New Documentation: An authoritative reference on the YM2612

Huge · Post by **Huge** » Wed Apr 18, 2012 3:50 pm

I have vgm2mid, a version that dates to 2001. Also a copy of gym2mid somewhere, and a handful of midis converted by either of them. (too lazy to clean them up)

Damn, the vgm format is 10+ year old already...

Nemesis · Post by **Nemesis** » Wed Apr 18, 2012 11:28 pm

Also, does anybody have any links to good resources for learning how to read IC schematics? Looking at all the photos of the YM2612 has made me want to be able to analyze these circuits for myself.

I'm interested in this too. I've got some old IC design books from the 1970's and 1980's which seem to be a good resource. I think at the end of the day, you need to learn as much about how to design these circuits as possible. If you understand what the principles of the design are, you should be able to more easily recognize what individial elements on the circuit are.

King Of Chaos · Post by **King Of Chaos** » Thu Apr 19, 2012 3:36 pm

I didn't see it mentioned on any of the last few pages, but this topic might be of interest: http://forums.sonicretro.org/index.php?showtopic=28589

andlabs · Post by **andlabs** » Thu Apr 19, 2012 3:44 pm

There's a topic on that elsewhere here.

Nemesis: which books do you have? I guess it's time to actually learn computer engineering =P

Mamizou · Post by **Mamizou** » Thu Apr 19, 2012 5:07 pm

Hi there! I've been monitoring this thread for awhile now and updating the Wikipedia article on the YM2612 as this has progressed. Seeing the OPN2 finally decapped is really amazing, and it's interesting comparing the full die image to those of the decapped OPL2 and OPL3. Seeing a decapped VDP would be amazing as well, even one from a Model 2 Genesis/MD. Anyway, I'm curious if it's possible to get a good clean image of the decapped OPN2 for use on the Wikipedia article, like this? That way I can include it on the YM2612 article.

(It'd be even more awesome if, at some point in the future, a YM2608 and YM2203 get decapped, just to see the progression of development of the OPN family in general. I wish I had the hardware/money to donate. T-T Wishful thinking, hahaha.)

Anyway, thanks for all of the amazing decapping and photographing work, and I can't wait for more info to surface on my favorite FM synth chip family.

Huge · Post by **Huge** » Thu Apr 19, 2012 7:30 pm

You should send a PM to HardwareMan if you want to see any more chips decapped and have the cash to pay for it.

ICEknight · Post by **ICEknight** » Mon Apr 23, 2012 10:43 am

As long as it's not illegal or anything, how about creating a Kickstarter for decapping as many Mega Drive chips as possible, from every motherboard revision?

If many people show interest and the initial goal is met, more systems could be added, even (Mega CD, Mega 32X, SEGA Nomad, Laseractive SEGA pac, SEGA Pico, or anything related).

Oh, and hi.

Huge · Post by **Huge** » Mon Apr 23, 2012 8:34 pm

Finances are the main problem I think.

I'd chip in if we are talking about decapping Saturn stuff, by the way. I think that's the one that needs it the most. The Mame team was supposedly trying to decap the SH1, but nothing came out of that but a partial, low-res die shot.

sega16 · Post by **sega16** » Mon Apr 23, 2012 8:35 pm

Why would there even be a chance that decapping an old ic chip be illegal I don't understand were that would have came from.

Nemesis · Post by **Nemesis** » Wed Apr 25, 2012 3:14 am

There's no legality issue, in fact, reverse engineering itself isn't illegal, and we aren't even doing that, just "modifying" a physical device we legally own.

I two would love to see the Saturn chips decapped. I don't know if I've ever mentioned this publically before, but once my Mega Drive emulator is finally complete and released, with MegaCD and 32x support, I'll be turning my sights on the Saturn in a major way.

All that said, my ability to contribute financially to this effort at this point is virtually nill. Mortgage + kids, plus all the financial stress in the world today, well, I'm sure a lot of other people are in the same situation. It would be good to know just how much a decap like this costed all up, for the decap plus the photos, that way we would know the kind of money that needs to be raised. I mean, $50 for a chip is probably doable. $150 kind of puts it a lot further out of reach...

sigMate · Post by **sigMate** » Wed Apr 25, 2012 7:48 am

Hi guys,

I think I made a nice hardware implementation which could be used to experiment a little (undocumented registers, etc...). I just posted this : viewtopic.php?t=1122

Please tell me what you think

Thanks a lot !

sig

Huge · Post by **Huge** » Wed Apr 25, 2012 3:02 pm

Nemesis wrote:I two would love to see the Saturn chips decapped. I don't know if I've ever mentioned this publically before, but once my Mega Drive emulator is finally complete and released, with MegaCD and 32x support, I'll be turning my sights on the Saturn in a major way.

Don't know if I mentioned this (only to AamirM, I think), but if anyone is doing anything Saturn related work, I'd be glad to help in any way I can - I've a full library of games to test in emulators (all full quality bin/cue dumps), and an almost full library of hardware revisions to test applications on (the evolution of the hardware is mighty interesting & actually breaks some games).

I want to write down all documentations, but I still need to properly photo the machines I have with the lightbox I built - oh, and I want to write a CMS behind this too (not gonna cave in and use wordpress, oh no).

edit: perhaps soon it will be time to open a Saturn board here.
And here's the partial SH1 die shot:
http://decap.mameworld.info/2010/06/29/the-big-picture/

It's the Hitachi HDS6437097F CDB106 from the saturn... there are actually multiple revisions of this too... an earlier one: HD6437097F20 CDB105, and a later one, 315-5932 YGR022, which is basically that chip and the otherwise separate YGR019 disc controller under one ASIC.

Chilly Willy · Post by **Chilly Willy** » Wed Apr 25, 2012 6:29 pm

Decapping the SH1 or SH2 is a waste of time and money. Both chips are fully documented by Hitachi/Renesas, and it's not like the Saturn use "undocumented" features of the chips like older systems did on older CPUs (like the undocumented opcodes on the 6502). Put that money into decapping the chips that matter, like the VDP1 and VDP2!

Huge · Post by **Huge** » Wed Apr 25, 2012 6:55 pm

The Saturn uses a custom "SH1 + other things" though, and it's supposed to hold the details over the copy protection, which remains poorly understood even today (there is talk about not one but 3 security rings, two being red herrings).
And the entire CD Block is a black box that you cannot interface with, only with BIOS calls, last time I checked. The copy protection details may be in the YGR chip though. Thing is, no one knows anything about either of them, AND there is no way to access or work with any of them.

Compared to that, the VDP1/2 is almost fully documented.

If we'd skip the cd block, I'd put my money on the SCU. The DSP part has poor to missing documentation. Even if it only used by a few dozen games, it is the biggest unknown. Plus the SCU itself is responsible for a lot of timing related things.

Anyway, this is going off topic.

Mamizou · Post by **Mamizou** » Wed Apr 25, 2012 10:09 pm

Huge wrote:If we'd skip the cd block, I'd put my money on the SCU. The DSP part has poor to missing documentation.

On the topic of sound chips, the SCSP would be very interesting since it does both FM and PCM (I guess an operator would just read from a PCM buffer instead of the quarter-sine ROM). I think the Sonic Jam games used it directly instead of emulating a YM2612, although I could be totally wrong.