SpritesMind.Net

And in C it's well known that FALSE is 0 and TRUE is not 0 :)
AFAIR, C defines TRUE as 1. But with 68k it's better to use -1 (#$FF, 255). Why? Because value -1 is invariant in logical and bitwise operators:
not( 0 ) = -1
not( FALSE ) = TRUE

68k has even special instructions Scc which set byte to ...

Once I learned programming in asm for Genesis reading these docs
http://www.ticalc.org/pub/text/68k/
You may read "68kguide.txt" first.

Everything okay:

When I wrote this they really didn't work

Bits 6 and 7 are very useful, as they allow the internal 14-bit output data from the operator unit to be read out over the status register

Awesome! I would like to know this in 2009 Regretfully, now I'm not very interested in all this stuff about YM.

Your links don't work...

Heh, in my tests I discovered that bit 5 in $2C makes channels silent. I tested Channel 1 in Alg 7. I didn't guess to test DAC mode :)

The FM test register at 0x21 has the upper 6 bits with a known function, with the lower 2 bits with no know function.
What are these functions excatly? I made ...

It's not released, it was compiled by MetalliC from emu-land
http://www.emu-land.net/forum/index.php ... #msg838076

This idea is not new Not hot-switching of carts nor "audio dumping". For example,
http://www.emu-land.net/forum/index.php ... 1753.0;all
(in Russian).

Hmm - does the FM channel still update even if it is set to PCM?
Yes, it does. FM Channel 6 and Reg $2A are independent of each other. "DAC mode" (Reg $2B) only chose output of one of them. If it is in "FM Mode" it outputs Channel 6 output. If it is in "DAC Mode", it outputs value stored in Reg ...

Hmm, my first thought was about real prescaler (regs $2D-2F)...

Example:
clock_rate = 7 670 000
channels = 6
operators per channel = 4
cycles per operator = 6
Channel_freq = 7670000 / 6 / 4 / 6 = 53263

If you increase number of channels without changing clock_rate, Channel_freq will decrease and ...

For values > 1 the formula should be
att += template_step * (not (att)) >> 4
Maybe it's a bit incorrect. It's hard to test increasings when template_step is 2 or more, because of pretty high values of AR to get it and pretty short periods between ATTACK steps.

template_step is taken from EG ...

they are paid at least $24K (in California, at least $34K, I believe)

For month or for year? In first case it's very high cost, but in second case it's not normal for such loaded work, imho.

EG uses next formula for ATTACK RATE = 1
att += not(att) asr 4

When I tried to get template of EG attack values on hardware I used modualtion of oscillator by sum of 2 operators (algo 5, as far as I remember). Modulators were at sin256 phase (maximum), frequency = 0, TL = 0, amplitude at start = 0 ...

I moved to another OS and another HDDs :) All in mess, so now I "lost" some of my tools.

Here is nlz of MUSHA by old version of Analyzer (which I found in one of my numerous folders; new version is somewhere in archives).
http://shedevr.org.ru/ghost/musha.rar
Simply disasm this ROM. My disassembler ...

No, we've discussed this. Glitch in DAC. Sine table is normal, I tested it using phase modulation.

Okay, I understand.
And if you run this test on MD1 you should see only 2 steps: from -1 to 0 and from 0 to +1. You won't see step from -2 from -1 because of glitch, so level "-1" will be last longer on X axis.
And step -1 to 0 will be HUGE (that is ladder effect).

Also, I noticed another glitch in ...