Mowgli, you missed the entire point of what I was saying. Here it is nice and simple:
The pressure of the gas is driving the impellor, true. However, when the air that is supplying that pressure is cooled, the pressure will drop. The longer you make that transition from the manifold to the turbo, the more the temperature will drop. Also, on a smaller scale, in a pipe that long the gas will be compressed slightly, resulting in a longer spool time. To keep the gases flowing smoothly, and to keep the pressure at an ideal value, the turbo should be mounted as close to the exhaust port as possible, providing the flow path is optimal.
I got my training on gas turbine engines, I used to work on GE LM2500 engines, and also Allison 501K17's. I started training on them in 1995.