Rather than answer particular messages, let me put forth a few facts that some of you may not know regarding component costs.
First of all, these drop wildly as quantities increase. For example, a pair of Toshiba's excellent 2SC5200/2SA1943 power devices (150W, 230V, 30 amp peak current, 30 MHz) cost far less than people are led to believe. I pay retail prices locally $3.50 for the pair - however, it's an unmatched pair. Now, to get proper performance, you need to match them to at least 5% tolerance, preferably 3%, my standard being 1% (but I'm a nut, it's hardly ever done commercially). To be able to match them, you need a whole lot of them, say 100+100 for any half serious run of 5-6 units. Of course, what fails the grade isn't thrown away, it can still merrily be used for many other applications which do not require close tolerances. Anyway, for 100 pairs, the price drops to $3.00.
So, using the usual commercial approach, which is to add another pair to an existing design and ramp up the power supply lines (or, more often than not, simply make them stiffer, with less sag) actually costs me a grand total of $6 per stereo amp. In view of the time, effort and instrumentation I have toi use to make the matches, this easily triples, but it's still just $18, and with two pairs, you can have a very reasonable 70-100W/8 ohms power amp, though for 100W/8 ohms, I prefer using three pairs - which adds still another $18 for a stereo amp.
You would also need a larger power transformer; if I used 2x200VA toroids for a 2x50W/8 ohms amp, I'd move on up to 2x500VA for a 2x100W/8 ohms amp, allowing a wide margin for low impedance loads (but again, I'm a nut). Anyway, locally, this would add like $60 for a pair, pushing my added cost to (36+60) $96. Larger heat sinks, etc, would push the price up to $150, if that.
So, to me, the price difference between a 50 and a 100 WPC amp, which includes greater quality (more current delivery, much improved power supplies) is around $150. Using the usual industry mark-up (manufacturer to distributor to retailer), this would typically be multiplied by a factor of 3, which means $450 by the time it got to you. For a manufacturer, this would be less, because I buy at retail prices, and they buy at wholesale prices.
So why the big differences, then? One answer is that because manufacturers are tripping over themselves to create all kinds of cases which would be striking in appearance, which often neccessitates special heat sinks - and that adds a hefty price premium. I on the other hand, use standard German made Fischer heat sinks, widely available, and thus pay much less even at retail prices.
Another answer is product perception. A 100 WPC power amp is assumed to be better than the 50 WPC power amp, whether this is so or not, and in my book, there is absolutely no correlation between sound quality and power in most cases - it becomes a serious factor as your nominal speaker impedance begins to turn evil and as the speaker effeciency starts to drop low, i.e. below say 87 dB/1W/1m. But most consumers automatically assume 100 W is better than 50 W, despite some obvious facts pointing in the other direction, such as, for example, that a 50 WPC amp can be made with just one pair of well matched power devices in a single ended push-pull (SEPP) configuration, as yet unbeaten in terms of timing by any multiple output device amp.
Second, you are handed a lot of bull daily that X's power devices are the best in the world, where X is often Sanken. I totally disagree with any "best" policy, there is no best, because one man's meat is another man's poison. It all boils down to how you use them, and especially to how far you plan to push them. But because of a wild ad campaign, the public BELIEVES the Sanken devices to be the best around, and they are then accordingly made to pay for the priviledge of owning them.
Same goes for other components as well, capacitors, resistors, you name it. The tragedy here is that some products ARE indeed far better to most others, but this is lost in the overall humdrum of manufacturers promoting their products.
Third, you are made to pay for what you may not really need, typically facilities in preamps and power outputs in power amps, only to have the unit half decently made. There were, and fortunately there still are, some notable exceptions to this rule, although there's no arguing that more money can mean better sound, as higher prices ease the designer's burden.
Off hand, I'd say the net price of components in quality made products from manufacturers (implying purchasing at wholesale prices) account for typically 15-20% of the product, excluding the case and power transformers (and here, they vary wildly in their approaches).
Lastly, many reputed products are made in the most unlikely of places. For example, how many audio companies from taiwan do you know? I don't know any, but it's there products from people like Rotel, NAD, Luxman, Proton and several others really come from. Mind you, this is not rebadging, this is subcontracting.
But there's a lot of rebadging too. Famous company X buys a great products from a totally unknown company or individual Y for peanuts, ramps up the production and charges wild prices for it. I remember McIntosh buying a design by Rajko Nestorovic in the mid-70ies, putting it into production and making heaps of money from it, although, in all truth, this enabled Rajko Nestorovic, a countryman of mine, to etsablish his Nestorovic Labs, still functional, making the Nestorovic drivers.
Today, they don't bother with this, they buy the complete manufactured boards and simply package it locally - they need to do that in order to whitewash the import and show that 51% of the value was added in the US, so they can stick "Made in U.S.A." on it, thus allowing for still higher prices.
Cheers,
DVV
