Index: trunk/MagicSoft/Mars/NEWS
===================================================================
--- trunk/MagicSoft/Mars/NEWS	(revision 8885)
+++ trunk/MagicSoft/Mars/NEWS	(revision 8889)
@@ -23,5 +23,6 @@
      events of some trees could have been skipped. Since this is only a
      very tiny part of the data no severe effect to your analysis results 
-     is expected.
+     is expected (an exception might be very short, i.e. singel runs, 
+     datasets).
 
    * Dataset file now allow to include the sequence file directly within
@@ -34,4 +35,9 @@
    * added code to process the sum-trigger flag
 
+   * the histogram classes MH3 and MHn now support profile histograms
+
+   * the Random Forest Train-classes (MJTRain*) now store the used datasets
+     in the output file
+
  ;merpp
 
@@ -50,4 +56,10 @@
     * Both pedestal cams now conatain the number of events processed
 
+    * In Mars V2.0 the extraction of the fundamental pedestal was broken, 
+      i.e. for the extraction of the calibration pulses basically the average
+      over all events of the first slice instead of all slices was used.
+      Since pedestal and calibration constants are continously recalculated
+      this only effected the very first events of every sequence.
+
  ;star
 
@@ -59,8 +71,26 @@
       than one sigma. Points with null or negative significances are omitted.
 
+    * DrawNicePlot now allows to set a projection range
+
  ;sponde
 
     * fixed resource files ('''Spectrum''' instead of '''MJSpectrum''')
 
+    * a new option "--force-runtime" is available. In case of very short
+      datasets (for light-curves) which are in the order of a few
+      minutes the calculated effective on-time is not very acurate because
+      it is quantizised in th eorder of a minute. Therefore you can switch
+      to using the real run-time instead (remakrt: this ignores any
+      dead-time!)
+
+    * a more analytical way is used to "fill" Monte Carlos into regions 
+      of impact parameters which have not been simulated because the
+      trigger efficiency is assumed to be 0.
+
+    * estimated sensitivity curves for hi and lo zenith angles are shown
+      scaled to the correct observation time and collection area
+
+    * the Monte Carlo events after cuts are now written to the output file
+      if an output file given