Snowy_River said:
Cool, thanks for the info! And good to see I wasn't too far off, it's been a while since I was into that material...
Do I pass, Professor?
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
IBM and 45nm Chip Development
- Thread starter MacRumors
- Start date
- Sort by reaction score
Cool, thanks for the info! And good to see I wasn't too far off, it's been a while since I was into that material...
Do I pass, Professor?
~Shard~ said:Cool, thanks for the info! And good to see I wasn't too far off, it's been a while since I was into that material...
Do I pass, Professor?
Oh, I think your response was close enough to merit a passing grade.
Tachyons
<offtopic>
General relativity tells us that no massful particle can be accelerated up to light speed and no massless particle that is already at light speed can travel faster than light speed. Light speed is a cosmological barrier. What enforces this limit? Why is the limit approximately 186,000 mps? Why are arbitrary speeds not allowed? We don't know, but physicists are continuing to search for a theory of everything (TOE) that may explain the limit, or it may not. If a theory of everything is based only on the structure and laws of "our" universe, it may not be sufficiently far reaching if it turns out that our universe is not the only universe. A number of strange theories (or ideas) have been proposed by leading cosmologists like Lee Smolin, Stephen Hawking, Mikio Kaku, Brian Greene, etc. Ideas like Quantum Cosmology, Cosmological Natural Selection, Superstring and its variants (M-Theory/Brane Theory), and others all contend that existence may be far grander than we ever imagined, with additional dimensions (10 spatial and 1 temporal according to Superstring), multiple branes (parallel worlds or universes), and big bangs taking place all the time (Quantum Cosmology, Cosmological Natural Selection). In any case, testability remains the biggest problem! If a theory cannot be tested, it is merely speculation. Science holds testability in extremely high regard, and for very good reasons.
Tachyons, then, are merely speculative, but the idea is interesting. It has to do with the speed of light barrier. If nothing on "this" side of the barrier can surpass the speed to light, then nothing from this side can cross the barrier. But what might be some properties of the other side, assuming there is one? Physicists speculate that time, for example, would be running backwards. And nothing on that side could move at or below the speed of light, but "particles" (i.e. tachyons) could, instead, travel at any speed faster than light, theoretically going up to infinite speed (unless, of course, there is a second speed bump somewhere way out there).
<offtopic>
General relativity tells us that no massful particle can be accelerated up to light speed and no massless particle that is already at light speed can travel faster than light speed. Light speed is a cosmological barrier. What enforces this limit? Why is the limit approximately 186,000 mps? Why are arbitrary speeds not allowed? We don't know, but physicists are continuing to search for a theory of everything (TOE) that may explain the limit, or it may not. If a theory of everything is based only on the structure and laws of "our" universe, it may not be sufficiently far reaching if it turns out that our universe is not the only universe. A number of strange theories (or ideas) have been proposed by leading cosmologists like Lee Smolin, Stephen Hawking, Mikio Kaku, Brian Greene, etc. Ideas like Quantum Cosmology, Cosmological Natural Selection, Superstring and its variants (M-Theory/Brane Theory), and others all contend that existence may be far grander than we ever imagined, with additional dimensions (10 spatial and 1 temporal according to Superstring), multiple branes (parallel worlds or universes), and big bangs taking place all the time (Quantum Cosmology, Cosmological Natural Selection). In any case, testability remains the biggest problem! If a theory cannot be tested, it is merely speculation. Science holds testability in extremely high regard, and for very good reasons.
Tachyons, then, are merely speculative, but the idea is interesting. It has to do with the speed of light barrier. If nothing on "this" side of the barrier can surpass the speed to light, then nothing from this side can cross the barrier. But what might be some properties of the other side, assuming there is one? Physicists speculate that time, for example, would be running backwards. And nothing on that side could move at or below the speed of light, but "particles" (i.e. tachyons) could, instead, travel at any speed faster than light, theoretically going up to infinite speed (unless, of course, there is a second speed bump somewhere way out there).
Snowy_River said:Oh, I think your response was close enough to merit a passing grade.
Thanks, you get an Apple next class... (pun intended)
RE: Tachyons
Yes, way off topic...
There is one loop hole in your explanation, though (and I glossed over it in my explanation, too). That is, while it is theoretically impossible for things that are traveling faster than light to interact with things that are traveling slower than light, it is, in fact, possible for both to interact with particles that are traveling at the speed of light - i.e. light! So, really, tachyons aren't merely speculative, as they can be tested for. However, all such tests have, to date, shown no results.
As for time moving backward, this is only one possible interpretation of the equations. Another interpretation is that there is another type of mass that is "imaginary", and all tachyons must be made of this type of mass. Now, as physicists don't really understand mass yet, this idea doesn't ruffle too many people's feathers.
The more common place for the idea of traveling back in time is actually the idea of antimatter. This is simply the idea that an anti-particle, say a positron (the antimatter equivalent of an electron), is simply an electron traveling backwards in time. This explanation actually fits a lot of aspects of quantum physics. In general, at the quantum level, looking at a reaction, you cannot tell which direction time travels in, assuming that particles that are traveling backward in time are anti-particles. When you expand out to the grand scheme of things, the only way you can tell which direction time travels in is by observing that the overall flow must be toward increased entropy (i.e. randomness).
Yes, way off topic...
ksz said:
There is one loop hole in your explanation, though (and I glossed over it in my explanation, too). That is, while it is theoretically impossible for things that are traveling faster than light to interact with things that are traveling slower than light, it is, in fact, possible for both to interact with particles that are traveling at the speed of light - i.e. light! So, really, tachyons aren't merely speculative, as they can be tested for. However, all such tests have, to date, shown no results.
As for time moving backward, this is only one possible interpretation of the equations. Another interpretation is that there is another type of mass that is "imaginary", and all tachyons must be made of this type of mass. Now, as physicists don't really understand mass yet, this idea doesn't ruffle too many people's feathers.
The more common place for the idea of traveling back in time is actually the idea of antimatter. This is simply the idea that an anti-particle, say a positron (the antimatter equivalent of an electron), is simply an electron traveling backwards in time. This explanation actually fits a lot of aspects of quantum physics. In general, at the quantum level, looking at a reaction, you cannot tell which direction time travels in, assuming that particles that are traveling backward in time are anti-particles. When you expand out to the grand scheme of things, the only way you can tell which direction time travels in is by observing that the overall flow must be toward increased entropy (i.e. randomness).
Actually I love this stuff. It makes my head hurt and my heart sing to think of the possibilities. You're right: we don't even understand mass. We don't really understand gravity either. And time is an even more bizarre concept. So are higher spatial dimensions.
Photons have no mass, but they still have momentum. And their path can be bent by gravity (or distortion in the spacetime fabric). Alan Guth's Inflation Theory suggests that mass may be explained as the resistance to acceleration (not resistance to constant velocity) imparted by a theoretical Higgs Field whose carrier particle is the Higgs Boson. Particles that interact with the Higgs Boson have mass, but those that don't, like the photon, have no mass. Photons, hence, experience no resistance to acceleration and accelerate to light speed instantaneously. Something else, however, prevents them from going any faster. What is that something?
Photons have no mass, but they still have momentum. And their path can be bent by gravity (or distortion in the spacetime fabric). Alan Guth's Inflation Theory suggests that mass may be explained as the resistance to acceleration (not resistance to constant velocity) imparted by a theoretical Higgs Field whose carrier particle is the Higgs Boson. Particles that interact with the Higgs Boson have mass, but those that don't, like the photon, have no mass. Photons, hence, experience no resistance to acceleration and accelerate to light speed instantaneously. Something else, however, prevents them from going any faster. What is that something?
i love listening to this stuff, i only wish i had something to contribute that wasnt already said.
Whatever they need to do to increase the speed and decrease the cost. Be competitive IBM. Actually be more than competitive.
ksz said:
Well, in fact, according to Quantum Field Theory, photos, acting as the mediator particles for the electromagnetic field, do have mass, and do travel both faster and slower than the speed of light. Of course, the farther you get from the EM field source, the lower the likelihood that you'll find a photo that has mass and is traveling at anything other than the speed of light, to the point that once you're a just a few atom diameters away from the source, the probability is diminishingly small. So things are even stranger still.
Why 45nm chips
All i can say is that transistors behave weird. Sometimes they don't work at higher dimensions and the same transistor (build with the same productionproces) works well at smaller dimensions like 45nm.
At my work (Amis) we work in a very old FAB (cleanroom) and we were told that we never could build transistors on silicium smaller than 1,2 micron (size from the gate) and we succesfully build chips with a gate from 0,7 micron!
We have a hightech FAB also and they could not build the same chip (0,7 micron-proces) with state of the art hightech machines.
It's possible that 45nm is realistic and that transistors failed in bigger dimensons , but will work just fine in smaller dimensions.
Building chips is very complex...And you can't believe how many factors there are that could have impact on yield.Not only engeneering , machines but also human and a much more.(Even luck)
Let's hope this is true , i can't wait to work with 45nm chip in a Mac
Greetings ,
Dokter_Mac
PS:Sorry for my poor english , we are not all Americans or English men
I can explain it much better in dutch.
All i can say is that transistors behave weird. Sometimes they don't work at higher dimensions and the same transistor (build with the same productionproces) works well at smaller dimensions like 45nm.
At my work (Amis) we work in a very old FAB (cleanroom) and we were told that we never could build transistors on silicium smaller than 1,2 micron (size from the gate) and we succesfully build chips with a gate from 0,7 micron!
We have a hightech FAB also and they could not build the same chip (0,7 micron-proces) with state of the art hightech machines.
It's possible that 45nm is realistic and that transistors failed in bigger dimensons , but will work just fine in smaller dimensions.
Building chips is very complex...And you can't believe how many factors there are that could have impact on yield.Not only engeneering , machines but also human and a much more.(Even luck)
Let's hope this is true , i can't wait to work with 45nm chip in a Mac
Greetings ,
Dokter_Mac
PS:Sorry for my poor english , we are not all Americans or English men
I can explain it much better in dutch.
You're making my head hurt again.Snowy_River said:
There are massless particles, aren't there? Force-exchange particles whether photons for mediating the EM force or gluons for mediating the strong nuclear force (correct me if I'm wrong) are usually listed as having zero mass. If I recall, the equation for momentum of a massful particle is:
p = mv (mass times velocity)
But for photons, momentum is:
p = E/c (energy of the photon divided by speed of light)
Wikipedia states that "massless objects such as photons also carry momentum..."
See: http://en.wikipedia.org/wiki/Momentum
So are photons massless? It is not impossible for a massful object to attain light speed?
ksz said:
Partly due to E=mc^2 though, mass increases as you approach the speed of light. If you look at it one way, if you broke the speed of light, you would have infinite mass!
Also, time dilation occurs as near-light speeds. In other words, time does in fact slow down as you approach the speed of light. Lots of fun discussion here!
ksz said:You're making my head hurt again.
There are massless particles, aren't there? Force-exchange particles whether photons for mediating the EM force or gluons for mediating the strong nuclear force (correct me if I'm wrong) are usually listed as having zero mass. If I recall, the equation for momentum of a massful particle is:
p = mv (mass times velocity)
But for photons, momentum is:
p = E/c (energy of the photon divided by speed of light)
Wikipedia states that "massless objects such as photons also carry momentum..."
See: http://en.wikipedia.org/wiki/Momentum
So are photons massless? It is not impossible for a massful object to attain light speed?
Pardon me, I should have spoken more clearly. Photons can have mass, both real and imaginary. When a photon is acting as a mediator particle for the EM field, it can travel at just about any speed. These are more commonly called virtual photons. The probability of it traveling at a speed other than the speed of light diminishes as it travels farther, though. When a photon is traveling at a speed other than the speed of light, that is when it has mass - real mass if it is traveling slower than the speed of light or imaginary mass if it is traveling faster than the speed of light. Now, this starts to sound like photons really act like the tachyons that we were discussing earlier, and therefore should exist. Well, yes and no. They are still just photons, but there are some processes that cannot be explained unless we assume that they are traveling faster (and in some cases much, much faster) than the speed of light, as least for short distances (and here I mean VERY short distances). This even gets into some of the quantum effects that make electronics at these very small process sizes so difficult. A 45nm process is drawing lines that are only about 400 atoms across. So effects that show up at distances around the diameter of the atom aren't necessarily being effectively canceled out when you get fewer and fewer atom diameters to work with. (Hey, how about that for bringing it back on topic!
)But, of course, your equations are perfectly correct for photons and massful particles that aren't acting within this particular area of quantum probability. However, there are plenty of examples of force mediator particles that aren't massless, so don't go down the wrong path thinking that all mediator particles must be massless. Just a few examples of massive mediator particles include the W, the Z, the gluon and the pion.
~Shard~ said:Partly due to E=mc^2 though, mass increases as you approach the speed of light. If you look at it one way, if you broke the speed of light, you would have infinite mass!
A common misrepresentation. It's not a matter of your mass changing, at least not the way you would think of it. Your mass, as we commonly think of it, is simply your rest mass, which is the 'm' in that lovely formula that you quoted. Now, the E in that formula is really your rest Energy. It also doesn't change. Now, what does change is your total energy. This includes kinetic energy (so it's no wonder that your energy would increase as you go faster).
The catch is that most of us are familiar with the Newtonian form of kinetic energy KE=1/2 * mv^2. However, this only applies if your not moving at some significant fraction of the speed of light. Once you get going really fast, then it's far easier just to look at total energy, which is
E=(mc^2)/SQRT(1-(v/c)^2)
So, as your speed approaches the speed of light, v/c approaches 1, and energy approaches a divide by zero error, or infinity. The other way to say this is that the amount of energy needed to get to 90% of the speed of light would be needed again to get from 90% to 97.5% of the speed of light, and that same amount again would be needed to get to 98.9% from 97.5% of the speed of light, and again from 98.9% to 99.4%. The closer you get the smaller the step you make for the same amount of energy.
How does this tie into time dilation? Well, from your point of view, you're constantly moving faster at a constant acceleration. If that were the case, then you'd get to the speed of light a some finite time. But, due to time dilation, your clock is slowing down, so even though you think you'll get there, you never will. It's said that a photon's clock is stopped, i.e. no time passes for a photon.
Anyway, I'll stop before I make ksz's head explode.
Snowy_River said:
Thanks for the feedback Snowy_River, that's great! As I said before, it's been a while since I touched this stuff, so it's great to have a refresher!
I'll just finish by saying that although Physics is truly fascinating, in the end, it is only a model. And it is a damn good model at that, but still, it is only a representation of how we see this universe and how we understand things to behave.
I think mankind will truly advance when we can go past physics. Without physical limitations, physical amounts of time and physical distances to travel, the possibilities are endless in a world not limited solely by physics, (which in my mind is how the universe truly is), a lot of these points and paradoxes become moot. I personally believe there is so much more out there, and physics is only half of it. For example, electricity, which is sort of the lowest level of understanding we have of things (i.e. down to the sub-atomic level), is just a by-product of a higher form of energy we do not even know about yet. Quantum physics fascinates me as well, with its notions of fixed probabilities, tunneling, and good old Schroedinger and his feline.
But, this is all best left for another thread. Perhaps one of these days Ill start a thread dedicated solely to this type of discussion, because I could go on for days.
poundsmack said:
If I start a thread on this topic (whenever I have some spare time that is!) I will be sure to PM you with its link - and Snowy_River as well.
~Shard~ said:If I start a thread on this topic (whenever I have some spare time that is!) I will be sure to PM you with its link - and Snowy_River as well.
sounds good. I will be taking a quantum physics class this comming semester, i find all this stuff so facinating. but everything i know has been said i am afriad.....must....learn.....more
Thanks for that explanation! These phenomena are still quite mind boggling and I think my head already exploded from the 14 hour flight to Taiwan yesterday. (Spending a few weeks at a 300mm fab that's ramping up its 65nm process.)Snowy_River said:Anyway, I'll stop before I make ksz's head explode.
A little more on topic. Have either of you heard anything about the advances in quantum computing? I remember seeing an interview with Michio Kaku where he was talking about the theoretical barriers of our current technologies and that scientists have even been able to do simple math using quantum particles.Snowy_River said:Pardon me, I should have spoken more clearly. Photons can have mass, both real and imaginary...Anyway, I'll stop before I make ksz's head explode.
Also I've heard a few things about bio computing like RNA processing and memory experiments at SDSU... Do you guys know anything about these? I'd love to hear more.
By the way I love these discussions, so if a thread is started on this topic let me know as well.
A good source for information about quantum computing is Scientific American. A number of recent issues have discussed everything from quantum encryption to nanotech. I attended a seminar last year in Sapporo (on Hokkaido island) given by a dean from the ETH, the Swiss Federal Institute of Technology in Zurich (alma mater of Albert Einstein). The talk began with conventional process technology up through the 22nm node then turned to a glimpse at the quantum future. This was quite a bit over my head, and judging by the glazed look in the audience, it went over everyone else's head too. Nevertheless, I'll ask the dean for a copy of that material.
It means they've hit some really hard problems that they have not hit before.graphyte said:What do you mean when you say the industry has hit a wall? Do you mean we have reached that point where you can't go further in chip manufacturing?
It is well known that you need smaller processes in order to implement increasing complexity at higher speeds - otherwise the chips get too large to be reliably manufactured and speed-of-light issues prevent signals from propagating through all the required circuits in time.
Historically, there have been problems with moving to smaller processes, but these problems were solved relatively quickly by the big chipmakers. When they moved to 90nm, the problems they hit were not solved that easily. It's forced all of the major chipmakers to do a lot more work than expected to get reliable chips produced. Even today, I don't think they've solved all of the problems.
This is probably the biggest reason why we haven't seen 3GHz G5 chips yet.
This doesn't mean the speed is impossible, but it does mean they haven't yet figured out how to reach that speed in a way that will allow production in large quantities at a reasonable price.
I don't think so. Apple's been hurt really badly by IBM's inability to produce 3GHz G5 chips. I'm sure Apple will release 3GHz systems as soon as possible once IBM is able to make the chips in sufficient quantities. I don't think they'll wait more than a short time for an appropriate expo to announce it at.graphyte said:a) Apple has got 3 Ghz G5 chips, and they are waiting for WWDC... (or all-new PowerBooks G5 nest Tuesday ?)
Probably not impossible, but difficult and maybe too expensive.graphyte said:b) It is imposible to make them... and we'll have to wait till Apple decides to implement new processors...
New designs are being developed all the time and it is likely that Apple is doing the necessary research to decide when/if it is appropriate to switch to one.
Of course, we, the customers, will be the last to know when/if they actually make such a decision.
One possibly useful book for people who can't get enough of this kind of physics is QED by Richard Feynman. It attempts to explain Quantum Electrodynamics in a way that mere mortals can understand.Snowy_River said:
It does a good job, to the extent that this subject can be made intelligeble by mere mortals in the first place.
It is a buit out of date (published in 1988), but I think it's probably a good introduction to the subject, nevertheless. And (IMO) if you can't understand this explanation, you're not likely to be able to understand the stuff that people are publishing today. Feynman was one of those few physicists that had a knack for making subjects understandable.
At least using current theories.Snowy_River said:
Many people look at the Einstein equations and assume that nothing can be accelerated beyond light speed. But that's not necessarily true. It just means that these equations can not be used to describe what happens if you do.
Newton didn't figure out the relativistic terms of the acceleration/energy equations because those terms are insignificant at slow speeds. It is possible that there are additional terms that Einstein (and modern physics) don't know about yet because we have not yet been able to accelerate any measuring equipment to the necessary speeds where those terms would become significant.
Maybe there are no other terms - in which case, lightspeed would be a very real barrier. (Although the effect of FTL travel might still be possible using concepts that are today mostly in the realm of science fiction.) But it is possible that there are other terms that are completely insignificant for the problem domains we're currently working in. If so, then exceeding lightspeed may actually be possible. But current theories (as far as I know) have not seen any need for these terms to exist, so this concept remains purely hypothetical.
(To summarize all this, it's important to not confuse reality with our current/best theories for explaining reality. Theories do change occasionally, as new discoveries are made, and those theories shape our concept of what is and is not possible.)
As an Amazon Associate, MacRumors earns a commission from qualifying purchases made through links in this post.
shamino said:
I couldn't agree more. That's an excellent book. I love Feynman's work.
shamino said:
Well, yes, but...
At this point we've successfully accelerated objects up to 99.9998% of the speed of light and they have consistently obeyed the Laws of Special Relativity. These are considered to be among the most tested equations in all of science. While there may be other terms, it's likely that these other terms won't come into play before we start dealing with particles on the scale of a Planc mass. (That's about 150,000,000,000,000,000,000 times the mass of the proton.) So, unfortunately, at this point we don't have any sneak peeks into what might allow FTL travel or communication, if it's at all possible within the physical laws that govern our universe.
As an Amazon Associate, MacRumors earns a commission from qualifying purchases made through links in this post.
I read QED recently and came away unsatisfied. While Feynman's lectures are very well written and thought out, and QED is a book rendered from a handful of short lectures, it doesn't go far enough. It describes and illustrates the "many paths" idea very well, but QED focuses on a tiny aspect of reality. For the layperson interested in a much broader introduction to cosmology, here are some recommendations:shamino said:
1. The Elegant Universe, Brian Greene.
2. The Fabric of the Cosmos, Brian Greene (somewhat recent release).
3. Parallel Worlds, Michio Kaku (recent release).
4. Hyperspace, Michio Kaku (so good, you have to read it twice).
5. Black Holes and Time Warps, Kip Thorne (more technical than the others).
As an Amazon Associate, MacRumors earns a commission from qualifying purchases made through links in this post.
ksz said:
Yes, but what it focuses on is what it says it focuses on: Quantum Electro Dynamics (QED). It's not meant to be a book about a broader introduction to cosmology. It's meant to be an easy to read explanation of the interactions of some of the smallest particles with each other. That's all, and it does an excellent job at what it sets out to do.