Compounds offer far and away the best turbo performance - best response with greater width and area under their boost curve at reasonable efficiency and utilising more of the heat energy in the exhaust gas.
However they take up more room, require more custom work and the fuel injection pump is hard pressed to provide enough fuel to take full advantage of their potential.
VNT turbos can provide better performance than waste gate singles, but as Dougal pointed out are difficult to find in a size that best suits a tuned 4BD1T - IMHO the GT2259V is still a little small to get the best out of a tuned 89 or later 4BD1T. I have a GT2256V that was on the 300Tdi in my old rangie and some performance tuners in the UK supply GT2259V for the 300Tdi where they perform well, which in my mind confirms my previous statement.
With the type of compressors (radial outflow) used in our turbos, boost pressure is produced dynamically - a function of the square of the speed that the air leaves the tips on the impellor, which is related to the impellor exducer (outlet) diameter squared and rpm. The air flow rate is dependent on and limited by the inducer (inlet) area (a function of diameter squared).
When smaller compressor wheels are driven faster to provide boost pressure outside their efficient operating range the heat produced in the air reduces the density, defeating much of the gain from higher boost pressure - also higher drive pressure is required. Larger diameter impellors can produce higher boost pressure more efficiently than small impellors. An analogy might be spinning tyres, burning rubber but failing to move the vehicle – with the small compressor, think of the tips of the impellor blades slipping against air flowing at lower speed. This explanation is rather simplistic and for a better understanding the reader should investigate how/why compressor trim affects efficiency – air in contact with the impellor blade for longer (distance and time) while accelerating up to the tip speed.
The value of boost pressure does not tell the full story. A better match turbo can provide better performance at lower boost pressure than some other turbo at higher boost. What we really are interested in is the air mass flow combined with lowest drive pressure (exhaust manifold pressure) and two different turbos providing 30 psi boost don't necessarily provide equivalent air mass flow and drive pressure.
Normal pressure gauges are designed to dampen pressure variations and can’t respond quick enough to detect transient (short time interval) pressures. Actual pressure peaks are greater and affected by complex shock waves, etc.. Higher drive pressure reduces the engine’s volumetric efficiency (VE) – inlet valves open before top dead centre, while the exhaust valve closes after top dead centre and unless the pressure in the inlet manifold is greater, fresh air can’t flow into the cylinder.
Both higher drive pressure and higher boost pressure increase pumping losses in the engine – power required to pump the inlet air and exhaust gas is lost.
c.h.e.i.f. you said that you changed injector nozzles. This may be the reason why you smoke problem is so bad. The stock nozzle is designed so that each of the 4 nozzles spray into the region of the 4 corners of the combustion chamber in the piston crown. The angle and orientation of the nozzle holes are made to suit the particular offset and angle of the injector to the combustion chamber.
A nozzle from another engine will most likely use a different arrangement of nozzle holes. If the spray is hitting the sides of the combustion chamber, or the piston crown then combustion will suffer badly (e.g. black smoke), and can cause other problems.
The stock nozzles are designed for stock engines and when the air density in the combustion chamber is increased significantly (higher boost pressure and better intercooler) the spray pattern doesn't reach as far into the combustion chamber, adversely affecting combustion efficiency.
