I'm confused about Intels 14nm process lead

[imported_ats]

On a semi-related topic; Im confused about transistors as a whole.

Like I get what a MOSFET transistor looks like, I know FinFET's, planar, SOI, etc. etc. al that stuff. I get how a single transistor works, but I dont understand how several transistors work together and stuff outside of that. Anyone have any sources?

I already tried reading tons of Anand stuff on it and watched videos, etc. but I'm clearly missing a lot. Anyone have any good resources to learn of this stuff?

a transistor is simply a switch (we'll stay at the logical level and not get into the analog realm). Logic gates are formed by using transistors to either pull up to Vdd or pull down to Vss. Vdd tends to be the voltage supply rail. Vss tends to be the ground rail.

CMOS is a complementary logic set. In a complementary logic set, you are pull ups and pull downs complimenting each other. Pull ups use PMOS gates which have inverted functionality (1 input is off, 0 input is on). Pull downs use NMOS gates which have direct functionality (1 is on, 0 is off). So as an example, we'll put together a simple NOT AND (aka NAND) gate.

In a NAND gate you you have 2 PMOS in parallel at the top half of the gate connected to VDD and 2 NMOS at the bottom in serial connected to VSS.

VDD->PMOS_A||PMOS_B->Output->NMOS_A->NMOS_B->VSS

So lets put together the logic chart:

In_A  In_B  P_A  P_B  N_A  N_B  Ouput
0     0     On   On   Off  Off  1
0     1     On   Off  Off  On   1
1     0     Off  On   On   Off  1
1     1     Off  Off  On   On   0

Now one thing to note is that either the PMOS side or the NMOS side can function on its own if we replace the other side with a simple resistor.

So that's the $.01 explanation, let me know if that helps.

 

[elemein]

a transistor is simply a switch (we'll stay at the logical level and not get into the analog realm). Logic gates are formed by using transistors to either pull up to Vdd or pull down to Vss. Vdd tends to be the voltage supply rail. Vss tends to be the ground rail.

CMOS is a complementary logic set. In a complementary logic set, you are pull ups and pull downs complimenting each other. Pull ups use PMOS gates which have inverted functionality (1 input is off, 0 input is on). Pull downs use NMOS gates which have direct functionality (1 is on, 0 is off). So as an example, we'll put together a simple NOT AND (aka NAND) gate.

In a NAND gate you you have 2 PMOS in parallel at the top half of the gate connected to VDD and 2 NMOS at the bottom in serial connected to VSS.

VDD->PMOS_A||PMOS_B->Output->NMOS_A->NMOS_B->VSS

So lets put together the logic chart:

In_A  In_B  P_A  P_B  N_A  N_B  Ouput
0     0     On   On   Off  Off  1
0     1     On   Off  Off  On   1
1     0     Off  On   On   Off  1
1     1     Off  Off  On   On   0

Now one thing to note is that either the PMOS side or the NMOS side can function on its own if we replace the other side with a simple resistor.

So that's the $.01 explanation, let me know if that helps.

Well that stuff I get. Boolean algebra is stuff I've done in a couple of my computer math courses

But... How does this:

http://orion.pt.tu-clausthal.de/atp/projects/images/mosfet2.png

Become this:

http://cdn2.wccftech.com/wp-content/uploads/2014/08/Intel-14nm-Interconnect-Path.jpg

And this just makes it like 100x more confusing:

http://download.intel.com/newsroom/kits/22nm/gallery/images/Intel-22nm_Transistor.jpg

And then theres even more stuff that makes it even more confusing, but I'll leave it at that... All those pics dont seem connected in any way shape or form... And yet they are?

 

[witeken]

On a semi-related topic; Im confused about transistors as a whole.

Like I get what a MOSFET transistor looks like, I know FinFET's, planar, SOI, etc. etc. al that stuff. I get how a single transistor works, but I dont understand how several transistors work together and stuff outside of that. Anyone have any sources?

I already tried reading tons of Anand stuff on it and watched videos, etc. but I'm clearly missing a lot. Anyone have any good resources to learn of this stuff?

Seen this one? http://forums.anandtech.com/showthread.php?t=2420424

This one takes it in the real world: https://www.youtube.com/watch?v=NGFhc8R_uO4

Good analogy: https://www.youtube.com/watch?v=lNuPy-r1GuQ

I honestly don't really know that much about what you're asking for. You can also search on IEEE Spectrum.

 

[Idontcare]

But... How does this:

http://orion.pt.tu-clausthal.de/atp/projects/images/mosfet2.png

Become this:

http://cdn2.wccftech.com/wp-content/uploads/2014/08/Intel-14nm-Interconnect-Path.jpg

And this just makes it like 100x more confusing:

http://download.intel.com/newsroom/kits/22nm/gallery/images/Intel-22nm_Transistor.jpg

I wondered that too while I was in college.

And then I learned the answer, which is something that I suppose could be boiled down into something succinct enough as a forum post (but doubtful, truly), but then I had the pleasure of going to work in the industry and not only getting to learn the answer first-hand but to also be part of the reason it exists node after node.

Here is my advice to an individual such as yourself who has just the right kind of curiosity - do what you can to seek out someone local, within driving distance, that you can meet face to face for coffee and perhaps a lunch or two. Someone at a nearby college or uni.

I loved teaching at a uni while I worked at TI because the students were full of curiosity and not quite jaded with the workplace cynicism that is embodied in Dilbert cartoons. And I had the pleasure of meeting a few students who had genuine, and relevant, curiosity the likes of which you exhibit.

You crave the kind of understanding and information that cannot be effectively passed along via 2D text and images, it can be attempted but it cannot supplant the experience of doing versus passively reading.

So I say to you - pursue that. Indulge that. It will take you places that others can't even begin to imagine exists.

Find someone you can interface with in the real world who is plugged in to this industry and let life take it from there.

 

[Abwx]

E.g. A SINGLE electron is traveling down a wire. So there is ONE, and only ONE electron, Yes ?
NO, not at the quantum level, because single electrons, can actually move, as if there was a second electron, near it.

Electrons dont move in the cables, at least not as assumed erroneously by the general public, and it is happy since the electrical current speed is about 200 000 Km/s, if electrons did actualy move at this speed cables would be torned apart by the kinetic energy of such formidable bullets...

Actualy electrons speed is about 0.02 mm/s, with an alternative current, like in processors switching circuits, they are about immobile, for instance at 1GHz they would move in a direction during 0.5 ns and get to the opposite one during the next 0.5 ns, that is at a 0.02 mm/s speed, what actualy move at 200 000 km/s is the electric potential, when you switch on a current it act the same way that when you push a end of a stick, whatever the length of the stick the other side will move at the same speed as the pushed end, the potential that is applied to an end of a cable is transmitted electrons by electrons up to the other end of the cable at speed comparable to light.

 

[SOFTengCOMPelec]

What I was trying to refer to, was stuff like this ..

Italian physicists Pier Giorgio Merli, Gian Franco Missiroli, and Giulio Pozzi repeated the experiment using single electrons, showing that each electron interferes with itself as predicted by quantum theory.

http://en.wikipedia.org/wiki/Double-slit_experiment

But I probably should NOT have mentioned the wire. So I accept that I was somewhat wrong, sorry.

 

[Tuna-Fish]

On a semi-related topic; Im confused about transistors as a whole.

Like I get what a MOSFET transistor looks like, I know FinFET's, planar, SOI, etc. etc. al that stuff. I get how a single transistor works, but I dont understand how several transistors work together and stuff outside of that. Anyone have any sources?

I already tried reading tons of Anand stuff on it and watched videos, etc. but I'm clearly missing a lot. Anyone have any good resources to learn of this stuff?

Paul DeMone's The Stuff Dreams Are Made Of is a reasonable primer for cmos logic.

 

[carop]

AFAIK, 2D routing is "easier" on the design teams, and given that TSMC is a foundry that needs to service design teams of all budgets, 2D routing is the way to go at the expense of yields.

Design teams can learn how to work with restricted design rules. The real issue for the semiconductor industry is the high cost of 1D layout by SAxP plus cut mask as well as the cost of 15% larger dies that come with 1D layouts.

The memory industry have been using 1D layouts for the past 6 generations now. Long before Intel adopted gridded design rules at N45.

Intel claims that 1D layouts are density neutral or have better density than 2D layouts, but the semiconductor industry tend to accept that 1D layouts have a 15% larger die cost.

 

[Idontcare]

What I was trying to refer to, was stuff like this ..

http://en.wikipedia.org/wiki/Double-slit_experiment

But I probably should NOT have mentioned the wire. So I accept that I was somewhat wrong, sorry.

Not sure how many people here have had the opportunity or exposure to a grad-level quant course...but quantum mechanics are really eye opening as well as "common sense" challenging.

There are a few popular memes, thanks to a certain alive/dead cat, but once you get past the whimsy of it you do find yourself immersed in an otherwise philosophically challenging environment.

I spent five long years grappling with the stuff. Learned a lot, struggled with a lot, felt overwhelmed with a lot.

Double-slit is great but it has one problem - it doesn't account for the fact that the slits themselves are quantum mechanical systems which are perturbed by, and thus interact with, the electrons that are passing through the slit.

Even worse (to our biologically simplistic minds) is the fact that electrons (and all fermions) are identical particles...meaning we actually have no idea where the electron came from when we "observe" it in a double-slit experiment.

We know we sent one in...and it interacted with a bunch of other electrons that form the evanescent wave at the surfaces of the slits, but we have no idea which electron in that total system equation contributed to the electron which our detectors indicate they have detected.

 

[SOFTengCOMPelec]

Not sure how many people here have had the opportunity or exposure to a grad-level quant course...but quantum mechanics are really eye opening as well as "common sense" challenging.

There are a few popular memes, thanks to a certain alive/dead cat, but once you get past the whimsy of it you do find yourself immersed in an otherwise philosophically challenging environment.

I spent five long years grappling with the stuff. Learned a lot, struggled with a lot, felt overwhelmed with a lot.

Double-slit is great but it has one problem - it doesn't account for the fact that the slits themselves are quantum mechanical systems which are perturbed by, and thus interact with, the electrons that are passing through the slit.

Even worse (to our biologically simplistic minds) is the fact that electrons (and all fermions) are identical particles...meaning we actually have no idea where the electron came from when we "observe" it in a double-slit experiment.

We know we sent one in...and it interacted with a bunch of other electrons that form the evanescent wave at the surfaces of the slits, but we have no idea which electron in that total system equation contributed to the electron which our detectors indicate they have detected.

The problem is (probably) going to be, that people like yourself (Process/chip designers), will sooner or later, probably have to deal with chips, whose feature size is so small, that it is operating at the quantum level. Throwing tons of our existing (Newtonian etc) Physics rule books, out of the window. And then bringing in quantum Physics, which is partly, more like art/religion/philosophy/randomness/non-scientific, than conventional Physics.

I've at least partly, read book(s) on it, as I find it amazing. But I expect the other 99.99999% of the population, to find it boring and/or way too difficult, to comprehend.
I can only partially understand it, but find it fascinating, anyway.

tl;dr When/if we primarily build chips, using quantum physics, we may have to throw out 12 monthly new chip release cycles (Intel), out of the window (especially when the feature size is dramatically reduced, every other generation, tick tock).

EDIT: Also, that is a very good/interesting point, about the slits themselves, causing a quantum effect, and potentially changing/complicating the results. I've never thought of it like that.

 

[witeken]

Well that stuff I get. Boolean algebra is stuff I've done in a couple of my computer math courses

But... How does this:

http://orion.pt.tu-clausthal.de/atp/projects/images/mosfet2.png

Become this:

http://cdn2.wccftech.com/wp-content/uploads/2014/08/Intel-14nm-Interconnect-Path.jpg

And this just makes it like 100x more confusing:

http://download.intel.com/newsroom/kits/22nm/gallery/images/Intel-22nm_Transistor.jpg

And then theres even more stuff that makes it even more confusing, but I'll leave it at that... All those pics dont seem connected in any way shape or form... And yet they are?

So the first image is obviously a FinFET schematic. The second is the interconnect stack. If you have a bunch of transistors, you can't do much with them if they aren't connected, so in that interconnect stack you can see 13 layers of Cu. The upper layers are used for signals that travel a long distance, while the first one connects with the transistor.

The third image is a photo of real life 22nm 3D transistors. I assume that one transistor consists of multiple fins: if you're working with FinFET, you can't change the fin length or width, thus if you want more drive current, you'll have to use multiple fins (it's quantized: https://www.google.be/search?q=finfet+quantized&oq=finfet+quantized). Not sure if that helps.

 

[witeken]

I loved teaching at a uni

Interesting, didn't know that. Thought you simply working in that industry.

I wondered that too while I was in college.

:s Am I simplifying too much?

 

[witeken]

And then bringing in quantum Physics, which is partly, more like art/religion/philosophy/randomness/non-scientific, than conventional Physics.

Quantum mechanics has the best tested theory in all of physics: QED.

 

[SOFTengCOMPelec]

Quantum mechanics has the best tested theory in all of physics: QED.

I'm not necessarily disagreeing with you here.

I probably did not explain what I meant, well enough.

I was trying to describe, the following, and the many other aspects of it, which are "different"..less/not-deterministic.

http://en.wikipedia.org/wiki/Quantum_mechanics

Philosophical implications[edit]
Main article: Interpretations of quantum mechanics
Since its inception, the many counter-intuitive aspects and results of quantum mechanics have provoked strong philosophical debates and many interpretations. Even fundamental issues, such as Max Born's basic rules concerning probability amplitudes and probability distributions, took decades to be appreciated by society and many leading scientists. Richard Feynman once said, "I think I can safely say that nobody understands quantum mechanics."[61] According to Steven Weinberg, "There is now in my opinion no entirely satisfactory interpretation of quantum mechanics."[62]
The Copenhagen interpretation - due largely to the Danish theoretical physicist Niels Bohr - remains the quantum mechanical formalism that is currently most widely accepted amongst physicists, some 75 years after its enunciation. According to this interpretation, the probabilistic nature of quantum mechanics is not a temporary feature which will eventually be replaced by a deterministic theory, but instead must be considered a final renunciation of the classical idea of "causality." It is also believed therein that any well-defined application of the quantum mechanical formalism must always make reference to the experimental arrangement, due to the conjugate nature of evidence obtained under different experimental situations.
Albert Einstein, himself one of the founders of quantum theory, disliked this loss of determinism in measurement. Einstein held that there should be a local hidden variable theory underlying quantum mechanics and, consequently, that the present theory was incomplete. He produced a series of objections to quantum theory, the most famous of which has become known as the Einstein–Podolsky–Rosen paradox. John Bell showed that this "EPR" paradox led to experimentally testable differences between quantum mechanics and local realistic theories. Experiments have been performed confirming the accuracy of quantum mechanics, thereby demonstrating that the physical world cannot be described by any local realistic theory.[63] The Bohr-Einstein debates provide a vibrant critique of the Copenhagen Interpretation from an epistemological point of view.
The Everett many-worlds interpretation, formulated in 1956, holds that all the possibilities described by quantum theory simultaneously occur in a multiverse composed of mostly independent parallel universes.[64] This is not accomplished by introducing some "new axiom" to quantum mechanics, but on the contrary, by removing the axiom of the collapse of the wave packet. All of the possible consistent states of the measured system and the measuring apparatus (including the observer) are present in a real physical - not just formally mathematical, as in other interpretations - quantum superposition. Such a superposition of consistent state combinations of different systems is called an entangled state. While the multiverse is deterministic, we perceive non-deterministic behavior governed by probabilities, because we can only observe the universe (i.e., the consistent state contribution to the aforementioned superposition) that we, as observers, inhabit. Everett's interpretation is perfectly consistent with John Bell's experiments and makes them intuitively understandable. However, according to the theory of quantum decoherence, these "parallel universes" will never be accessible to us. The inaccessibility can be understood as follows: once a measurement is done, the measured system becomes entangled with both the physicist who measured it and a huge number of other particles, some of which are photons flying away at the speed of light towards the other end of the universe. In order to prove that the wave function did not collapse, one would have to bring all these particles back and measure them again, together with the system that was originally measured. Not only is this completely impractical, but even if one could theoretically do this, it would have to destroy any evidence that the original measurement took place (including the physicist's memory). In light of these Bell tests, Cramer (1986) formulated his transactional interpretation.[65] Relational quantum mechanics appeared in the late 1990s as the modern derivative of the Copenhagen Interpretation.

 

[witeken]

I think philosophy of QM is quite meaningless. Science can only describe (mathematically) how the universe works. We are not omniscient. Any insight in "why" questions at the most fundamental level we gain is just bonus points. We can't look into "God"'s mind, so to speak.

What we can do however, is for example measure the energy of the universe. If it's zero, we get the (imo quite unnecessary) confirmation that the existence of the universe doesn't violate any laws of nature. If that's the case, we should adapt our notions of nothing, not the other way around.

Just some thought, I've only read your bolded text.

 

[SOFTengCOMPelec]

I think philosophy of QM is quite meaningless. Science can only describe (mathematically) how the universe works. We are not omniscient. Any insight in "why" questions at the most fundamental level we gain is just bonus points. We can't look into "God"'s mind, so to speak.

What we can do however, is for example measure the energy of the universe. If it's zero, we get the (imo quite unnecessary) confirmation that the existence of the universe doesn't violate any laws of nature. If that's the case, we should adapt our notions of nothing, not the other way around.

Just some thought, I've only read your bolded text.

It probably sounds rather weird to me (Quantum Mechanics), as I don't know that much about it. Perhaps if I did, I would consider it less/not weird.

Googling similar terms to "philosophy of QM", seems to indicate that peoples opinions, vary, as to what extent (if any), "philosophy of QM" applies to Quantum Mechanics.

Attempting to go back on topic:
If we do have future, Quantum mechanic feature sized integrated circuits, then reading about them, will probably be like reading a science fiction novel.

E.g. Theories involving multiple Universes, Cats that are both alive and dead at the same time and other weirdness (to me, at least).

 

[witeken]

It probably sounds rather weird to me (Quantum Mechanics), as I don't know that much about it. Perhaps if I did, I would consider it less/not weird.

Please take my comments with an appropriate grain of salt, since I only know a few things about QM through Wikipedia, Youtube, etc. It is indeed quite weird: we humans are quite used to the fact that you can know pretty much everything about a macroscopic system, but in QM, there's no way you can know what a particle is doing, etc. This probabilistic nature has lead people to the conclusion of parallel universes: https://www.youtube.com/watch?v=ZacggH9wB7Y. I'm not so sure about this (but then again, I'm not a physicist), and I'd argue that there's no basis for such a hypothesis. You can't even falsify it. So I'd say you just have to accept what the equations tell us.

Attempting to go back on topic:
If we do have future, Quantum mechanic feature sized integrated circuits, then reading about them, will probably be like reading a science fiction novel.

E.g. Theories involving multiple Universes, Cats that are both alive and dead at the same time and other weirdness (to me, at least).

Quantum mechanics is the thing that has stopped Dennard scaling. Quantum computers are way above my knowledge, but the QT transistor is quite interesting, imo. http://spectrum.ieee.org/semiconductors/devices/the-tunneling-transistor

 

[III-V]

Attempting to go back on topic:
If we do have future, Quantum mechanic feature sized integrated circuits, then reading about them, will probably be like reading a science fiction novel.

Tunnel FETs make use of quantum tunneling. There's also quantum dots... we're at the very beginning of commercializing quantum mechanics.

 

[SOFTengCOMPelec]

Tunnel FETs make use of quantum tunneling. There's also quantum dots... we're at the very beginning of commercializing quantum mechanics.

Yes, also tunnel diodes "negative resistance", and the fact that can be used, to make extremely high frequency amplifiers (if I remember, correctly), are interesting devices.

Strictly speaking, electronics devices, are probably working at the Quantum level, anyway. It is just that we are only gradually (probably not yet, but I am not sure), being able to control the creation of devices, at such a small level.

 

[GreenChile]

So the first image is obviously a FinFET schematic. The second is the interconnect stack. If you have a bunch of transistors, you can't do much with them if they aren't connected, so in that interconnect stack you can see 13 layers of Cu. The upper layers are used for signals that travel a long distance, while the first one connects with the transistor.

The third image is a photo of real life 22nm 3D transistors. I assume that one transistor consists of multiple fins: if you're working with FinFET, you can't change the fin length or width, thus if you want more drive current, you'll have to use multiple fins (it's quantized: https://www.google.be/search?q=finfet+quantized&oq=finfet+quantized). Not sure if that helps.

That first schematic is a planar transistor. You are correct about the other two.

 

[SOFTengCOMPelec]

Please take my comments with an appropriate grain of salt, since I only know a few things about QM through Wikipedia, Youtube, etc. It is indeed quite weird: we humans are quite used to the fact that you can know pretty much everything about a macroscopic system, but in QM, there's no way you can know what a particle is doing, etc. This probabilistic nature has lead people to the conclusion of parallel universes: https://www.youtube.com/watch?v=ZacggH9wB7Y. I'm not so sure about this (but then again, I'm not a physicist), and I'd argue that there's no basis for such a hypothesis. You can't even falsify it. So I'd say you just have to accept what the equations tell us.


Quantum mechanics is the thing that has stopped Dennard scaling. Quantum computers are way above my knowledge, but the QT transistor is quite interesting, imo. http://spectrum.ieee.org/semiconductors/devices/the-tunneling-transistor

Thanks. The last link, has made a very interesting and informative read for me. All I need now, is for an Intel i7-99,999,6771 from 2026 to accidentally Quantum Tunnel back in time, to my current space/time coordinates. Then I can appreciate its 99,999,999 cores at 9.99GHz, and its AVX9, 8192 bit instruction set.

Tunnel FETs make use of quantum tunneling. There's also quantum dots... we're at the very beginning of commercializing quantum mechanics.

Quotes like

Research into TFETs made from III-V materials

make me wonder why no one has made it their forum id.

 

[lopri]

The performance of a chip is (roughly speaking) determined by its architecture, while its power consumption is by the process node. Silvermont is about on par or so in terms of IPC, but from the little data I've got, I'd say that Silvermont crushes Krait in terms of PC.

I don't know, man.. I checked out a new Dell tablet at a Bestbuy last weekend. I had a high hope (relatively speaking), and the tablet was a pleasant surprise.. for the first 10 seconds or so. It felt incredibly light, almost as if it's hollow inside.

Playing with it a little while, it became clear the performance is not there. At least not at $400 it asks for. You tell me you will buy that tablet over iPads and Galaxys, or even Nexus tablets. I certainly wouldn't.

 

[III-V]

make me wonder why no one has made it their forum id.

TFET? They're anti-enthusiast, people around here wouldn't touch those with a ten-foot pole

They're not very good for high performance. Very good for low power, though.

 

[elemein]

Seen this one? http://forums.anandtech.com/showthread.php?t=2420424

This one takes it in the real world: https://www.youtube.com/watch?v=NGFhc8R_uO4

Good analogy: https://www.youtube.com/watch?v=lNuPy-r1GuQ

I honestly don't really know that much about what you're asking for. You can also search on IEEE Spectrum.

Oh wow I have a lot of replies. I'll do the best I can to reply

I havent read the first link, Ill take a look.

I have seen the second link, a couple times and it didn't help me as much as I wished it did :/

Ill take a peek at the third link later as well

I wondered that too while I was in college.

And then I learned the answer, which is something that I suppose could be boiled down into something succinct enough as a forum post (but doubtful, truly), but then I had the pleasure of going to work in the industry and not only getting to learn the answer first-hand but to also be part of the reason it exists node after node.

Here is my advice to an individual such as yourself who has just the right kind of curiosity - do what you can to seek out someone local, within driving distance, that you can meet face to face for coffee and perhaps a lunch or two. Someone at a nearby college or uni.

I loved teaching at a uni while I worked at TI because the students were full of curiosity and not quite jaded with the workplace cynicism that is embodied in Dilbert cartoons. And I had the pleasure of meeting a few students who had genuine, and relevant, curiosity the likes of which you exhibit.

You crave the kind of understanding and information that cannot be effectively passed along via 2D text and images, it can be attempted but it cannot supplant the experience of doing versus passively reading.

So I say to you - pursue that. Indulge that. It will take you places that others can't even begin to imagine exists.

Find someone you can interface with in the real world who is plugged in to this industry and let life take it from there.

This would be very good advice, besides the fact that I don't really have much idea of where to go to find someone whose directly linked to the industry who has precisely the information that I am looking for. Due to living in Southern Ontario, Canada, while we have lots of good educational facilities (my college is great obviously), I'm not sure of where I'd go to ask about IC design on a modern, practical level :/

Paul DeMone's The Stuff Dreams Are Made Of is a reasonable primer for cmos logic.

Ill take a look!

So the first image is obviously a FinFET schematic. The second is the interconnect stack. If you have a bunch of transistors, you can't do much with them if they aren't connected, so in that interconnect stack you can see 13 layers of Cu. The upper layers are used for signals that travel a long distance, while the first one connects with the transistor.

The third image is a photo of real life 22nm 3D transistors. I assume that one transistor consists of multiple fins: if you're working with FinFET, you can't change the fin length or width, thus if you want more drive current, you'll have to use multiple fins (it's quantized: https://www.google.be/search?q=finfet+quantized&oq=finfet+quantized). Not sure if that helps.

I thought the first image is of a planar transistor? Isn't FinFET like so: http://m.eet.com/media/1181899/finfet fig 1.jpg

 

[TuxDave]

This would be very good advice, besides the fact that I don't really have much idea of where to go to find someone whose directly linked to the industry who has precisely the information that I am looking for. Due to living in Southern Ontario, Canada, while we have lots of good educational facilities (my college is great obviously), I'm not sure of where I'd go to ask about IC design on a modern, practical level :/

Is there an Electrical Engineering program at your university? VLSI layout is typically an undergraduate 4th year class and those students should be able to answer your question. Maybe it's time to make some friends and engineers definitely don't have enough.