ROM chip onto my ZIF And it is working. So at least that's something positive for today. :( -Gerry

buggus Joined: 24 Oct 2009 Posts: 23	Posted: Fri Sep 24, 2010 12:27 pm
buggus Joined: 24 Oct 2009 Posts: 23	The Mapper is inside of what you call the ROM. For space and cost saving it is combined into a single custom chip. More then common on Sega systems. I know Ni****do allmost allways stayed with separated paging chips, so maybe that confused you. And, /OE should be /RD, not MREQ.

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Fri Sep 24, 2010 2:09 pm
	buggus wrote The Mapper is inside of what you call the ROM. For space and cost saving it is combined into a single custom chip. More then common on Sega systems. I know Ni****do allmost allways stayed with separated paging chips, so maybe that confused you. And, /OE should be /RD, not MREQ. Interesting how certain Carts will use the /RD connection (On my Diagram Cart Edge TAB# 26) Like the SONIC cart PCB I used above, the Original ROM chip does not interface to it at all. However the GARFIELD Cart PCB Seen in the Picture I've attached here below, does use the /RD TAB. Although it connects to the Mapper IC. As you can see in the Photo there is a 315-5912 TQFP44 Mapper chip present. Would the ROM on this particular cart still use additional Mapper integration? or do you think that this is possibly a normal MASK ROM? Perhaps I should try this Cart PCB instead. Any thoughts? -Gerry sega_gg_garfield_442.jpg (170.03 KB) GARFIELD CART PCB

Maxim Site Admin Joined: 19 Oct 1999 Posts: 14798 Location: London	Posted: Fri Sep 24, 2010 2:50 pm
	Discrete mapper hardware on GG carts has not been investigated very deeply. The image you show certainly looks a lot like a discrete mapper (the 315- chip) and an ordinary mask ROM (the MPR- chip). You'd need to analyse the connections (hopefully they mostly go connector <-> mapper <-> ROM) and figure out the meaning of any extra lines. Then you need to figure out the capabilities - I guess it must support 1MB ROMs, but maybe only slot 2 mapping, and no save RAM. If you have the capabilities, you can try desoldering the ROM and dumping it by trying common pinouts. That'd get you most of the way to your goal.

buggus Joined: 24 Oct 2009 Posts: 23	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Fri Sep 24, 2010 4:10 pm
buggus Joined: 24 Oct 2009 Posts: 23	Gerry_MAN wrote As you can see in the Photo there is a 315-5912 TQFP44 Mapper chip present. Would the ROM on this particular cart still use additional Mapper integration? or do you think that this is possibly a normal MASK ROM? Could you please scan the bottom side if this cart aswell?

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Fri Sep 24, 2010 4:43 pm
	Maxim wrote Discrete mapper hardware on GG carts has not been investigated very deeply. The image you show certainly looks a lot like a discrete mapper (the 315- chip) and an ordinary mask ROM (the MPR- chip). You'd need to analyse the connections (hopefully they mostly go connector <-> mapper <-> ROM) and figure out the meaning of any extra lines. Then you need to figure out the capabilities - I guess it must support 1MB ROMs, but maybe only slot 2 mapping, and no save RAM. If you have the capabilities, you can try desoldering the ROM and dumping it by trying common pinouts. That'd get you most of the way to your goal. Actually I did this with the Original SONIC Mask ROM (Photo Below) I wanted to make sure I didn't have any shorts present in my wiring from the ZIF socket. Using the Original ROM and connecting it to the ZIF would verify this. Which worked and the Sonic Game played fine on the GG, so no shorts in my soldering work to worry about. But of course the Game image dump I performed with the Willem didn't work when I tried it with MEKA EMU. I didn't think it would as there were too many uncertainties. I tried setting the Willem to an ATMEL AT49F020 a 256KB size Flash chip. I knew the Game Sonic was of this size so it seemed somewhat doable. Anyone know of a better setting to use with the Willem to perform a chip read? Now if the reason the Dump failed is due to the Embedded Mapper in the Sonic IC or most likely my Pin connections I'm not sure. I'll try it with the Garfield ROM and see if there is a Difference. I'm thinking of rigging the original ROM chips to the Logic analyzer now that I'll have them connected to the DIP 32 Adapters and have access to the Pins while in operation. Comparing the Waveforms between the MASK ROMs and my Target Flash chip of Choice. I'll monitor the ATMEL Flash chip while running a Flashed game on the Gameboy. This might show something we could not see otherwise. Although still not sure as they are two different Platforms. But I'm mainly interested in looking at the I/O pins to make sure I have those right. The Address & Data pins will no doubt be very different from one and other. Regardless, I see it as the more info I can collect the better. Any thoughts anyone? -Gerry mpr_sonic_002_669.jpg (75.79 KB) Sonic TSOP to DIP gg_flash_cart_003_478.jpg (124.4 KB) My GG Flash Cart mpr_sonic_004_651.jpg (134.79 KB) Willem - Sonic ROM Dump

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Fri Sep 24, 2010 5:02 pm
	I just de-soldered the MASK ROM but from looking at the Back side, it looks like I will have to remove the Mapper IC as well. There are a few through holes on the underside of the Mapper chip. I think I'll do that later tonight. -Gerry gg_garf_cart_001_151.jpg (102.73 KB) gg_garf_cart_002_836.jpg (86.25 KB)

Maxim Site Admin Joined: 19 Oct 1999 Posts: 14798 Location: London	Posted: Fri Sep 24, 2010 8:25 pm
	The ROM+mapper only exposes 16 address lines. You need to write to address $ffff to select the 16KB chunk to expose in the $8000-$bfff region. Your ROM dumper may not be able to do that... The Garfield ROM is more likely to work.

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Fri Sep 24, 2010 9:27 pm
	Maxim wrote The ROM+mapper only exposes 16 address lines. You need to write to address $ffff to select the 16KB chunk to expose in the $8000-$bfff region. Your ROM dumper may not be able to do that... The Garfield ROM is more likely to work. Ahh Ok, That is motivating. I'll try that out first. Now in my Diagram I had " NC" displayed for the ROM Pins that were literally Unconnected and had not trace lines connected to them at all. But on this particular Cartridge they are all being used. They are probably the additional address pins right? Any idea what they would be? -Gerry

buggus Joined: 24 Oct 2009 Posts: 23	Posted: Fri Sep 24, 2010 10:49 pm
buggus Joined: 24 Oct 2009 Posts: 23	I checked, there is NO standard mask rom with 64mbit / 1mbyte in a 32pin package. Judgeing on the needed address lines and the pcb layout, i suspect this is again a custom chip, which looks like it obeys mostly the pinout of the rom with mapper chip. They seem to use the NC pins AND some of the pins previously used by the internal mapper as address lines. I would begin tracing the lines from the edge connector to the rom chip, ruling out as much pins as possible. whats then left unknown AND is ONLY connected to the suspected mapper chip can only be address lines and an /OE line. Find out the /OE pin (i reckon pin 22), and try to read the Chip. From the shifting in data blocks in the dump you can get the layout of the left address lines, compared to the known good dump. This makes sense because the paging window of SMS/GG is 16k, i see 7 lines going from mapper to rom, would count as 14 lines directly from the sms plus 6 from the mapper (7th is /OE), being 20 -> 1024kByte / 1Mbyte. But then, the chip marking starts with "MPR", maybe giving the impression its more then a rom...

Maxim Site Admin Joined: 19 Oct 1999 Posts: 14798 Location: London	Posted: Sat Sep 25, 2010 10:48 am
	All Sega ROMs (with and without integrated mappers) have MPR numbers, unique across all systems/arcade/etc from the beginning of time for a particular ROM image.

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Mon Sep 27, 2010 5:52 am
	Here are the Scans of the Garfield Cartridge PCB with all the chips removed. I'll be connecting the ROM to the Logic Analyzer tomorrow if all goes well. And I should probably put together a Schematic of this PCB also. One step at a time. :) -Gerry gg_garf_cart_no_ic_180.jpg (156.75 KB) gg_garf_cart_002_172.jpg (86.25 KB)

buggus Joined: 24 Oct 2009 Posts: 23	Posted: Mon Sep 27, 2010 9:55 pm
buggus Joined: 24 Oct 2009 Posts: 23	I suggest you a pinout you may want to try with a flash chip: 22 - /OE 29 - A14 30 - A15 31 - A17 01 - A19 02 - A18 03 - A16 There are now 5 pins left which go to the cut-out short circuits. Wonder what the mapper may be capable of...

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Tue Sep 28, 2010 5:43 pm
	Hey buggus, That's a great suggestion. After I mapped out all the PCB traces from this Garfield cart PCB, and along with your pin suggestions I came up with this Pinout diagram for the SEGA 315-5912 (TQFP44) Mapper Chip. Now certain Cartridge edge TAB's are connected by traces to both the Mapper and the ROM contacts, and some are not. There are some traces that will go from the Mapper chip to only the Cart edge TAB's or only to the ROM/Flash chip Pins. So I have labeled those accordingly in the Diagram. I labeled certain contacts with "Z80" to designate the connections that go only to the Cartridge edge TAB connections. In any case, It looks quite promising. I'll be rigging this up tonight with my Flash chip and hopefully confirm this diagram. I'll see how it goes. -Gerry SEGA 315-5912 - Mapper.jpg (183.76 KB)

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Wed Sep 29, 2010 3:05 pm
	Still non functional. Dang namit!! Tested all my wiring connections again and tried a new Flash chip. But still Zip!! I was waiting on getting my Logic Analyzer POD back from a friend so I just picked it up this morning and I'm going to be running some scans on the original ROM chips and hopefully I'll be able to see some logical solution. No Pun intended. :) I'll keep you all posted. Later Folks, -Gerry

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Fri Oct 01, 2010 1:11 am
	Hi Everyone, I just got another 1 MByte GG Cartridge in from a friend. The cart is entitled "Virtua Fighter Animation". It is exactly the same PCB as the Garfield 1MByte cart. PCB 171-7000A and it also has the 315-5912 Mapper chip. I'm getting another 1MB cartridge "Sonic Blast" in next week. I'm assuming this will be the same PCB as well. In any case, I just wanted to share that info. I'll post my progress as I go. Cheers Folks, -Gerry sega_vf_315_5912_004_133.jpg (90.66 KB) sega_vf_315_5912_003_208.jpg (157.96 KB) sega_vf_315_5912_002_138.jpg (96.03 KB)

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Mon Oct 04, 2010 3:21 pm Last edited by Gerry_MAN on Fri Oct 08, 2010 5:09 pm; edited 1 time in total
	Hey Folks, I got my Logic Analyzer up and running this Morning. I had some problems with the POD interface wires being too short so I built a 60 pin extension with some old ribbon cable and an old Genesis Cartridge to hold the connector PCB. I etched the PCB for the 60 pin connectors yesterday. It has worked out nicely so far. I'll be interfacing to various ROM chips today to check for pattern similarities in hopes to see something that may shed some light on how to get this sucker to work with a Flash chip. The photos Below are some shots of the Logic Analyzer and the connecting wires interfaced to one of the ROMs. I used a 32 Pin DIP socket and a pin array to give the POD interface connectors something to grip onto, and also to protect the delicate ROM TSOP pins from damage. Anyhow, I'll post my Scan results hopefully over the next couple of days. Later folks, -Gerry logic_analyzer_004_198.jpg (85.05 KB) logic_analyzer_003_214.jpg (86.48 KB) logic_analyzer_006_691.jpg (105.96 KB) gg_rom_2_la_001_530.jpg (53.86 KB) gg_rom_2_la_008_199.jpg (185.07 KB) gg_rom_2_la_013_201.jpg (170.92 KB)

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Thu Oct 07, 2010 2:50 pm
	I just got the "Sonic Blast" GG Cartridge in Today and as I suspected, the cartridge is identical to the other 1MB cartridges. You can see the Images below. -Gerry sonic_b_gg_cart_006_971.jpg (84.31 KB) sonic_b_gg_cart_004_529.jpg (108.41 KB) sonic_b_gg_cart_007_717.jpg (100.21 KB)

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Fri Oct 08, 2010 5:00 pm Last edited by Gerry_MAN on Mon Oct 11, 2010 7:54 am; edited 1 time in total
	Hey Folks, So I got my Logic Analyzer software configured and captured the first 4096 pulse samples off the Original ROM chip from the "Garfield" 1MByte Cartridge. I put all the Trace Data into separate fields. Such as "Address", "Data" and also had the designated /OE and /CE lines displayed. But I also separated all the Pins and show their Logic Waveforms individually. Now remember this is only the first 4096 Pulses off all the pins on the ROM chip so that is essentially a fraction of a second of activity. So now the task is to try and analyze the individual Waveform patterns and see if they make sense as the pin Labels they have been given. /OE, /CE and such specifically. So now the Pictures below are the 4096 Samples in sections and I took Screen captures as I scrolled across the Waveforms. Observe the Scroll bar indicator at the bottom of the Captured images to see the location of the samples in view. The First image I have zoomed in closely so you can see the detail in the Pulses and I also show the Address and DATA Bus's in HEX. The other waveform images are more of a wider view so they detail is less but still able to view. In any case I'm still playing with the Analyzer unit and will be taking more captures and also performing some Pattern Recognition scans. Take a look at the Images and let me know your thoughts. Regards, -Gerry ******************************************************************************** Analyzer Capture Images Re-posted further down with corrections and additional Pin information. ********************************************************************************

Charles MacDonald Joined: 28 Sep 1999 Posts: 1200	Posted: Fri Oct 08, 2010 10:36 pm
Charles MacDonald Joined: 28 Sep 1999 Posts: 1200	There's so much information missing here that it's hard to tell what the goal is. Are the signal numbers ROM pins 1 through 32? Shouldn't the signal names list what Z80 or mapper chip pin/signal they connect to as well? I see the ROM, but where is the mapper chip? If it isn't present what did you connect the unknown ROM pins to? There's no way to tell what portions of the dump correspond to cartridge access versus Z80 RAM or I/O access. The signal that is ROM CE from the cartridge port should be labeled to indicate that. Is /RD the same as Z80 /RD? Is /OE also /RD, if so why is it included?

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Sat Oct 09, 2010 1:03 am Last edited by Gerry_MAN on Sun Oct 10, 2010 6:10 am; edited 1 time in total
	Charles MacDonald wrote There's so much information missing here that it's hard to tell what the goal is. Are the signal numbers ROM pins 1 through 32? Shouldn't the signal names list what Z80 or mapper chip pin/signal they connect to as well? I see the ROM, but where is the mapper chip? If it isn't present what did you connect the unknown ROM pins to? There's no way to tell what portions of the dump correspond to cartridge access versus Z80 RAM or I/O access. The signal that is ROM CE from the cartridge port should be labeled to indicate that. Is /RD the same as Z80 /RD? Is /OE also /RD, if so why is it included? Hi Charles, I guess I jumped the Gun here. I just got the Logic analyzer working and that's where I was at. I'll add that Pin info for you and perform a new scan. All the Data here is off the ROM chip connections from the 1MByte Cartridge "Garfield". My goal is to confirm the Designated Pins for the ROM contacts on the GG cart. As you know Datasheets cease to exist to confirm the pins. I have used the Provided GG Cartridge Pinout diagrams from SMSPower and backtraced to the ROM and derived a ROM Pinout like most of us do. I posted the diagram previously. Now regardless of my attempts to interface a Flash or EEPROM in place of the ROM, it still does not function. As seen in the previous posts some people are speculating that the ROM chips on GG carts are a ROM and Mapper chip integrated together into one. I was hoping to try and determine if this was the case with the 1MByte cartridges such as "Garfield" which use the 315-5912 Mapper chip. I had hopes that it was a mapper and ROM completely separate from each other. But still not sure. And yes, Pins 1-32 are the ROM pins individually. But I also displayed them together as a BUS labeled as ADDRBUS, DATABUS (these are displayed in HEX.) I also show /OE, /CE of the ROM. My Goal is a GG Flash cart of course, and hoping to use an official GG cartridge and Modify it by simply interfacing a Flash chip. As opposed to etching my own PCB and using other Mapper chips. But I may end up having to do this if all GG carts are in fact all designed around using a ROM+Mapper integrated Chip. So Charles, Thanks so much for the input here. Sorry for the lack of info though. I'll make sure I add that before the next post. I guess I'll get back to the Lab. Cheers! -Gerry

RetroSpark Joined: 08 Dec 2005 Posts: 488 Location: Melbourne, Australia	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Sat Oct 09, 2010 2:13 am
	Quote 1MByte cartridges such as "Garfield" which use the 315-5912 Mapper chip. I had hopes that it was a mapper and ROM completely separate from each other. It seems highly likely to me that where the 315-5912 is present, the other chip is just a normal ROM. Quote My goal is to confirm the Designated Pins for the ROM contacts on the GG cart. OK, let's assume the other chip is indeed a normal ROM. In "Garfield", it will be a 32-pin 1MByte ROM chip. This means that there will be 20 address lines and 8 data lines, leaving just 4 unknown pins. It should be straightforward to trace the address and data lines to derive a mostly-complete ROM pinout. I would do it, but it seems you already have: Quote I have [...] derived a ROM Pinout like most of us do. I posted the diagram previously. Maybe I've missed something, but I can't see where you've posted a proposed ROM pinout - only one for the 315-5912.

RetroSpark Joined: 08 Dec 2005 Posts: 488 Location: Melbourne, Australia	Posted: Sat Oct 09, 2010 6:39 am
	I took the time to trace the lines on your PCB images. You'll be pleased to hear that I agree with your proposed pinout for the 315-5912. However, I would say that we have no way of knowing which is which among the connections which do not go to the cartridge connector - those in red in your diagram. Do you agree that the ROM pinout implied by the 315-5912 pinout is as follows? The 7 pins marked "???" are A14-A19 and /OE (in some order): 1: ??? 2: ??? 3: ??? 4: A12 5: A7 6: A6 7: A5 8: A4 9: A3 10: A2 11: A1 12: A0 13: D0 14: D1 15: D2 16: GND 17: D3 18: D4 19: D5 20: D6 21: D7 22: ??? 23: A10 24: /RD 25: A11 26: A9 27: A8 28: A13 29: ??? 30: ??? 31: ??? 32: Vcc I assume you have created a flash cartridge using this pinout, with A14-A19 and /OE as in your diagram? And this flash cartridge is not working? If so, the cause may well be that the "???" connections are in the wrong order. Here's what I would do to try and figure out the correct order: The first step is to find out which ROM pin is /OE. To do this, try taking the Garfield ROM chip and grounding A0-A13 and /RD. Attach the "???" pins to Vcc. In other words, connect up the ROM as follows: GND: Pins 4-12 (inclusive), 16, 23-28 Vcc: Pins 1-3, 22, 29-32 Then, watch what happens to the data lines (13-15 and 17-21) as you try grounding pins 1, 2, 3, 22, 29, 30, 31, one at a time. One of these combinations should produce a valid byte on the data lines. Which combination is it? What is the output byte? Maxim wrote Then you need to figure out the capabilities - I guess it must support 1MB ROMs, but maybe only slot 2 mapping, and no save RAM. Using the MEKA debugger, we can see that Garfield uses mapping in slots 1 and 2, so the 315-5912 presumably supports at least those slots. I guess once Gerry_MAN has his flash cart working, we might be able to find out about slot 0 mapping using PFR Detect. As for save RAM, clearly it is not supported on this PCB, but the 315-5912 may still support it - perhaps this is the purpose of (some of) the unconnected pins 21 to 25?

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Sat Oct 09, 2010 10:01 pm Last edited by Gerry_MAN on Mon Oct 11, 2010 2:51 pm; edited 1 time in total
	Paul Baker wrote I took the time to trace the lines on your PCB images. You'll be pleased to hear that I agree with your proposed pinout for the 315-5912. However, I would say that we have no way of knowing which is which among the connections which do not go to the cartridge connector - those in red in your diagram. Do you agree that the ROM pinout implied by the 315-5912 pinout is as follows? The 7 pins marked "???" are A14-A19 and /OE (in some order): 1: ??? 2: ??? 3: ??? 4: A12 5: A7 6: A6 7: A5 8: A4 9: A3 10: A2 11: A1 12: A0 13: D0 14: D1 15: D2 16: GND 17: D3 18: D4 19: D5 20: D6 21: D7 22: ??? 23: A10 24: /RD 25: A11 26: A9 27: A8 28: A13 29: ??? 30: ??? 31: ??? 32: Vcc I assume you have created a flash cartridge using this pinout, with A14-A19 and /OE as in your diagram? And this flash cartridge is not working? If so, the cause may well be that the "???" connections are in the wrong order. Here's what I would do to try and figure out the correct order: The first step is to find out which ROM pin is /OE. To do this, try taking the Garfield ROM chip and grounding A0-A13 and /RD. Attach the "???" pins to Vcc. In other words, connect up the ROM as follows: GND: Pins 4-12 (inclusive), 16, 23-28 Vcc: Pins 1-3, 22, 29-32 Then, watch what happens to the data lines (13-15 and 17-21) as you try grounding pins 1, 2, 3, 22, 29, 30, 31, one at a time. One of these combinations should produce a valid byte on the data lines. Which combination is it? What is the output byte? Maxim wrote Then you need to figure out the capabilities - I guess it must support 1MB ROMs, but maybe only slot 2 mapping, and no save RAM. Using the MEKA debugger, we can see that Garfield uses mapping in slots 1 and 2, so the 315-5912 presumably supports at least those slots. I guess once Gerry_MAN has his flash cart working, we might be able to find out about slot 0 mapping using PFR Detect. As for save RAM, clearly it is not supported on this PCB, but the 315-5912 may still support it - perhaps this is the purpose of (some of) the unconnected pins 21 to 25? Hi Paul, The Pinouts you provided for the Garfield ROM chip are exactly as I had laid out. Here is my Pinout Diagram for the GG Garfield ROM chip (MPR-18201-S) that I neglected to post. Sorry about that. I've been posting in many forums so I got mixed up and assumed I had posted it earlier. http://www.digital-circuitry.com/IMAGES/My%20LAB/Sega/GameGear/1200%20X%20960/GG_TSOP_ROM_001.jpg Quote The 7 pins marked "???" are A14-A19 and /OE (in some order): Yes I agree. And I have been trying various combinations with these pins but have yet to find any that work. I will try to determine the ??? pins for the ROM chip by the procedures you gave.Thanks for that suggestion. :) I also just updated the 315-5912 Mapper Pinout diagram. Here is the link below: http://www.digital-circuitry.com/IMAGES/My%20LAB/Sega/GameGear/1200%20X%20960/SEGA%20315-5912%20-%20Mapper_003.jpg I'll hopefully have some time in the Lab later tonight. I'll try and a few things and also update my Logic analyzer Scan images with more detailed Info. Again I really appreciate the help and advice here from everyone. And if you see any inconsistencies with my diagrams let me know. Talk to you all later, -Gerry *********************************************************************** Pinout Diagram for the GG Garfield ROM chip (MPR-18201-S) HAS BEEN UPDATED. *********************************************************************** SEGA 315-5912 - Mapper_003.png (65.83 KB) GG_TSOP_ROM_001.jpg (119.4 KB)

RetroSpark Joined: 08 Dec 2005 Posts: 488 Location: Melbourne, Australia	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Sun Oct 10, 2010 2:07 am
	Gerry_MAN wrote The Pinouts you provided for the Garfield ROM chip are exactly as I had laid out. Good. This means that the pinouts we have so far (for both ICs) are probably correct, since two of us have come to the same results independently. Now, to find out what the ??? pins are... Quote I will try to determine the ??? pins for the ROM chip by the procedures you gave.Thanks for that suggestion. :) I hope it works. That is, I hope you get a good data byte from one of the combinations. That will tell us which ROM pin is /OE. From there, I can put together a similar procedure to find out the exact order of address lines. Quote And if you see any inconsistencies with my diagrams let me know. I spotted a few small issues with GG_TSOP_ROM_001.jpg: - Pin 24, /RD, should be marked with a * - it connects to both the Z80 and the 315-5912. - The description of the green pins should probably read "Data Output" rather than "Data I/O". - Presumably the copyright date should be 2010 instead of 2001?

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Sun Oct 10, 2010 6:17 am
	Updated the GG_TSOP_ROM_001.jpg: image link. Thanks Paul. -Gerry GG_TSOP_ROM_001.jpg (119.4 KB)

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Mon Oct 11, 2010 9:10 am
	Hey Folks, So I have performed a new scan with the Logic Analyzer and I added some more information regarding the PIN connections. Figured you all might want to have a look at my progress just for kicks. This scan is once again taken from the 1Mbyte Garfield ROM. Pins 1-32. All pins are scanned using positive logic. I also have the Address Bus and Data Bus displayed as its own field in HEX format. Also, the scans I take are only 4096 samples and is roughly around 800uS of activity and I only show a small portion of this capture in the photos. Now I have the pins labeled as such: 5_R_A7* - 5 being the PIN # - R indicating it as a ROM pin - A7 being the PIN function (Address BUS 7) - the ( * ) indicates the pin connects to both the Cartridge edge (Z80) and the 315-5912 Mapper IC. - The ( ** ) indicates the PIN connects only to the 315-5912 Mapper IC only. These pins are of course the ?? unknown pins as we have no reference to identify them. I still have to go through the steps Paul suggested to identify the ?? pins but figured I would post the new updated scans first for you all to have a look. The First 3 capture images are zoomed in quite close for added detail. The last image is a zoomed out view, to give you a wider perspective of the ROM activity. I remarked how the ?? PIN 31 is in a constant logic HIGH state. This could possibly be a /WR function could it not? Although ROMs of course do not have this feature. It just caught my eye and it made me think. I'll be posting Scans of the ROM activity during the Test that Paul suggested and get some feedback from everyone. Hopefully I'll have some time to play in the Lab tomorrow. Talk to you later, -Gerry GG_ROM_1MB_002A.jpg (355.58 KB) GG_ROM_1MB_002B.jpg (355.55 KB) GG_ROM_1MB_002C.jpg (354.67 KB) GG_ROM_1MB_002D.jpg (433.56 KB)

Charles MacDonald Joined: 28 Sep 1999 Posts: 1200	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Tue Oct 12, 2010 12:34 am
Charles MacDonald Joined: 28 Sep 1999 Posts: 1200	If you have a device programmer I would recommend connecting the ROM to a socket with the known address signals, CS, and OE hooked up, and then tie the unknown pins to sequential high-order address lines, each through a series resistor (say 4.7K) if you want to be careful. If then you then dump the ROM as whatever type has that many address lines (27C080), you'll exercise all possible states of the unknown pins. That way you (or any of us) could examine the dump and figure out which pin corresponds to what.

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Tue Oct 12, 2010 4:56 am
	Charles MacDonald wrote If you have a device programmer I would recommend connecting the ROM to a socket with the known address signals, CS, and OE hooked up, and then tie the unknown pins to sequential high-order address lines, each through a series resistor (say 4.7K) if you want to be careful. If then you then dump the ROM as whatever type has that many address lines (27C080), you'll exercise all possible states of the unknown pins. That way you (or any of us) could examine the dump and figure out which pin corresponds to what. Hi Charles, I appreciate the help. You guys know allot more than I do when it comes to the "code" so I'll let you all take a shot at examining the DUMP data. I'm in the midst of trying a few things but as soon as I'm done, I'll hook the chip up to the Universal Programmer and download the data to a .rom file for your review. Hopefully soon we will have this 1MByte Programmable cartridge up and running. One step at a time. :) Regards, -Gerry

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Tue Oct 12, 2010 5:57 am
	Charles MacDonald wrote If you have a device programmer I would recommend connecting the ROM to a socket with the known address signals, CS, and OE hooked up, and then tie the unknown pins to sequential high-order address lines, each through a series resistor (say 4.7K) if you want to be careful. If then you then dump the ROM as whatever type has that many address lines (27C080), you'll exercise all possible states of the unknown pins. That way you (or any of us) could examine the dump and figure out which pin corresponds to what. In one of the previous posts "buggus" had made this same suggestion. Quote I checked, there is NO standard mask rom with 64mbit / 1mbyte in a 32pin package. Judgeing on the needed address lines and the pcb layout, i suspect this is again a custom chip, which looks like it obeys mostly the pinout of the rom with mapper chip. They seem to use the NC pins AND some of the pins previously used by the internal mapper as address lines. I would begin tracing the lines from the edge connector to the rom chip, ruling out as much pins as possible. whats then left unknown AND is ONLY connected to the suspected mapper chip can only be address lines and an /OE line. Find out the /OE pin (i reckon pin 22), and try to read the Chip. From the shifting in data blocks in the dump you can get the layout of the left address lines, compared to the known good dump. This makes sense because the paging window of SMS/GG is 16k, i see 7 lines going from mapper to rom, would count as 14 lines directly from the sms plus 6 from the mapper (7th is /OE), being 20 -> 1024kByte / 1Mbyte. But then, the chip marking starts with "MPR", maybe giving the impression its more then a rom... I got so wrapped up with the Logic analyzer that I forgot to try this. :) I'm thinking I may build a 512KB Flash cart first while continuing to work on this 1MByte version. And I'll go with using the SMS mappers like Maxim suggested in one of the first posts on this topic. Now by using a Mapper such as the 315-5235 would exclude some ROMs from functioning correct ?...... but having 512Kb of memory to play with is still pretty good. Still lots of fun if your playing around with your own code. But regardless, a 1Mbyte version is much more attractive so my money is on this one. I'm really curious if the Garfield cart is using a ROM+Mapper integrated chip and also the 315-5912 Mapper all in unison to achieve the 1MByte memory size? Guess I need to spend some more time in the Lab. Talk to you all later, -Gerry

buggus Joined: 24 Oct 2009 Posts: 23	Posted: Mon Nov 29, 2010 11:25 pm
buggus Joined: 24 Oct 2009 Posts: 23	Gerry_MAN wrote Guess I need to spend some more time in the Lab. Nothing new?

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Re: RE: SEGA GameGear Flash Cart - Custom Electronics Posted: Tue Nov 30, 2010 1:08 am
	buggus wrote Gerry_MAN wrote Guess I need to spend some more time in the Lab. Nothing new? I've been going insane fixing up my Mom's house as she is trying to sell. And also going nuts lining things up for the coming College semester. I hate dealing with the Red tape. Needless to say I've been busy with other things. However, I have been working on something and I will be placing a post in the coming weeks before Xmas with all the details. :) Nothing too crazy but still SMSPower worthy. I'll keep you posted. -Gerry

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	RE:SEGA GameGear Flash Cart - Custom Electronics Posted: Mon Dec 13, 2010 3:28 am
	Gerry_MAN wrote buggus wrote Gerry_MAN wrote Guess I need to spend some more time in the Lab. Nothing new? I've been going insane fixing up my Mom's house as she is trying to sell. And also going nuts lining things up for the coming College semester. I hate dealing with the Red tape. Needless to say I've been busy with other things. However, I have been working on something and I will be placing a post in the coming weeks before Xmas with all the details. :) Nothing too crazy but still SMSPower worthy. I'll keep you posted. -Gerry Well, as I mentioned earlier..... I would be making a Posting before the holidays so here it is. A 512Kb Flash cartridge for the SEGA Game Gear. Here is the New Topic thread if you haven't seen it yet: http://www.smspower.org/forums/viewtopic.php?t=12734 I'll continue to work on this 1 MegaByte version over the holidays. But this should keep us busy for the time being. :) Cheers! -Gerry

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	SEGA GG Cartridge PCB Scans Posted: Sat Feb 12, 2011 2:52 am
	Just adding another Rare GG PCB scan for reference. This PCB has RAM interconnected to the board, along with a Battery backup. I'm now able to officially confirm certain connections on the ROM by having access to available datasheets for the RAM chip that is interconnected. See the pictures below: -Gerry GG_ROM+RAM_ 007.jpg (217.9 KB) GG_ROM+RAM_ 003.jpg (175.41 KB) GG_ROM+RAM_ 005.jpg (191.29 KB)

Snow_Cat Joined: 07 Jan 2011 Posts: 47 Location: Ontario	Posted: Sun Feb 13, 2011 4:45 am
Snow_Cat Joined: 07 Jan 2011 Posts: 47 Location: Ontario	docs.google.com2010.09.20.wip.zip updated: 315-5912 8M mapper added Round testpad block of 2 and 3 added Slot drill and pad sizes reduced by 30% small drill footprint added for DIL42-6.07

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	I Finally Did it! Posted: Wed Feb 23, 2011 4:22 am
	Victory at last. A 1 Megabyte DIY SEGA Game Gear Flash cart is now possible !! :) Check out the details on the new Topic (Link Posted below:) http://www.smspower.org/forums/viewtopic.php?t=12931 Kind Regards, Gerry O'Brien

django Joined: 24 Oct 2005 Posts: 29	Posted: Sun Apr 24, 2011 6:32 pm
django Joined: 24 Oct 2005 Posts: 29	a bit late , but i want you to show my reengeneering method just using Paint. This three Pictures are just examples, can´t find a finished Sonic Blast PCB at the moment maybe i´ve done that but i think i hoped for having a better scan. As you can see, it would be helpful if the scans are hrizontally and vertically good and the lightning should be optimized before taking a photo. personally I never got good results with a scanner, too many light. first step is to save the picture with 256 colors, than try it with 16 colors. If you have good contrast with 16 colors you can use the fill option and are finish within minutes. If not, it´s helpfull to redraw the wires with different colors, ending with 2 colors. I prefer to use eagle from cadsoft to do my pcbs, it´s the most used hobbyist ecad software used here in germany, do their free version. eagle has a bitmap import funtion so you open a new board, make a new layer and give it a dark color e.g. dark blue for the bottom layer. than you can add the parts and the edge connector to the schematic just use the "ic-packages" make the nets and route them with your imported bitmap as background. Sonic Blast GG Top-paint.PNG (161.59 KB) Sonic Blast GG Bottom-paint.PNG (129.36 KB) gg_romram_003_297-256colors-paint.PNG (291.6 KB)

Gerry_MAN Joined: 11 Dec 2009 Posts: 204 Location: Canada	Posted: Sun Apr 24, 2011 7:01 pm
	Nicely Done! I agree, the Cadsoft Eagle software is my current PCB design software of choice. Although, I am starting to use the National Instruments PCB layout software. The main advantage for this software is having a seamless transition from the Schematic design entry and Simulation, using the NI "Multisim" simulator; and then going directly to the PCB layout all in one software package. This will save you allot of time overall and is simply a much more efficient way of doing things in my opinion. So I'm in the midst of switching over to the NI software. Anyhow, If you get a finished PCB layout for this cart, please post it here. I'd love to see it. Cheers! -Gerry

Agent24 Joined: 28 Mar 2011 Posts: 88	Posted: Sun Apr 24, 2011 10:22 pm
Agent24 Joined: 28 Mar 2011 Posts: 88	I think you need to desolder everything so you can be sure where the tracks go?? I don't like Eagle because it will let you put tracks too close together, violating the clearances you have set and you won't know about it until afterwards when you run DRC. I prefer KiCAD as DRC runs in realtime, if you try to lay a track too close to the other it will stop you doing it - Also it has an outline display to show where the clearance is so you know exactly what's going on.

Forums

View topic - SEGA GameGear Flash Cart - Custom Electronics