Thanks for the FYI. I'll be sure to let all of my colleagues know that we're idiots.
I can find quite a few ways to waste that much ram right now, but realistically it's not going to be used for anything useful anytime soon. The biggest benefit I see from this is that it will allow you to be lazier and spend less time optimizing for memory, allowing more focus on features.
That will still be years down the road.
What is the most expensive part of developing a game? Time. The same as any large project. Especially those involving a sizable staff where any one segment can create a bottleneck for everyone else.
You say that you understand the technical side but do you know the first thing about management? Spending less time optimizing isn't being lazy, it's being economical with what actually matters. Man hours.
It's a pretty different situation. MS stated publicly that bandwidth of the 360 EDRAM was 256GB/s at first, but that was proven to be incorrect. This time around we've got an 'official' 102GB/s (that can be confirmed by looking at the specs), but that's been revised up to 192GB/s.
Remember that according to the article this is coming from developer sources, and I don't think it would do MS any good to misrepresent the eSRAM bandwidth in that situation.
256GB/s was correct, for one specific part of the hardware. That wasn't the pipe from the EDRAM to the GPU, but that didn't stop MS from acting like it was pre-release.
Now the 192GB/s rumor is coming from unnamed sources via the gaming press with even less clarification of what exact part of the data pipeline it's referring to. It is in every way less meaningful than the 256GB/s claim from last generation.
But the reason they got there was dumb fucking luck. Microsoft though they were going to have a memory advantage from day one and got unlucky.
Why do you assume it was dumb luck? You think that Sony's lead hardware designer, working in Japan, didn't have frequent discussion with key people from Samsung and Hynix (both South Korean companies, so literally just a few hours away) about their long term road map for GDDR5? If both told him that they expected 4-Gbit chips to be ready for full production this spring why wouldn't he design the system around that?
I can't imagine Cerny and co. hadn't been painted this very road map a year or more ago by both manufacturers of GDDR5 RAM. Initially playing it safe with the 4GB allotment was sensible at the time as the ramp up was something that could happen late in the console's life without significant negative impacts on software development or hardware production. In short, Sony had an ace in the hole and were just waiting to play it when they knew it would deliver.