Crosswords4 mins ago
Ever increasing computer speed?
How is computer speed able to keep on increasing by so much every few months? I've been told in the past it has something to do with increasing compression efficiency...but the person didn't really explain it very well! If this were the case, why can't we just do it in one big whollop and have a massively powerful computer?
Thanks.
Answers
No best answer has yet been selected by perm. Once a best answer has been selected, it will be shown here.
For more on marking an answer as the "Best Answer", please visit our FAQ.As to how computer speed keeps increasing, it is down to several factors, including number of transistors on chip, architecture of chip, clock frequency of chip...and more stuff I can't think of right now. Doing it in "one big whollop" wouldn't really be practical or economically sensible - what would the processor manufacturers do with millions of obsolete chips?
Just because you have no clue whatsoever about how they increase processor performance, doesnt mean you have to make up a stupid conspiracy theory...
Solving the problems of high clock speed, designing a better architecture, trying to reduce the size of the transistors again and again...
That doesnt happen in a day. It takes quite a lot of research.
Anyway, they do do "big whollops" now and then, every time they release a new chip architecture.
The ATI Radeon 9800 series cards all carry virtually identical GPUs. The XT standard models have no difference except that the XT's capability is fully utilised, wheras the standard has been castrated. This is so they can manufacture the same chip, and sell it to different markets i.e. those who have lots of money, and those who have lots more.
I am doing a Computer Science degree. The people who tell me these things are not "stupid", they are the experts.
http://www.intel.com/research/silicon/mooreslaw.htm
During the production of a GPU, there are batches of silicon wafer that are used in the production of a model. Relatively speeking, only a small percentage of a given batch of wafer usually has the ability to yield a top end clock speed safely (or within a given tolerance level - linked to realistic reliability and stability factors).
What happens then is that they'll use a small amount of the silicon from a given batch to make the best version, say, the X800XT PE (Platinum Edition) and the rest will be used in lesser components with lower clockspeed. Whilst the chips in the lesser components might be capable of reaching the performance of the better component, they often cannot yield the same performance safely without additional cooling or careful monitoring.
Related Questions
Sorry, we can't find any related questions. Try using the search bar at the top of the page to search for some keywords, or choose a topic and submit your own question.