Sign in with
Sign up | Sign in
Your question

Why are CPUs stuck at 3Ghz

Tags:
  • CPUs
  • Core
  • Overclocking
Last response: in Overclocking
Share
August 26, 2009 3:52:02 PM

I am curious why CPU manufacturers are stuck at 3Ghz?
I understand we are utilizing multi core technology, increased the bus width and better use caching to give us the added horse power but that is only useful if your OS and Application can utilize it.

More about : cpus stuck 3ghz

a b à CPUs
a b K Overclocking
August 26, 2009 3:57:30 PM

who is stuck @ 3Ghz?

your question is kinda confusing......
August 26, 2009 4:24:06 PM

can you rephrase your question into a better understandable question? also you are talking about stock speeds right?
Related resources
a b à CPUs
August 26, 2009 5:04:58 PM

we are basically stuck around the 3.0 ghz viscinity because the more ghz a cpu uses, the more heat it outputs and because air is used to coolmostof the worlds processors the heat will be too much to handle and even liquid cooling has its limits, so since they are no longer able to simply compete ghz for ghz, they must find new ways for cpus to get faster while maintaining an acceptable thermal envalope
a b à CPUs
August 26, 2009 5:06:57 PM

hopefully amd will be releasing a 3.6ghz quad when they figure out how to get the 965s tdp to 125 w
August 26, 2009 5:55:42 PM

"viscinity" is not a word.
seeing how you cannot spell a simple word, it seems your whole paragraph is meaningless and untrustworthy
a c 172 à CPUs
a c 197 K Overclocking
August 26, 2009 6:20:08 PM

overshocks said:
"viscinity" is not a word.
seeing how you cannot spell a simple word, it seems your whole paragraph is meaningless and untrustworthy

"viscinity" "visinity" "vicsinty" "viscinty""vicinty" "vicinity" is hardly a simple word. :) 

And I don't know for certain, but keeping in mind that this is an international website, there's a pretty good chance that Inglish English may not be his first language.
August 26, 2009 7:04:53 PM

overshocks said:
"viscinity" is not a word.
seeing how you cannot spell a simple word, it seems your whole paragraph is meaningless and untrustworthy

I've been reading Tom's for more than ten years. Your ability to spell is amazing. Why waste your talent here?
a b à CPUs
a b K Overclocking
August 26, 2009 8:37:45 PM

Proper capitalization, punctuation, spelling and grammar; so disappointingly and typically beyond the ambitions of the "text" generation. :pfff: 
a b à CPUs
a b K Overclocking
August 26, 2009 10:23:01 PM

^Good to see you again man.
a b à CPUs
a b K Overclocking
August 26, 2009 11:36:21 PM

Icandealwiththeoccasionalmisspellingbutthelackofpunctuationjustdrivesmeinsane.
a b à CPUs
a b K Overclocking
August 26, 2009 11:42:08 PM

That's what happens when you run your manifold pressure too high, your RPM too low, and your mixture too lean.

Incidentally, nice Avatar. P51 Charlie?
a b à CPUs
a b K Overclocking
August 27, 2009 12:09:24 AM

I personally kind of liked the Avatar in the visnicinity of 5 posts above^
a b à CPUs
a b K Overclocking
August 27, 2009 12:20:43 AM

Verry vivasceous! :D 
a b à CPUs
a b K Overclocking
August 27, 2009 12:25:10 AM

lol......

this has turned into a comedy show.......
a b à CPUs
a b K Overclocking
August 27, 2009 12:37:13 AM

CompuTronix said:
That's what happens when you run your manifold pressure too high, your RPM too low, and your mixture too lean.

Incidentally, nice Avatar. P51 Charlie?


Yup. I actually got to fly it last week. I'm told its the only one left modified for 2-up. I was only allowed to touch pedals and stick :) 
a b à CPUs
a b K Overclocking
August 27, 2009 12:48:50 AM

You lucky bast*** :bounce:  ! A once-in-a-lifetime experience! :love: 
a b à CPUs
a b K Overclocking
August 27, 2009 2:01:46 AM

AYe! I've had flying a P51 as a dream since I was a kid building them as plastic models. It was the one thing on my list that I had reconciled never to do.

Then my wife went to Wawa for coffee, saw a Flyer advertising this little air show, and the next day I drove to the airport and bought an hour.

http://www.collingsfoundation.org/menu.htm
a b à CPUs
a b K Overclocking
August 27, 2009 3:41:56 AM

Fort Lauderdale in January. Thanks!
August 27, 2009 4:30:01 AM

texastailspin said:
I am curious why CPU manufacturers are stuck at 3Ghz?
I understand we are utilizing multi core technology, increased the bus width and better use caching to give us the added horse power but that is only useful if your OS and Application can utilize it.


I will get this back on subject.

A single transistor can switch in the Terahertz range by itself and Intel has shown this off before (google it), however the copper wires and the distances between the transistors heavily limit us. We are hitting the limit of how fast electrons can flow through copper. Even though electrons can travel near the speed of light, as they pass through materials such as copper, they hit the nuclei of the copper atoms and bounce back and forth making the distance they actually travel forward somewhat slow. With all these collisions happening it creates heat due to the friction. Heat is a measure of how much an atom is vibrating back and forth. Because the copper atoms are vibrating so much it further increases the friction making even more heat while even more-so slowing down the electrons. This is why we try to keep our processors cool so we can overclock higher. With less collisions the electrons move onwards faster which can yield a higher clock speed. Also by making processors smaller it decreases the distance between the transistors which is 1 of the many reasons we are still slightly getting faster.

Processors will not get a significant clock speed increase until we use different materials. I personally think it would be great if we could build a transistor that could switch off of light. Then we should be able to hit terahert frequencies.

I am basing this off of what I know and it is kind of my theory, I did not research this so don't crucify me if I got something wrong.

Also I wanted to say I am amazed that some of you had to ask the op to rephrase and elaborate on the question. Really, is it that hard to understand? O ya, this is rhetorical, I don't actually care to read why.
August 27, 2009 2:42:29 PM

I wonder if silver would make a noticable increase if used instead of copper in todays PC?

Edit: as a conducter
August 27, 2009 3:04:01 PM

yeah i think they're messing with new materials but i honestly don't see the consumer market getting these high-tech cpus even if they happen. Intel and AMD are pretty lazy and i think they'll just stick with increasing the number of cores rather than having faster ones.
a b à CPUs
August 27, 2009 4:27:13 PM

I think we are the in viscinicsincinity of a troll. :p 
August 27, 2009 4:34:49 PM

whose the troll?
a b à CPUs
August 27, 2009 6:55:47 PM

every1s a trole
August 28, 2009 12:21:27 AM

knotknut said:
I wonder if silver would make a noticable increase if used instead of copper in todays PC?

Edit: as a conducter



I have thought about that also. But silver is only slightly more conductive than copper. It will make an improvement, but it probably will not outweigh the added cost.
August 28, 2009 12:35:38 AM

derek2006 said:
I have thought about that also. But silver is only slightly more conductive than copper. It will make an improvement, but it probably will not outweigh the added cost.


What's about Carbon. EDIT: I meant instead of SILICON based transistors.

Regards.

Hope someone actually answers ;)  .
a b à CPUs
a b K Overclocking
August 28, 2009 12:47:18 AM

overshocks said:
"viscinity" is not a word.
seeing how you cannot spell a simple word, it seems your whole paragraph is meaningless and untrustworthy

O rly? http://www.tomshardware.com/forum/253708-29-best-overcl...

xaira said:
what about platinum?


Platinum would probably prevent Intel keeping up with Moore's Law. Remember that Moore's Law now says (after it was refined decades ago) that the number of transistors that can be placed on an IC to achieve the lowest cost per transistor doubles every two years. Most people forget about that middle part.
August 28, 2009 12:57:03 AM

Quote:
randomizer wrote:

Platinum would probably prevent Intel keeping up with Moore's Law. Remember that Moore's Law now says (after it was refined decades ago) that the number of transistors that can be placed on an IC to achieve the lowest cost per transistor doubles every two years. Most people forget about that middle part.



What's about my last post ;)  .

Regards.
a b à CPUs
a b K Overclocking
August 28, 2009 1:11:09 AM

Carbon would be cheaper but I don't know if they'll end up using it.
August 28, 2009 1:21:17 AM

Additionally platinum by it self does not conduct electricity all that well. Here is a list of elements and their conductivities.

http://environmentalchemistry.com/yogi/periodic/electri...

Also I wanted to say that the distances and the interconnects to each transistor is not the only factor slowing us down and creating heat. I kind of made it sound that way. But transistors also create heat when switching and can be much more efficient when using different materials such as when we switched to High-k dielectrics.

This time I actually did go back and research some of the things I posted last night. The transistor that Intel made uses different materials from the ones we currently use. I want to correct myself and say that a single transistor that is in your processor cannot operate at 1THz, however it will still operate at a much higher speed by itself. The University of Illinois also made a 604GHZ transistor but again with different materials. If we can find a way to use these transistors along with different interconnects we could get very fast cpu's using electrons yet.

Also I do not know much about how carbon can be used as a replacement over silicon. Hopefully someone else does.
a b à CPUs
a b K Overclocking
August 28, 2009 1:27:14 AM

I assume when we're talking about carbon that we're talking about carbon nanotubes.
August 28, 2009 2:47:41 AM

derek2006 said:
Additionally platinum by it self does not conduct electricity all that well. Here is a list of elements and their conductivities.

http://environmentalchemistry.com/yogi/periodic/electri...

Also I wanted to say that the distances and the interconnects to each transistor is not the only factor slowing us down and creating heat. I kind of made it sound that way. But transistors also create heat when switching and can be much more efficient when using different materials such as when we switched to High-k dielectrics.

This time I actually did go back and research some of the things I posted last night. The transistor that Intel made uses different materials from the ones we currently use. I want to correct myself and say that a single transistor that is in your processor cannot operate at 1THz, however it will still operate at a much higher speed by itself. The University of Illinois also made a 604GHZ transistor but again with different materials. If we can find a way to use these transistors along with different interconnects we could get very fast cpu's using electrons yet.

Also I do not know much about how carbon can be used as a replacement over silicon. Hopefully someone else does.



I'll see what i can Find.

Regards.
August 28, 2009 3:42:50 AM

lol the viscinity guy's answer was good, don't rip on him for misspelling a word dude. www.englishpage.com if you want to discuss punctuation etc.
!