Saturn CD Block ROM dumped

I'll have to get back to you on 55/56, I haven't untangled all the relevant yet, and have no hardware to test on here. Meanwhile, in CDB105:

FF00: returns CR3:CR4 as CR1:2, CR3:4 (no side effects)
FF01: returns 8 RAM bytes from F000000 (start of SH-1 SRAM)
else: returns zeroes

Presumably it used to do more. The first half of the RAM locations (CR1:CR2) are likely blank, as the location referenced is only used in debug mode (which prevents multitasking/command mode from even starting). The latter half (CR3:CR4) contain a ROM checksum (DWORDwise sum), which would be interesting to catalogue if one couldn't dump the ROM. By my calculations (`perl -e 'undef$/;printf"%08X\n",unpack("%32L>*",<>)'`) a CDB105 has a sum of zero, and a YGR022 has 0x00000080.

 

Great work! Keep it up! Deunan is working on a drive-replacement for the Saturn. I bet he would love to see this information aswell.

 

Prototype board (with USB and microSD) is taped out today, and I've got some of the CDB strapped into an emulator rig now. Progress...

 

I think Deunan has a different approach than jhk since he didn't dump the CD block ROM but he is only analyzing the Saturn CD-ROM interface in a way similar to what this guy did some time ago:

http://web.archive.org/web/20121103092509/http://www.crazynation.org/SEGA/Saturn/cd_tech.htm

Anyway it's really exciting to see which developer will release his device for first, IMHO jhl will win the challenge since he started more ahead than others.

 

It's not a race... there are at least 3 different devices with quite different characteristics in the mix here. It makes me very happy to see such a healthy ecosystem. It does seem curious that all this has appeared in the last year after a fairly long hiatus, though - am I just missing what's been happening in between?

 

Probably the Sega Saturn is getting very older and need really new organs...
Joking part, I think this flourishing is due the introduction of programmable logical devices like CPLD and FPGA.

 

x

 

Alright, no problem. By the way, if you need something tested, let me know. I'm setup for code testing and I have code for issuing out custom CD block commands, etc.

So I'm trying to understand 0xFF, so basically with a 0xFF00 write to CR1, and CR3-CR4 write as well? Then data returned to CR1 and CR3? Or am I misunderstanding? And I guess 0xFF01 doesn't take any additional data from the other command registers.

Interesting, I guess the checksum comes in handy for doing QA testing.

Yeah, I don't see why people think there needs to be some kind of competition. If anything, I think it's great that there's so many people still seriously working on Saturn hacking. I won't deny I was pretty discouraged to see how big of a ghost town places like SegaXtreme or Yabause's forum have gotten lately. Nice to see there's people still busy. Also, I think it tends to come in cycles. I noticed already way back when I first got into Saturn hacking (in like 2001 or so) you'd see several talented people come in, do some awesome stuff, then move on. I'd say the cycles tend to be around 2-4 years for most.

 

You forgot :
3. the people who have the necessary knowledge for such projects are busy with family or real work



To jhl:
Do you know if it is possible to load custom code from custom MPEG cartridge to SH1 CPU without Saturn motherboard modification ?

 

Yes, that's what's going on in this picture:



A real MPEG card is combined with a USB interface to take over the SH1 (and then, in this case, dump it over USB). The card is just in the MPEG slot, the only motherboard connections are me debugging.

In essesnce, that's the board I'm putting together now, just with lots more shiny features

 

Basically, is the board on the picture as described on the diagram below ?

And can you remap SH-1 ROM to another address while mapping USB instead of SH-1 ROM ??
Please let me know if I am completely wrong

 

x

 

It is more or less that board (though that diagram doesn't match the Saturn exactly), and you can map in as much as you like. There are three available chip selects before you even need glue.


That box, if it's meant to represent the MPEG card, should have ~512kb ROM, 512kb RAM, an MPEG audio and an MPEG video chip.

 

x

 

today hardware is easy way to emu CD-Laser streaming
than just do samething look like Saturn dev , use retail hardware to mod , it should work

 

Really? Interesting. The Saturn itself checks the 512kb before allowing the MPEG ROM to boot, actually. Luckily they're crap at random numbers...

 

It'd be very appealing to plug something in and be able to run games off SD or some other type of storage. Maybe it would motivate more translations of games.

An exploit in how discs are authenticated and loaded would also be impressive. It reminds me of the PS2 and the exploit where you patch games to look like DVD movies.

 

That would be awesome. Off-topic: What would help with translations is better tooling for debugging (CyberWarriorX has made major contributions). Booting without having to burn a CD is a bonus.

The mere fact that we can now improve the documentation for CD-block is amazing.

 

x

 

I have a JVC RG-VC3. How I can check the ROM and RAM size?