Why does it need two GPUs anyway? The high performance once should be able to downclock when not needed and save power that way.
It's funny how this brings me back to my car days and the way a 4-barrel carb works, which would only seem natural. In a 4-barrel carb the primary's are the energy savers, they're basically two flaps that only allow in so much fuel. When you step on the gas even further the second (secondary's) set of flaps open to allow for more fuel = giving you that great burst of power. When you don't need it, let off the gas. There's no flipping a switch, no restarting a car, just a natural way of providing more energy and of course using, wasting?, more.
Sorry to ramble, and I know that computers are a different bag of nuts, but it just seems there should be a natural way of doing things and Apple has definitely missed the mark here.
I'll wait for revB also.
I don't quite get the point in both gpus on the macbook pro! An extra hour on battery!? or so they claim...brilliant...not!
What am I missing here? Most everyone seems to be assuming that Apple deliberately decided that they needed a notebook with two GPUs, and designed the new MacBook Pro accordingly.
Seems to me that exactly the opposite occurred. I suspect that if the MacBook Pro had been designed in a vacuum, only one graphics solution would have been used, and we'd all be runnning a single discrete GPU all of the time.
But the MBP wasn't designed in a vacuum, it was designed alongside the MacBook.
If you're looking for economies of scale, you use as many common parts across as many products as possible. Take a look at the MB disassembly pics that are already starting to show up around the 'net. It appears that the MB and MBP use the same basic logic board arrangement. I strongly suspect that Apple recognized that it would be lower cost to have one basic board solution with one CPU and GPU/integrated Northbridge chip package for all MB and MBP, rather than having to design to suit two different Northbridges (one with integrated GPU, one without).
In other words, from Apple's point of view, Apple *knows* going into the MBP design that the GPU integrated into the Northbridge chip package is a *complete waste* of a GPU, but that their *overall* costs are lower because it's cheaper to waste that GPU than to design and procure two separate Northbridge parts and design and procure two different logic boards to accommodate them.
So, as Apple moves toward product release, what do they do? They tell their MBP customers about a "benefit": that the discrete GPU can be turned off and the otherwise useless, along-for-the-ride integrated GPU can be used to extend battery life (basically, an extra "feature" that comes along for "free"). They never claim that the two GPUs can be used together (at least not yet, maybe they have something up their collective sleeves for Snow Leopard, but for the time being, that's irrelevant to today's discussion).
So, looking at this from a manufacturing standpoint (the logical thing to do if you want to try to figure out what's driving Apple's design decisions), I don't see what basis there is for *any* complaints. Basically, the new MBP owners get something for nothing (a power conservation mode they'll likely never use) and maybe something which will bring them more performance when Snow Leopard comes out.
So enough with the Apple name-calling for not enabling some form of co-processing which was never intended in the first place.
$0.02 deposited,
Mark
