4 ms·

is randomness not good? you aren't going to be driving around on a rail irl

• nitrogen 13y ago
  Not when you're trying to make an objective comparison with a small number of trials.
  • simonh 13y ago
    If the test was always exactly the same, designers could optimize the car design for that precise impact force at that precise point.
    • nitrogen 13y ago
      The ability to compare different cars is probably more important than preventing designers from sticking a five foot thick plate of metal on the precise diagonal of the crash test, which would be obvious. A major part of crash testing has to be the forces felt by the test dummy, so there's not a whole lot designers could do to reduce those forces in exactly one direction without also reducing them in other directions.