I split my 8 gallon batches into 3 x 3 gallon car boys and pitch US05, Nottingham, and some other liquid British strain (been changing one up each time) separately. Then I recombine at kegging time. This is the only beer I routinely use dry yeast in, because I know I can cheaply get a big pitch with a couple packets in each. The third strain comes from a slurry from an immediately prior batch of pale ale. Last year I bottled 187ml samples of each unblended. The US05 accented the roast more, giving a drier perception (same final gravity as the Notty). The Notty was a little fruitier, and richer tasting. My third yeast was WLP007. It finished a point or two higher from memory, and tasted somewhere in between the other two. So I think it helps add another layer of complexity.