those tests seem poor at best to determine the sword's full ability. chopping a soft sapling, a rotten log, or a rusty tin barrel isn't impressive. the scratches it takes are even less so. I've demolished a metal barrel like that with a 1080 sword and didn't have a knick on it.
however, I'll concede that he probably had it sharpened to hell, when chisel-sharp is all you really need. too sharp means the edge will fold, chip, knick, and break easier, challenging the structural integrity of the blade and creating weaknesses.
"tempering" or "heat treating" aligns and compresses iron molecules. the heating excites the molocules, and quenching it in water rapidly cools it, causing the molocules to compress, creating a tighter and more orderly molecular construct, lending greater strength to the overall shape. temperature control is vital, and consistancy is key. temprature extremes ruin the effect. the fire must be a controlled temprature, level at around 1440 degrees F. the water temprature varies as well, and legends tell of japanese smiths cutting off an apprentices hand for checking the water temprature without the master's consent, but something very cold like ice or snow (ala conan) could result in shattering.
also, remember that the FORM and FUNCTION of a sword are equally as vital to its success as its MATERIAL properties. smaller blades like knives and daggers can enjoy the benifit of stainless tool-steel or even a titanium alloy without comprimising its stability. larger blades can benifit from things like fullers, a diamond profile, tapering, and other techniques to lend it strength and durability while minimizing weight and altering balance and center of percussion to build further resistance to breakage.