Forums

Sega Master System / Mark III / Game Gear
SG-1000 / SC-3000 / SF-7000 / OMV
Home - Forums - Games - Scans - Maps - Cheats - Credits
Music - Videos - Development - Hacks - Translations - Homebrew

View topic - rom 32K 48K QUESTION

Reply to topic
Author Message
  • Joined: 13 Aug 2021
  • Posts: 25
Reply with quote
rom 32K 48K QUESTION
Post Posted: Tue May 10, 2022 7:46 pm
PEOPLE RECENTLY SAYS A 32K 48K CARD WITHOUT MAPPER TO RUN GAMES. BUT I NOTE THAT ONLY 32K WORKS. 48K ROM NEITHER LIGHTING CANDLE BLACK BLUE RED START

I TESTED SOME STAY HERE ON THE FORUM NOTHING WORKED
SMS-basic-cart.jpg (13.93 KB)
SMS-basic-cart.jpg
48kcart_115.jpg (29.71 KB)
48kcart_115.jpg

  View user's profile Send private message
  • Joined: 29 Mar 2015
  • Posts: 40
  • Location: Japan
Reply with quote
Post Posted: Wed May 11, 2022 12:04 am
darleiv wrote
PEOPLE RECENTLY SAYS A 32K 48K CARD WITHOUT MAPPER TO RUN GAMES. BUT I NOTE THAT ONLY 32K WORKS. 48K ROM NEITHER LIGHTING CANDLE BLACK BLUE RED START

I TESTED SOME STAY HERE ON THE FORUM NOTHING WORKED

What SMS console model and region are you using to test?
If it is a non-Japanese console with a BIOS, is there a valid ROM header and checksum?

I would make sure to declare a 32K ROM size in the header. Even though the header can specify a ROM size of 48k, this size is documented as buggy here:
https://www.smspower.org/Development/ROMHeader
  View user's profile Send private message Visit poster's website
  • Joined: 13 Aug 2021
  • Posts: 25
Reply with quote
Post Posted: Fri May 13, 2022 2:59 pm
raphnet wrote
darleiv wrote
PEOPLE RECENTLY SAYS A 32K 48K CARD WITHOUT MAPPER TO RUN GAMES. BUT I NOTE THAT ONLY 32K WORKS. 48K ROM NEITHER LIGHTING CANDLE BLACK BLUE RED START

I TESTED SOME STAY HERE ON THE FORUM NOTHING WORKED

What SMS console model and region are you using to test?
If it is a non-Japanese console with a BIOS, is there a valid ROM header and checksum?

I would make sure to declare a 32K ROM size in the header. Even though the header can specify a ROM size of 48k, this size is documented as buggy here:
https://www.smspower.org/Development/ROMHeader


but 32k rom doesn't use header. connects if mo-7 in oe rom and ready - but rom 48k does not work
  View user's profile Send private message
  • Joined: 29 Mar 2015
  • Posts: 40
  • Location: Japan
Reply with quote
Post Posted: Tue May 17, 2022 12:16 am
darleiv wrote
but 32k rom doesn't use header. connects if mo-7 in oe rom and ready - but rom 48k does not work

So the 32k ROM you tested that works does not have the text "TMR SEGA" at address 7FF0? Could you confirm what SMS console model you have tested on then?

M0_7 is low from 0000-7FFF (32k) and M8_B is low from 8000-BFFF (16k). For a 48k ROM to work (really a 64k chip with the last 16k non-accessible) your wiring needs to also take M8_B into account, to enable the ROM for the full 48k address range of 0000-BFFF. (you could for instance combine M0_7 and M8_B together using a AND gate, so when either goes low the output is low)
  View user's profile Send private message Visit poster's website
  • Joined: 06 Mar 2022
  • Posts: 100
  • Location: London, UK
Reply with quote
Post Posted: Wed May 18, 2022 7:37 am
If I'm reading the sketch in the original post correctly, the OP does indeed and the M0-7 and M8-B lines, however, they also appear to feed into another and gate with the CE for the cart slot, which I suspect will have the effect of enabling the (presumably 64k) ROM at the same time as the onboard RAM.

The CE line needs to be fed into an or gate or an invertor or into another CS pin on the ROM if there's one available I think.
  View user's profile Send private message Visit poster's website
  • Joined: 13 Aug 2021
  • Posts: 25
Reply with quote
Post Posted: Thu May 19, 2022 2:57 am
[quote="raphnet"]

I'm from Brazil . this is sega tectoy
sega rom 32k the TMR is ignored when using the MO-7 even with bios - the 48k rom already made the circuit in the photo. i get a black screen even with tmr
  View user's profile Send private message
  • Joined: 05 Nov 2014
  • Posts: 424
  • Location: Auckland - NZ
Reply with quote
Post Posted: Thu May 19, 2022 4:10 am
The circuit you have looks to be active when ever /CE is active, so it will be conflicting with RAM. I think you might need to add an OR gate to fix that.

OR /CE with M0-7, OR /CE with M8-B.

AND the outputs of both OR gates and feed the output of the AND gate to ROM /CE.

If its a standard EPROM your using, you dont need to connect /WR to the rom at all. Just connect the /WE pin on the ROM to +5v.
  View user's profile Send private message
  • Joined: 14 Aug 2000
  • Posts: 684
  • Location: Adelaide, Australia
Reply with quote
rom 32K 48K QUESTION
Post Posted: Thu May 19, 2022 6:41 am
The logic circuit driving the EPROM /CE looks sound to me. It won’t conflict with RAM if these are three 2-input NAND gates.

What brand and model of EPROM are you using? This can matter because some EPROMs will not output data if the programming pins are not in the correct state, whereas some EPROMs don't care.
  View user's profile Send private message
  • Joined: 06 Mar 2022
  • Posts: 100
  • Location: London, UK
Reply with quote
Post Posted: Thu May 19, 2022 7:44 am
Ah yeah you're right async, they must all be nands - it wouldn't even get close to working if they were and gates!

Hey darliev, not sure if it's useful but maybe you can post a photo so we can see chip numbers and your circuit?
  View user's profile Send private message Visit poster's website
  • Joined: 13 Aug 2021
  • Posts: 25
Reply with quote
Post Posted: Tue May 24, 2022 5:55 pm
willbritton wrote
Ah yeah you're right async, they must all be nands - it wouldn't even get close to working if they were and gates!

Hey darliev, not sure if it's useful but maybe you can post a photo so we can see chip numbers and your circuit?


don't notice the bunch of jumpers, the board is for testing.
32k rom works perfectly
eprom sst39sf040-70/40 or w27c512
some don't even work
note that 48k rom only runs with sega mapper at least with me
schemaco.png (157.12 KB)
schemaco.png

  View user's profile Send private message
  • Joined: 14 Aug 2000
  • Posts: 684
  • Location: Adelaide, Australia
Reply with quote
rom 32K 48K QUESTION
Post Posted: Wed May 25, 2022 1:12 am
You have a 74HC08 chip where the board is silkscreened "7400". Is this intentional?

74HC08 is quad 2-input AND gates.
74HC00 is quad 2-input NAND gates.

Also, there should be at least a 0.1uF ceramic capacitor close to the EPROM and a 1uF~10uF electrolytic capacitor on the board for the power rail.
  View user's profile Send private message
  • Joined: 13 Aug 2021
  • Posts: 25
Reply with quote
Post Posted: Wed Jun 01, 2022 7:37 pm
asynchronous wrote
You have a 74HC08 chip where the board is silkscreened "7400". Is this intentional?

74HC08 is quad 2-input AND gates.
74HC00 is quad 2-input NAND gates.

Also, there should be at least a 0.1uF ceramic capacitor close to the EPROM and a 1uF~10uF electrolytic capacitor on the board for the power rail.


yes I tested it with the 2 , 7400 7408, so as the board is for testing I didn't put any filters. if you need to solder an axial one.
  View user's profile Send private message
  • Joined: 06 Mar 2022
  • Posts: 100
  • Location: London, UK
Reply with quote
Post Posted: Thu Jun 02, 2022 2:27 pm
It definitely won't work with a 74HC08, you need to use a 74HC00.

The schematic's moderately confusing, but from my understanding, to get it working as a 48K you need:

1. To install a 74HC00 on the reverse of the board - PROBLEM: you seem to have a 74HC08 instead

2. To select OE1 mode with the OE1/CE1 jumper on the reverse of the board underneath the 7400 - PROBLEM: you appear to have not selected _either_ of these options. You have soldered a jumper wire which appears disconnected.

3. To jump the CE1/GND connection on the front of the board - CORRECT: you have done this

4. To connect WE to +5V - CHECK THIS: I can't see the silkscreen clearly, but looks like on the front of the board, next to pin 1 of the 27C1001. Be sure you are connecting WE to +5V here!

5. To connect A15 to A15 on the jumper front of board, far left - CORRECT: you appear to have this connected.

6. To clear all other jumpers - PROBLEM: you appear to have connected M07 to OE1, front of board, top right hand corner. I think this is going to give you a problem, you are now driving OE1 from two places = bad news.

I will post a picture of what I think you need to fix. Please note, I'm very much guessing here based on the schematic and the photos, so no promises this is correct!

Also, on the capacitors as async points out, you might be very lucky and find it works without, but from experience you're going to want to install those straight away: it's so easy to do and you will be grateful to eliminate one more source of potential problems.
  View user's profile Send private message Visit poster's website
  • Joined: 06 Mar 2022
  • Posts: 100
  • Location: London, UK
Reply with quote
Post Posted: Thu Jun 02, 2022 2:50 pm
Suggested corrections attached.

I've not indicated where the caps go, because I'm not sure enough, but I rather suspect the unlabelled pair of pads next to pin 32 of 27C1001 is a space for a surface mount decoupling cap (0.1uF).

Async's right, you should also have a bulk cap (1uF) for the whole board, not sure where's best, you could put a through-hole one between pins 21 and 35 of the edge connector perhaps, although it might be a bit of a stretch for the legs...or else assuming the SMT pads near pin 28 of the 27C512 are also across the supply rails you might put something there.
corrections.jpg (629.26 KB)
corrections.jpg

  View user's profile Send private message Visit poster's website
Reply to topic



Back to the top of this page

Back to SMS Power!