are we going to see the first terahertz cpu in the future

[tedjani]

are we going to see the first terahertz or even petahertz in the future or are we just going to add more cores

[LIGISTX]

moar corzzzzzzzzzzzzz. probably... parallel processing should result in better performance for just about everything if the program is coded well. 

[Emkryan]

Both numbers will go up. But we don't even know if we'll stick with this type of CPU architecture for long enough to reach that level. What if tomorrow we start using quantum computers or find some other way of designing CPU architectures that don't use the concept of cores or even don't run at a frequency like today? Like a new CPU that runs at 500Mhz but it's way more powerful than any i7 out there etc. You never know how/when technology is going to change

[megadarkwood]

Maybe. When we are all dead.

[Excon]

Companies are already experimenting with materials like graphite I believe. Using such materials allows them to achieve amazing clock speeds.

[lmao234]

If you make this, will you get the Noble Peace Prize?

[brat137]

At some point in the future it is almost guaranteed that we will. Unless everyone suddenly becomes content with what we have and do not require anything to be faster or more efficient.

[Mooshi]

After a certain point, clock speed alone can only do so much. It's cores that will be the way of the future.

[CoolBeans]

We have gone from 2GHZ to 5GHZ in 5-7 years, so I doubt it

[wtfxmitchell]

Maybe. When we are all dead.

i like your way of thinking

[Mooshi]

We have gone from 2GHZ to 5GHZ in 5-7 years, so I doubt it

 

Don't remind me..I went on a 5ghz run again on this 2 year old chip and the voltage + small heatsink with x2 92mm fans..let's say that didn't go very well.

[ttam]

There won't be a need to go that high.

In the past 10 years we've grown a lot. 14 years ago I remember the 1ghz processor that came out.

[TheCrazyCryptonian]

My guess is that we will have quantum computers before we even get near to those figures.

[TheCrazyCryptonian]

And we will have no need for those sorts of speeds as we will reach a point where we will find no noticeable difference in speed.

[Mooshi]

And we will have no need for those sorts of speeds as we will reach a point where we will find no noticeable difference in speed.

 

It's getting that way now.

 

With all the performance increases in chips, a cpu that is a few years old can match it's replacement's performance if it shares the same number of cores/threads when overclocked enough. (Kinda like Gen 1 chip at 4.6ghz is the same performance of Gen 2 chip at 4.2ghz) That isn't a very big improvement in my eyes. It's only when you start adding more cores on top of that is when performance starts to shine.

[Jumper118]

More bits not cores, a 128bit cpu should be more efficient than a 64 bit cpu.

[Glenwing]

are we going to see the first terahertz or even petahertz in the future

 

Well, we're sure not going to see it at any other time.

[megadarkwood]

i like your way of thinking

 

Well, we're sure not going to see it at any other time.

This guy. I like him.

[Mentalguy]

are we going to see the first terahertz or even petahertz in the future or are we just going to add more cores

At one gigahertz, light can only move about a foot, per clock(1 light-nanosecond). with 10ghz you are reaching a clock speed so fast light(or electrons) cant cant reach the other end of the chip before the next cycle. that is about the limit. Unless we move to another form of computing, we will be limited by this.

 

I see, in the near future, more cores/threads being able to make up for it, but there is only so many you can put onto one chip.

[swifty7]

Companies are already experimenting with materials like graphite I believe. Using such materials allows them to achieve amazing clock speeds.

actually it's called graphene and it's going to change everything if they let it that is.

[Mentalguy]

actually it's called graphene and it's going to change everything if they let it that is.

not really, still limited by the speed of light

[juretrn]

not really, still limited by the speed of light

And speed of light matters how? Graphene will probably enable smaller processes and much greater frequencies (x10) compared to silicon, and that will be pretty significant. 

[Mentalguy]

And speed of light matters how? Graphene will probably enable smaller processes and much greater frequencies (x10) compared to silicon, and that will be pretty significant. 

but, there will be a limitation on how fast the processor can retrieve and send information. when you hit the physical limit on a processors capability to send information across it, higher clock speed is not going to help

[A/C]

but, there will be a limitation on how fast the processor can retrieve and send information. when you hit the physical limit on a processors capability to send information across it, higher clock speed is not going to help

But what about those processors that would use light instead of electricity to transfer data? I can't remember what they're called, photon processors? Anyway, like I said, it's a processor which would use light to process data instead of electricity. That would eliminate crosstalk interference, and heat output. And would enable use of more light beams parallel. That's all I remember, it's a concept, but a very plausable one.

[juretrn]

but, there will be a limitation on how fast the processor can retrieve and send information. when you hit the physical limit on a processors capability to send information across it, higher clock speed is not going to help

That limit is VERY hypothetical at this point in time, not getting there anytime soon.