I FIGURED OUT WHY!!!!
(Btw if someone said this already, sorry...)
So the Wii U uses a Power based CPU. These types of CPU's are designed for servers, AND as many of you may know, Server CPU's are clocked slightly lower than typical CPU's.
Server CPU: Roughly 2.1 GHz to 3.0 GHz
Typical CPU: Roughly 3.0 GHz to 3.9 GHz
AND while the server CPU is slightly lower-clocked, speed isn't the primary execution it was designed for. Instead, it was to handle better Multi-Tasking and System Stability, and considering the Wii U has to handle a MASSIVE online infrastructure,and Support for Multiple Monitors (Wii U Gamepads), it will be AMAZINGLY more POWERFUL and STABLE for developers to work with over Xbox and Playstation, regardless of Clock-Speed.
Oh, and a +, he noted that it's only SLIGHTLY underclocked. Meaning that it will be one of the fastest Power CPU models, and the best performance.
Kinda. While their is definately a huge difference in priority between the POWER,line and the xustom powermac power pc's used,in xenon and cell, Nintendos not using a straight POWER cpu. They are having a customized cpu based off of the POWER technology.
I dont think nintendo is going this route because they want a server system for miiverse... But because the POWER line has strong real world performance.
Heres the deal with the 360, and kinda the ps3 as well sense they actually.share the same powerpc core cpu.
IBM advertised a cpu core and pitched it based on the concept of a through the roof theoretical flop performance per dollar.
So microsoft and sony, using the same core cpu engaged in a flop war, Microsoft went tricore sony went with 7 spus.
So the flop war began, 1 tflop 2 tflops... But aside from flops general perfomance of those cpus sucked.
The problem is the core itself is a very narrow
2-issue in-order execution core, featuring a 64KB L1 cache (32K
instruction/32K data) and either a 1MB or 512KB L2 cache (for Xenon or Cell,
respectively). Supporting SMT, the core can execute two threads
simultaneously similar to a Hyper Threading enabled Pentium 4. The Xenon
CPU is made up of three of these cores, while Cell features just one.
Now, stuff like this is great for ripping out flops.... But not much else.
But people or SOOOO fixated on flops. Theyre practically meaningless.
What about all those Flops?
I hear it over and over again people are obsessed with the peak theoretical performance of the CPU. Ever since the
announcement of the Xbox 360 and PS3 hardware way back when, people have been set on
comparing Microsoft's figure of 1 trillion floating point operations per
second to Sony's figure of 2 trillion floating point operations per second
(TFLOPs). And its still happening. Everyone should know for a fact that these numbers
are meaningless, but just in case you need some reasoning for why, let's
look at the facts.
First and foremost, a floating point operation can be anything; it can be
adding two floating point numbers together, or it can be performing a dot
product on two floating point numbers, it can even be just calculating the
complement of a fp number. Anything that is executed on a FPU is fair game
to be called a floating point operation. So those operations used,to benchmark that theoretical peak flop could be COMPLETELY,useless to real world gaming.
Floating point power doesnt mean diddly squat if you want to run general purpose code.
When a manufacturer is talking about peak floating point performance
there are a few things that they aren't taking into account. Being able to
process billions of operations per second depends on actually being able to
have that many floating point operations to work on. That means that you
have to have enough bandwidth to keep the FPUs fed, no mispredicted
branches, no cache misses and the right structure of code to make sure that
all of the FPUs can be fed at all times so they can execute at their peak
rates. Not to mention that the requirements for
hitting peak theoretical performance are always ridiculous; caches are only
so big and thus there will come a time where a request to main memory is
needed, and you can expect that request to be fulfilled in a few hundred
clock cycles, where no floating point operations will be happening at all.
The hard cold reality is the xenon and cell had fantastic looking peak performance, they couldnt get anywhere near it in real world performance.
The wiiu being slightly lower clocked then xenon or cell is meaningless.
The wii u cpu has been confirmed to be an out of order processor. Thats a huge deal. That alone already means its going to destroy xenon and cell in real world performance. But its not going to perform like a narrow in order processor. Code optimized for that is simply not going to run as well.