I don't have much of excuse for the PS3 since they seem to be purposely making it less reliable.

The Wii however just isn't nearly as advanced. It isn't fair to compare it to HD consoles. It's well built but it isn't advanced technology and its expensive.

Even if you take out RROD, the two HD consoles fail at a 1 in 10 rate. That's quite bad. The Wii has a fairly normal failure rate.

It's perfectly fair to compare the Wii to the HD consoles. This is a gaming/sales site. Reliability matters to consumers, and the HD consoles are not as reliable as consumer electronics devices should be.

A 12,000 sample size might be accurate for, say, 5 million products.

For 100 million I'd rather go with manufacturer figures. This is marginally more accurate than the GamePro poll.

The ratio of sample size:install base is roughly 1:8000. This is equivalent to an install base of 80,000, and asking TEN people.

That's not how it works actually. Yes, as the population increases the size of the necessary sample increases, but it doesn't do so proportionally. As the population increases, the margin necessary for the sample to still be accurate diminishes.

For a population of say 4 million voters for example, the necessary sample size is a little less than 500. Increase the population to 200 million, and while the population has grown by 50 times, the sample size only needs to be about twice as big.

Statustics coesn't work that way. Sample size does not need to scale with population size. All you have to do is have a sufficient large number such that you are almost guaranteed (95-99% chance) to get a "representative sample." Supposing there is no selection bias 12,000 would be more than enough to accuratley give a representative sample from a population of 100 million. You have absolutely no idea what you are talking about. Please go read a statistics book.

It's true I know nothing about this, and I'm probably just being dim.

But seeing 12,000 and 100 million, I wonder how that can possibly be a good sample size and how, Gnizmo, picking 12,000 from 1 million is the same as picking 12,000 from 100 million.

But I will submit to those who know more than me about this sort of thing.

Well statistics is an interesting subject, and it's good to have at least a decent layman's understanding of it (it's relevant in a lot of subjects, like this). I studied it in second year of University, and I certainly wouldn't recommend that, but there are books available that are interesting enough to not feel like work and give a decent understanding.

Its not the same, but it gives the same result. I am too tired to go into the specifics, so on some of this I ask you to take me at my word. Failing that you should be able to find some mathematical proofs that back me up with a bit of googling. Sorry for the laziness, but I am running dead tired lately.

What you are looking for with sample is to get, essentially, a small scale version of the larger population. For example, if the population is 10% musicisans then you would want 10 people to be musicians in a sample of 100. This does not always work out as probability does not work over the short term. You can flip a coin 10 times and easily have it land on the same side each time. This poses a large potential problem that kind of corrects itself.

As it turns out over a large number you start seeing the laws of probability kick in. Flip a coin a hundred times and it will be very close to 50%. On a small scale it is easy for a few improbably picks to skew the results dramatically. A sample size of hundred is thrown off 1% for every odd answer. Increase it to just 500 though and it is down to 0.2%. As a result you get a slower growth for sample size needed to iron out a possible random selection bias. You just are not that likely to get 1 of the 6,000 people with one eye in a population of 20 million with two eyes. You do want to increase the size to account for the increase in general population, but not even close to a perfect correlation.

Hopefully this isn't too confused. I am sleep deprived, in pain, and hopped up on codine so it takes a lot of effort to think clearly. I am certain Demotruk will clarify anything I leave vague.

If it was old technology, it wouldn't be so small. Your problem is that you are only measuring a 'newer technology' by increases in power when it can be looked at from a different value perspective.

The Wii isn't the issue though. The Wii is fairly normal, it's no more reliable than the other devices I own (including my gaming PC...). It's the other two that are unreliable.