diff --git a/config.in b/config.in
index 809eada..61b6bf3 100644
--- a/config.in
+++ b/config.in
@@ -8,6 +8,7 @@ comment "General Setup"
 	choice 'Target MCU'           \
  	  "ATmega8        atmega8     \
 	   ATmega32       atmega32    \
+	   ATmega32U4     atmega32u4  \
 	   ATmega328      atmega328   \
 	   ATmega328p     atmega328p  \
 	   ATmega644      atmega644   \
diff --git a/profiles/LoL-Shield_Leonardo b/profiles/LoL-Shield_Leonardo
new file mode 100644
index 0000000..3edf168
--- /dev/null
+++ b/profiles/LoL-Shield_Leonardo
@@ -0,0 +1,125 @@
+#
+# Automatically generated by make menuconfig: don't edit
+#
+
+#
+# General Setup
+#
+MCU=atmega32u4
+FREQ=16000000
+
+#
+# Borg Hardware
+#
+NUM_ROWS=9
+NUM_COLS=14
+NUMPLANE=3
+BORG_HW=HW_LOLSHIELD
+
+#
+# lolshield setup
+#
+USER_TIMER0_FOR_WAIT=1
+BRIGHTNESS=127
+
+#
+# Features
+#
+RANDOM_SUPPORT=y
+# LAP_TIME_EXTENSION is not set
+SCROLLTEXT_SUPPORT=y
+SCROLLTEXT_FONT=FONT_C64
+SCROLLTEXT_BUFFER_SIZE=128
+SCROLL_X_SPEED=20
+SCROLL_Y_SPEED=20
+SCROLLTEXT_TEXT="</#Borgware 2D"
+# RFM12_SUPPORT is not set
+JOYSTICK_SUPPORT=y
+# PARALLEL_JOYSTICK_SUPPORT is not set
+# NES_PAD_SUPPORT is not set
+# RFM12_JOYSTICK_SUPPORT is not set
+# HC165_JOYSTICK_SUPPORT is not set
+NULL_JOYSTICK_SUPPORT=y
+# CAN_SUPPORT is not set
+# MENU_SUPPORT is not set
+
+#
+# Games
+#
+GAME_TETRIS_CORE=y
+# GAME_TETRIS is not set
+# GAME_BASTET is not set
+# GAME_TETRIS_FP is not set
+# GAME_SPACE_INVADERS is not set
+# GAME_SNAKE is not set
+GAME_BREAKOUT=y
+
+#
+# Animations
+#
+ANIMATION_SCROLLTEXT=y
+ANIMATION_SPIRAL=y
+SPIRAL_DELAY=5
+ANIMATION_JOERN1=y
+ANIMATION_SNAKE=y
+SNAKE_CYCLE_DELAY=100
+SNAKE_TERMINATION_DELAY=60
+SNAKE_MAX_LENGTH=64
+SNAKE_MAX_APPLES=10
+ANIMATION_CHECKERBOARD=y
+ANIMATION_FIRE=y
+FIRE_S=30
+FIRE_N=5
+FIRE_DIV=44
+FIRE_DELAY=50
+FIRE_CYCLES=800
+ANIMATION_MATRIX=y
+MATRIX_STREAMER_NUM=30
+MATRIX_CYCLES=500
+MATRIX_DELAY=60
+ANIMATION_RANDOM_BRIGHT=y
+ANIMATION_STONEFLY=y
+ANIMATION_FLYINGDOTS=y
+ANIMATION_GAMEOFLIFE=y
+GOL_DELAY=100
+GOL_CYCLES=360
+ANIMATION_BREAKOUT=y
+# ANIMATION_MHERWEG is not set
+ANIMATION_MOIRE=y
+ANIMATION_LTN_ANT=y
+# ANIMATION_TIME is not set
+TIME_MASTER_ADDR=00
+TIME_UPDATE_TIMEOUT=50
+ANIMATION_BMSCROLLER=y
+# ANIMATION_LABORLOGO is not set
+ANIMATION_AMPHIBIAN=y
+# ANIMATION_LOGO_OOS is not set
+ANIMATION_FAIRYDUST=y
+
+#
+# Fixed-point math patterns
+#
+ANIMATION_PLASMA=y
+FP_PLASMA_DELAY=10
+ANIMATION_PSYCHEDELIC=y
+FP_PSYCHO_DELAY=25
+ANIMATION_BLACKHOLE=y
+ANIMATION_SQUARES=y
+# ANIMATION_TESTS is not set
+# ANIMATION_OFF is not set
+
+#
+# small Animations
+#
+# SMALLANIMATION_ROWWALK is not set
+SMALLANIMATION_ROWWALK_SPEED=50
+SMALLANIMATION_ROWWALK_COUNT=10
+# SMALLANIMATION_COLWALK is not set
+SMALLANIMATION_COLWALK_SPEED=50
+SMALLANIMATION_COLWALK_COUNT=10
+# SMALLANIMATION_ROWBOUNCE is not set
+SMALLANIMATION_ROWBOUNCE_SPEED=50
+SMALLANIMATION_ROWBOUNCE_COUNT=10
+# SMALLANIMATION_COLBOUNCE is not set
+SMALLANIMATION_COLBOUNCE_SPEED=50
+SMALLANIMATION_COLBOUNCE_COUNT=10
diff --git a/profiles/LoL-Shield b/profiles/LoL-Shield_Uno-Duemilanove-Diavolino
similarity index 99%
rename from profiles/LoL-Shield
rename to profiles/LoL-Shield_Uno-Duemilanove-Diavolino
index ee35113..90b017b 100644
--- a/profiles/LoL-Shield
+++ b/profiles/LoL-Shield_Uno-Duemilanove-Diavolino
@@ -20,6 +20,7 @@ BORG_HW=HW_LOLSHIELD
 # lolshield setup
 #
 USER_TIMER0_FOR_WAIT=1
+BRIGHTNESS=127
 
 #
 # Features
diff --git a/src/borg_hw/borg_hw_lolshield.c b/src/borg_hw/borg_hw_lolshield.c
index a0705a7..237be9c 100644
--- a/src/borg_hw/borg_hw_lolshield.c
+++ b/src/borg_hw/borg_hw_lolshield.c
@@ -34,79 +34,615 @@
 #include <avr/io.h>
 #include <avr/wdt.h>
 #include <avr/pgmspace.h>
+#if NUMPLANE >= 8
+#	include <math.h>
+#endif
 #include "borg_hw.h"
 
 // buffer which holds the currently shown frame
 unsigned char pixmap[NUMPLANE][NUM_ROWS][LINEBYTES];
 
+// Number of ticks of the prescaled timer per cycle per frame, based on the
+// CPU clock speed and the desired frame rate.
 #define FRAMERATE 80UL
+#define	TICKS (F_CPU + 6 * (FRAMERATE << SLOWSCALERSHIFT)) / (12 * (FRAMERATE << SLOWSCALERSHIFT))
 #define	CUTOFF(scaler)	((128 * 12 - 6) * FRAMERATE * scaler)
 
-// BEWARE: only the Diavolino Kit is supported at the moment
-// other platforms don't work, yet
-const uint8_t
 #if defined (__AVR_ATmega168__) || defined (__AVR_ATmega48__) || defined (__AVR_ATmega88__) || defined (__AVR_ATmega328P__) || defined (__AVR_ATmega1280__) || defined (__AVR_ATmega2560__) || defined (__AVR_ATmega8__)
-#   if F_CPU < CUTOFF(8)
-        fastPrescaler = _BV(CS20),				// 1
-        slowPrescaler = _BV(CS21);				// 8
-#       define SLOWSCALERSHIFT 3
-#       define FASTSCALERSHIFT 3
-#   elif F_CPU < CUTOFF(32)
-        fastPrescaler = _BV(CS21),				// 8
-        slowPrescaler = _BV(CS21) | _BV(CS20);			// 32
-#       define SLOWSCALERSHIFT 5
-#       define FASTSCALERSHIFT 2
-#   elif F_CPU < CUTOFF(64)
-        fastPrescaler = _BV(CS21),				// 8
-        slowPrescaler = _BV(CS22);				// 64
-#       define SLOWSCALERSHIFT 6
-#       define FASTSCALERSHIFT 3
+#	if F_CPU < CUTOFF(8)
+#		define FASTPRESCALER (_BV(CS20))                          // 1
+#		define SLOWPRESCALER (_BV(CS21))                          // 8
+#		define FASTSCALERSHIFT 3
+#		define SLOWSCALERSHIFT 3
+#	elif F_CPU < CUTOFF(32)
+#		define FASTPRESCALER (_BV(CS21))                          // 8
+#		define SLOWPRESCALER (_BV(CS21) | _BV(CS20))              // 32
+#		define FASTSCALERSHIFT 2
+#		define SLOWSCALERSHIFT 5
+#	elif F_CPU < CUTOFF(64)
+#		define FASTPRESCALER (_BV(CS21))                          // 8
+#		define SLOWPRESCALER (_BV(CS22))                          // 64
+#		define FASTSCALERSHIFT 3
+#		define SLOWSCALERSHIFT 6
 #   elif F_CPU < CUTOFF(128)
-        fastPrescaler = _BV(CS21) | _BV(CS20),			// 32
-        slowPrescaler = _BV(CS22) | _BV(CS20);			// 128
-#       define SLOWSCALERSHIFT 7
-#       define FASTSCALERSHIFT 2
-#   elif F_CPU < CUTOFF(256)
-        fastPrescaler = _BV(CS21) | _BV(CS20),			// 32
-        slowPrescaler = _BV(CS22) | _BV(CS21);			// 256
-#       define SLOWSCALERSHIFT 8
-#       define FASTSCALERSHIFT 3
+#		define FASTPRESCALER (_BV(CS21) | _BV(CS20))              // 32
+#		define SLOWPRESCALER (_BV(CS22) | _BV(CS20))              // 128
+#		define FASTSCALERSHIFT 2
+#		define SLOWSCALERSHIFT 7
+#	elif F_CPU < CUTOFF(256)
+#		define FASTPRESCALER (_BV(CS21) | _BV(CS20))              // 32
+#		define SLOWPRESCALER (_BV(CS22) | _BV(CS21))              // 256
+#		define FASTSCALERSHIFT 3
+#		define SLOWSCALERSHIFT 8
 #   elif F_CPU < CUTOFF(1024)
-        fastPrescaler = _BV(CS22) | _BV(CS20),			// 128
-        slowPrescaler = _BV(CS22) | _BV(CS21) | _BV(CS20);	// 1024
-#       define SLOWSCALERSHIFT 10
-#       define FASTSCALERSHIFT 3
-#   else
-#       error frame rate is too low
-#   endif
+#		define FASTPRESCALER (_BV(CS22) | _BV(CS20))              // 128
+#		define SLOWPRESCALER (_BV(CS22) | _BV(CS21) | _BV(CS20))  // 1024
+#		define FASTSCALERSHIFT 3
+#		define SLOWSCALERSHIFT 10
+#	else
+#		error frame rate is too low
+#	endif
+#elif defined (__AVR_ATmega32U4__)
+#	if F_CPU < CUTOFF(8)
+#		define FASTPRESCALER (_BV(WGM12) | _BV(CS10))             // 1
+#		define SLOWPRESCALER (_BV(WGM12) | _BV(CS11))             // 8
+#		define FASTSCALERSHIFT 3
+#		define SLOWSCALERSHIFT 3
+#	elif F_CPU < CUTOFF(64)
+#		define FASTPRESCALER (_BV(WGM12) | _BV(CS11))             // 8
+#		define SLOWPRESCALER (_BV(WGM12) | _BV(CS11) | _BV(CS10)) // 64
+#		define FASTSCALERSHIFT 3
+#		define SLOWSCALERSHIFT 6
+#	elif F_CPU < CUTOFF(256)
+#		define FASTPRESCALER (_BV(WGM12) | _BV(CS11) | _BV(CS10)) // 64
+#		define SLOWPRESCALER (_BV(WGM12) | _BV(CS12))             // 256
+#		define FASTSCALERSHIFT 2
+#		define SLOWSCALERSHIFT 8
+#	elif F_CPU < CUTOFF(1024)
+#		define FASTPRESCALER (_BV(WGM12) | _BV(CS12))             // 256
+#		define SLOWPRESCALER (_BV(WGM12) | _BV(CS12) | _BV(CS10)) // 1024
+#		define FASTSCALERSHIFT 2
+#		define SLOWSCALERSHIFT 10
+#	else
+#		error frame rate is too low
+#	endif
 #else
 #   error no support for this chip
 #endif
 
 
-ISR(TIMER2_OVF_vect) {
-	// For each cycle, we have potential planes to display. Once every plane has
-	// been displayed, then we move on to the next cycle.
-	// NOTE: a "cycle" means a subset of LEDs
+#ifndef BRIGHTNESS
+#	define BRIGHTNESS 127 /* full brightness by default */
+#elif BRIGHTNESS < 0 || BRIGHTNESS > 127
+#	error BRIGHTNESS must be between 0 and 127
+#endif
 
-	// 12 Cycles of Matrix
-	static uint8_t cycle = 0;
+#define BRIGHTNESSPERCENT ((BRIGHTNESS * BRIGHTNESS + 8ul) / 16ul)
+#define M (TICKS << FASTSCALERSHIFT) * BRIGHTNESSPERCENT /*10b*/
+#define C(x) ((M * (unsigned long)(x * 1024) + (1 << 19)) >> 20) /*10b+10b-20b=0b*/
 
-	// planes to display
-	// NOTE: a "plane" in the Borgware is the same as a "page" in Jimmie's lib
-	static uint8_t plane = 0;
+#if NUMPLANE < 8
+uint8_t const prescaler[NUMPLANE + 1] = {
+	FASTPRESCALER,
+#	if NUMPLANE >= 2
+	FASTPRESCALER,
+#	endif
+#	if NUMPLANE >= 3
+	FASTPRESCALER,
+#	endif
+#	if NUMPLANE >= 4
+	FASTPRESCALER,
+#	endif
+#	if NUMPLANE >= 5
+	FASTPRESCALER,
+#	endif
+#	if NUMPLANE >= 6
+	FASTPRESCALER,
+#	endif
+#	if NUMPLANE >= 7
+	FASTPRESCALER,
+#	endif
+	SLOWPRESCALER
+};
+#else
+uint8_t prescaler[NUMPLANE + 1] = {0};
+#endif
 
+uint8_t counts[NUMPLANE + 1] = {0};
+
+/**
+ *  Set the overall brightness of the screen from 0 (off) to 127 (full on).
+ */
+static void setBrightness()
+{
+	/*   ---- This needs review! Please review. -- thilo  */
+	// set up page counts
+	uint8_t i;
+
+	// NOTE: The argument of C() is calculated as follows:
+	// pow((double)x / (double)NUMPLANE, 1.8) with 0 <= x <= NUMPLANE
+	// Changing the scale of 1.8 invalidates any tables above!
+#if NUMPLANE < 8
+	int const temp_counts[NUMPLANE + 1] = {
+		0.000000000000000000000000000,
+#	if NUMPLANE == 2
+		C(0.287174588749258719033719),
+#	elif NUMPLANE == 3
+		C(0.138414548846168578011273),
+		C(0.481987453865643789008288),
+#	elif NUMPLANE == 4
+		C(0.082469244423305887448095),
+		C(0.287174588749258719033719),
+		C(0.595813410589956848895099),
+#	elif NUMPLANE == 5
+		C(0.055189186458448592775827),
+		C(0.192179909437029006191722),
+		C(0.398723883569384374148115),
+		C(0.669209313658414961523135),
+#	elif NUMPLANE == 6
+		C(0.039749141141812646682574),
+		C(0.138414548846168578011273),
+		C(0.287174588749258719033719),
+		C(0.481987453865643789008288),
+		C(0.720234228706005730202833),
+#	elif NUMPLANE == 7
+		C(0.030117819624378608378557),
+		C(0.104876339357015443964904),
+		C(0.217591430058779483625031),
+		C(0.365200625214741059210155),
+		C(0.545719579451565794947498),
+		C(0.757697368024318751444923),
+#	endif
+		C(1.000000000000000000000000),
+	};
+#else
+#	warning "NUMPLANE >= 8 links floating point stuff into the image"
+	// NOTE: Changing "scale" invalidates any tables above!
+	const float scale = 1.8f;
+	int temp_counts[NUMPLANE + 1] = {0};
+
+	for (i = 1; i < (NUMPLANE + 1); i++) {
+		temp_counts[i] = C(pow(i / (float)(NUMPLANE), scale));
+	}
+#endif
+
+	// Compute on time for each of the pages
+	// Use the fast timer; slow timer is only useful for < 3 shades.
+	for (i = 0; i < NUMPLANE; i++) {
+		int interval = temp_counts[i + 1] - temp_counts[i];
+		counts[i] = 256 - (interval ? interval : 1);
+#if NUMPLANE >= 8
+		prescaler[i] = FASTPRESCALER;
+#endif
+	}
+
+	// Compute off time
+	int interval = TICKS - (temp_counts[i] >> FASTSCALERSHIFT);
+	counts[i] = 256 - (interval ? interval : 1);
+#if NUMPLANE >= 8
+		prescaler[i] = SLOWPRESCALER;
+#endif
+}
+
+/**
+ * Distributes the framebuffer content among current cycle pins.
+ * @param cycle The cycle whose pattern should to be composed.
+ * @param plane The plane ("page" in LoL Shield lingo) to be drawn.
+ */
+static void compose_cycle(uint8_t const cycle, uint8_t plane) {
 	// pointer to corresponding bitmap
-	uint8_t *p = &pixmap[plane][0][0];
-
-	// tune timer values (fiddle around with them to adjust brightness)
-	static uint8_t const prescaler[NUMPLANE + 1] = {3, 3, 3, 6};
-	TCCR2B = prescaler[plane];
-	static uint8_t const counts[NUMPLANE + 1] = {241, 198, 148, 255};
-	TCNT2  = counts[plane];
+	uint8_t *const p = &pixmap[plane][0][0];
 
+#if defined (__AVR_ATmega1280__) || defined (__AVR_ATmega2560__)
+#	warning "BEWARE: Borgware-2D has not been tested on Arduino Mega 1280/2560!"
 	// Set sink pin to Vcc/source, turning off current.
-	static uint8_t sink_b, sink_d = 0;
+	static uint8_t sink_b = 0, sink_e = 0, sink_g = 0, sink_h = 0;
+	PINB = sink_b;
+	PINE = sink_e;
+	PING = sink_g;
+	PINH = sink_h;
+
+	DDRB &= ~0xf0;
+	DDRE &= ~0x38;
+	DDRG &= ~0x20;
+	DDRH &= ~0x78;
+
+	static uint8_t const sink_b_cycle[] =
+		{0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x10, 0x20, 0x40, 0x80};
+	static uint8_t const sink_e_cycle[] =
+		{0x10, 0x20, 0x00, 0x08, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00};
+	static uint8_t const sink_g_cycle[] =
+		{0x00, 0x00, 0x20, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00};
+	static uint8_t const sink_h_cycle[] =
+		{0x00, 0x00, 0x00, 0x00, 0x08, 0x10, 0x20, 0x40, 0x00, 0x00, 0x00, 0x00};
+
+	uint8_t pins_b = sink_b = pgm_read_byte(&sink_b_cycle[cycle]);
+	uint8_t pins_e = sink_e = pgm_read_byte(&sink_e_cycle[cycle]);
+	uint8_t pins_g = sink_g = pgm_read_byte(&sink_g_cycle[cycle]);
+	uint8_t pins_h = sink_h = pgm_read_byte(&sink_h_cycle[cycle]);
+
+	// convert framebuffer to LoL Shield cycles on Arduino Mega 1280/2560
+	// (I could have done this with a lookup table, but that would be slower as
+	// non-constant bit shifts are quite expensive on AVR)
+	// NOTE: (0,0) is UPPER RIGHT in the Borgware realm
+	if (plane < NUMPLANE) {
+		switch(cycle) {
+		case 0:
+			pins_b |= (0x02u & p[ 0]) << 6; // x= 1, y= 0, mapped pin D13
+			pins_b |= (0x02u & p[ 2]) << 5; // x= 1, y= 1, mapped pin D12
+			pins_b |= (0x02u & p[ 4]) << 4; // x= 1, y= 2, mapped pin D11
+			pins_b |= (0x02u & p[ 6]) << 3; // x= 1, y= 3, mapped pin D10
+			pins_e |= (0x02u & p[16]) << 2; // x= 1, y= 8, mapped pin D5
+			pins_h |= (0x02u & p[ 8]) << 5; // x= 1, y= 4, mapped pin D9
+			pins_h |= (0x02u & p[10]) << 4; // x= 1, y= 5, mapped pin D8
+			pins_h |= (0x02u & p[12]) << 3; // x= 1, y= 6, mapped pin D7
+			pins_h |= (0x02u & p[14]) << 2; // x= 1, y= 7, mapped pin D6
+			break;
+		case 1:
+			pins_b |= (0x08u & p[ 0]) << 4; // x= 3, y= 0, mapped pin D13
+			pins_b |= (0x08u & p[ 2]) << 3; // x= 3, y= 1, mapped pin D12
+			pins_b |= (0x08u & p[ 4]) << 2; // x= 3, y= 2, mapped pin D11
+			pins_b |= (0x08u & p[ 6]) << 1; // x= 3, y= 3, mapped pin D10
+			pins_e |= (0x08u & p[16]);      // x= 3, y= 8, mapped pin D5
+			pins_h |= (0x08u & p[ 8]) << 3; // x= 3, y= 4, mapped pin D9
+			pins_h |= (0x08u & p[10]) << 2; // x= 3, y= 5, mapped pin D8
+			pins_h |= (0x08u & p[12]) << 1; // x= 3, y= 6, mapped pin D7
+			pins_h |= (0x08u & p[14]);      // x= 3, y= 7, mapped pin D6
+			break;
+		case 2:
+			pins_b |= (0x20u & p[ 0]) << 2; // x= 5, y= 0, mapped pin D13
+			pins_b |= (0x20u & p[ 2]) << 1; // x= 5, y= 1, mapped pin D12
+			pins_b |= (0x20u & p[ 4]);      // x= 5, y= 2, mapped pin D11
+			pins_b |= (0x20u & p[ 6]) >> 1; // x= 5, y= 3, mapped pin D10
+			pins_e |= (0x20u & p[16]) >> 2; // x= 5, y= 8, mapped pin D5
+			pins_h |= (0x20u & p[ 8]) << 1; // x= 5, y= 4, mapped pin D9
+			pins_h |= (0x20u & p[10]);      // x= 5, y= 5, mapped pin D8
+			pins_h |= (0x20u & p[12]) >> 1; // x= 5, y= 6, mapped pin D7
+			pins_h |= (0x20u & p[14]) >> 2; // x= 5, y= 7, mapped pin D6
+			break;
+		case 3:
+			pins_b |= (0x20u & p[ 1]) << 2; // x=13, y= 0, mapped pin D13
+			pins_b |= (0x20u & p[ 3]) << 1; // x=13, y= 1, mapped pin D12
+			pins_b |= (0x20u & p[ 5]);      // x=13, y= 2, mapped pin D11
+			pins_b |= (0x20u & p[ 7]) >> 1; // x=13, y= 3, mapped pin D10
+			pins_e |= (0x01u & p[16]) << 4; // x= 0, y= 8, mapped pin D2
+			pins_e |= (0x04u & p[16]) << 3; // x= 2, y= 8, mapped pin D3
+			pins_g |= (0x10u & p[16]) << 1; // x= 4, y= 8, mapped pin D4
+			pins_h |= (0x20u & p[ 9]) << 1; // x=13, y= 4, mapped pin D9
+			pins_h |= (0x20u & p[11]);      // x=13, y= 5, mapped pin D8
+			pins_h |= (0x20u & p[13]) >> 1; // x=13, y= 6, mapped pin D7
+			pins_h |= (0x20u & p[15]) >> 2; // x=13, y= 7, mapped pin D6
+			break;
+		case 4:
+			pins_b |= (0x10u & p[ 1]) << 3; // x=12, y= 0, mapped pin D13
+			pins_b |= (0x10u & p[ 3]) << 2; // x=12, y= 1, mapped pin D12
+			pins_b |= (0x10u & p[ 5]) << 1; // x=12, y= 2, mapped pin D11
+			pins_b |= (0x10u & p[ 7]);      // x=12, y= 3, mapped pin D10
+			pins_e |= (0x01u & p[14]) << 4; // x= 0, y= 7, mapped pin D2
+			pins_e |= (0x04u & p[14]) << 3; // x= 2, y= 7, mapped pin D3
+			pins_e |= (0x20u & p[17]) >> 2; // x=13, y= 8, mapped pin D5
+			pins_g |= (0x10u & p[14]) << 1; // x= 4, y= 7, mapped pin D4
+			pins_h |= (0x10u & p[ 9]) << 2; // x=12, y= 4, mapped pin D9
+			pins_h |= (0x10u & p[11]) << 1; // x=12, y= 5, mapped pin D8
+			pins_h |= (0x10u & p[13]);      // x=12, y= 6, mapped pin D7
+			break;
+		case 5:
+			pins_b |= (0x08u & p[ 1]) << 4; // x=11, y= 0, mapped pin D13
+			pins_b |= (0x08u & p[ 3]) << 3; // x=11, y= 1, mapped pin D12
+			pins_b |= (0x08u & p[ 5]) << 2; // x=11, y= 2, mapped pin D11
+			pins_b |= (0x08u & p[ 7]) << 1; // x=11, y= 3, mapped pin D10
+			pins_e |= (0x01u & p[12]) << 4; // x= 0, y= 6, mapped pin D2
+			pins_e |= (0x04u & p[12]) << 3; // x= 2, y= 6, mapped pin D3
+			pins_e |= (0x10u & p[17]) >> 1; // x=12, y= 8, mapped pin D5
+			pins_g |= (0x10u & p[12]) << 1; // x= 4, y= 6, mapped pin D4
+			pins_h |= (0x08u & p[ 9]) << 3; // x=11, y= 4, mapped pin D9
+			pins_h |= (0x08u & p[11]) << 2; // x=11, y= 5, mapped pin D8
+			pins_h |= (0x10u & p[15]) >> 1; // x=12, y= 7, mapped pin D6
+			break;
+		case 6:
+			pins_b |= (0x04u & p[ 1]) << 5; // x=10, y= 0, mapped pin D13
+			pins_b |= (0x04u & p[ 3]) << 4; // x=10, y= 1, mapped pin D12
+			pins_b |= (0x04u & p[ 5]) << 3; // x=10, y= 2, mapped pin D11
+			pins_b |= (0x04u & p[ 7]) << 2; // x=10, y= 3, mapped pin D10
+			pins_e |= (0x01u & p[10]) << 4; // x= 0, y= 5, mapped pin D2
+			pins_e |= (0x04u & p[10]) << 3; // x= 2, y= 5, mapped pin D3
+			pins_e |= (0x08u & p[17]);      // x=11, y= 8, mapped pin D5
+			pins_g |= (0x10u & p[10]) << 1; // x= 4, y= 5, mapped pin D4
+			pins_h |= (0x04u & p[ 9]) << 4; // x=10, y= 4, mapped pin D9
+			pins_h |= (0x08u & p[13]) << 1; // x=11, y= 6, mapped pin D7
+			pins_h |= (0x08u & p[15]);      // x=11, y= 7, mapped pin D6
+			break;
+		case 7:
+			pins_b |= (0x02u & p[ 1]) << 6; // x= 9, y= 0, mapped pin D13
+			pins_b |= (0x02u & p[ 3]) << 5; // x= 9, y= 1, mapped pin D12
+			pins_b |= (0x02u & p[ 5]) << 4; // x= 9, y= 2, mapped pin D11
+			pins_b |= (0x02u & p[ 7]) << 3; // x= 9, y= 3, mapped pin D10
+			pins_e |= (0x01u & p[ 8]) << 4; // x= 0, y= 4, mapped pin D2
+			pins_e |= (0x04u & p[ 8]) << 3; // x= 2, y= 4, mapped pin D3
+			pins_e |= (0x04u & p[17]) << 1; // x=10, y= 8, mapped pin D5
+			pins_g |= (0x10u & p[ 8]) << 1; // x= 4, y= 4, mapped pin D4
+			pins_h |= (0x04u & p[11]) << 3; // x=10, y= 5, mapped pin D8
+			pins_h |= (0x04u & p[13]) << 2; // x=10, y= 6, mapped pin D7
+			pins_h |= (0x04u & p[15]) << 1; // x=10, y= 7, mapped pin D6
+			break;
+		case 8:
+			pins_b |= (0x01u & p[ 1]) << 7; // x= 8, y= 0, mapped pin D13
+			pins_b |= (0x01u & p[ 3]) << 6; // x= 8, y= 1, mapped pin D12
+			pins_b |= (0x01u & p[ 5]) << 5; // x= 8, y= 2, mapped pin D11
+			pins_e |= (0x01u & p[ 6]) << 4; // x= 0, y= 3, mapped pin D2
+			pins_e |= (0x02u & p[17]) << 2; // x= 9, y= 8, mapped pin D5
+			pins_e |= (0x04u & p[ 6]) << 3; // x= 2, y= 3, mapped pin D3
+			pins_g |= (0x10u & p[ 6]) << 1; // x= 4, y= 3, mapped pin D4
+			pins_h |= (0x02u & p[ 9]) << 5; // x= 9, y= 4, mapped pin D9
+			pins_h |= (0x02u & p[11]) << 4; // x= 9, y= 5, mapped pin D8
+			pins_h |= (0x02u & p[13]) << 3; // x= 9, y= 6, mapped pin D7
+			pins_h |= (0x02u & p[15]) << 2; // x= 9, y= 7, mapped pin D6
+			break;
+		case 9:
+			pins_b |= (0x01u & p[ 7]) << 4; // x= 8, y= 3, mapped pin D10
+			pins_b |= (0x80u & p[ 0]);      // x= 7, y= 0, mapped pin D13
+			pins_b |= (0x80u & p[ 2]) >> 1; // x= 7, y= 1, mapped pin D12
+			pins_e |= (0x01u & p[ 4]) << 4; // x= 0, y= 2, mapped pin D2
+			pins_e |= (0x01u & p[17]) << 3; // x= 8, y= 8, mapped pin D5
+			pins_e |= (0x04u & p[ 4]) << 3; // x= 2, y= 2, mapped pin D3
+			pins_g |= (0x10u & p[ 4]) << 1; // x= 4, y= 2, mapped pin D4
+			pins_h |= (0x01u & p[ 9]) << 6; // x= 8, y= 4, mapped pin D9
+			pins_h |= (0x01u & p[11]) << 5; // x= 8, y= 5, mapped pin D8
+			pins_h |= (0x01u & p[13]) << 4; // x= 8, y= 6, mapped pin D7
+			pins_h |= (0x01u & p[15]) << 3; // x= 8, y= 7, mapped pin D6
+			break;
+		case 10:
+			pins_b |= (0x40u & p[ 0]) << 1; // x= 6, y= 0, mapped pin D13
+			pins_b |= (0x80u & p[ 4]) >> 2; // x= 7, y= 2, mapped pin D11
+			pins_b |= (0x80u & p[ 6]) >> 3; // x= 7, y= 3, mapped pin D10
+			pins_e |= (0x01u & p[ 2]) << 4; // x= 0, y= 1, mapped pin D2
+			pins_e |= (0x04u & p[ 2]) << 3; // x= 2, y= 1, mapped pin D3
+			pins_e |= (0x80u & p[16]) >> 4; // x= 7, y= 8, mapped pin D5
+			pins_g |= (0x10u & p[ 2]) << 1; // x= 4, y= 1, mapped pin D4
+			pins_h |= (0x80u & p[ 8]) >> 1; // x= 7, y= 4, mapped pin D9
+			pins_h |= (0x80u & p[10]) >> 2; // x= 7, y= 5, mapped pin D8
+			pins_h |= (0x80u & p[12]) >> 3; // x= 7, y= 6, mapped pin D7
+			pins_h |= (0x80u & p[14]) >> 4; // x= 7, y= 7, mapped pin D6
+			break;
+		case 11:
+			pins_b |= (0x40u & p[ 2]);      // x= 6, y= 1, mapped pin D12
+			pins_b |= (0x40u & p[ 4]) >> 1; // x= 6, y= 2, mapped pin D11
+			pins_b |= (0x40u & p[ 6]) >> 2; // x= 6, y= 3, mapped pin D10
+			pins_e |= (0x01u & p[ 0]) << 4; // x= 0, y= 0, mapped pin D2
+			pins_e |= (0x04u & p[ 0]) << 3; // x= 2, y= 0, mapped pin D3
+			pins_e |= (0x40u & p[16]) >> 3; // x= 6, y= 8, mapped pin D5
+			pins_g |= (0x10u & p[ 0]) << 1; // x= 4, y= 0, mapped pin D4
+			pins_h |= (0x40u & p[ 8]);      // x= 6, y= 4, mapped pin D9
+			pins_h |= (0x40u & p[10]) >> 1; // x= 6, y= 5, mapped pin D8
+			pins_h |= (0x40u & p[12]) >> 2; // x= 6, y= 6, mapped pin D7
+			pins_h |= (0x40u & p[14]) >> 3; // x= 6, y= 7, mapped pin D6
+			break;
+		}
+	}
+
+	// Enable pullups (by toggling) on new output pins.
+	PINB = PORTB ^ pins_b;
+	PINE = PORTE ^ pins_e;
+	PING = PORTG ^ pins_g;
+	PINH = PORTH ^ pins_h;
+
+	// Set pins to output mode; pullups become Vcc/source.
+	DDRB |= pins_b;
+	DDRE |= pins_e;
+	DDRG |= pins_g;
+	DDRH |= pins_h;
+
+	// Set sink pin to GND/sink, turning on current.
+	PINB = sink_b;
+	PINE = sink_e;
+	PING = sink_g;
+	PINH = sink_h;
+#elif defined (__AVR_ATmega32U4__)
+	// Set sink pin to Vcc/source, turning off current.
+	static uint8_t sink_b = 0, sink_c = 0, sink_d = 0, sink_e = 0;
+	PINB = sink_b;
+	PINC = sink_c;
+	PIND = sink_d;
+	PINE = sink_e;
+
+	DDRB &= ~0xF0;
+	DDRC &= ~0xC0;
+	DDRD &= ~0xD3;
+	DDRE &= ~0x40;
+
+	static uint8_t const PROGMEM sink_b_cycle[] =
+		{0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x10, 0x20, 0x40, 0x80, 0x00, 0x00};
+	static uint8_t const PROGMEM sink_c_cycle[] =
+		{0x00, 0x00, 0x00, 0x40, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x80};
+	static uint8_t const PROGMEM sink_d_cycle[] =
+		{0x02, 0x01, 0x10, 0x00, 0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x40, 0x00};
+	static uint8_t const PROGMEM sink_e_cycle[] =
+		{0x00, 0x00, 0x00, 0x00, 0x00, 0x40, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00};
+
+	uint8_t pins_b = sink_b = pgm_read_byte(&sink_b_cycle[cycle]);
+	uint8_t pins_c = sink_c = pgm_read_byte(&sink_c_cycle[cycle]);
+	uint8_t pins_d = sink_d = pgm_read_byte(&sink_d_cycle[cycle]);
+	uint8_t pins_e = sink_e = pgm_read_byte(&sink_e_cycle[cycle]);
+
+	// convert Borgware-2D framebuffer to LoL Shield cycles on Arduino Leonardo
+	// (I could have done this with a lookup table, but that would be slower as
+	// non-constant bit shifts are quite expensive on AVR)
+	// NOTE: (0,0) is UPPER RIGHT in the Borgware realm
+	if (plane < NUMPLANE) {
+		switch(cycle) {
+		case 0:
+			pins_b |= (0x02u & p[ 4]) << 6; // x= 1, y= 2, mapped pin D11
+			pins_b |= (0x02u & p[ 6]) << 5; // x= 1, y= 3, mapped pin D10
+			pins_b |= (0x02u & p[ 8]) << 4; // x= 1, y= 4, mapped pin D9
+			pins_b |= (0x02u & p[10]) << 3; // x= 1, y= 5, mapped pin D8
+			pins_c |= (0x02u & p[ 0]) << 6; // x= 1, y= 0, mapped pin D13
+			pins_c |= (0x02u & p[16]) << 5; // x= 1, y= 8, mapped pin D5
+			pins_d |= (0x02u & p[ 2]) << 5; // x= 1, y= 1, mapped pin D12
+			pins_d |= (0x02u & p[14]) << 6; // x= 1, y= 7, mapped pin D6
+			pins_e |= (0x02u & p[12]) << 5; // x= 1, y= 6, mapped pin D7
+			break;
+		case 1:
+			pins_b |= (0x08u & p[ 4]) << 4; // x= 3, y= 2, mapped pin D11
+			pins_b |= (0x08u & p[ 6]) << 3; // x= 3, y= 3, mapped pin D10
+			pins_b |= (0x08u & p[ 8]) << 2; // x= 3, y= 4, mapped pin D9
+			pins_b |= (0x08u & p[10]) << 1; // x= 3, y= 5, mapped pin D8
+			pins_c |= (0x08u & p[ 0]) << 4; // x= 3, y= 0, mapped pin D13
+			pins_c |= (0x08u & p[16]) << 3; // x= 3, y= 8, mapped pin D5
+			pins_d |= (0x08u & p[ 2]) << 3; // x= 3, y= 1, mapped pin D12
+			pins_d |= (0x08u & p[14]) << 4; // x= 3, y= 7, mapped pin D6
+			pins_e |= (0x08u & p[12]) << 3; // x= 3, y= 6, mapped pin D7
+			break;
+		case 2:
+			pins_b |= (0x20u & p[ 4]) << 2; // x= 5, y= 2, mapped pin D11
+			pins_b |= (0x20u & p[ 6]) << 1; // x= 5, y= 3, mapped pin D10
+			pins_b |= (0x20u & p[ 8]);      // x= 5, y= 4, mapped pin D9
+			pins_b |= (0x20u & p[10]) >> 1; // x= 5, y= 5, mapped pin D8
+			pins_c |= (0x20u & p[ 0]) << 2; // x= 5, y= 0, mapped pin D13
+			pins_c |= (0x20u & p[16]) << 1; // x= 5, y= 8, mapped pin D5
+			pins_d |= (0x20u & p[ 2]) << 1; // x= 5, y= 1, mapped pin D12
+			pins_d |= (0x20u & p[14]) << 2; // x= 5, y= 7, mapped pin D6
+			pins_e |= (0x20u & p[12]) << 1; // x= 5, y= 6, mapped pin D7
+			break;
+		case 3:
+			pins_b |= (0x20u & p[ 5]) << 2; // x=13, y= 2, mapped pin D11
+			pins_b |= (0x20u & p[ 7]) << 1; // x=13, y= 3, mapped pin D10
+			pins_b |= (0x20u & p[ 9]);      // x=13, y= 4, mapped pin D9
+			pins_b |= (0x20u & p[11]) >> 1; // x=13, y= 5, mapped pin D8
+			pins_c |= (0x20u & p[ 1]) << 2; // x=13, y= 0, mapped pin D13
+			pins_d |= (0x01u & p[16]) << 1; // x= 0, y= 8, mapped pin D2
+			pins_d |= (0x04u & p[16]) >> 2; // x= 2, y= 8, mapped pin D3
+			pins_d |= (0x10u & p[16]);      // x= 4, y= 8, mapped pin D4
+			pins_d |= (0x20u & p[ 3]) << 1; // x=13, y= 1, mapped pin D12
+			pins_d |= (0x20u & p[15]) << 2; // x=13, y= 7, mapped pin D6
+			pins_e |= (0x20u & p[13]) << 1; // x=13, y= 6, mapped pin D7
+			break;
+		case 4:
+			pins_b |= (0x10u & p[ 5]) << 3; // x=12, y= 2, mapped pin D11
+			pins_b |= (0x10u & p[ 7]) << 2; // x=12, y= 3, mapped pin D10
+			pins_b |= (0x10u & p[ 9]) << 1; // x=12, y= 4, mapped pin D9
+			pins_b |= (0x10u & p[11]);      // x=12, y= 5, mapped pin D8
+			pins_c |= (0x10u & p[ 1]) << 3; // x=12, y= 0, mapped pin D13
+			pins_c |= (0x20u & p[17]) << 1; // x=13, y= 8, mapped pin D5
+			pins_d |= (0x01u & p[14]) << 1; // x= 0, y= 7, mapped pin D2
+			pins_d |= (0x04u & p[14]) >> 2; // x= 2, y= 7, mapped pin D3
+			pins_d |= (0x10u & p[ 3]) << 2; // x=12, y= 1, mapped pin D12
+			pins_d |= (0x10u & p[14]);      // x= 4, y= 7, mapped pin D4
+			pins_e |= (0x10u & p[13]) << 2; // x=12, y= 6, mapped pin D7
+			break;
+		case 5:
+			pins_b |= (0x08u & p[ 5]) << 4; // x=11, y= 2, mapped pin D11
+			pins_b |= (0x08u & p[ 7]) << 3; // x=11, y= 3, mapped pin D10
+			pins_b |= (0x08u & p[ 9]) << 2; // x=11, y= 4, mapped pin D9
+			pins_b |= (0x08u & p[11]) << 1; // x=11, y= 5, mapped pin D8
+			pins_c |= (0x08u & p[ 1]) << 4; // x=11, y= 0, mapped pin D13
+			pins_c |= (0x10u & p[17]) << 2; // x=12, y= 8, mapped pin D5
+			pins_d |= (0x01u & p[12]) << 1; // x= 0, y= 6, mapped pin D2
+			pins_d |= (0x04u & p[12]) >> 2; // x= 2, y= 6, mapped pin D3
+			pins_d |= (0x08u & p[ 3]) << 3; // x=11, y= 1, mapped pin D12
+			pins_d |= (0x10u & p[12]);      // x= 4, y= 6, mapped pin D4
+			pins_d |= (0x10u & p[15]) << 3; // x=12, y= 7, mapped pin D6
+			break;
+		case 6:
+			pins_b |= (0x04u & p[ 5]) << 5; // x=10, y= 2, mapped pin D11
+			pins_b |= (0x04u & p[ 7]) << 4; // x=10, y= 3, mapped pin D10
+			pins_b |= (0x04u & p[ 9]) << 3; // x=10, y= 4, mapped pin D9
+			pins_c |= (0x04u & p[ 1]) << 5; // x=10, y= 0, mapped pin D13
+			pins_c |= (0x08u & p[17]) << 3; // x=11, y= 8, mapped pin D5
+			pins_d |= (0x01u & p[10]) << 1; // x= 0, y= 5, mapped pin D2
+			pins_d |= (0x04u & p[ 3]) << 4; // x=10, y= 1, mapped pin D12
+			pins_d |= (0x04u & p[10]) >> 2; // x= 2, y= 5, mapped pin D3
+			pins_d |= (0x08u & p[15]) << 4; // x=11, y= 7, mapped pin D6
+			pins_d |= (0x10u & p[10]);      // x= 4, y= 5, mapped pin D4
+			pins_e |= (0x08u & p[13]) << 3; // x=11, y= 6, mapped pin D7
+			break;
+		case 7:
+			pins_b |= (0x02u & p[ 5]) << 6; // x= 9, y= 2, mapped pin D11
+			pins_b |= (0x02u & p[ 7]) << 5; // x= 9, y= 3, mapped pin D10
+			pins_b |= (0x04u & p[11]) << 2; // x=10, y= 5, mapped pin D8
+			pins_c |= (0x02u & p[ 1]) << 6; // x= 9, y= 0, mapped pin D13
+			pins_c |= (0x04u & p[17]) << 4; // x=10, y= 8, mapped pin D5
+			pins_d |= (0x01u & p[ 8]) << 1; // x= 0, y= 4, mapped pin D2
+			pins_d |= (0x02u & p[ 3]) << 5; // x= 9, y= 1, mapped pin D12
+			pins_d |= (0x04u & p[ 8]) >> 2; // x= 2, y= 4, mapped pin D3
+			pins_d |= (0x04u & p[15]) << 5; // x=10, y= 7, mapped pin D6
+			pins_d |= (0x10u & p[ 8]);      // x= 4, y= 4, mapped pin D4
+			pins_e |= (0x04u & p[13]) << 4; // x=10, y= 6, mapped pin D7
+			break;
+		case 8:
+			pins_b |= (0x01u & p[ 5]) << 7; // x= 8, y= 2, mapped pin D11
+			pins_b |= (0x02u & p[ 9]) << 4; // x= 9, y= 4, mapped pin D9
+			pins_b |= (0x02u & p[11]) << 3; // x= 9, y= 5, mapped pin D8
+			pins_c |= (0x01u & p[ 1]) << 7; // x= 8, y= 0, mapped pin D13
+			pins_c |= (0x02u & p[17]) << 5; // x= 9, y= 8, mapped pin D5
+			pins_d |= (0x01u & p[ 3]) << 6; // x= 8, y= 1, mapped pin D12
+			pins_d |= (0x01u & p[ 6]) << 1; // x= 0, y= 3, mapped pin D2
+			pins_d |= (0x02u & p[15]) << 6; // x= 9, y= 7, mapped pin D6
+			pins_d |= (0x04u & p[ 6]) >> 2; // x= 2, y= 3, mapped pin D3
+			pins_d |= (0x10u & p[ 6]);      // x= 4, y= 3, mapped pin D4
+			pins_e |= (0x02u & p[13]) << 5; // x= 9, y= 6, mapped pin D7
+			break;
+		case 9:
+			pins_b |= (0x01u & p[ 7]) << 6; // x= 8, y= 3, mapped pin D10
+			pins_b |= (0x01u & p[ 9]) << 5; // x= 8, y= 4, mapped pin D9
+			pins_b |= (0x01u & p[11]) << 4; // x= 8, y= 5, mapped pin D8
+			pins_c |= (0x01u & p[17]) << 6; // x= 8, y= 8, mapped pin D5
+			pins_c |= (0x80u & p[ 0]);      // x= 7, y= 0, mapped pin D13
+			pins_d |= (0x01u & p[ 4]) << 1; // x= 0, y= 2, mapped pin D2
+			pins_d |= (0x01u & p[15]) << 7; // x= 8, y= 7, mapped pin D6
+			pins_d |= (0x04u & p[ 4]) >> 2; // x= 2, y= 2, mapped pin D3
+			pins_d |= (0x10u & p[ 4]);      // x= 4, y= 2, mapped pin D4
+			pins_d |= (0x80u & p[ 2]) >> 1; // x= 7, y= 1, mapped pin D12
+			pins_e |= (0x01u & p[13]) << 6; // x= 8, y= 6, mapped pin D7
+			break;
+		case 10:
+			pins_b |= (0x80u & p[ 4]);      // x= 7, y= 2, mapped pin D11
+			pins_b |= (0x80u & p[ 6]) >> 1; // x= 7, y= 3, mapped pin D10
+			pins_b |= (0x80u & p[ 8]) >> 2; // x= 7, y= 4, mapped pin D9
+			pins_b |= (0x80u & p[10]) >> 3; // x= 7, y= 5, mapped pin D8
+			pins_c |= (0x40u & p[ 0]) << 1; // x= 6, y= 0, mapped pin D13
+			pins_c |= (0x80u & p[16]) >> 1; // x= 7, y= 8, mapped pin D5
+			pins_d |= (0x01u & p[ 2]) << 1; // x= 0, y= 1, mapped pin D2
+			pins_d |= (0x04u & p[ 2]) >> 2; // x= 2, y= 1, mapped pin D3
+			pins_d |= (0x10u & p[ 2]);      // x= 4, y= 1, mapped pin D4
+			pins_d |= (0x80u & p[14]);      // x= 7, y= 7, mapped pin D6
+			pins_e |= (0x80u & p[12]) >> 1; // x= 7, y= 6, mapped pin D7
+			break;
+		case 11:
+			pins_b |= (0x40u & p[ 4]) << 1; // x= 6, y= 2, mapped pin D11
+			pins_b |= (0x40u & p[ 6]);      // x= 6, y= 3, mapped pin D10
+			pins_b |= (0x40u & p[ 8]) >> 1; // x= 6, y= 4, mapped pin D9
+			pins_b |= (0x40u & p[10]) >> 2; // x= 6, y= 5, mapped pin D8
+			pins_c |= (0x40u & p[16]);      // x= 6, y= 8, mapped pin D5
+			pins_d |= (0x01u & p[ 0]) << 1; // x= 0, y= 0, mapped pin D2
+			pins_d |= (0x04u & p[ 0]) >> 2; // x= 2, y= 0, mapped pin D3
+			pins_d |= (0x10u & p[ 0]);      // x= 4, y= 0, mapped pin D4
+			pins_d |= (0x40u & p[ 2]);      // x= 6, y= 1, mapped pin D12
+			pins_d |= (0x40u & p[14]) << 1; // x= 6, y= 7, mapped pin D6
+			pins_e |= (0x40u & p[12]);      // x= 6, y= 6, mapped pin D7
+			break;
+		}
+	}
+
+	// Enable pullups (by toggling) on new output pins.
+	PINB = PORTB ^ pins_b;
+	PINC = PORTC ^ pins_c;
+	PIND = PORTD ^ pins_d;
+	PINE = PORTE ^ pins_e;
+
+	// Set pins to output mode; pullups become Vcc/source.
+	DDRB |= pins_b;
+	DDRC |= pins_c;
+	DDRD |= pins_d;
+	DDRE |= pins_e;
+
+	// Set sink pin to GND/sink, turning on current.
+	PINB = sink_b;
+	PINC = sink_c;
+	PIND = sink_d;
+	PINE = sink_e;
+#else
+	// Set sink pin to Vcc/source, turning off current.
+	static uint8_t sink_b = 0, sink_d = 0;
 	PIND = sink_d;
 	PINB = sink_b;
 
@@ -114,168 +650,169 @@ ISR(TIMER2_OVF_vect) {
 	DDRD = 0;
 	DDRB = 0;
 
-	static uint8_t const sink_d_cycle[] =
+	static uint8_t const PROGMEM sink_d_cycle[] =
 		{0x04, 0x08, 0x10, 0x20, 0x40, 0x80, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00};
-	static uint8_t const sink_b_cycle[] =
+	static uint8_t const PROGMEM sink_b_cycle[] =
 		{0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x02, 0x04, 0x08, 0x10, 0x20};
 
-	uint8_t pins_d = sink_d = sink_d_cycle[cycle];
-	uint8_t pins_b = sink_b = sink_b_cycle[cycle];
+	uint8_t pins_d = sink_d = pgm_read_byte(&sink_d_cycle[cycle]);
+	uint8_t pins_b = sink_b = pgm_read_byte(&sink_b_cycle[cycle]);
 
-	// convert Borgware-2D framebuffer to LoL Shield cycles
+	// convert Borgware-2D framebuffer to LoL Shield cycles on Diavolino
 	// (I could have done this with a lookup table, but that would be slower as
 	// non-constant bit shifts are quite expensive on AVR)
+	// NOTE: (0,0) is UPPER RIGHT in the Borgware realm
     if (plane < NUMPLANE) {
 		switch(cycle) {
 		case 0:
-			pins_b |= (0x02u & p[ 0]) << 4; // x=12, y= 0, high=13
-			pins_b |= (0x02u & p[ 2]) << 3; // x=12, y= 1, high=12
-			pins_b |= (0x02u & p[ 4]) << 2; // x=12, y= 2, high=11
-			pins_b |= (0x02u & p[ 6]) << 1; // x=12, y= 3, high=10
-			pins_b |= (0x02u & p[ 8]);      // x=12, y= 4, high= 9
-			pins_b |= (0x02u & p[10]) >> 1; // x=12, y= 5, high= 8
-			pins_d |= (0x02u & p[12]) << 6; // x=12, y= 6, high= 7
-			pins_d |= (0x02u & p[14]) << 5; // x=12, y= 7, high= 6
-			pins_d |= (0x02u & p[16]) << 4; // x=12, y= 8, high= 5
+			pins_b |= (0x02u & p[ 0]) << 4; // x= 1, y= 0, mapped pin D13
+			pins_b |= (0x02u & p[ 2]) << 3; // x= 1, y= 1, mapped pin D12
+			pins_b |= (0x02u & p[ 4]) << 2; // x= 1, y= 2, mapped pin D11
+			pins_b |= (0x02u & p[ 6]) << 1; // x= 1, y= 3, mapped pin D10
+			pins_b |= (0x02u & p[ 8]);      // x= 1, y= 4, mapped pin D9
+			pins_b |= (0x02u & p[10]) >> 1; // x= 1, y= 5, mapped pin D8
+			pins_d |= (0x02u & p[12]) << 6; // x= 1, y= 6, mapped pin D7
+			pins_d |= (0x02u & p[14]) << 5; // x= 1, y= 7, mapped pin D6
+			pins_d |= (0x02u & p[16]) << 4; // x= 1, y= 8, mapped pin D5
 			break;
 		case 1:
-			pins_b |= (0x08u & p[ 0]) << 2; // x=10, y= 0, high=13
-			pins_b |= (0x08u & p[ 2]) << 1; // x=10, y= 1, high=12
-			pins_b |= (0x08u & p[ 4]);      // x=10, y= 2, high=11
-			pins_b |= (0x08u & p[ 6]) >> 1; // x=10, y= 3, high=10
-			pins_b |= (0x08u & p[ 8]) >> 2; // x=10, y= 4, high= 9
-			pins_b |= (0x08u & p[10]) >> 3; // x=10, y= 5, high= 8
-			pins_d |= (0x08u & p[12]) << 4; // x=10, y= 6, high= 7
-			pins_d |= (0x08u & p[14]) << 3; // x=10, y= 7, high= 6
-			pins_d |= (0x08u & p[16]) << 2; // x=10, y= 8, high= 5
+			pins_b |= (0x08u & p[ 0]) << 2; // x= 3, y= 0, mapped pin D13
+			pins_b |= (0x08u & p[ 2]) << 1; // x= 3, y= 1, mapped pin D12
+			pins_b |= (0x08u & p[ 4]);      // x= 3, y= 2, mapped pin D11
+			pins_b |= (0x08u & p[ 6]) >> 1; // x= 3, y= 3, mapped pin D10
+			pins_b |= (0x08u & p[ 8]) >> 2; // x= 3, y= 4, mapped pin D9
+			pins_b |= (0x08u & p[10]) >> 3; // x= 3, y= 5, mapped pin D8
+			pins_d |= (0x08u & p[12]) << 4; // x= 3, y= 6, mapped pin D7
+			pins_d |= (0x08u & p[14]) << 3; // x= 3, y= 7, mapped pin D6
+			pins_d |= (0x08u & p[16]) << 2; // x= 3, y= 8, mapped pin D5
 			break;
 		case 2:
-			pins_b |= (0x20u & p[ 0]);      // x= 8, y= 0, high=13
-			pins_b |= (0x20u & p[ 2]) >> 1; // x= 8, y= 1, high=12
-			pins_b |= (0x20u & p[ 4]) >> 2; // x= 8, y= 2, high=11
-			pins_b |= (0x20u & p[ 6]) >> 3; // x= 8, y= 3, high=10
-			pins_b |= (0x20u & p[ 8]) >> 4; // x= 8, y= 4, high= 9
-			pins_b |= (0x20u & p[10]) >> 5; // x= 8, y= 5, high= 8
-			pins_d |= (0x20u & p[12]) << 2; // x= 8, y= 6, high= 7
-			pins_d |= (0x20u & p[14]) << 1; // x= 8, y= 7, high= 6
-			pins_d |= (0x20u & p[16]);      // x= 8, y= 8, high= 5
+			pins_b |= (0x20u & p[ 0]);      // x= 5, y= 0, mapped pin D13
+			pins_b |= (0x20u & p[ 2]) >> 1; // x= 5, y= 1, mapped pin D12
+			pins_b |= (0x20u & p[ 4]) >> 2; // x= 5, y= 2, mapped pin D11
+			pins_b |= (0x20u & p[ 6]) >> 3; // x= 5, y= 3, mapped pin D10
+			pins_b |= (0x20u & p[ 8]) >> 4; // x= 5, y= 4, mapped pin D9
+			pins_b |= (0x20u & p[10]) >> 5; // x= 5, y= 5, mapped pin D8
+			pins_d |= (0x20u & p[12]) << 2; // x= 5, y= 6, mapped pin D7
+			pins_d |= (0x20u & p[14]) << 1; // x= 5, y= 7, mapped pin D6
+			pins_d |= (0x20u & p[16]);      // x= 5, y= 8, mapped pin D5
 			break;
 		case 3:
-			pins_b |= (0x20u & p[ 1]);      // x= 0, y= 0, high=13
-			pins_b |= (0x20u & p[ 3]) >> 1; // x= 0, y= 1, high=12
-			pins_b |= (0x20u & p[ 5]) >> 2; // x= 0, y= 2, high=11
-			pins_b |= (0x20u & p[ 7]) >> 3; // x= 0, y= 3, high=10
-			pins_b |= (0x20u & p[ 9]) >> 4; // x= 0, y= 4, high= 9
-			pins_b |= (0x20u & p[11]) >> 5; // x= 0, y= 5, high= 8
-			pins_d |= (0x01u & p[16]) << 2; // x=13, y= 8, high= 2
-			pins_d |= (0x04u & p[16]) << 1; // x=11, y= 8, high= 3
-			pins_d |= (0x10u & p[16]);      // x= 9, y= 8, high= 4
-			pins_d |= (0x20u & p[13]) << 2; // x= 0, y= 6, high= 7
-			pins_d |= (0x20u & p[15]) << 1; // x= 0, y= 7, high= 6
+			pins_b |= (0x20u & p[ 1]);      // x=13, y= 0, mapped pin D13
+			pins_b |= (0x20u & p[ 3]) >> 1; // x=13, y= 1, mapped pin D12
+			pins_b |= (0x20u & p[ 5]) >> 2; // x=13, y= 2, mapped pin D11
+			pins_b |= (0x20u & p[ 7]) >> 3; // x=13, y= 3, mapped pin D10
+			pins_b |= (0x20u & p[ 9]) >> 4; // x=13, y= 4, mapped pin D9
+			pins_b |= (0x20u & p[11]) >> 5; // x=13, y= 5, mapped pin D8
+			pins_d |= (0x01u & p[16]) << 2; // x= 0, y= 8, mapped pin D2
+			pins_d |= (0x04u & p[16]) << 1; // x= 2, y= 8, mapped pin D3
+			pins_d |= (0x10u & p[16]);      // x= 4, y= 8, mapped pin D4
+			pins_d |= (0x20u & p[13]) << 2; // x=13, y= 6, mapped pin D7
+			pins_d |= (0x20u & p[15]) << 1; // x=13, y= 7, mapped pin D6
 			break;
 		case 4:
-			pins_b |= (0x10u & p[ 1]) << 1; // x= 1, y= 0, high=13
-			pins_b |= (0x10u & p[ 3]);      // x= 1, y= 1, high=12
-			pins_b |= (0x10u & p[ 5]) >> 1; // x= 1, y= 2, high=11
-			pins_b |= (0x10u & p[ 7]) >> 2; // x= 1, y= 3, high=10
-			pins_b |= (0x10u & p[ 9]) >> 3; // x= 1, y= 4, high= 9
-			pins_b |= (0x10u & p[11]) >> 4; // x= 1, y= 5, high= 8
-			pins_d |= (0x01u & p[14]) << 2; // x=13, y= 7, high= 2
-			pins_d |= (0x04u & p[14]) << 1; // x=11, y= 7, high= 3
-			pins_d |= (0x10u & p[13]) << 3; // x= 1, y= 6, high= 7
-			pins_d |= (0x10u & p[14]);      // x= 9, y= 7, high= 4
-			pins_d |= (0x20u & p[17]);      // x= 0, y= 8, high= 5
+			pins_b |= (0x10u & p[ 1]) << 1; // x=12, y= 0, mapped pin D13
+			pins_b |= (0x10u & p[ 3]);      // x=12, y= 1, mapped pin D12
+			pins_b |= (0x10u & p[ 5]) >> 1; // x=12, y= 2, mapped pin D11
+			pins_b |= (0x10u & p[ 7]) >> 2; // x=12, y= 3, mapped pin D10
+			pins_b |= (0x10u & p[ 9]) >> 3; // x=12, y= 4, mapped pin D9
+			pins_b |= (0x10u & p[11]) >> 4; // x=12, y= 5, mapped pin D8
+			pins_d |= (0x01u & p[14]) << 2; // x= 0, y= 7, mapped pin D2
+			pins_d |= (0x04u & p[14]) << 1; // x= 2, y= 7, mapped pin D3
+			pins_d |= (0x10u & p[13]) << 3; // x=12, y= 6, mapped pin D7
+			pins_d |= (0x10u & p[14]);      // x= 4, y= 7, mapped pin D4
+			pins_d |= (0x20u & p[17]);      // x=13, y= 8, mapped pin D5
 			break;
 		case 5:
-			pins_b |= (0x08u & p[ 1]) << 2; // x= 2, y= 0, high=13
-			pins_b |= (0x08u & p[ 3]) << 1; // x= 2, y= 1, high=12
-			pins_b |= (0x08u & p[ 5]);      // x= 2, y= 2, high=11
-			pins_b |= (0x08u & p[ 7]) >> 1; // x= 2, y= 3, high=10
-			pins_b |= (0x08u & p[ 9]) >> 2; // x= 2, y= 4, high= 9
-			pins_b |= (0x08u & p[11]) >> 3; // x= 2, y= 5, high= 8
-			pins_d |= (0x01u & p[12]) << 2; // x=13, y= 6, high= 2
-			pins_d |= (0x04u & p[12]) << 1; // x=11, y= 6, high= 3
-			pins_d |= (0x10u & p[12]);      // x= 9, y= 6, high= 4
-			pins_d |= (0x10u & p[15]) << 2; // x= 1, y= 7, high= 6
-			pins_d |= (0x10u & p[17]) << 1; // x= 1, y= 8, high= 5
+			pins_b |= (0x08u & p[ 1]) << 2; // x=11, y= 0, mapped pin D13
+			pins_b |= (0x08u & p[ 3]) << 1; // x=11, y= 1, mapped pin D12
+			pins_b |= (0x08u & p[ 5]);      // x=11, y= 2, mapped pin D11
+			pins_b |= (0x08u & p[ 7]) >> 1; // x=11, y= 3, mapped pin D10
+			pins_b |= (0x08u & p[ 9]) >> 2; // x=11, y= 4, mapped pin D9
+			pins_b |= (0x08u & p[11]) >> 3; // x=11, y= 5, mapped pin D8
+			pins_d |= (0x01u & p[12]) << 2; // x= 0, y= 6, mapped pin D2
+			pins_d |= (0x04u & p[12]) << 1; // x= 2, y= 6, mapped pin D3
+			pins_d |= (0x10u & p[12]);      // x= 4, y= 6, mapped pin D4
+			pins_d |= (0x10u & p[15]) << 2; // x=12, y= 7, mapped pin D6
+			pins_d |= (0x10u & p[17]) << 1; // x=12, y= 8, mapped pin D5
 			break;
 		case 6:
-			pins_b |= (0x04u & p[ 1]) << 3; // x= 3, y= 0, high=13
-			pins_b |= (0x04u & p[ 3]) << 2; // x= 3, y= 1, high=12
-			pins_b |= (0x04u & p[ 5]) << 1; // x= 3, y= 2, high=11
-			pins_b |= (0x04u & p[ 7]);      // x= 3, y= 3, high=10
-			pins_b |= (0x04u & p[ 9]) >> 1; // x= 3, y= 4, high= 9
-			pins_d |= (0x01u & p[10]) << 2; // x=13, y= 5, high= 2
-			pins_d |= (0x04u & p[10]) << 1; // x=11, y= 5, high= 3
-			pins_d |= (0x08u & p[13]) << 4; // x= 2, y= 6, high= 7
-			pins_d |= (0x08u & p[15]) << 3; // x= 2, y= 7, high= 6
-			pins_d |= (0x08u & p[17]) << 2; // x= 2, y= 8, high= 5
-			pins_d |= (0x10u & p[10]);      // x= 9, y= 5, high= 4
+			pins_b |= (0x04u & p[ 1]) << 3; // x=10, y= 0, mapped pin D13
+			pins_b |= (0x04u & p[ 3]) << 2; // x=10, y= 1, mapped pin D12
+			pins_b |= (0x04u & p[ 5]) << 1; // x=10, y= 2, mapped pin D11
+			pins_b |= (0x04u & p[ 7]);      // x=10, y= 3, mapped pin D10
+			pins_b |= (0x04u & p[ 9]) >> 1; // x=10, y= 4, mapped pin D9
+			pins_d |= (0x01u & p[10]) << 2; // x= 0, y= 5, mapped pin D2
+			pins_d |= (0x04u & p[10]) << 1; // x= 2, y= 5, mapped pin D3
+			pins_d |= (0x08u & p[13]) << 4; // x=11, y= 6, mapped pin D7
+			pins_d |= (0x08u & p[15]) << 3; // x=11, y= 7, mapped pin D6
+			pins_d |= (0x08u & p[17]) << 2; // x=11, y= 8, mapped pin D5
+			pins_d |= (0x10u & p[10]);      // x= 4, y= 5, mapped pin D4
 			break;
 		case 7:
-			pins_b |= (0x02u & p[ 1]) << 4; // x= 4, y= 0, high=13
-			pins_b |= (0x02u & p[ 3]) << 3; // x= 4, y= 1, high=12
-			pins_b |= (0x02u & p[ 5]) << 2; // x= 4, y= 2, high=11
-			pins_b |= (0x02u & p[ 7]) << 1; // x= 4, y= 3, high=10
-			pins_b |= (0x04u & p[11]) >> 2; // x= 3, y= 5, high= 8
-			pins_d |= (0x01u & p[ 8]) << 2; // x=13, y= 4, high= 2
-			pins_d |= (0x04u & p[ 8]) << 1; // x=11, y= 4, high= 3
-			pins_d |= (0x04u & p[13]) << 5; // x= 3, y= 6, high= 7
-			pins_d |= (0x04u & p[15]) << 4; // x= 3, y= 7, high= 6
-			pins_d |= (0x04u & p[17]) << 3; // x= 3, y= 8, high= 5
-			pins_d |= (0x10u & p[ 8]);      // x= 9, y= 4, high= 4
+			pins_b |= (0x02u & p[ 1]) << 4; // x= 9, y= 0, mapped pin D13
+			pins_b |= (0x02u & p[ 3]) << 3; // x= 9, y= 1, mapped pin D12
+			pins_b |= (0x02u & p[ 5]) << 2; // x= 9, y= 2, mapped pin D11
+			pins_b |= (0x02u & p[ 7]) << 1; // x= 9, y= 3, mapped pin D10
+			pins_b |= (0x04u & p[11]) >> 2; // x=10, y= 5, mapped pin D8
+			pins_d |= (0x01u & p[ 8]) << 2; // x= 0, y= 4, mapped pin D2
+			pins_d |= (0x04u & p[ 8]) << 1; // x= 2, y= 4, mapped pin D3
+			pins_d |= (0x04u & p[13]) << 5; // x=10, y= 6, mapped pin D7
+			pins_d |= (0x04u & p[15]) << 4; // x=10, y= 7, mapped pin D6
+			pins_d |= (0x04u & p[17]) << 3; // x=10, y= 8, mapped pin D5
+			pins_d |= (0x10u & p[ 8]);      // x= 4, y= 4, mapped pin D4
 			break;
 		case 8:
-			pins_b |= (0x01u & p[ 1]) << 5; // x= 5, y= 0, high=13
-			pins_b |= (0x01u & p[ 3]) << 4; // x= 5, y= 1, high=12
-			pins_b |= (0x01u & p[ 5]) << 3; // x= 5, y= 2, high=11
-			pins_b |= (0x02u & p[ 9]);      // x= 4, y= 4, high= 9
-			pins_b |= (0x02u & p[11]) >> 1; // x= 4, y= 5, high= 8
-			pins_d |= (0x01u & p[ 6]) << 2; // x=13, y= 3, high= 2
-			pins_d |= (0x02u & p[13]) << 6; // x= 4, y= 6, high= 7
-			pins_d |= (0x02u & p[15]) << 5; // x= 4, y= 7, high= 6
-			pins_d |= (0x02u & p[17]) << 4; // x= 4, y= 8, high= 5
-			pins_d |= (0x04u & p[ 6]) << 1; // x=11, y= 3, high= 3
-			pins_d |= (0x10u & p[ 6]);      // x= 9, y= 3, high= 4
+			pins_b |= (0x01u & p[ 1]) << 5; // x= 8, y= 0, mapped pin D13
+			pins_b |= (0x01u & p[ 3]) << 4; // x= 8, y= 1, mapped pin D12
+			pins_b |= (0x01u & p[ 5]) << 3; // x= 8, y= 2, mapped pin D11
+			pins_b |= (0x02u & p[ 9]);      // x= 9, y= 4, mapped pin D9
+			pins_b |= (0x02u & p[11]) >> 1; // x= 9, y= 5, mapped pin D8
+			pins_d |= (0x01u & p[ 6]) << 2; // x= 0, y= 3, mapped pin D2
+			pins_d |= (0x02u & p[13]) << 6; // x= 9, y= 6, mapped pin D7
+			pins_d |= (0x02u & p[15]) << 5; // x= 9, y= 7, mapped pin D6
+			pins_d |= (0x02u & p[17]) << 4; // x= 9, y= 8, mapped pin D5
+			pins_d |= (0x04u & p[ 6]) << 1; // x= 2, y= 3, mapped pin D3
+			pins_d |= (0x10u & p[ 6]);      // x= 4, y= 3, mapped pin D4
 			break;
 		case 9:
-			pins_b |= (0x01u & p[ 7]) << 2; // x= 5, y= 3, high=10
-			pins_b |= (0x01u & p[ 9]) << 1; // x= 5, y= 4, high= 9
-			pins_b |= (0x01u & p[11]);      // x= 5, y= 5, high= 8
-			pins_b |= (0x80u & p[ 0]) >> 2; // x= 6, y= 0, high=13
-			pins_b |= (0x80u & p[ 2]) >> 3; // x= 6, y= 1, high=12
-			pins_d |= (0x01u & p[ 4]) << 2; // x=13, y= 2, high= 2
-			pins_d |= (0x01u & p[13]) << 7; // x= 5, y= 6, high= 7
-			pins_d |= (0x01u & p[15]) << 6; // x= 5, y= 7, high= 6
-			pins_d |= (0x01u & p[17]) << 5; // x= 5, y= 8, high= 5
-			pins_d |= (0x04u & p[ 4]) << 1; // x=11, y= 2, high= 3
-			pins_d |= (0x10u & p[ 4]);      // x= 9, y= 2, high= 4
+			pins_b |= (0x01u & p[ 7]) << 2; // x= 8, y= 3, mapped pin D10
+			pins_b |= (0x01u & p[ 9]) << 1; // x= 8, y= 4, mapped pin D9
+			pins_b |= (0x01u & p[11]);      // x= 8, y= 5, mapped pin D8
+			pins_b |= (0x80u & p[ 0]) >> 2; // x= 7, y= 0, mapped pin D13
+			pins_b |= (0x80u & p[ 2]) >> 3; // x= 7, y= 1, mapped pin D12
+			pins_d |= (0x01u & p[ 4]) << 2; // x= 0, y= 2, mapped pin D2
+			pins_d |= (0x01u & p[13]) << 7; // x= 8, y= 6, mapped pin D7
+			pins_d |= (0x01u & p[15]) << 6; // x= 8, y= 7, mapped pin D6
+			pins_d |= (0x01u & p[17]) << 5; // x= 8, y= 8, mapped pin D5
+			pins_d |= (0x04u & p[ 4]) << 1; // x= 2, y= 2, mapped pin D3
+			pins_d |= (0x10u & p[ 4]);      // x= 4, y= 2, mapped pin D4
 			break;
 		case 10:
-			pins_b |= (0x40u & p[ 0]) >> 1; // x= 7, y= 0, high=13
-			pins_b |= (0x80u & p[ 4]) >> 4; // x= 6, y= 2, high=11
-			pins_b |= (0x80u & p[ 6]) >> 5; // x= 6, y= 3, high=10
-			pins_b |= (0x80u & p[ 8]) >> 6; // x= 6, y= 4, high= 9
-			pins_b |= (0x80u & p[10]) >> 7; // x= 6, y= 5, high= 8
-			pins_d |= (0x01u & p[ 2]) << 2; // x=13, y= 1, high= 2
-			pins_d |= (0x04u & p[ 2]) << 1; // x=11, y= 1, high= 3
-			pins_d |= (0x10u & p[ 2]);      // x= 9, y= 1, high= 4
-			pins_d |= (0x80u & p[12]);      // x= 6, y= 6, high= 7
-			pins_d |= (0x80u & p[14]) >> 1; // x= 6, y= 7, high= 6
-			pins_d |= (0x80u & p[16]) >> 2; // x= 6, y= 8, high= 5
+			pins_b |= (0x40u & p[ 0]) >> 1; // x= 6, y= 0, mapped pin D13
+			pins_b |= (0x80u & p[ 4]) >> 4; // x= 7, y= 2, mapped pin D11
+			pins_b |= (0x80u & p[ 6]) >> 5; // x= 7, y= 3, mapped pin D10
+			pins_b |= (0x80u & p[ 8]) >> 6; // x= 7, y= 4, mapped pin D9
+			pins_b |= (0x80u & p[10]) >> 7; // x= 7, y= 5, mapped pin D8
+			pins_d |= (0x01u & p[ 2]) << 2; // x= 0, y= 1, mapped pin D2
+			pins_d |= (0x04u & p[ 2]) << 1; // x= 2, y= 1, mapped pin D3
+			pins_d |= (0x10u & p[ 2]);      // x= 4, y= 1, mapped pin D4
+			pins_d |= (0x80u & p[12]);      // x= 7, y= 6, mapped pin D7
+			pins_d |= (0x80u & p[14]) >> 1; // x= 7, y= 7, mapped pin D6
+			pins_d |= (0x80u & p[16]) >> 2; // x= 7, y= 8, mapped pin D5
 			break;
 		case 11:
-			pins_b |= (0x40u & p[ 2]) >> 2; // x= 7, y= 1, high=12
-			pins_b |= (0x40u & p[ 4]) >> 3; // x= 7, y= 2, high=11
-			pins_b |= (0x40u & p[ 6]) >> 4; // x= 7, y= 3, high=10
-			pins_b |= (0x40u & p[ 8]) >> 5; // x= 7, y= 4, high= 9
-			pins_b |= (0x40u & p[10]) >> 6; // x= 7, y= 5, high= 8
-			pins_d |= (0x01u & p[ 0]) << 2; // x=13, y= 0, high= 2
-			pins_d |= (0x04u & p[ 0]) << 1; // x=11, y= 0, high= 3
-			pins_d |= (0x10u & p[ 0]);      // x= 9, y= 0, high= 4
-			pins_d |= (0x40u & p[12]) << 1; // x= 7, y= 6, high= 7
-			pins_d |= (0x40u & p[14]);      // x= 7, y= 7, high= 6
-			pins_d |= (0x40u & p[16]) >> 1; // x= 7, y= 8, high= 5
+			pins_b |= (0x40u & p[ 2]) >> 2; // x= 6, y= 1, mapped pin D12
+			pins_b |= (0x40u & p[ 4]) >> 3; // x= 6, y= 2, mapped pin D11
+			pins_b |= (0x40u & p[ 6]) >> 4; // x= 6, y= 3, mapped pin D10
+			pins_b |= (0x40u & p[ 8]) >> 5; // x= 6, y= 4, mapped pin D9
+			pins_b |= (0x40u & p[10]) >> 6; // x= 6, y= 5, mapped pin D8
+			pins_d |= (0x01u & p[ 0]) << 2; // x= 0, y= 0, mapped pin D2
+			pins_d |= (0x04u & p[ 0]) << 1; // x= 2, y= 0, mapped pin D3
+			pins_d |= (0x10u & p[ 0]);      // x= 4, y= 0, mapped pin D4
+			pins_d |= (0x40u & p[12]) << 1; // x= 6, y= 6, mapped pin D7
+			pins_d |= (0x40u & p[14]);      // x= 6, y= 7, mapped pin D6
+			pins_d |= (0x40u & p[16]) >> 1; // x= 6, y= 8, mapped pin D5
 			break;
 		}
     }
@@ -289,8 +826,40 @@ ISR(TIMER2_OVF_vect) {
 	// Set sink pin to GND/sink, turning on current.
 	PIND = sink_d;
 	PINB = sink_b;
+#endif
+}
 
-	plane++;
+#if !defined (__AVR_ATmega32U4__)
+ISR(TIMER2_OVF_vect) {
+#else
+ISR(TIMER1_COMPA_vect) {
+#endif
+	// For each cycle, we have potential planes to display. Once every plane has
+	// been displayed, then we move on to the next cycle.
+	// NOTE: a "cycle" is a subset of LEDs that can be driven at once.
+
+	// 12 Cycles of Matrix
+	static uint8_t cycle = 0;
+
+	// planes to display
+	// NOTE: a "plane" in the Borgware is the same as a "page" in Jimmie's lib
+	static uint8_t plane = 0;
+
+#if defined (__AVR_ATmega168__) || defined (__AVR_ATmega48__) || defined (__AVR_ATmega88__) || defined (__AVR_ATmega328P__) || defined (__AVR_ATmega1280__) || defined (__AVR_ATmega2560__)
+	TCCR2B = prescaler[plane];
+#elif defined (__AVR_ATmega8__) || defined (__AVR_ATmega128__)
+	TCCR2 = prescaler[page];
+#elif defined (__AVR_ATmega32U4__)
+	TCCR1B = prescaler[plane];
+#endif
+#if !defined (__AVR_ATmega32U4__)
+	TCNT2 = counts[plane];
+#else
+	TCNT1 = counts[plane];
+#endif
+
+	// distribute framebuffer contents among current cycle pins
+	compose_cycle(cycle, plane++);
 
 	if (plane >= (NUMPLANE + 1)) {
 		plane = 0;
@@ -303,20 +872,49 @@ ISR(TIMER2_OVF_vect) {
 }
 
 void borg_hw_init() {
-	// prepare Timer2
+
+#if defined (__AVR_ATmega168__) || defined (__AVR_ATmega48__) || defined (__AVR_ATmega88__) || defined (__AVR_ATmega328P__) || defined (__AVR_ATmega1280__) || defined (__AVR_ATmega2560__)
 	TIMSK2 &= ~(_BV(TOIE2) | _BV(OCIE2A));
 	TCCR2A &= ~(_BV(WGM21) | _BV(WGM20));
 	TCCR2B &= ~_BV(WGM22);
 	ASSR &= ~_BV(AS2);
+#elif defined (__AVR_ATmega8__)
+	TIMSK &= ~(_BV(TOIE2) | _BV(OCIE2));
+	TCCR2 &= ~(_BV(WGM21) | _BV(WGM20));
+	ASSR &= ~_BV(AS2);
+#elif defined (__AVR_ATmega128__)
+	TIMSK &= ~(_BV(TOIE2) | _BV(OCIE2));
+	TCCR2 &= ~(_BV(WGM21) | _BV(WGM20));
+#elif defined (__AVR_ATmega32U4__)
+	// The only 8bit timer on the Leonardo is used by default, so we use the 16bit Timer1
+	// in CTC mode with a compare value of 256 to achieve the same behaviour.
+	TIMSK1 &= ~(_BV(TOIE1) | _BV(OCIE1A));
+	TCCR1A &= ~(_BV(WGM10) | _BV(WGM11));
+	OCR1A = 256;
+#endif
 
-    // Then start the display
-    TIMSK2 |= _BV(TOIE2);
-    TCCR2B = fastPrescaler;
+	setBrightness();
 
-    // interrupt ASAP
-    TCNT2 = 0xff;
+	// Then start the display
+#if defined (__AVR_ATmega168__) || defined (__AVR_ATmega48__) || defined (__AVR_ATmega88__) || defined (__AVR_ATmega328P__) || defined (__AVR_ATmega1280__) || defined (__AVR_ATmega2560__)
+	TIMSK2 |= _BV(TOIE2);
+	TCCR2B = FASTPRESCALER;
+#elif defined (__AVR_ATmega8__) || defined (__AVR_ATmega128__)
+	TIMSK |= _BV(TOIE2);
+	TCCR2 = FASTPRESCALER;
+#elif defined (__AVR_ATmega32U4__)
+	// Enable output compare match interrupt
+	TIMSK1 |= _BV(OCIE1A);
+	TCCR1B = FASTPRESCALER;
+#endif
+	// interrupt ASAP
+#if !defined (__AVR_ATmega32U4__)
+		TCNT2 = 255;
+#else
+		TCNT1 = 255;
+#endif
 
 	// activate watchdog timer
 	wdt_reset();
-	wdt_enable(0x00); // 17ms watchdog
+	wdt_enable(WDTO_500MS); // 17ms watchdog
 }
diff --git a/src/borg_hw/config.in b/src/borg_hw/config.in
index 9aabae5..7c1d244 100644
--- a/src/borg_hw/config.in
+++ b/src/borg_hw/config.in
@@ -14,7 +14,7 @@ choice 'Hardware Driver'			\
     Andre-Borg             HW_BORG_ANDRE \
     Laufschrift-Borg	   HW_BORG_LS \
     Laufschrift-Borg-MH    HW_BORG_MH \
-    LoL-Shield/Diavolino   HW_LOLSHIELD \
+    LoL-Shield             HW_LOLSHIELD \
     FFM-Jochen             HW_BORG_LSJO \
     FFM-LedBrett           HW_LEDBRETT \
     Borg-Mini              HW_BORG_MINI \
diff --git a/src/borg_hw/config_lolshield.in b/src/borg_hw/config_lolshield.in
index ceaf309..fd99bed 100644
--- a/src/borg_hw/config_lolshield.in
+++ b/src/borg_hw/config_lolshield.in
@@ -3,4 +3,6 @@ comment "lolshield setup"
 
 define_int USER_TIMER0_FOR_WAIT 1
 
+uint "Brightness (0-127)"  BRIGHTNESS  127
+
 endmenu
diff --git a/src/util.c b/src/util.c
index bdd2a29..3b608e9 100644
--- a/src/util.c
+++ b/src/util.c
@@ -21,7 +21,6 @@ extern jmp_buf newmode_jmpbuf;
 
 
 void wait(int ms){
-
 /* Always use Timer1 except for the Arduino/LoL Shield platform. */
 #ifndef USER_TIMER0_FOR_WAIT
 	TCCR1B = _BV(WGM12) | _BV(CS12); //CTC Mode, clk/256
@@ -36,7 +35,6 @@ void wait(int ms){
 	OCR0A = (F_CPU/256000);	//1000Hz
 #endif
 
-
 	for(;ms>0;ms--){
 
 #ifdef CAN_SUPPORT
@@ -57,7 +55,7 @@ void wait(int ms){
 		}
 #endif
 
-#if defined (__AVR_ATmega644P__) || defined (__AVR_ATmega644__) || defined (__AVR_ATmega328__) || defined (__AVR_ATmega328P__) || (__AVR_ATmega1284P__) || defined (__AVR_ATmega1284__)
+#if defined (__AVR_ATmega32U4__) || defined (__AVR_ATmega644P__) || defined (__AVR_ATmega644__) || defined (__AVR_ATmega328__) || defined (__AVR_ATmega328P__) || (__AVR_ATmega1284P__) || defined (__AVR_ATmega1284__)
 /* Timer1 for the masses */
 #	ifndef USER_TIMER0_FOR_WAIT
 		while(!(TIFR1 & _BV(OCF1A))); //wait for compare match flag