Typically intel processor clock aggressiveness is managed in software by a technology called speedstep, and in general terms this is controlled by how demanding your application is being. If an application has really bursty loads, the CPU will ramp up only for as long as the task is live. On really heavy duty tasks like big renders, you’ll find the CPU clocks as high as possible as long as it can until it hits thermal limits. My guess from the product release page information is that the processor will clock up to 60W immediately, then it’ll take a few seconds to fully saturate the heat sink and block’s thermal mass before throttling lower to 28W. This means that for the first case of a bursty load, you’ll spike to 60W, then the task will finish, and the CPU will drop to very low package power, allowing the cooler to get well below its thermal limits, before the next task comes along. For the second case of a long-term task, you’re not likely to hit 60W again until that task is over, as you’re operating at the thermal limit set in your processor.
To make a long story longer, you should be able to play around with tuning how long it takes with tools like ThrottleStop or Intel Processor Power Management Utility (the former being my personal preference). You play with a couple primary variables, A) the speedstep multiplier to tune how aggressively your CPU behaves, B) undervolting your CPU to tune just how much processing power you get per watt, and finally C) you can also adjust the thermal limit on your processor, allowing it to get hotter than the factory setting. This typically means you can dump a lot more heat out of the heat sink (in other words, that heat sink’s heat transfer is proportional to the temperature delta), but you might sacrifice some life span for your CPU.