HappySqurriel said:
Squilliam said:
HappySqurriel said:
Squilliam said:
You do realise than the countless reasons for the Wii being the size it is, are going to be just as important in the next generation? They were targeting a specific size/shape and since they've been so successful I believe they will target it again.
The reason why I suspect the Wii HD won't be quite as powerful is because the Xbox 360/PS3 were that far ahead of the power curve over last generation and they produced far more heat. Heat and power consumption are not going down as fast as chipsizes are.
|
The Wii HD may also need to be a small low energy consumption system, but there is nothing saying it has to be as small as the Wii or as low in energy use and there are ways to keep it (about) the same size while increasing energy consumption without running into heat problems.
Consider that a laptop DVD drive is less than half the thickness of a regular DVD drive, and if you chose to use a laptop drive in the Wii HD you could greatly increase the airflow around the CPU and GPU which should enable you to run using warmer components; if you combined that with more active cooling you could run even warmer components without any concern for overheating.
Being that it is likely that you will see slim-line XBox 360/PS3 systems late 2009/early 2010 as they move to the 32nm process which will not be much larger than a Wii, and Nintendo has used the exact same companies to design their console as Microsoft has, why wouldn't Nintendo be able to surpass that feat in 2012 using a 22nm process?
|
This is what I figure the "next" Wii will be targetted as.
- Unified CPU/GPU architecture with fast ED-Ram or equivelent cache on board to save bandwidth.
- Simplified south/northbridge also intergrated onto the CPU die/multichip module.
- 2 x 2gbit GDDR 5 chips on a 64 bit bus yielding roughly the same bandwidth/memory of the Xbox 360
- High density next generation optical drive running at 4/6 speed to yield faster transfer speeds so they don't have to use a HDD which are expensive.
- 720P is the target rendering resolution.
- Internal components in the range of 20-30W total power consumption.
- 4-8gb Flash onboard for storage.
So it will be cheap, easy to manufacture and small, fitting into the form factor of the current Wii, without being obtrusive so it will use quiet fans etc. Why change a manufacturing style which works? They don't need to offer anything more than the current Xbox 360/PS3 really and therefore can benefit from the engine design work thats already been done.
Btw, they will likely be using the 32nm process in 2011/2012 as the 22nm process is generally reserved for cutting edge designs and consoles tend to be behind the process curve.
|
Just a few thoughts, IBM's current roadmap has the 32nm process being ready by the end of 2009 and is planning to have a 28nm process out "6 months after the 32nm process" ... In the next 2.5 years I would expect them to have the 22nm process out. Beyond that the statement "consoles tend to be behind the process curve." is not true; just looking back on the previous generation, both the Gamecube and XBox used the 180nm process pretty early in its life. A more accurate statement would be that "Consoles tend to be produced with the process the console manufacturing company's technolgy partners are ready for" which may or may not be the newest process, typically depending on their long term strategy.
Anyways, people assumptions about what Nintendo's next generation hardware will be is heavily dependant on their assumptions of why Nintendo produced the hardware they did in the current generation. Personally, I think a combination of uncertainty about the success of the Wii, the exploding development costs (and developers inability to controll themself), and the expense of hardware that could produce graphics with advanced texture effects at high resolutions were some of the main factors ... In the next generation they will have less to worry about in terms of the success of their system, developers will have learned to produce games at a reasonable price (for expected sales) or have gone out of business, and hardware capable of advanced texture effects at high resolution should be fairly inexpensive (for example, even low end graphics cards will be able to run a game like Crysis at high detail at above 1080p and over 60fps).
|
They'll likely target the process node which has a good combination of power efficiency, yields, and price. It isn't always cheaper to target the smallest process nodes especially when considering power efficiency, and yields are likely two of the most important considerations.
In all I would say the main limiting factor for Nintendo to produce a fast console is not price of materials rather its heat. If you consider the massive increase in power requirements from the last generation to the current one, or the fact that even a midrange GPU will use more power than an Extreme-edition quad core CPU.
