Whether the heads/lower runner was redesigned to fit the injector pattern or the injector to fit the runner design is not know.
The engineers explained to me that the purpose was to solve a fuel puddling [recondensing from vapor] on a section of cold [cooler from fuel evaporation] wall inside the head.
Whether they had to change the software to match the injector is unknown. Probably!
Injectors have dead time and slope and shut off bounce. there is a time lag between the instant application of voltage, when the pintle begins to move up and the rate of change of the pintle position till it is fully open. Based on mass flow the rule of thumb is 66% of the expected flow during the opening time segment.The non standard flow period varies from injector type to injector type. The first 1.3-1.5 milliseconds. At idle the total time is 2.2 millisecs..........66% is non standard flow. [confirms that they used the rule of thumb].
The battery voltage is critical as the slope changes and the total ramp up time. The ecu monitors battery voltage and corrects the open time............especially important at cranking and beyond till alternator replenishes the amount in the first few seconds......the battery may drop to 9-10 volts in a cold crank.
You can see this clearly by seeing the current waveform on the injector fuse line..........you can see the current start to flow, the ramp up, a little squiggle when the pintle moves off its seat [0.6-0.7 millisecs], and the final current which is faily level until the shut off, when it drops to zero, then there is a bounce when the back emf collapses and the pintle spring unloads its energy.
http://www.asashop.org/autoinc/dec2000/mech.htm
By observing the current peak and ramp up and pintle opening point at the fuse you can see all 8 injectors simultaneously and compare the averages.
A 35 #[370cc/min] flows [34psi/43.4psi=.78341 squareroot=0.8851 x 370= 327.5 cc/min or 5.4581 cc per second or 0.0054581 cc per millisecond.
At cruise the O2/ecu would fix errors, WOT and acceleration probably wouldn't matter as everything is super rich.
We would have to compare 2 Q [perfect 90 and perfect 94] at exactly the same rpm /coolant temp/idle fuel pressure ----side to side and read with the Consult, the injector open time at idle to determine if they made software changes.
See why the power balance is so accurate in determing injector health.
A 0.1 millisec at 2.2 millisecs is about a 6% flow change, at WOT high rpm it [the same 0.1 millisecond] is 1%.
The ecu sets injector opening time at idle by rpm....[tries to maintain a certain rpm]....once everything is warmed up.
If you have the time and energy probably a good experiment to see what if anything happens with 94 injectors on a 90-93. I WOULDN'T do the converse using 90-93 injectors on a 94-96.
Obd2 [1996 Q and later] would be much more sensitive to type of injectors.
Obd2 reads the fuel temperature to correct for viscosity changes in fuel and flow rate thru the injector........very very fine tuned system.