--- a/pirelli/audio	Sat Mar 22 05:53:02 2014 +0000
+++ b/pirelli/audio	Tue Mar 25 08:51:32 2014 +0000
@@ -18,6 +18,34 @@
 Finally on L8 they go to the Winbond chip!  The speaker connection pins appear
 to be the two leftmost ones in the top row of 9 pins.
 
+Winbond chip connections
+
+Chip orientation: as viewed in steve-m's depopulated L8 picture, the pinout
+is a L/R flip of the diagram in the datasheet (top and bottom match up, left
+and right flipped).
+
+VREF appears to go to a cap to GND like in the reference schem.
+VDD has a fat trace going to a cap; there must be some via(s) to L7 for the
+actual power source, but nothing is visible.  Coord range to search:
+x between 3786 and 3975, y between 1205 and 1387.  Might be a bump at
+(3880,1380).  Probably connects to the "fat finger" trace on L7.
+
+LED output goes to exposed TP, no visible signs of going anywhere else.
+IRQ output goes to exposed TP at (4174,1316), look for it on L7.  Couldn't
+find it - maybe it isn't connected after all?  Will try programming the
+chip w/o interrupts.
+
+RST input: visible microvia at (4078,1357), found it on L7, goes to (4232,1515),
+looks like a microvia back to L8.  Found it on L8: it's an exposed test point
+with the microvia in its centre, and the trace from it goes to SPCA552E ball
+J7!  From the Nokia schem, it looks like this SPCA pin is an LCM output that
+can also be repurposed as a GPIO.
+
+CS input: visible microvia at (4325,1105) inside an exposed TP. On L7 the long
+trace ends at (1760,1385), looks like a possible microvia back to L8.
+Found it on L8: it connects to a pin on one of the little ICs under the LCD,
+suspected chip select decoder.
+
 Calypso-Iota Voice Band interface
 
 Tracing the Calypso voice output signal, starting from Calypso ball P14 (VDX).