But then surly the argument comes in where it would cost more to manufacturer two different products.
I am sure it is cheaper for them to just disable modules than it is to produce a different CPU without the module.
I agree that by doing it the way that they are doing it now, that they are wasting resources. at the end of the day though it all comes down in what is cheaper for them to produce
It costs SIGNIFICANTLY less to produce 2 module chips to sell as 2 module chips than cut 4 module chips in half.
There are exceptions and points in production that better is true or not. Early on, it isn't always true simply because you WILL have failed cores so earlier on there should be a decent supply of 4 module chips with only 2 or 3 modules working. While yields aren't great it can sometimes be better to make only 4 module chips and sell off the non working ones smaller, because the margins on 4 module chips are much larger and offset lower production numbers.
Think about it like this, a two module chip at say 157.5mm2, or a 4 module at 315mm2, the bigger chips would AT BEST produce just under half as many chips per wafer. The wafer cost is pretty much irrelevant, call it £10k, and call it 500 2 module chips, meaning its $20 per core, and you can sell them for £95, thats 500 chips at £95 = £47.5k.
Now take 240 chips off the same wafer £41 a chip, but you sell them for £240 a chip, you're making £57.6k.
Basically its not a huge difference, but generally speaking more people are interested in the higher end models. Though this ignores yields, (well mostly) and how much you lose on 4 module chips that only work as 3/2 module chips and native 2 module chips that only have 1 module working..... there would be little to no interest these days from anyone for a 1 module chip, the performance of a not epic dual core is, well, meh these days and the kinds of computers you might want them in, low end laptops and maybe that performance level in a tablet, can be achieved with FAR better low power optimised architectures.
This ignores the fact that, Trinity will be more profitable than a 2 module Bulldozer, which is also a dead end core soon to be replaced. Producing the masks alone for a 2 module bulldozer would cost a few million, and they want to push trinity, not a 2 module bulldozer, so its essentially a waste.
They're already supply constrained enough as it is .... making several separate production lines which are unnecessary would pretty much put them out of the desktop processor market, at the moment.
It's something they could consider once the Gulf money for GF eventuates in more production facilities.
TSMC don't have any suitable processes for the bigger desktop CPUs ... besides, they have horrific problems with sub 40nm themselves, and will have zero spare capacity for the foreseeable future.
Essentially, not really, its not more production lines, just using a different mask set on the wafers, they'd be done on the same lines. The problem is, or not problem the thing is they have no interest long term in making 2 module Bulldozers, as 2 module Piledrivers + a gpu will be both faster, and make a higher margin and be of more interest in all the markets they want to sell volume in. Every wafer start of a 2 module Bulldozer, is one less wafer of Trinty's being made, and that is the product laptop makers, desktop builders want.
If we see 2 module Piledrivers towards the end of 2012 when we see 8 core piledrivers as well, who knows, potentially.
AMD are aware, but like I stated, for whatever reason they don't want to push their brand.
I live in the UK, and UK is a big market. I've never seen any adverts on the telly for AMD (that I can remember).
Advertising and marketing is always a risk. If you spend £1M on marketing, you may or may not get that back in additional profit. There is no magic formula. You have to try and see.
At present Intel are their main competitors. They advertise nationally. They are HUGE. Perhaps AMD should go down the route of pushing their brand. It may benefit them, but they won't know, until they try.
As always, if you can make 30mil chips a quarter AND you can sell 30mil chips per quarter, every cent you spend on marketing is lost profits. AMD's production has been borderline insane from their fab, they push their fab harder than most.
Think about the recession a couple years back, Intel shut down 2 fabs as demand was down, AMD didn't and remained at full capacity, because they have one fab, not 5 and they run it at near 100% capacity, with demand far above that. While Intel run their fabs at a lower capacity, so it was cheaper to run a few fabs at higher capacity than all their fabs at 70-80% capacity. This also means Intel can generate more sales from marketing as they have spare capacity, AMD can't.
When AMD have hugely more capacity available to them, its possible we'll see more marketing, the other problem is of course that Intel are STUPID rich, cash rich, and can spend an extra couple billion on marketing if they want, AMD are still in debt, meaning spending on marketing isn't as easy.
If Intel spend say 500mil a year now and AMD spend next to nothing, if AMD spent 500mil, Intel could spend 2billion and simply out advertise them into oblivion.
)
