OsmocomBB

h2. Winbond W56964

A "datasheet":https://engineering.purdue.edu/ece477/Webs/F06-Grp04/Datasheets%20sept24/W56964%20Docs/Document/DesignGuide/W56964%20CPU%20interface%20and%20registers%20rev.%20A1.pdf is available in a Purdue University "project report":https://engineering.purdue.edu/ece477/Webs/F06-Grp04/index.html. It is a bit discouraging that they apparently didn't get it to work.

There used to be application notes in "Google's cache":http://www.google.com/#q=W569+site:uglyduck.ath.cx. That "page":http://www.google.com/#q=CCChien0+site:uglyduck.ath.cx also had what Winbond calls _middleware_, i.e., the driver.

h2. Sunplus SPMA100B

No datasheet available and registers are different although the chip is pin-compatible to the above. Traces generated with the original firmware and register dumps during operation suggest:

<pre>

#define SPMA_SEQUENCER_FIFO	0xc5

#define SPMA_SEQUENCER_SPACE	0xc7	// decreases with every written byte

#define SPMA_SEQUENCER_STATUS	0xd6

#define SPMA_SEQ_FIFO_EMPTY	0x80

#define SPMA_SEQ_FINISHED	0x40

#define SPMA_SEQ_FAILED		0x20

0xe1

Power controls. No sound or clicks if sequence (0xff -> 0x7e -> 0x1e -> 0x04 -> 0x00) is wrong.

0xe3 ... 0xe6

Volume registers where 0=Mute, 1=-30dB ... 31=0dB in 1dB steps (equal to W56964)

</pre>

h2. Macronix MX92U832AZCG

Seems to be a product of "Modiotek":http://www.modiotek.com.tw/. No further information available.

h2. OKI ML2864

The "datasheet":http://read.pudn.com/downloads119/sourcecode/others/505180/SRTP/documents/datasheet/ML2864_000002_E.pdf explains the concepts a bit more detailed. The _Exclusive Format_ might solve the mystery of the data send to the SPMA100B.