No, it is merely a simple way to compare as a ROM.
No. It's flat-out
wrong. The PSU does not dictate how much a system
does draw, it simply sets an approximate upper bound on how much it
can draw.
Hence my very deliberate choice of not using horsepower as an analogy.
Second, the iMac has been the perennial punching bag for being 'wimpy' because of its low power components, and the rallying point of the enthusiasts for apple to sell a headless iMac ... aka xMac. For the past year, the new i7 CPU has been the poster child for what Apple should be building this xMac around, and for a point of reference, Dell's Studio XPS has a 360w PSU.
You are comparing apples to oranges. An i7 machine is more properly compared to a Mac Pro. It's ridiculous to compare its power supply to an iMac.
Third, please note ... as has been stated before ... that the generic desktop PC's PSU doesn't also drive the display, whereas for the iMac it does. This system architecture difference should be resolved and IIRC, a 24" LCD burns around 75W of power, so a value of roughly this magnitude would need to be added to the ratings in order to make a fair comparison of equals.
Which is utterly irrelevant to my estimates, because I'm estimating how much the whole system draws, rather than reading some maximum numbers of spec sheet and making wholly incorrect assumptions about usage.
To say nothing of the further dishonesty of assuming a typical PC has a 24" LCD connected to it.
Thus, for whatever your now agreed upon PSU revision is (300w, or perhaps 360w for the Dell i7), add to that a budget for the external display (eg, ~75w for a 24" LCD) and see if you've now reached the conclusion of being in the same ballpark as the original 400W value.
Except the original conclusion is completely wrong, because the maximum rating of the power supplies does not dictate how much power the system draws - it merely sets an upper bound of how much it can draw.
YMMV how reasonable is a guesstimate when it uses utility rates that are 40% lower than their local actual, even before we see that their peak utilization rate was assumed to be a mere 3%. Dare I say "LowBall"?
It has nothing to do with utility rates and everything to do with your ridiculous assumptions about power draw and usage patterns.
The fallacy with this approach is that it is claiming that the OEM's 300W (and larger) PSUs are technically unnecessary overkill. But given the cutthroat competition in the PC marketplace, then why haven't the OEMs figured this out on their own and cut their costs by going to a non-oversized PSU?
There is no such claim. OEMs spec their power supplies for the maximum possible power draw that system could pull with all options added. For a maxed-out configuration - even on an entry-level machine - that could be 2-3 hard or optical drives, a beefy video card, 8GB+ RAM and a quad-core CPU.
I don't see how it would make sense for the OEMs to so grossly oversize a component and subsequently lose an opportunity to improve their razor-thin profit margins...so please enlighten us.
Because it's (obviously) cheaper to have a single standard PSU across the dozens of different configurations the typical PC can be bought in, than try to exactly match a given configuration to a specific PSU. To say nothing of having vastly more physical space to work with.
All of your assumptions in this comparison are ridiculous. From the PSU running at 100%, through 24/7 operation at that load, to the hardware configurations. Added to which, you are arguing with people who a) actually understand the technology, and b) have actually taken measurements.
You are wrong. A typical PC (plus monitor) will draw something in the range of 75-100W, depending on its use. Marginally more than an iMac ? Probably. Significantly more than an iMac ? Not a chance. Even a relatively high-end gaming machine, under peak load, will only draw around 250-300W. Only the highest-end machines, with dual quad-core CPUs, multiple SLIed video cards, four or more hard disks and 16GB+ RAM might start to push over the 500-600W mark, and even then only under heavy load (in typical usage they'll be down around the 100W mark, like other machines).
Your fundamental mistake is assuming that the capacity of the PSU is a reliable indicator of how much power a system actually draws. You've made some other bad assumptions as well, but they mostly stem from that initial error.
to Eidorian). Dell calculates $19/year in power (mini-tower only), where you'd get $401. Only different by a factor of 21.
