Intel has had years to release an atom that wasnt a total disgrace. Their refusal has allowed both apple and qualcomm to rake in multiple billions at their expense.
First and foremost Atom wasn't a
total disgrace. Atom created the netbook market in 2008 and ruled without any kind of competition until 2011 when AMD launched Brazos and the old 2008 core is still competitive with ARM A15.
It's not hard to understand Intel decision in 2008. There was no tablet in 2008 and power consumption targets for Atom wouldn't make it viable, and smartphones were a developing market, far from the dominant position they have now. Intel wanted a only foothold on this market. Intel didn't have a product either. Only now, on a very mature 32nm process is that Atom reached power consumption levels compared to ARM, and this in a moment where ARM power budgets are going up, not down. And Intel wouldn't take resources from Core, the netburst debacle isn't something easy to forget, it should be impregnated on the corporate DNA there.
But in hindsight Intel should really focused on Atom, bringing it to tick-tock cadence since day 0 , but in 2008/2009 when those decisions were being made those things weren't clear, and the decision wasn't really straight. Focusing on Atom without adequate resources could harm Core, and regardless of what some analysts say Core is still a huge cash cow, bigger than Qualcomm for that matter.
But with 22nm Atom and Android working on x86 and digital radio circuitry I don't see any reason on why Intel won't grab a significant share of the tablet market. At the very least they can crater margins and use their old fabs to undercut ARM manufacturers on price.