I thought I should post this here as it's something which is likely to be a factor in one's purchasing decisions.
The iDSD Diablo is specced as having 4.98W @ 32 Ohm. There is no further info given on the product page other than a voltage value of 12.6V which at 32 Ohms is indeed 4.98W.
The industry standard for measuring power is to find the maximum value an amplifier can continuously output without going above 1% THD. This is how almost all power specs for amps you see are obtained.
Sometimes manufacturers might use 10% instead, though this isn't ideal and if this is the value given it should be clearly stated so as not to mislead consumers.
As well as continuous or RMS power, some manufacturers will provide a peak power value. Which is the amount of power an amp can supply for short durations, but not continuously. This is specified by the CEA 2006 standard as 20 'on' cycles of a 1khz sine followed by 480 'rest' cycles (signal 20dB lower than the 'on' cycle).
A peak power measurement is useful to have, ideally in addition to the continuous power value, but again, it needs to be specified that it is a peak measurement, as it's going to be a higher number than the RMS value in most cases and cannot be compared to the continuous/RMS power spec of other amps.
The iDSD diablo has a spec of 4.98W, however when I tested it I was only able to get up to about 0.8W.
5.2V @ 32 Ohm is 0.845W and the Diablo distorts heavily at this level (regardless of being plugged in or not, fully charged, turbo mode etc):
I contacted iFi about this, and they told me that the power spec is a peak power spec, not an RMS spec. This is not made clear anywhere on the product page.
They also said that they did the test using a SINGLE sine cycle, not the 20 that the CEA 2006 spec requires.
I repeated their test, and yeah, it will do it for a single cycle, but trying the actual CEA 2006 test, by the third cycle it's already distorting massively. So even as a peak value 4.98W is inflated and incorrect according to industry standard testing.
I've asked iFi if they are specifying power on other products this way and if they're going to update the specs to make clear that this is a 'max instantaneous' value, and NOT an accurate peak or RMS value. At the moment they've said they will add this info to their FAQ page but have not said anything about updating the product page itself.
I'm not particularly happy about this as this methodology significantly inflates the power spec of the Diablo in a way that is misleading to consumers and not even updating the product page to at LEAST say 'peak' or something (even though as explained above the figure is still arguably incorrect even as a peak measurement) is in my opinion quite irresponsible ESPECIALLY given as power is one of the main selling points of the Diablo.
The actual power output of the Diablo that can be compared to other products is about 0.8W, NOT 4.98W.