Welcome to Librarium Online!
Recently, in a Popular Science magazine, I have read something about how processor capabilities are reaching their limit. What is the reason for this? What are they doing to expand capabilities? Whats the next step in processor technology?
**Life is an undefined equation**
Only two things are infinite, the universe and human stupidity and I'm not sure about the former.
--- Albert Einstein
98% of the teen population has smoked weed, if you are the 2% that hasn't, put this in your signature.
I've heard this same complaint about every 5 years for the last 20. Don't worry, something new will certainly come along. They were saying the same thing about 6 years ago, and then they invented hyperthreading. I wouldn't worry about it, innovation never stands still.
Marion: You're not the man I knew ten years ago.
Indiana: It's not the years, honey, it's the mileage.
For the last time, there are
NO FEMALE SPACE MARINES!!!!!
i thought the next step involved super conduction? as in getting the processor so cold it works thousands of times faster.
I still hate Twiglets!
They're moving on to 64-bit architectures, these days - so while the actual Ghz may come to a halt until something new comes along, this is not the only determinant of a processor's speed.
Also Reduced Instruction Set Computers usually operate faster than a Complex Instruction Set Computer of the same Mhz (while apples "appear" to be slower than PC's they are usually much faster). Buses, caches and branch prediction all play their part to make a processor faster.
Eventually all these "tricks" will exhaust themselves, of course, and other solutions through physics will have to be found.
There should be a wealth of info if you do a search for "Computer Architecture".
Having an army and not owning a rulebook is like owning a car with no steering wheel.Originally Posted by amishcellphone
There's the problem that we don't buy really high tech stuff. Nanotechnology will be pretty available in the near future, but possibly won't go to the broad market because its cost is too high.
Take a look at PC's vs macs. Macs pwn PC's, but being about 4 times the price more, don't have the market PC's do.
Four times the price? Umm no thats just wrong. What are you looking at, top of the line powermacs?Originally Posted by onlainari
Now, show me a PC that matches an iBook. Spec for spec, but evan if you do, guess what? it wont run OS.
And now, back to the show! Proc speeds arn't everything, there are many other things that go into making them work cooler and more efficiently.
Last edited by Marlinspike; August 14th, 2005 at 14:07.
Processor speeds will probably top out in the near future. Unless a stable trinary system comes along. But the major advances will be in how to get the most out of that speed. 64 bit architecture is one such advance. Better and faster dedicated processors are another (eg graphics cards, physics cards). But the one thing which will boost the speed of the average computer will be cheap coprocessor mother boards. That is 2 processors (or more, but currently two) working in serial, and it gives big speed boosts, especially if the processors are already fast.
Also, the reason that Macs don't have the market share of PC's is not down to price. It is down to Mr Gates getting to market 6 months before Mr Jobs. Everyone bought the product which was out at the time, and then stayed with it because changing over would make their initial purchase, all of their related purchases, and all of the work they had done since useless (back then a PC couldn't even read a mac floppy disc, and vice versa).
Last edited by Corianis; August 15th, 2005 at 00:49.
Trinary systems? Are you trying to sugest an evolution of binary state systems? that would be analogue, Binary exists because the electricity is flowing or not, the use of trasistors/diodes etc to create logic gates (jeez MOSFETs and JFETS, how long ago was that?) can convert the binary state system (as above electricity is there or not) into a logical processing system.
By keeeping the number of instructions that such a system can process down you can keep the thing running faster (hence RISC architecture became proliferate in early 80s - but this has never transfered to the comercial PC)
but if you some how started to detect the different levels of current, you would not start processing in base 3 (your trinary), but would actually become an analogue state processor, and would doubtless be a big step backwards because of years spent condencing logic gates in binary state systems
Binary is High Voltage/Low Voltage (On, Off, Yes, No).
Trinary is a possibility (in that it has neither been proven nor disproved) which utilises High Voltage/Medium Voltage/Low Voltage (On, Partly On, Off; Yes, Maybe, No) circuitry. It is similar in theory to Quantum State processors, but not as inherently unstable.
The quantum state theory, which has been around for over 10 years as a theory, relies on instability so I would hope not.
All modern logical operators (AND/OR/NOT and the inclusive/exclusive versions of OR) are based on 2 stat devices having a maybe state is one of 2 things, underpinned by 2 multiple 2 state devices acting as 3 state (the use of EXLUSIVE OR gates to determine if 2 lines are in a different state) or would require radical re-engineering of all existing Computer hardware and software - financially a non starter for a comercial company - with no really hope of improved preformance (RISC proves that actually simplyfying improves this)