+     nevries
From: (Nico E. de Vries)
Newsgroups: sci.crypt

Subject: IBM-PC random generator, source included
Message-ID: <>
Date: 22 Jun 92 19:40:06 GMT
Organization: Academic Computer Centre Utrecht
Lines: 154


I have to give up. I hoped the number of requests for my random
generator would fade away a bit but the oposite is true. Therefore
here I post the source. Its on the bottom of this posting. 

It is a hardware based random generator (the hardware is the IBM-PC)
based on the jitter of crystals. To be able to "measure" it two crystals
are needed. Some extra tricks are needed to convert the partly deterministic
results of this fase to completely undeterministic data.

The generator generates 128 bits/second.

I welcome all comments, complaints, suggestions etc.

Nico E. de Vries
_ _
O O  USENET  FIDO 2:281/708.1  COMPUSERVE "soon" (tm)
 o   This text reflects MY opinions, not that of my employer BITECH.      
\_/  This text is supplied 'AS IS', no waranties of any kind apply.      
     Don't waste your time on complaining about my hopeless typostyle.

"Unfortunately, the current generation of mail programs do not have checkers
 to see if the sender knows what he is talking about" (A.S. Tanenbaum)


Source has been compiled with BC++ 3.0.


// (C) 1992 BITECH, Nico de Vries, all rights reserved.
// BITECH and Nico de Vries are not responsible for any damages resulting
// from use of this program. Use it AS IS or don't use it.
// Use of this source or derivates of this source is allowed only
//   - if this head is not changed in any way
//   - the SOURCE is not sold in any way
//   - if the name BITECH is not used in the executable or its docs
//   - something like "random number generator by Nico de Vries" is mentioned
//     in the executable or its docs
//   - if BITECH/Nico de Vries, is notified of its use in a specific program
//   - in case of commercial use at least $100 is donated to Greenpeace
//   - BITECH/Nico de Vries gets improvements/problem reports
//   - improved versions can be used by BITECH/Nico de Vries without
//     restrictions
// The main goal of this routine is to allow for better and more reliable
// data encryption "for the masses". Please keep this in mind when using
// the routine.


// onld interrupt handler
static void interrupt (*vivOld)(void);

// needed to make clock run OK
static int iIntCtr;

// storage for temporary information
static volatile int iActive;
static volatile unsigned uCtr;
static int iPtr;
static unsigned puArray[32];

static void interrupt Handler (void){
   // 0.86 ms timeslice
   outportb (0x43,0x36);
   outportb (0x40,0x0);
   outportb (0x40,0x04);

   if (iActive){
      if (iPtr==32){
      } else {
         puArray[iPtr] = uCtr;

   if (++iIntCtr==64){
      (*vivOld)(); // 18.2 times per second call
      outportb (0x20, 0x20);

// remove handler
void RemHan (void){
   setvect (0x08, vivOld);

   // set timer to 18.2 times/second
   outportb (0x43,0x36);
   outportb (0x40,0);
   outportb (0x40,0);

// install handler
void InstHan (void){
   vivOld = getvect (0x08); // timer
   setvect (0x08, Handler);

// generate true random number
unsigned long Random (void){
   int i,j;
   unsigned long res=0;
   unsigned long tmp;
   // to be on the safe side we intialize uCtr with the timer
   uCtr = peek(0x40,0x6c) & 0xFF;

   // repeat 8 times to change random "bursts" into random numbers
   for (j=0;j<8;j++){
      while (iActive){
         while (uCtr) uCtr--;
         uCtr = 0xFF;
      for (i=0;i<32;i++)
         tmp = (tmp<<1) ^ (puArray[i] & 1);
   return res;


void main (void){
   int i,j;
   unsigned long r;
   printf ("RANDOM [1.0] 'generates 16 true random numbers'\n");
   printf ("(C) Copyright 1992, All rights reserved, BITECH.\n");
   printf ("'AS IS' use allowed\n");
   printf ("BITECH is not responsible for any damages resulting from use of this  program.\n\n");
   for (i=0;i<16;i++){
      r = Random();
      printf ("%08lX ",r);
      for (j=0;j<32;j++){
         printf ("%d",(r&(1L<<(31-j)))!=0);
      printf ("\n");