I know this is one of the reasons I’m still waiting to get an Xbox 360. It’ll be the most inconvenient item in my AV setup, especially with an external HD-DVD unit…

When you build a beta unit you are using a smaller run of circuit boards, produced at one time. You are using a smaller stock of parts, from (usually) your choice of suppliers, and you are very careful about how the units are produced.

Then you test. And the units go through changes if needed, and you test again. Eventually you wind up with a design and assembly process that passes your tests.

But things (sometimes small and unavoidable things) change as you scale up production. Perhaps a part is not available from a certain supplier, so you switch suppliers. Same part, same specs, different realities. And small differences make all the difference. Especially if you are running your hardware hard. Maybe the test batch had a thermal compound that worked just that much better then the spec said. But in full production you had to switch to a different supplier (due to cost, or availability, or production requirements) and the new stuff is right on the spec, and you lose that buffer zone that the original test units had.

That kind of stuff is extremely hard to predict, because sometimes there is very little explanation for the new failure.

And for the record: Xbox 360 hardware failure has very little to do with mechanical failure, usually it’s a heat issue affecting the Motherboard, Graphics Processor, or CPU. And as far as analogies go, the expensive Kitchenaid mixer would be a $2500 dollar Alienware computer. The Xbox 360 is a high end console, but not the be-all-end-all of computing power. It’s more like a Cuisinart mixer (and yeah, those fail sometimes).

You can concoct theories all you like about how it’s simply impossible to avoid problems like this, but the simple fact is that we don’t see this kind of pain and failure rates in many other manufactured products. If I go buy a Kitchenaid mixer, I don’t expect it’s going to blow up within two years, and they have plenty of moving parts.

What you’re saying is really that reliability costs money. That’s not surprising. I claim that the answer shouldn’t be “consumers should suck it up so manufacturers can make more money”.

Nat- When was the last time a mass produced product was stress tested after a full-scale batch was assembled, but before shipping to distributors? I would venture: never. You create a prototype, you use a turn-key solution to generate a beta batch. You stress test the daylights out of the beta. Then you have to implement full-scale production. But the means, and sometimes the suppliers and parts themselves, are different from the small scale batch. Yes, you can do some testing with the mass produced units, but you won’t be able to do the same stress testing that the beta units went through. You can’t afford to, because that’s a lot of money sitting around not being turned into profit.

My bet would be that the units they stress tested were a lot hardier units simply because they were made in a smaller batch, with choice parts. They were also more expensive units. As much as 2-10 times as expensive (depending on parts and assembly). Not a cost you’d want to pay as a consumer. These are simply the realities of mass production. The test unit is usually a sturdier unit, and you can’t predict all of the minor changes that affect the unit when it shifts to mass production.

Once they hit the shelves you just have to deal with it.

I mean, why should manufacturers go through the time and expense and effort of real QA and testing if we’re just going to shrug our shoulders and say “eh, these things just break, that’s how it is” when their poorly-engineered pieces of crap inevitably fail?

To hell with that. In almost any other part of the market, the idea that a FOUR HUNDRED DOLLAR purchase shouldn’t reasonably be expected to be reliable would be completely insane.

We should expect better than this.