Really? Dig up past threads regarding world records. There are always moaners like me harping on about how liquid nitrogen and helium clocks are irrelevant and quite yawnworthy. They don't indicate real world performance or overclockability....so who gives a rats ass.
THerein lies the problem, they do exactly what you claim they don't.
They establish several things, they establish the architecture scales up to 2v very well, they establish there isn't a cold bug, you might not use phase change but quite a few people do. The LN2 cooling could have resulted in a chip that topped out at 6Ghz, didn't scale beyond 1.6V, and blew up at 1.7v....... it didn't, thats ALL very relevant information.
Lets say they ONLY did crap watercooling, and lets say at 1.5v they hit 5Ghz and couldn't go higher....... then stopped.
What stopped them, voltage, cooling, heat, clock speed, who knows, take your pick. By establishing exactly what the architecture can do, it answers all those questions, with better cooling more voltage will indeed allow higher clock speeds for those with better cooling, from a better high end air cooler, to high end water.
Thats the problem, you're completely wrong, it is relevant and you repeated ignore that and all the other info gather at the event. AMD can't help how other sites report it, nor is it their "fault" that the chip broke the record, it may have only hit 8Ghz and the headlines might have been 5.5Ghz on air, woo.
The event gave us plenty of useful info, and this has pretty much been pointed out by several people several times over. Now if this record was done in 6 months, when we already knew exactly how Bulldozer scaled on air, water, what its voltage limits were, then it would be pretty irrelevant. But most of the time when people tell you "X voltage is safe as long as temps are below Y", comes from people who test chips to their limit and kill them. This is the kind of overclocking that helps establish whats safe, the overclockers will continue to play with some of those chips for weeks, if some die and they were used at 1.8v, thats useful info, if the ones used at 2v are still going strong in 2 months at silly overclocks in 24/7 rigs, thats useful info.
When you come back with what amounts to "yeah, but its crap, its only two cores" comes across less as a genuine expression of opinion and far more as "hahah, Intel are better". When you repeatedly also ignore that Sandy can't come close to that record fully enabled, the previous record was a heavily crippled Intel chip and thats simple how world records are done, only around 30% of the chip was actually disabled and it would be more a case of available amps from the mobo than that the chip itself wasn't capable of doing it with all cores enabled. For instance, they didn't try each module individually, maybe one of the other 3 modules, or all of the were faster and they by chance got the least fast module.
Never been at bleeding edge before 
.
- To many haters jumping the gun without reading the facts tbh 
