Problems with F-Zero and Super Mario World

A place that you can discuss reproduction of classic titles or "licensed-for-reproduction" homebrew for personal use.

Moderators: B00daW, Moderators

Forum rules
1. NO BLATANT PIRACY. This includes reproducing homebrew less than 10 years old, with the exception of free software.
2. No advertising your reproductions, with the exception of free software.
3. Be nice. See RFC 1855 if you aren't sure what this means.
Post Reply
soniccd123
Posts: 6
Joined: Sat Jun 27, 2020 12:01 pm

Problems with F-Zero and Super Mario World

Post by soniccd123 » Sat Jun 27, 2020 12:25 pm

Hello,

I'm a new user at NESdev but been reading the forums for some time. I've been making SNES and Sega Genesis repros for myself its been aproximately a year with sucess, but theres two games i'm having problems since the beginning and besides looking all over the internet, this site, data sheets, trying diferent roms, schematics and even cartridges photos, i've not been able to make them work with the cartridges i've made: F-Zero and Super Mario World.

For cartridges, i've been using a custom made lorom board that i designed, using a 27c322 EPROM as the maskrom and two 74hc245 bus transceivers to convert the 16bit data to two bytes. It doesnt have SRAM yet, and i thought that could be the cause of the problem, as some games like that use SRAM like Super Metroid boot in this configuration but have problems during gameplay (you can't go past the save selection screen).

That was what i've thought till i actually bought a Super Metroid repro and tried reprograming other games in it. The game played like a charm but after i finished it, i've decided to open the cartridge and see how the repro was made. It uses a SHVC-1A3M-30 as donor board and has a 29L3211 Flash installed with a adapter over the maskrom. I reprogramed the 29L3211 with other games and they all worked well, tried Actraiser and it plays and saves without problems. But then i decided to try SMW and F-Zero to see if the presence of SRAM in this cart would made them work.

It didn't.

As for what is happening when i boot the games, F-Zero just show a black screen and no audio. Super Mario World works as far as Mario hits a block or gets a coin, when either happen, the game halts with the screen full of yellow graphical glitches and the music continue to play.

I've programed the Flash memory a few times to see if maybe it was some corrupted data, even dumped the cart and played the dump in a emulator, which worked just fine.

I've been searching all over the internet to see if anyone have had the same problem, i've found two posts on gamefaqs of people complaining of the same error on SMW with original cartridges, but without a solution. I've thought maybe my SNES is dying, but all other games work fine with no bugs, even those that use mode 7 (which i though could be the problem for F-zero).

Has anyone experienced something like this? I'm sorry to bother with this as forum newbie, but I've came to a point that i don't know what can be happening. I've searched in the NESdev forum but found no information regarding similar problems

Thanks

Soniccd123

lidnariq
Posts: 9417
Joined: Sun Apr 13, 2008 11:12 am
Location: Seattle

Re: Problems with F-Zero and Super Mario World

Post by lidnariq » Sat Jun 27, 2020 12:42 pm

soniccd123 wrote:
Sat Jun 27, 2020 12:25 pm
has a 29L3211 Flash installed with a adapter over the maskrom.
Be very careful with this part. That is a 3V part, and it requires extra hardware to safely use in a SNES. Without the extra hardware, you could find yourself destroying your SNES.
to see if the presence of SRAM in this cart would made them work.
Some SNES games care about the exact size of the RAM on the cart, and will generate an error if too much RAM is present, or buggy behavior if too little is present.

soniccd123
Posts: 6
Joined: Sat Jun 27, 2020 12:01 pm

Re: Problems with F-Zero and Super Mario World

Post by soniccd123 » Sat Jun 27, 2020 1:18 pm

lidnariq wrote:
Sat Jun 27, 2020 12:42 pm
soniccd123 wrote:
Sat Jun 27, 2020 12:25 pm
has a 29L3211 Flash installed with a adapter over the maskrom.
Be very careful with this part. That is a 3V part, and it requires extra hardware to safely use in a SNES. Without the extra hardware, you could find yourself destroying your SNES.
to see if the presence of SRAM in this cart would made them work.
Some SNES games care about the exact size of the RAM on the cart, and will generate an error if too much RAM is present, or buggy behavior if too little is present.
Yes, i found the 29L3211 quite strange, i'm actually surprised that it works, the adapter has a voltage regulator to supply the chip 3.3V, but its strange that no level shifter is used in the chip I/O and the chip has not died.

I might try to pull up the last SRAM adresses so the games think it has the right amount of SRAM, this cart has 64k SRAM and the F-Zero/SMW uses 16.

lidnariq
Posts: 9417
Joined: Sun Apr 13, 2008 11:12 am
Location: Seattle

Re: Problems with F-Zero and Super Mario World

Post by lidnariq » Sat Jun 27, 2020 2:25 pm

soniccd123 wrote:
Sat Jun 27, 2020 1:18 pm
Yes, i found the 29L3211 quite strange, i'm actually surprised that it works, the adapter has a voltage regulator to supply the chip 3.3V, but its strange that no level shifter is used in the chip I/O and the chip has not died.
It takes a while. And it's not only a problem for the chip, but also the parts inside the SNES. The overvoltage diodes on the flash are only rated for 0.5V (so if the regulator is actually 3.3V, 3.8V), and the bus drivers on the SNES's address and data bus aren't perfect (and sag some as a function of amount of current). The problem is less that the flash can be damaged (although it probably is), and more that the bus drivers in the SNES aren't rated for this kind of usage, and so it's them that can fail, and SNES parts aren't trivially replaced.

soniccd123
Posts: 6
Joined: Sat Jun 27, 2020 12:01 pm

Re: Problems with F-Zero and Super Mario World

Post by soniccd123 » Sat Jun 27, 2020 3:29 pm

lidnariq wrote:
Sat Jun 27, 2020 2:25 pm
soniccd123 wrote:
Sat Jun 27, 2020 1:18 pm
Yes, i found the 29L3211 quite strange, i'm actually surprised that it works, the adapter has a voltage regulator to supply the chip 3.3V, but its strange that no level shifter is used in the chip I/O and the chip has not died.
It takes a while. And it's not only a problem for the chip, but also the parts inside the SNES. The overvoltage diodes on the flash are only rated for 0.5V (so if the regulator is actually 3.3V, 3.8V), and the bus drivers on the SNES's address and data bus aren't perfect (and sag some as a function of amount of current). The problem is less that the flash can be damaged (although it probably is), and more that the bus drivers in the SNES aren't rated for this kind of usage, and so it's them that can fail, and SNES parts aren't trivially replaced.
Understood, i will stop using this cart then.

soniccd123
Posts: 6
Joined: Sat Jun 27, 2020 12:01 pm

Re: Problems with F-Zero and Super Mario World

Post by soniccd123 » Fri Jul 10, 2020 5:44 pm

Hello, i just wanted to report that i've tested the carts on another SNES and they worked. My SNES has a SHVC-CPU-01 motherboard and the other one a SNS-CPU-RGB-01. I don't know if this has any influence in the problem and really don't know what is happening, maybe my SNES is dying? I've tried the Burn-In cartridge and it passed all the tests, so i find this problem really bizarre. Only F-Zero and SMW have problems in this SNES unit, all other games i've tested (arround 20 games) have worked.

Just wanted to report, maybe it can help someone somehow. Thanks for all the replies.

calima
Posts: 1130
Joined: Tue Oct 06, 2015 10:16 am

Re: Problems with F-Zero and Super Mario World

Post by calima » Sat Jul 11, 2020 12:51 am

F-zero uses an expansion chip, maybe it pulls more power than your other carts. SMW doesn't use one though. Are your other 20 carts plain or with chips?

edit: Oops, remembered wrong. F-zero wasn't a DSP game.

Post Reply