Hacker News Reader

Comment by jleahy | Hacker News Reader

What nonsense, an extra 9ft of cable is ~14ns (given it's completely passive). A human would struggle to detect 14ms (6 orders of magnitude difference) and it'll be completely swamped by polling latency.

View on HN

Gamers' lack of understanding of the speed of data along a copper wire has made a lot of money for accessory manufacturers

Would you please calculate the drift velocity and the propagation velocity of the wave and the interrupt delay on the loop sense?

Before people waste time with these calculations can you explain how they relate to signals on a wire? You have mixed up some phenomena while reading up on electronics.

1A current in a standard 10 gauge wire has a drift velocity of 23m/s, the wave propagation is dependant on quite a few other factors. The SNES controller is on a 20mA loop, with short to ground IIRC and the wire is likely to be 18 or 20 gauge, the velocity is still not nearly as high as you think, since the loop is interrupted you also have a delay on the sense line. Overall a 30ft cable adds about 2-3ms of latency to the controller...

Drift velocity is irrelevant. Changes in the electric field propagate at nearly the speed of light. The fact that it's happening in a wire slows it down a little bit, but it's still measured in nanoseconds, not milliseconds.

Drift velocity is unrelated to signal propagation. Otherwise it'd take entire days for packets to cross the country to get to you, which is clearly not the case. If we want to take an analogy, signal propagation is like a sound wave in the electromagnetic field inside the conductor; in real air, the speed of sound is clearly faster than the wind.

> in real air, the speed of sound is clearly faster than the wind

I love when HN users give such simple, real world analogies that are so easy to visualize or relate to. For those of us from completely different fields, it really does nail down the concept. always appreciated :)

You are off by several orders of magnitude.

There's a very handy calculator here http://hyperphysics.phy-astr.gsu.edu/hbase/electric/miccur.h...

If you put in the diameter of a 10AWG wire (2.588mm), and a current of 1A, it gives a drift velocity of 5.026cm/h.

For 20mA in a 20AWG wire (0.812mm), that reduces to 1.02cm/h.

Drift velocity is absolutely irrelevant to the propagation delay, which is on the order of magnitude of the speed of light.

Reply on news.ycombinator.com