/*
 *  test_spike_poisson_ps.sli
 *
 *  This file is part of NEST.
 *
 *  Copyright (C) 2004 The NEST Initiative
 *
 *  NEST is free software: you can redistribute it and/or modify
 *  it under the terms of the GNU General Public License as published by
 *  the Free Software Foundation, either version 2 of the License, or
 *  (at your option) any later version.
 *
 *  NEST is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *  GNU General Public License for more details.
 *
 *  You should have received a copy of the GNU General Public License
 *  along with NEST.  If not, see <http://www.gnu.org/licenses/>.
 *
 */


/* BeginDocumentation
Name: testsuite::test_spike_poisson_ps - tests if spikes independent of resolution

Synopsis: (test_spike_poisson_ps) run -> compare with results at other resolutions

Description:

The difference in spike times is exactly 0.0 between 
all resolutions. 
 test_spike_poisson_ps_precision.sli
already verified that spikes are produced and recorded with
double resolution.

This script will give identical results almost always, even if the resolution
is not representable as a double. The main reason for this is that the 
poisson_generator_ps generates spikes without attention to resolution. This
means that exactly the same sequence of spike times is generated internally
in the generator independent of resolution. Rounding errors occur, if at all,
only when the spike time is split into a multiple of the resolution and the 
offset. In this calculation, the resolution enters as a fixed multiple of tics.
The calculation is reversed in the spike detector, when step-multiple and offset
are recombined. This splitting and recombination apparently gives reproducible 
results. The only deviation observed was for two spikes at very early times (<0.01 ms) 
and coarse resolution (0.1 ms); total of 388.533 spikes tested. Therefore I
introduced the check against the 1e-15 relative error limit below.

If this test fails go back to
 test_spike_poisson_ps_precision.sli
to check whether poisson_generator_ps can emit spike times at
double precision or whether spike times are limited to the 
precision of a tic.

FirstVersion: February 2007
Author: Plesser, Diesmann
*/


/unittest (6688) require
/unittest using

M_ERROR setverbosity

/Transmission
{
/h Set

% end of stimulation period
/stimtime 30.0 def

% make simulation time long enough so that
% at least one full min-delay passes after
% end of stimulation even if resolution does 
% not divide 1.0ms. Otherwise, not all spikes
% will reach the spike_detector.
/simtime stimtime 2.0 add def 

ResetKernel

0 << /resolution h >> SetStatus


/spike_detector Create /sd Set  
sd  << /precise_times true >> SetStatus
 

/poisson_generator_ps Create /pg Set
pg << /rate 128921.3 /stop stimtime >> SetStatus

pg sd Connect

simtime Simulate 

sd [/events /times] get cva 
} def



{
  [0.1 0.03 0.02 0.01 0.001]
  { Transmission } Map 

  % compute absolute relative error
  dup First /refval Set
  Rest { refval sub refval div { abs } Map } Map
  Flatten { 1e-15 gt } Select [] eq
}
assert_or_die