Fixing the PARATA netlist in nl_dribling.cpp

[aovestdipaperino]

@cuavas
I found a way to fix the PARATA nelist in nl_dribling.cpp.
The sound now matches what I recorded with my phone. The issue is when the netlist is not used standalone it performs very slowly, so I guess it would be a good candidate to use FRONTIER. But I can't seem to be able to get it working.

This is the parata.cpp file:

// Parata circuit
#include "netlist/devices/net_lib.h"

// The schematic incorrectly labels two JFETS in the PARATA circuit as 2N3812.  They're actually 2N3819.
// JFET transistors not supported, but this should do the trick; but not for this game.
#define Q_2N3819(name) MOSFET(name, "NMOS(VTO=-3 KP=0.001 CAPMOD=0)")

#define STANDALONE_PARATA

// Simplified LM339 model - uses high-level simulation of differential input stage.
static NETLIST_START(LM339)
{
	// * CONNECTIONS:   NON-INVERTING INPUT
	// *                | INVERTING INPUT
	// *                | | POSITIVE POWER SUPPLY
	// *                | | | NEGATIVE POWER SUPPLY
	// *                | | | | OPEN COLLECTOR OUTPUT
	// *                | | | | |
	// *                1 2 3 4 5
	NET_MODEL("LM339_QO NPN(BF=200)")

	AFUNC(CMP, 2, "min(max(A1 - A0, 0), 1e-6)")
	VCCS(QB, 80)
	QBJT_EB(QO, "LM339_QO")
	ANALOG_INPUT(XGND, 0)
	RES(RDUMMY, RES_M(1000))

	NET_C(CMP.Q, QB.IP)
	NET_C(QB.OP, QO.B)
	NET_C(XGND, QB.IN)
	NET_C(QB.ON, QO.E, RDUMMY.2)

	ALIAS(1, CMP.A0)
	ALIAS(2, CMP.A1)
	ALIAS(3, RDUMMY.1)
	ALIAS(4, QO.E)
	ALIAS(5, QO.C)
}

// The schematic employes an extra JFET to amplify the output which, without it, is attenuated.
static NETLIST_START(parata)
{
#ifdef STANDALONE_PARATA
	SOLVER(Solver, 1000)
	PARAM(Solver.DYNAMIC_TS, 1)
	PARAM(Solver.DYNAMIC_MIN_TIMESTEP, 1e-5)
    TTL_INPUT(I_PARATA, 0)
	NET_C(GND, I_PARATA.GND)
	NET_C(I_V6, I_PARATA.VCC)
    NET_C(INPUT, I_PARATA)
#endif // STANDALONE_PARATA
    NET_MODEL("BC309 PNP(IS=1.0366E-15 ISE=1.6295E-14 ISC=9.4971E-17 XTI=3 BF=80 BR=16.116 IKF=0.47497 IKR=0.012081 XTB=0 VAF=8.486 VAR=9.5149 VJE=0.84456 VJC=1.0282 RE=1.9597 RC=1.1393 RB=3.2133 CJE=2.0636E-14 CJC=2.6904E-12 XCJC=0.075977 FC=0.54298 NF=1.03136 NR=1.2907 NE=1.3702 NC=1.038 MJE=0.68352 MJC=0.5401 TF=5.4303E-11 TR=3.4233E-10 EG=1.11 IRB=0.00046855 RBM=2.0822 XTF=0.27447 VTF=0.19311 ITF=0.0030573 PTF=0 CJS=0 VJS=0.75 MJS=0)")

	ANALOG_INPUT(I_V5, 5)
    ANALOG_INPUT(I_V6, 6)

	QBJT_EB(Q1, "BC309")
	Q_2N3819(Q2)

	RES(R5, RES_K(1))
	RES(R6, RES_K(150))
	RES(R7, RES_K(220))

	RES(R9, RES_K(2.2))
	CAP(C5, CAP_U(2.2))
	CAP(C6, CAP_U(1))
	DIODE(D1, "1N914")
	DIODE(D2, "1N914")

	SUBMODEL(LM339, IC_B9)
#ifdef STANDALONE_PARATA
	NET_C(GND, IC_B9.4)
#else // STANDALONE_PARATA
	ALIAS(GND, IC_B9.4)
#endif // STANDALONE_PARATA
	ALIAS(INPUT, R5.1)

	NET_C(R5.2, Q1.B)
	NET_C(Q1.E, I_V5)
	NET_C(Q1.C, C5.2, R7.2, R6.1)
	NET_C(C5.1, GND)
	NET_C(R7.1, GND)
	NET_C(R6.2, IC_B9.1, D1.A)
	NET_C(D1.K, R9.1, IC_B9.5, Q2.G, Q2.S)
	NET_C(IC_B9.2, C6.2, Q2.D)
	NET_C(C6.1, GND)
	NET_C(R9.2, I_V5)
	NET_C(IC_B9.3, I_V5)

	NET_C(D2.A, Q2.G)
	NET_C(D2.K, Q2.S)

	Q_2N3819(Q3)
	RES(R8, RES_K(100))
	DIODE(D3, "1N914")
	NET_C(Q2.D, Q3.G)
	NET_C(Q3.D, I_V5)
	NET_C(D3.A, Q3.G)
	NET_C(D3.K, Q3.S)	
	NET_C(Q3.S, R8.2)
	NET_C(R8.1, GND)
	ALIAS(OUTPUT, Q3.S)
    LOG(REAL_PARATA, OUTPUT)
}

input.csv file:

0,I_PARATA.IN,0
0.5,I_PARATA.IN,1
1.5,I_PARATA.IN,0

Launch it with: ./nltool --cmd=run -i input.csv -t 2 -n parata parata.cpp

[aovestdipaperino]

...and I found a solution.