There were 4 rooms for my first C++ class at uni: 3 with brand new Pentium II's and 1 with the old (and slow) 486's. No wonder, nobody wanted to be in that latter room for the exam at the end of the semester. So the teacher had to randomly dispatch people in those 4 rooms.
Funny surprise: the results from the 486's room were slightly better (than the results in other rooms). Hypotheses: people in the 486's room were thinking twice before launching the slow compilation (leading to less compilation errors), as opposed to launching it often in other rooms (with probably more compilation errors). Bottom line: increasing the focus reduces waste of time.
This phenomena is mentioned in The Mythical Man Month. An interactive debug session is three times more productive than batch submissions. However, when access to computational power becomes too easy, people fiddle aimlessly without a directed plan.
Funny surprise: the results from the 486's room were slightly better (than the results in other rooms). Hypotheses: people in the 486's room were thinking twice before launching the slow compilation (leading to less compilation errors), as opposed to launching it often in other rooms (with probably more compilation errors). Bottom line: increasing the focus reduces waste of time.