• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Any purely-serial processors faster than 1,000 THz?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Status
Not open for further replies.

Green Xenon

New Member
Joined
Jan 11, 2012
Hi:

Does there exist any CPU is today's world that possesses ALL the following specs?:

1. Clock-frequency at or higher than 1,000 THz
2. Single core
3. Limited to only 1 instruction per cycle

In theory, what is the maximum clock rate possible if the cooling system is sufficient? Higher frequencies generate more heat, and thus require more intense cooling.


Thanks,

GX
 
1. No (not that I know of, not a single processor at that speed)
2. See #1
3. See #1

I would imagine there is a single core CPU out, but its not limited to 1 IPC, and nowhere close to 1,000Thz.
 
Seen this account post the same stuff on other forums, just a heads up...not sure whats going on here.
 
Good to know, Fureh.

The fastest "processor" that I am aware of is actually a transistor designed at Georgia Tech University in collaboration with IBM. It had a switching frequency of 350 GHz at room temp and 500 GHz with liquid helium. That was five years ago and he is asking about a full production processor operating at or above 1 petahertz. :shock: He is either trolling or has absolutely no clue what he is asking.
 
Didnt know about that...though till I see those kinda numbers on air...guess its about as useful as the fusion power tests their doing...and ending up in the hole energy and materials wise.
 
why would we have a 1000thz CPU if writers thought of something crazy high for the 2400s

that was my point
 
Good to know, Fureh.

The fastest "processor" that I am aware of is actually a transistor designed at Georgia Tech University in collaboration with IBM. It had a switching frequency of 350 GHz at room temp and 500 GHz with liquid helium.

Helium is rather expensive. Can nitrogen substitute for helium?
 
Good to know, Fureh.

The fastest "processor" that I am aware of is actually a transistor designed at Georgia Tech University in collaboration with IBM. It had a switching frequency of 350 GHz at room temp and 500 GHz with liquid helium.

What is the highest-frequency processor that is available to the general public? How much does it cost?
 
What is the highest-frequency processor that is available to the general public? How much does it cost?

A little over 3 GHz. That is no typo. Three. Gigahertz. Nearly a million times (literally, as 1000 THz is one million gigahertz) slower than what you asked about in your original post.
 
A little over 3 GHz. That is no typo. Three. Gigahertz. Nearly a million times (literally, as 1000 THz is one million gigahertz) slower than what you asked about in your original post.

He has obviously seen alien technology far superior to what we have on earth.

In fact, he might need a 1000THz processor to finish rebuilding his ship so he can return home. :rofl:
 
Scientists at the University of Rochester have come up with a new "ballistic computing" chip design that could lead to 3,000-gigahertz — that's 3-terahertz — processors that produce very little heat.

Marc Feldman, professor of computer engineering at the university, characterizes the design, the Ballistic Deflection Transistor (BDT), as radical. "There's a real problem for standard transistors to keep shrinking," he says. The BDT doesn't have a capacitance layer that becomes problematic at very small scales the way current transistor designs do.
Quentin Diduck, the graduate student at the University who came up with the idea, describes the BDT as the next step on the evolutionary track after relays, tubes, and semiconductors.

The BDT, according to the University of Rochester, "[bounces] the electrons into their chosen trajectories — using inertia to redirect for 'free,' instead of wrestling the electrons into place with brute energy." It functions more as an intersection for electrons than as a device that expends energy to stop and start them. Because of this approach, far less power is required.

The BDT relies on a layer of a semiconductor material called a "2D electron gas," which facilitates the transit of electrons without the interference of impurities.

Feldman says he expects the BDT to put out very little heat. "We don't have the mathematics to predict how small this is going to be," he says. "But the currents it would take are very small. So the power has to be small."

The heat generation for early versions of the design should be around a few microwatts per transistor, Feldman estimates, orders of magnitude less than current high-frequency transistors. "Now that's without doing any tricks to cut down the power," he says. "There are great opportunities for low-power design. But that's the future."

Companies that maintain large, energy-hungry data centers like Google, Microsoft, and Yahoo, to name a few, are no doubt looking forward to that day.

The BDT still has a ways to go before it ends up in PCs and servers, however. "Up to now this has been a one-graduate-student effort," explains Feldman. "We don't have any transistor behavior yet, but probably soon."

The National Science Foundation has just granted the University of Rochester team $1.1 million to develop a prototype.

______

So basically, using pulses of light to send transistors back and forth, less heat, less energy.
 
The heat generation for early versions of the design should be around a few microwatts per transistor, Feldman estimates, orders of magnitude less than current high-frequency transistors. "Now that's without doing any tricks to cut down the power," he says. "There are great opportunities for low-power design. But that's the future."
1 microwatt/transistor * 1 billion transistors = 1000 watts. :mad:

Lets hope the physics allow for an order of magnitude improvement. Those kinds of speeds at bearable power levels would be nice. (I suppose we *could* cut back on the number of transistors, but with most of them currently going to cache, we'd just wind up seriously bottlenecked by RAM...)

JigPu
 
Scientists at the University of Rochester have come up with a new "ballistic computing" chip design that could lead to 3,000-gigahertz — that's 3-terahertz — processors that produce very little heat.

Marc Feldman, professor of computer engineering at the university, characterizes the design, the Ballistic Deflection Transistor (BDT), as radical. "There's a real problem for standard transistors to keep shrinking," he says. The BDT doesn't have a capacitance layer that becomes problematic at very small scales the way current transistor designs do.
Quentin Diduck, the graduate student at the University who came up with the idea, describes the BDT as the next step on the evolutionary track after relays, tubes, and semiconductors.

The BDT, according to the University of Rochester, "[bounces] the electrons into their chosen trajectories — using inertia to redirect for 'free,' instead of wrestling the electrons into place with brute energy." It functions more as an intersection for electrons than as a device that expends energy to stop and start them. Because of this approach, far less power is required.

The BDT relies on a layer of a semiconductor material called a "2D electron gas," which facilitates the transit of electrons without the interference of impurities.

Feldman says he expects the BDT to put out very little heat. "We don't have the mathematics to predict how small this is going to be," he says. "But the currents it would take are very small. So the power has to be small."

The heat generation for early versions of the design should be around a few microwatts per transistor, Feldman estimates, orders of magnitude less than current high-frequency transistors. "Now that's without doing any tricks to cut down the power," he says. "There are great opportunities for low-power design. But that's the future."

Companies that maintain large, energy-hungry data centers like Google, Microsoft, and Yahoo, to name a few, are no doubt looking forward to that day.

The BDT still has a ways to go before it ends up in PCs and servers, however. "Up to now this has been a one-graduate-student effort," explains Feldman. "We don't have any transistor behavior yet, but probably soon."

The National Science Foundation has just granted the University of Rochester team $1.1 million to develop a prototype.

______

So basically, using pulses of light to send transistors back and forth, less heat, less energy.

Cool. We might start to see it implemented within 10 years, maybe be affordable for the average consumer in 20...
 
How many transistors would be necessary in an average BDT processor though? I mean, with current processes, my 40nm technology graphics card utilizes just over 2.1 billion transistors while remaining under 200 watts of power consumption. If you go by Jigpu's math, we're talking heat outputs in excess of 2000 watts. Granted, the processing power is exponentially more (around 80 orders of magnitude faster), BUT we're only mentioning 1 microwatt/transistor and only 1 billion transistors. Throw 2 microwatts/transistor and 3 billion transistors onto a die and suddenly you have a space heater..
 
Given the size of a typical IC transistor, I'm wondering is that might be a typo. Maybe they meant nanowatt? Microwatt seems awfully high.
 
Status
Not open for further replies.
Back