[Mono-bugs] [Bug 43597][Wis] Changed - When Random () is started with a seed 0, it always returns 0 for NextValue

[bugzilla-daemon@rocky.ximian.com]

Please do not reply to this email- if you want to comment on the bug, go to the
URL shown below and enter your comments there.

Changed by bmaurer@users.sf.net.

http://bugzilla.ximian.com/show_bug.cgi?id=43597

--- shadow/43597	Wed May 28 22:43:38 2003
+++ shadow/43597.tmp.11331	Wed May 28 22:44:08 2003
@@ -5,13 +5,13 @@
 OS Details: 
 Status: NEW   
 Resolution: 
 Severity: 001 One hour
 Priority: Wishlist
 Component: CORLIB
-AssignedTo: mono-bugs@ximian.com                            
+AssignedTo: bmaurer@users.sf.net                            
 ReportedBy: giuseppe.greco@agamura.com               
 QAContact: mono-bugs@ximian.com
 TargetMilestone: ---
 URL: 
 Cc: giuseppe.greco@agamura.com,spouliot@videotron.ca
 Summary: When Random () is started with a seed 0, it always returns 0 for NextValue
@@ -50,6 +50,19 @@
 We need to re-think the method we use on the RNG.
 
 ------- Additional Comments From spouliot@videotron.ca  2003-05-28 22:43 -------
 Current algorithm is standard but as a "small type" clause:
 "Set *idum (to be modified) to any nonzero integer value to 
 initialize the sequence."
+
+------- Additional Comments From bmaurer@users.sf.net  2003-05-28 22:44 -------
+It looks like Microsoft uses an alg from Knuth.
+
+It is written in Java here:
+http://db-www.aist-nara.ac.jp/members/shingo-f/zakki/java01.html.
+
+I think we should switch to this alg so that we have no bugs in our
+implementation that MS does not have. Java does something similar,
+their RNG is standerdized.
+
+I will translate the code. It will need to be reviewed (for accuracy,
+security).