06-23-2011, 07:13 AM
id assume 1st it needs to be hacked in order to see how it even reads and writes data(on all levels) so give it till it gets hacked then add 5 years
06-23-2011, 07:13 AM
06-26-2011, 09:37 AM
06-26-2011, 09:44 AM
06-26-2011, 12:13 PM
06-26-2011, 12:42 PM
06-26-2011, 03:56 PM
06-26-2011, 08:26 PM
06-26-2011, 08:45 PM
06-26-2011, 09:32 PM
06-27-2011, 06:43 AM
Now I'm gonna try word this in a way that doesn't result in me being ripped to pieces. I think you guys may be overexaggerating how long it will take, don't get me wrong I know it'll be a while but some of you are saying around 10 years which seems a tad high to me so it made me think about how far we've came in 10 years.
Just comparing intel CPU's back in 2001 we had single core pentium 4's ranging from 1.3GHz to 2GHz. These days we have quad cores from 2.8GHz to 3.4GHz with automatic overclocking by a small amount and six core CPU's from 3.2GHz to 3.47GHz. By rounding to 3 decimal places we can see pure clock speed has went up by more than 2.153x for the low end and about 1.735x for the high clock speeds but core count has increased to 4x for mainstream and 6x for enthusiast markets.
Now I'm not making any predictions about what we'll have in 10 years, I'll just make an estimate. If the same rate of growth continues then we could expect to have 6.02GHz 16-core cpu's for mainstream intel products to ironically 6.02GHz 36-core cpu's for enthusiast markets (again merely trying to estimate depending on the changes we've seen). That doesn't even take into account the architectual improvements over the period, for example Nehalem was 10-30% better at single threaded tasks clock for clock compared to penryn. Sandy Bridge is around 20% better clock for clock than Nehalem and Ivy Bridge is said to be 20% better than Sandy Bridge. There's HUGE gains to be had from architectual improvements and if graphene is made to be capable of making CPU's the power we have in our hands has the potential to be drastically better than silicon.
Now let's take a look at what is required to emulate the wii for example the i3-2100 3.1GHz dual core, it has a clock speed 3.444x higher than the wii and 2x the amount of cores. So let's assume the Wii U uses a 3.5GHz triple core IBM cpu like the rumours suggest, going by the performance requirement for the wii it would take a 12.05GHz 6 core CPU. Which is no doubt years away but assuming the emulator was very well threaded it wouldn't take a such a high clock speed when you consider Intel and AMD have been concentrating more on core count instead of raw clock speed for years so naturally we can expect to it to be much more likely to have over 30 cores by the time we're at 12GHz.
So in conclusion to my long winded completely inaccurate post I'd say 5 years after the console's release we'll start seeing the emulation scene getting started and by about 8 years I would expect to run most if not all games at full speed. Maybe I'm being optimistic but 10 years is a very long time in technological terms, anything can happen in that space of time so don't be so quick to write what we're capable of just yet
Just comparing intel CPU's back in 2001 we had single core pentium 4's ranging from 1.3GHz to 2GHz. These days we have quad cores from 2.8GHz to 3.4GHz with automatic overclocking by a small amount and six core CPU's from 3.2GHz to 3.47GHz. By rounding to 3 decimal places we can see pure clock speed has went up by more than 2.153x for the low end and about 1.735x for the high clock speeds but core count has increased to 4x for mainstream and 6x for enthusiast markets.
Now I'm not making any predictions about what we'll have in 10 years, I'll just make an estimate. If the same rate of growth continues then we could expect to have 6.02GHz 16-core cpu's for mainstream intel products to ironically 6.02GHz 36-core cpu's for enthusiast markets (again merely trying to estimate depending on the changes we've seen). That doesn't even take into account the architectual improvements over the period, for example Nehalem was 10-30% better at single threaded tasks clock for clock compared to penryn. Sandy Bridge is around 20% better clock for clock than Nehalem and Ivy Bridge is said to be 20% better than Sandy Bridge. There's HUGE gains to be had from architectual improvements and if graphene is made to be capable of making CPU's the power we have in our hands has the potential to be drastically better than silicon.
Now let's take a look at what is required to emulate the wii for example the i3-2100 3.1GHz dual core, it has a clock speed 3.444x higher than the wii and 2x the amount of cores. So let's assume the Wii U uses a 3.5GHz triple core IBM cpu like the rumours suggest, going by the performance requirement for the wii it would take a 12.05GHz 6 core CPU. Which is no doubt years away but assuming the emulator was very well threaded it wouldn't take a such a high clock speed when you consider Intel and AMD have been concentrating more on core count instead of raw clock speed for years so naturally we can expect to it to be much more likely to have over 30 cores by the time we're at 12GHz.
So in conclusion to my long winded completely inaccurate post I'd say 5 years after the console's release we'll start seeing the emulation scene getting started and by about 8 years I would expect to run most if not all games at full speed. Maybe I'm being optimistic but 10 years is a very long time in technological terms, anything can happen in that space of time so don't be so quick to write what we're capable of just yet