Lots of willy-waving and 'certainty' about an unannounced Apple product but let's give them the benefit of the doubt for the moment.
So let's say the new MBA really has a processor that is twice as fast as the current one and let's assume it will really have an i5 or i7 (ie the same is in the current MacBook Pros (ha!ha!).
First- I'm not sure that an MBA with 30 minutes battery life will be that useful and secondly I very much doubt for *any* use that the MBA (or any current generation machine, desktop or laptop) is processor bound- the graphics processor, mass storage, memory, network, system bus are where the bottlenecks lie- not in the processor (yes, I know, changing architectures allows you to improve some -but not all- of those areas).
So to return to my original point -which most of you have missed whilst quoting your fantasy performance figures- is that even if all the rumours are true (which they won't be) the new MBA (assuming there will be one this month) will be a relatively minor incremental improvement of the previous generation: exactly the same as all the other MBAs have been over the revision before them.
In fact I'll take it farther than that. If, by some miracle, the new MBA has performance comparable to the current MBPs and maintains -through some magical improvement in battery technology we haven't seen in decades and hasn't even been rumoured- to maintain the same battery life with the vastly increased power drain- then you're looking at a radically new machine.
And Apple's first attempts at anything: OSX, MBP, MP, MBA; always have serious design flaws or quality issues.
So to re-iterate: the new MBA (if it even exists) will either be a minor incremental improvement of the current MBA or it will be a wholly new product which will inevitably be plagued with problems and quality control issues.
So if you've sold your 2010 MBA in anticipation of the new MBA then, well, you've given me a good laugh (and the smart people a cheap MBA
.