russotto@eng.umd.edu (Matthew T. Russotto) (05/22/91)
How come LocalTalk works even when you switch the two wires? Shouldn't this
cause the bits to all be flipped (because the compares will all be reversed?)
--
Matthew T. Russotto russotto@eng.umd.edu russotto@wam.umd.edu
.sig under construction, like the rest of this campus.jackb@MDI.COM (Jack Brindle) (05/23/91)
In article <1991May22.032940.18413@eng.umd.edu> russotto@eng.umd.edu (Matthew T. Russotto) writes: > >How come LocalTalk works even when you switch the two wires? Shouldn't this >cause the bits to all be flipped (because the compares will all be reversed?) >-- >Matthew T. Russotto russotto@eng.umd.edu russotto@wam.umd.edu > .sig under construction, like the rest of this campus. The information on AppleTalk is encoded in the bit transitions, not the levels. This means that the SCC's manchester DPLL decoder looks for incoming bit transitions instead of specific levels to base its decoding decisions. Another way to look at it is that the LocalTalk interface boxes contain a transformer which does not maintain the data levels it passes through. A second year EE student hopefully should tell you that transformers work because of state transitions in the voltages being applied at its input. The Localtalk box is the same way. So the short answer is, it does not matter which way the wires are connected, just so they are. It may even work if only one is connected (just so you have a common ground path), but you will lose a lot of the advantages of the differential circuit operation. I hope you are not overly confused :-). Jack Brindle ham radio: wa4fib/7