Hello Mr. ThF,
I downloaded the midi file for Scarborough Fair to my hard drive, and then loaded it into Digital Performer 9 to see what was going on:
It turns out the Controller #64 (sustain pedal) was turned on and off at the following times:
Measure #1, Beat #1, 002 Pedal On
Measure #2, Beat #1, 001 Pedal Off, followed IMMEDIATELY by Pedal On on event 002 in Measure #2, Beat #1.
Similar events occurred in Measures #3, #4 and #5 -- as follows:
Measure #3, Beat #1, 001 Pedal Off, followed IMMEDIATELY by Pedal On on event 002 in Measure #3, Beat #1.
Measure #4, Beat #1, 001 Pedal Off, followed IMMEDIATELY by Pedal On on event 002 in Measure #4, Beat #1.
Measure #5, Beat #1, 001 Pedal Off, followed IMMEDIATELY by Pedal On on event 002 in Measure #5, Beat #1.
In Measure #6, Beat #1, event 001, the Pedal was OFF for the entire measure.
In Measure #7, Beat #1, event 002, the Pedal was turned ON for the entire measure.
In Measure #8, Beat #1, event 001, the Pedal was OFF for the entire measure.
In Measure #9, Beat #1, event 002, the Pedal was turned ON for the entire measure.
Finally, in Measure #10, Beat #1, the pedal was shut off.
* * * * * * * * * *
So here is why the sustain pedal "appeared" not to be working in the early measures:
As soon as it was turned off (Controller #64 was sent a value of 0) in event 001, one 480th of a measure later (event 002), it was turned back on!!!
You need to give a few ticks for the sustain pedal to come off, before turning it on again. If you are using an automated program such as Musescore, then you need to delay the onset of the Pedal, at least the value of a 16th note (semiquaver) or an 8th note (quaver) before sending on a Controller !64 value of 127. That will solve your problem.
Cheers,
Joe