It's not that loads has changed - the days since we had the solar panels installed have been predominantly grey - well, this is an English winter, after all. However, the system is responding differently already. Even on the greyest day, the panels seem to be generating power from around 9.30 to around 3.30, and we rarely seem to have the "zero output" days that we had for the first few weeks.
There hasn't been a huge change in the sun. In the first month after the winter solstice, the noon position of the sun only gets three degrees higher in the sky, and the day gets less than an hour longer. However, it is evidently enough to take it closer to the light threshold at which the panels start to generate.
We have discovered that the panels weren't optimally connected to the inverter (the south and the west facing panels should have been connected as separate "strings" into the inverter, and they were in fact all linked in series), so Rayotec are coming back to rewire a section of the system. I'm hoping that they are going to find out how much it would cost to install the PC interface module at the same time. I don't know whether this will have a noticeable effect on the performance of the system. In theory, the output of the two different sections of panels can be optimised separately by the inverter: I don't know whether this is likely to give a 1% or a 10% improvement.
The peak output of the system so far has been just over 1000 W - it has reached that on four days, the first after the solstice being 9th January - and the highest total output we've seen has been about 2 KWh. The total output of the system since installation is a mere 13 units, though we lost a couple of days which ought to have been good as a result of our messing around with the system.