Despite my assertions I was muddling up a couple of the coordinate spaces - my core VS was fine, but I hacked in a couple of extra lines at the last minute and forgot to transform them correctly [embarassed]
The ward model is an anisotropic one, thus you can effectively change the shape of the specular highlight my messing with the parameters. I've forgotten to annotate the above image, but the columns and rows are 0.0 to 1.0 in 0.2 steps - with the top-left being the 0.0,0.0 origin.
I'm not entirely convinced its correct, but I never seem to be happy. At least it's not broken - so if I have time later on I can always come back to it.
A few of my resources make mentions of the Ward parameters being quite sensitive, and that values less than or around 0.2 work best - which would correspond with the results I have above.
The idea, given its an anisotropic model, that I can represent various metalic and similar manufactured surfaces:
Reckon I could pass that off as brushed metal?