Intel: 1 billion transistor 20GHz chips possible by 2007
The increased Hz is great, but what we need are better opcode sets and a bigger departure from the 80886 support. What we have now are really powerful processor(s) (1.7+ GHz) in our machines (Heck I think my graphics controller could school my first three computer's CPUs combined) but no good use of the power. If we had optimised opcodes for our CPUs we would have blazing fast speed at lower Hz.
9 out of 10 time-sensitive embedded devices use RISC not CISC. MIPS processors not x86 class. Such as your palm, digital cable set-top box, routers, etc...
I guess in car terms, we already have enough RPMs we just have no torque. BTW, this not Intel's fault. They are supplying what the market demands... maybe we should start to think about what we ask for...
Intel and AMD will be offering improved IPC through diversification of the x86 CISC operating set with later 32-bit generation processors, which will extend through the upcoming 3-4 years before 64-bit is commercialized enough for mass introduction. Just because the processoers we have now feature a relatively poor IPC value as compared to some higher performance RISC solutions (UltraSparc-III or G4 for example), do not expect technology to stand still while manufacturing processes improve. x86 is nearing the end of its viable likfe cycle, but I feel it could survive for atleast ten more years through proper modification for improved efficiency at both the hardware and software layers.
I do have a real need for higher frequencies, as even a Thunderbird 1200 will likely prove inadaquate within the next 12-18 months for those needing true workstation/server class performance. While this is a small segment of the overall computing community, I believe the high-performance market will continue to expand due to the driving need for better productivity and the spreading of the "we need it yesterday" rationalization. Try running a realtime linear video editing application on a even a dual Athlon MP 1.2 GHz, then you'll start screaming for the need for more MHz.
Catch ya' later,
I know this is not really related but after reading this I'm left wondering when the industry will get some cojones and make serious research efforts towards applying quantum technology into PC architecture. By applying this we can move into the arena of true fuzzy logic and achieve speeds and processing power unmatchable with todays technology. I try to keep an ear to the ground for these things because they interset me, but applied quantum technology within pc architecture seems to only be discussed in classrooms and theoretically at that. As far as a 20ghz processor is concerned.. Nice idea.. But how about a 200Thz quantum processor? Now that would be nice..
hmm.. by all the speed i still have one question:
what about the HF problems those high frequencies would cause? ist it possible that the computers of the future will be embedded in a case that you can't open anymore because the interferrence from other devices or maybe even from the surrounding nature wuold cause the cpu to breakdown?
just an idea...
From what I understand (mind you its all from theoretical disussions in class) a computer of the future based on quantum technology (using all 32 electrical states of an atom) would be highly susceptable to magnetic interference and would require some sort of sheilding. As to wether or not we will see one in our lifetimes, I'd say its very probable that we will. It is just going to take a complete revamping of the way computers are built and programmed from the ground up.
New Security Features Planned for Firefox 4
Another Laptop Theft Exposes 21K Patients' Data
Oracle Hits to Road to Pitch Data Center Plans
Microsoft Preps Array of Windows Patches
Microsoft Nears IE9 Beta With Final Preview
Simplified Analytics Improve CRM, BI Tools
Android Passes RIM as Top Mobile OS in 2Q
VMware Updates Hyperic System Management
File Monitoring Key to Enterprise Security
LinkedIn Snaps Up SaaS Player mSpoke