diff --git a/code/b-parasite/config/prst_config.h b/code/b-parasite/config/prst_config.h
index 2fc7349..816320f 100644
--- a/code/b-parasite/config/prst_config.h
+++ b/code/b-parasite/config/prst_config.h
@@ -14,8 +14,7 @@
 #define PRST_LED_PIN NRF_GPIO_PIN_MAP(0, 28)
 
 // Deep sleep.
-// #define PRST_DEEP_SLEEP_IN_SECONDS 2
-#define PRST_DEEP_SLEEP_IN_SECONDS 2 * 3600
+#define PRST_DEEP_SLEEP_IN_SECONDS 300
 
 // Analog to digital converter (ADC).
 // Prints out ADC debug info, such as the values read for battery and soil
@@ -31,7 +30,7 @@
 // 1. Two most significant bits are set to 1;
 // 2. The remaining bits should not _all_ be set to 0;
 // 2. The remaining bits should not _all_ be set to 1;
-#define PRST_BLE_MAC_ADDR "f0:ca:f0:ca:00:10"
+#define PRST_BLE_MAC_ADDR "f0:ca:f0:ca:01:01"
 #define PRST_BLE_ADV_NAME "prst"
 // Total time spend advertising.
 #define PRST_BLE_ADV_TIME_IN_MS 1000
@@ -67,7 +66,6 @@
 // Whether to produce debug messages for the LDR
 #define PRST_ADC_PHOTO_DEBUG 0
 
-// #endif  // End of version-specific configuration.
 #elif defined(PRST_VERSION_1_2_X)
 
 #define PRST_HAS_PHOTOTRANSISTOR 1
@@ -75,7 +73,7 @@
 #define PRST_PHOTO_V_PIN NRF_GPIO_PIN_MAP(0, 29)
 #define PRST_PHOTO_OUT_PIN NRF_GPIO_PIN_MAP(0, 2)
 
-#define PRST_ADC_PHOTO_DEBUG 1
+#define PRST_ADC_PHOTO_DEBUG 0
 
 #endif  // End of version-specific configuration.
 
diff --git a/code/b-parasite/src/main.c b/code/b-parasite/src/main.c
index 99549df..821da77 100644
--- a/code/b-parasite/src/main.c
+++ b/code/b-parasite/src/main.c
@@ -69,13 +69,9 @@ static void rtc_callback() {
   uint16_t lux = 0;
 #if PRST_HAS_LDR || PRST_HAS_PHOTOTRANSISTOR
   nrf_gpio_pin_set(PRST_PHOTO_V_PIN);
-
-  // Just while debugging.
-  while (1) {
-    nrf_delay_ms(1000);
-    prst_adc_photo_sensor_t photo_read = prst_adc_photo_read(batt_read.voltage);
-    lux = photo_read.brightness;
-  }
+  nrf_delay_ms(50);
+  prst_adc_photo_sensor_t photo_read = prst_adc_photo_read(batt_read.voltage);
+  lux = photo_read.brightness;
   nrf_gpio_pin_clear(PRST_PHOTO_V_PIN);
 #endif
 
diff --git a/code/b-parasite/src/prst/adc.c b/code/b-parasite/src/prst/adc.c
index fe1fe99..86a5781 100644
--- a/code/b-parasite/src/prst/adc.c
+++ b/code/b-parasite/src/prst/adc.c
@@ -130,16 +130,10 @@ prst_adc_soil_moisture_t prst_adc_soil_read(double battery_voltage) {
 
 prst_adc_photo_sensor_t prst_adc_photo_read(double battery_voltage) {
   nrf_saadc_value_t raw_photo_output =
-      sample_adc_channel(PRST_ADC_PHOTO_CHANNEL);
-
-  if (raw_photo_output < 0) {
-    raw_photo_output = 0;
-  }
-
+      MAX(0, sample_adc_channel(PRST_ADC_PHOTO_CHANNEL));
   prst_adc_photo_sensor_t ret;
   ret.raw = raw_photo_output;
   ret.voltage = (3.6 * raw_photo_output) / (1 << PRST_ADC_RESOLUTION);
-  // ret.voltage = (2.4 * raw_photo_output) / (1 << PRST_ADC_RESOLUTION);
 
 #if PRST_HAS_LDR
   // The photo resistor forms a voltage divider with a 10 kOhm resistor.
@@ -165,37 +159,11 @@ prst_adc_photo_sensor_t prst_adc_photo_read(double battery_voltage) {
   // its two terminals is controlled by how much light there is in the ambient.
   // We measure that current by calculating the voltage across a resistor that
   // is connected in series with the phototransistor.
-  // Some infor:
-  // - Not all lights are the same. The ALS-PT19 has different current
-  // responses for incandescent and fluorescent lights and it shows no values
-  // for sunlight. We have to make some compromises here, aiming for a
-  // "reasonable" middle ground through calibration.
-  // - ALS-PT19' minimum voltage is 2.5 V. Our CR2032 battery has a theoretical
-  // range of 2.0 V to 3.0 V, but in our usage pattern it seems to spend most of
-  // its life around 2.6 V (https://github.com/rbaron/b-parasite/issues/1). In
-  // my manual tests it seems we can usually go lower than that. So while we're
-  // pushing it a bit, we should be okay most of the time.
-  //
-  // In order to design the value of the series resistor, we assume Vcc = 2.5V,
-  // and we want the upper limit of light intensity to correspond to a value of
-  // Vout < 2.5 V - 0.4 V (saturation) = 2.1 V. Let's call this the "direct
-  // sunlight" voltage across our resistor. This direct sunlight voltage will
-  // appear when the phototransistor outputs the direct sunlight current.
-  //
-  // In short, what we want:
-  // A value of R_L such that Vout is < 2.1 V (but close) when the sensor is in
-  // direct sunlight. While Vcc is 2.5 V, R_L = 470 Ohm outputs Vout ~ 1.7V, so
-  // there's still some wiggle room for even more sunnier days.
-  //
-  // Another caveat: the datasheet shows that the current in the transistor is
-  // relatively constant when we vary Vcc (Fig.4). This seems to be true for low
-  // currents (the datsheet uses 100 lx in Fig.4), but not for larger currents.
   const float phototransistor_resistor = 470.0f;
   const float current_sun = 3.59e-3f;
   // Assuming 10000 lux for the saturation test. Calibration with a proper light
   // meter would be better.
   const float lux_sun = 10000.0f;
-
   const float current = ret.voltage / phototransistor_resistor;
   ret.brightness = MAX(0, MIN(lux_sun * current / current_sun, UINT16_MAX));