Might be cool if we did a tradeoff with Greg. 1/4 as much power loss, but added electromigration. Some materials (like copper) would be largely unaffected, while others (like Silver) would decay each time a packet crosses, so that after, say, 32,000 packets it would become an 'old tin cable' and experience 2x the loss, then repeat with a 'worn tin cable' at 4x loss, and finally a 'ruined tin cable' with 8x (twice the current loss). After that it would just dissapear/break.
I know that Aluminium is rather susceptible to electromigration. Some quick research shows:
Worst: Silver, Aluminium (1)
Bad: Copper, Zinc, Lead (50)
Average: Tin (250)
Good: Iron, Nickel, Antimony (750+)
Best: Gold, Palladium, Platinum (>5000)
Totally immune: Cuprate-perovskite (A copper anion and CaTO3 alloy that is supercritical above 90K, used for superconductors.) If you want your head to explode, go to http://en.wikipedia.org/wiki/Phonon for info. Alternatively, be hypnotised by the vibration modes gif. I figure just 'superconductor wire' in general would be the 'ultimate material for longevity, voltage, amperage, and losslessness'
* = Depends on steel
Also, Greg, might I suggest some recipes for RL-superconductors?
First are the two Copper Anion superconductors, tetrachlorocuprate, which gives a use for chlorine. CuCl4-2 (I imagine you could make CuCl4 and then put it into an electrolyzer to produce the negative charge)
The other is YBa2CU3O7 which seems delightfully painful to make, appropriately for a lossless high-yield wire.
There are also organocuprates ([Cu(CH3)2]-)
My vote goes for YBa2Cu3O7, as it would be a real pain to make, requiring rare Yttrium and Barium. I imagine it would take a decent smelter along with the chemical reactor, made into a wire and wire into a coil ala the blast furnace coil blocks, and one coil block inside some sort of sheathing and rubber finally makes one superconductor wire.
Edit: Fixed research a bit. Numbers indicate the frequency of electromigration compared to baseline (aluminium). Adding 2-3% copper to aluminium increases electromigration resistance by 50x or more. So this is my rough chart.
Would be neat to add a maintenance mechanic like this.