I heard that there were many variable's that lead to a pc's ultimate speed. Once the CPU hit that speed, it would begin to emit "microwave" radiation. Will we ever pass this barrier? Is there any current technology that would help us pass this barrier that we know of?
So there's a limit to how fast a pc's can be?
- Thread starter Coldkilla
- Start date
Heard it from my professor at college. Saying that the charge inside a cpu can only go so fast before it starts to emit radiation. Unless we put CPU's in microwave radiation containers lol.
Unless he's a dousche bag because I asked him about some Unix/windows relation and their speeds with dual cores.. and he said he didn't know much about dual cores... get with the times I wanted to say.. but he's my professor.. i didn't pick em
I wouldn't doubt it. Seems all they do is increase RPMs. Eventually, if disks arent made of titainium, they'll shatter into bits and pieces.
Say, you had a 1 Terabyte CPU (not realistic but whatever), could this work? Say not Dual Core or a smaller processor core. Could he be talking about if you amp up the GHz while keeping the core inside the processor the same size?
What exactly, in terms of advances in technology allow us to continue forward towards faster and faster CPU's without any bordered limitations? I want to fire back at the teacher and make him look dumb in class lol, and im the one supposed to be learning from him
Say, you had a 1 Terabyte CPU (not realistic but whatever), could this work? Say not Dual Core or a smaller processor core. Could he be talking about if you amp up the GHz while keeping the core inside the processor the same size?
What exactly, in terms of advances in technology allow us to continue forward towards faster and faster CPU's without any bordered limitations? I want to fire back at the teacher and make him look dumb in class lol, and im the one supposed to be learning from him
Frequency Spectrum
Is this what he means?
Once they start hitting those kind of frequencies the become a different type of energy (Infra-red, Ultraviolet)?
Is this what he means?
Once they start hitting those kind of frequencies the become a different type of energy (Infra-red, Ultraviolet)?
myocardia
Diamond Member
- Jun 21, 2003
- 9,291
- 30
- 91
All of that only applies to radio waves. A cpu doesn't use radio waves. Never have, never will. BTW, what will eventually limit the frequencies that cpu's will be able to attain is the size of the "process". Intel is at a 65nm process now, but they're saying that they don't believe they'll ever be able to go much smaller than 45nm, because then it will start being too weak.
Yes there is technology they are working on right now that will allow for much faster computing. There are differnt materials that will allow computer to run at insane speeds I've read about them before. Also they are working on reverse hyperthreading right now. If they are able to do that, then the sky is the limit. They could have an 8 core CPU all cores running at 3.0 ghz and with reverse hyperthreading they would be like having a single 24 ghz chip.
Anything electronic that turns on and off emmits electromagnetic noise. I can't remember the formula for i think it's a combination of speed and amplitude that determine it's frequency. I think he's talking rubbish, even if they do start to emmit microwaves you just need to put in some sheilding.
Moore's laws are not laws, they are observations.
Oh yeah, reverse hyperthreading is not all that great, in most roles it'd produce only an incremental improvement in performance.
Moore's laws are not laws, they are observations.
Oh yeah, reverse hyperthreading is not all that great, in most roles it'd produce only an incremental improvement in performance.
In a few short years from now all PC storage will be on SSD. I would love to see a NB chipset support a PCI Bus slot for a 150X 4Gb SD card. that would be a perfect location for a page data file (no heat, no cabling, low power draw, and lifetime warranty).
He's probably talking about the actual clock speed of the CPU being limited, rather than performance. In any case, I haven't heard of any such problem too. The fastest "clock" is already in the hundreds of GHz IIRC.
This seems more like a "Highly Technical" forum question, but here I go.
A few years ago Stanford University was hosting video files on all types of technical subjects. One of them was computers. Intel engineers gave a lecture somewhere about the barriers they face. Here is a highly butchered and marginally inaccurate summary of the video ( Its been a couple of years since ive seen it).
Quantum tunneling:
As you know electrons flow through a circuit and cross gates. These gates either allow electrons to pass or they dont. As microprocessor dies have shrank, so has the width of these gates. The die shrinks, and more transistors are added and gate width shrinking is just a natural step. Now. quantum physics says electrons can "tunnel" or randomly jump from one place to another over very short distances. Gate lengths are so short now that they do feel the effect of these short jumps. Voltage leaks and the circuit shorts itself out.
Engineers do have solutions for this, but the smaller we get ( and hence faster), the more its going to happen
Source of this info
Photo masking:
Forgive me here, but im paraphrasing a very complicated concept from memory. But the principle is the important thing. All circuits are etched onto the wafer by various means. Typically a very short wavelength laser. As the traces, ie. the width of the circuit the electrons flow down get smaller so does the laser width/wavelength. Once that wavelength hits a certain point, you cant garauntee the trace will work. Something about the heisenberg uncertainty principle and the traces needing to be smaller than the shortest wavelength laser we have.
So there are 2 obstacles that need addressing for current and future CPU's.
If any forum members remember those video lectures from 3+ years ago please ring in here, they were really interesting.
A few years ago Stanford University was hosting video files on all types of technical subjects. One of them was computers. Intel engineers gave a lecture somewhere about the barriers they face. Here is a highly butchered and marginally inaccurate summary of the video ( Its been a couple of years since ive seen it).
Quantum tunneling:
As you know electrons flow through a circuit and cross gates. These gates either allow electrons to pass or they dont. As microprocessor dies have shrank, so has the width of these gates. The die shrinks, and more transistors are added and gate width shrinking is just a natural step. Now. quantum physics says electrons can "tunnel" or randomly jump from one place to another over very short distances. Gate lengths are so short now that they do feel the effect of these short jumps. Voltage leaks and the circuit shorts itself out.
Engineers do have solutions for this, but the smaller we get ( and hence faster), the more its going to happen
Source of this info
Photo masking:
Forgive me here, but im paraphrasing a very complicated concept from memory. But the principle is the important thing. All circuits are etched onto the wafer by various means. Typically a very short wavelength laser. As the traces, ie. the width of the circuit the electrons flow down get smaller so does the laser width/wavelength. Once that wavelength hits a certain point, you cant garauntee the trace will work. Something about the heisenberg uncertainty principle and the traces needing to be smaller than the shortest wavelength laser we have.
So there are 2 obstacles that need addressing for current and future CPU's.
If any forum members remember those video lectures from 3+ years ago please ring in here, they were really interesting.
Originally posted by: Pabster
Most professors fit that category.
I would ask someone from Intel/AMD. They actually do something useful in the world.
Smaller gates do mean more leakage and I'd also imagine that there is a limitation to how fast the electrons can react to the switching field, and a limitation to how long it takes the switching field to get strong enough to pull the electrons. A photon computer would be able to achieve infinite speed/zero time, but then you're not dealing with anything physical anymore. The move to multiple cores does indicate that at least with this current MOSFET technology, the biggest percentage gains from gate shrinks are over.
i was reading some computer magazine about a month ago, im thinking it was computer science but im not sure, i was sick and in the waiting room of the hospital, anyways i was reading that the cores of processors are going to be diamonds and other gems or some crystalize material and we will be able to reach extreme speeds, with low amounts of power, and very little heat. I am actually gonna go try to find that magazine now lol
ooh edit, link clicky
meh just search diamond CPU on google(redundent?)
ooh edit, link clicky
meh just search diamond CPU on google(redundent?)
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 22K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
News NVIDIA and Intel to Develop AI Infrastructure and Personal Computing Products- Started by poke01
- Replies: 403
-
Facebook
Twitter