The main reason clock speed isn't increasing as fast as transistor density is cache. The transistors are just more useful as fast, on-chip memory. The PA-RISC computer I'm typing this on (HP B-2600) has 4MB on on-chip cache! Now back to work!
Keep in mind
that we're using, by and large, the exact same semi-conductor process that was invented by TI back in 1954. There have been thousands or even
millions of refinements in the process, but we haven't switched to a non-silicon substrate, moved to light based computing, quantum computing,
or anything else.
Actually the original transistors used by TI were Germanium not silicon. And they were Bipolar Junction transistors, not the CMOS transistors used in most chips today. And lastly, there have been huge changes to manufacturing, such as self-aligned gate technology, thermal oxide deposition, etc. etc.
I'm sure all these "conveniences" will suck when they actually come out. I can just imagine: "my dryer sending me 5000 emails because it's notification program crashes... etc. etc.
All useful semiconductors are actually semi-insulators: you need doping and bias to promote electrons to the conduction band. In fact, semiconductors are sometimes characterized by their "bandgap voltage" which means how much voltage is required to promote an electron from the valence band, where it is tightly held to the lattice, and the conduction band, where it can move freely as part of an electrical current. The thing that makes diamond withstand high temperatures and voltages, its high bandgap, is also on of the things that makes it hard to work with.
I've said this already, but the article was confusing. They meant diamond has a can withstand higher temp. and voltage, not that it was higher resistance in an electrical sense. Their use of resistance had nothing to do with Ohm's Law. I guess the author didn't know about the way the phrase is generally used in electronics.
Also, they are talking about semiconductors, not superconductors. Very different beasts.
There is (at least) one key advantage silicon has over diamond (or any semiconductor except Germanium). It has a self-repairing crystal lattice. When dopant atoms (phosphorous, arsenic, etc.) are injected into the bulk silicon wafer using ion implantation (diffusion not used in practice too much anymore) they cause structural damage to the crystal lattice which would hurt circuit performace. However, Silicon has this magical property that if you heat it up to the right temperature (several hundred degrees Celcius) the lattice begins to reorganize itself to incorporate the dopant atoms without damage. Tis process is called Annealing and it is one of the key reasons Silicon became the dominant semiconductor (the other was the availability of a good thermal oxide, SiO2).
Diamond does not have this desirable property, so a lot of research will have to go into maintaining the quality of the crystal lattice.
You are correct, the artical wasn't very clear from a technical standpoint. C has better heat conductance so it can "resist" high temperatures and voltages more easily than silicon. They didn't mean resistance in an electrical context.
They mean it can withstand higher voltages, not that it has higher electrical resistance. Silicon is an excellent insulator (extremely high electrical resistance) unless impurities (dopants) are added. That is key to the operation of a MOSFET.
I'm not so amazed by China's progression. I'm an integrated circuit designer and I can tell you from experience that some of the best designers I've ever met are from the People's Republic of China. Once more of China's IC designers decide to stay in China instead of emigrating to the USA and Canada, we've got Trouble.
Also, once an architecture has been out for a while, there is a lot of information available which can be used to redesign it. Lastly, while 260 MHz was pushing the technology in 1997, it isn't that big a deal in 2002. Does anyone know what feature size the chip is fabbed in?
Doesn't hold a candle to FastHack'em for the C=64!!
Word. One time, I tried to copy Spy Hunter on the C-64 using an early version of FastHake'em and when I loaded the copy up, instead of saying "(c) Midway Games" on the bottom of the screen it said "We Prosecute Pirates". Me and my brother got scared the FBI was gonna show up! (I was 9 years old so don't laugh).
A Watermark is non-audibly / visually distinguishable pattern overlayed onto the signal. An example would be a 50Hz once-per-second spike added to heavy
rock music. Your ears would never notice it, but it would be in the analog signal, and digitally-resampling the sound would just put it straight back into the
digital form.
This is why the article talks about D/A (Digital to Analog) and A/D (Analog to Digital) converters.
Then filter it! If you can't hear it then filtering it out wont hurt anything.
Is Plaid Panty still around? Damn. What about Bob's Handy-Panty on Milwaukee Blvd? I have some good memories of hanging around the slushy machine during all 4 or 5 sunny days in the summer...
However one of the interesting aspects of the physics funding situation is the way that there is a seemingly inexhaustible supply of funds for large scale particle physics colliders which have no practical application I am aware of while fusion research has to scramble for every dollar while trying to solve the energy problem. This is in large part due to the fission mafia's attempt to sink every competing energy source the way they killed public funding for alternative energy research in the 80s (see Salter's duck).
Two points of issue here. First, particle physics research over the last fifty years is what made the solid-state revolution (and chips and affordable computers) possible. Only through linear accelerators was the crystal structure and properties of semiconductors properly worked out. Also, many technologies have come out of high-energy physics (particularly in the RF and signal-processing areas) in much the same way the Apollo program was helpful: Such difficult problems require new invention which can have useful application far beyond its intended use. I designed several integrated circuits for the BaBar detector at the Stanford Linear Accelerator (SLAC) and I can tell you that the physicists on that project were top-notch.
As for your fusion power comment, I must say that, for one thing, there hasn't been too much funding because there hasn't been too much progress. Second, the National Ignition Facility at LLNL sure cost a pretty penny...
Carl
Re:Just to remind people why more bits is good..
on
AMD's 64-bit Plot
·
· Score: 2
For those of you who are saying, "what about 64 bits? Will 64 bits be enough?" 2^64 is 32 orders of magnitude bigger than 2^32. 2^32 is roughly 4.5 billion (unsigned). 2^64 unsigned is 18,446,744,073,709,551,616, or roughly 2220 * 8309 trillion. 4.5 billion goes into that number 4.5 billion times.
(Btw, I am not suggesting that Windows is the very platform that should be standardized on. How many widespread virus out there have had that effect on Solaris?}
Ever heard of rtm? Brought our SunOS network to its knees about 10 years ago.
There is a fundamental difference between an analog modem and a device that sends digital data like a cable-modem or isdn router. Sheesh.
Sure, they are different, but they are both modems, and they use some form of modulation (time or frequency domain) to send digital data over an analog channel. All modern modems (including 56k, cable, etc.) are mixed-signal devices including an analog front-end along with digital processing. We haven't had purely analog modems since the 1200 bps days. Sheesh.
The chip will enable laptop users to connect directly to cellular networks without the need of a modem in the same way that PCs in a local network connect with each other
Give me a break. If it is wireless, I guarantee that there is modulation and demodulation involved. That means MODEM! The news here is that it is supposedly a monolithic solution and so it does in a chip what before was done on a board.
This reminds me of an argument I had once with an "expert" who tried to explain to me that a cable modem wasn't really a modem. Sheesh.
Theoreticly, a chip with twice the amount of transistors but running at the same clock speed, would be twice as powerfull as the model with half the transistors.
Why do you think this? The "power" of digital circuits is almost entirely determined by the feature size, given adequate integration. Doubling the transistors on a 0.5 micron process is not the same as using a 0.25 micron process. If you're thinking "put two processors on the 0.5 micron chip", you're mistaken, because it is well known how difficult multiprocessor systems are, one processor with twice the power is much easier to use.
Besides, most transistors in processors today are used in cache. Doubling the cache most certainly does not double performance (see any computer architecture text)
GaAs has been used for chips for years. Cost is of course a problem but there are others that make it very unlikely this will be used in general purpose microprocessors. The first problem is GaAs has a much higher defect density than silicon because it is a superlattice of gallium and arsenic and not a single crystal like silicon. For this reason GaAs chips have MUCH less yield than silicon chips so the number of transistors that can be integrated in GaAs in much less, even if it is put on a silicon substrate.
The second problem is the lack of a good thermal oxide in the GaAs material system. Silicon uses SiO2 which is an excellent insulator and more importantly has an extremely clean interface with silicon, so there are very few traps at the oxide-si interface. Because GaAs doesn't have a good oxide, MOS field-effect transistors (MOSFETS) are impossible and so digital GaAs chips use MESFETS, which are FETs without the oxide. It turns out the good oxide in silicon makes a lot of things possible that are impossible in GaAs. For example, the si oxide makes for a very high input impedance for Si transistors so they can be used to make dense RAM and very simple registers that rely on a high impedenence node. This structures are not possible in GaAs so more complicated, higher power circuits are required in GaAs to achieve the same functionality.
Unions exist to give job security to unskilled laborers. That's a Good Thing, but if it applies to the IT field, we're all in a lot of trouble.
That is a very arrogant and uninformed attitude. So you think machinists, electricians, teachers, police officers, firefighters, craftspeople, and nurses are unskilled? Come on, many working people are much more skilled than the average office worker. If fact, most unskilled laborers aren't in unions anyway, they're the guys you see in the morning hanging around 7-11 waiting for someone to pick them up for a day's work.
Unions exist to protect workers from exploitation from their employers and to promote a more equitible split of the fruits of their labor, pure and simple. And if you think unions are bad because they let working people earn a reasonable living, well, think of the alternative where our middle class evaporates and we have a somewhat sizeable well-to-do group and a HUGE poor population. Doesn't bode too well for the future, does it?
Even if you aren't in a union you benefit from their existance because they tend to normalize the employment market and keep corporations from acting too avariciously. Do you enjoy only working 5 days a week? How about getting medical insurance and retirement support? Do you like paid vacation? Thank the union movement.
Specifically, the free ride, the pardon and the lottery ticket. Depending on the situation, the some of the others may be as well.
No, they aren't. Irony is when what something seems at face value is actually opposite from the truth. Here are two simple examples of irony:
1. A giant man named "Tiny".
2. The song "Little Things" by Good Charlotte, in which they describe life-changing events, such as their Mother's mental problems, their father abandoning them, and the social humiliation of being poor as "little things".
By the way, "dramatic irony" is when the audience knows something the characters don't which enables the audience to interpret events differently from the characters.
In the article, the author states that the algorithm adapts to a form that can distinguish between two tones and uses a lot fewer gates than conventional designs. In almost the same breath, he says that there are five seemingly unconnected gates that, if disturbed, cause the system to fail. I'm a hardware designer, and where I come from that is called a BAD DESIGN.
We had some yahoo give a seminar about this at my uni and he was gushing about how it had come up with a patented voltage regulator topology and how it would revolutionize analog design. Well, an analog design that depends on its environment is simply a sucky analog design. In practice, analog circuits have to work in the harshest, most variable environments. I don't think that us "experts" are useless quite yet.
Slashdot Mirror
The main reason clock speed isn't increasing as fast as transistor density is cache. The transistors are just more useful as fast, on-chip memory. The PA-RISC computer I'm typing this on (HP B-2600) has 4MB on on-chip cache! Now back to work!
Keep in mind
that we're using, by and large, the exact same semi-conductor process that was invented by TI back in 1954. There have been thousands or even
millions of refinements in the process, but we haven't switched to a non-silicon substrate, moved to light based computing, quantum computing,
or anything else.
Actually the original transistors used by TI were Germanium not silicon. And they were Bipolar Junction transistors, not the CMOS transistors used in most chips today. And lastly, there have been huge changes to manufacturing, such as self-aligned gate technology, thermal oxide deposition, etc. etc.
I'm sure all these "conveniences" will suck when they actually come out. I can just imagine: "my dryer sending me 5000 emails because it's notification program crashes... etc. etc.
That game sure sounds like a lot of work. I think I'll smoke crack instead... not as much investment required.
All useful semiconductors are actually semi-insulators: you need doping and bias to promote electrons to the conduction band. In fact, semiconductors are sometimes characterized by their "bandgap voltage" which means how much voltage is required to promote an electron from the valence band, where it is tightly held to the lattice, and the conduction band, where it can move freely as part of an electrical current. The thing that makes diamond withstand high temperatures and voltages, its high bandgap, is also on of the things that makes it hard to work with.
I've said this already, but the article was confusing. They meant diamond has a can withstand higher temp. and voltage, not that it was higher resistance in an electrical sense. Their use of resistance had nothing to do with Ohm's Law. I guess the author didn't know about the way the phrase is generally used in electronics.
Also, they are talking about semiconductors, not superconductors. Very different beasts.
There is (at least) one key advantage silicon has over diamond (or any semiconductor except Germanium). It has a self-repairing crystal lattice. When dopant atoms (phosphorous, arsenic, etc.) are injected into the bulk silicon wafer using ion implantation (diffusion not used in practice too much anymore) they cause structural damage to the crystal lattice which would hurt circuit performace. However, Silicon has this magical property that if you heat it up to the right temperature (several hundred degrees Celcius) the lattice begins to reorganize itself to incorporate the dopant atoms without damage. Tis process is called Annealing and it is one of the key reasons Silicon became the dominant semiconductor (the other was the availability of a good thermal oxide, SiO2).
Diamond does not have this desirable property, so a lot of research will have to go into maintaining the quality of the crystal lattice.
You are correct, the artical wasn't very clear from a technical standpoint. C has better heat conductance so it can "resist" high temperatures and voltages more easily than silicon. They didn't mean resistance in an electrical context.
They mean it can withstand higher voltages, not that it has higher electrical resistance. Silicon is an excellent insulator (extremely high electrical resistance) unless impurities (dopants) are added. That is key to the operation of a MOSFET.
I'm not so amazed by China's progression. I'm an integrated circuit designer and I can tell you from experience that some of the best designers I've ever met are from the People's Republic of China. Once more of China's IC designers decide to stay in China instead of emigrating to the USA and Canada, we've got Trouble.
Also, once an architecture has been out for a while, there is a lot of information available which can be used to redesign it. Lastly, while 260 MHz was pushing the technology in 1997, it isn't that big a deal in 2002. Does anyone know what feature size the chip is fabbed in?
Doesn't hold a candle to FastHack'em for the C=64!!
Word. One time, I tried to copy Spy Hunter on the C-64 using an early version of FastHake'em and when I loaded the copy up, instead of saying "(c) Midway Games" on the bottom of the screen it said "We Prosecute Pirates". Me and my brother got scared the FBI was gonna show up! (I was 9 years old so don't laugh).
Carl
A Watermark is non-audibly / visually distinguishable pattern overlayed onto the signal. An example would be a 50Hz once-per-second spike added to heavy
rock music. Your ears would never notice it, but it would be in the analog signal, and digitally-resampling the sound would just put it straight back into the
digital form.
This is why the article talks about D/A (Digital to Analog) and A/D (Analog to Digital) converters.
Then filter it! If you can't hear it then filtering it out wont hurt anything.
Thank you for the education... I surrender.
Is Plaid Panty still around? Damn. What about Bob's Handy-Panty on Milwaukee Blvd? I have some good memories of hanging around the slushy machine during all 4 or 5 sunny days in the summer...
However one of the interesting aspects of the physics funding situation is the way that there is a seemingly inexhaustible supply of funds for large scale particle physics colliders which have no practical application I am aware of while fusion research has to scramble for every dollar while trying to solve the energy problem. This is in large part due to the fission mafia's attempt to sink every competing energy source the way they killed public funding for alternative energy research in the 80s (see Salter's duck).
Two points of issue here. First, particle physics research over the last fifty years is what made the solid-state revolution (and chips and affordable computers) possible. Only through linear accelerators was the crystal structure and properties of semiconductors properly worked out. Also, many technologies have come out of high-energy physics (particularly in the RF and signal-processing areas) in much the same way the Apollo program was helpful: Such difficult problems require new invention which can have useful application far beyond its intended use. I designed several integrated circuits for the BaBar detector at the Stanford Linear Accelerator (SLAC) and I can tell you that the physicists on that project were top-notch.
As for your fusion power comment, I must say that, for one thing, there hasn't been too much funding because there hasn't been too much progress. Second, the National Ignition Facility at LLNL sure cost a pretty penny...
Carl
Wouldn't that be 9.5 orders of magnitude?
Hey, is this the Dave Chen who went to College Park High School? If so send me an email to carl_r_grace@yahoo.com!
(Btw, I am not suggesting that Windows is the very platform that should be standardized on. How many widespread virus out there have had that effect on Solaris?}
Ever heard of rtm? Brought our SunOS network to its knees about 10 years ago.
There is a fundamental difference between an analog modem and a device that sends digital data like a cable-modem or isdn router. Sheesh.
Sure, they are different, but they are both modems, and they use some form of modulation (time or frequency domain) to send digital data over an analog channel. All modern modems (including 56k, cable, etc.) are mixed-signal devices including an analog front-end along with digital processing. We haven't had purely analog modems since the 1200 bps days. Sheesh.
The chip will enable laptop users to connect directly to cellular networks without the need of a modem in the same way that PCs in a local network connect with each other
Give me a break. If it is wireless, I guarantee that there is modulation and demodulation involved. That means MODEM! The news here is that it is supposedly a monolithic solution and so it does in a chip what before was done on a board.
This reminds me of an argument I had once with an "expert" who tried to explain to me that a cable modem wasn't really a modem. Sheesh.
Theoreticly, a chip with twice the amount of transistors but running at the same clock speed, would be twice as powerfull as the model with half the transistors.
Why do you think this? The "power" of digital circuits is almost entirely determined by the feature size, given adequate integration. Doubling the transistors on a 0.5 micron process is not the same as using a 0.25 micron process. If you're thinking "put two processors on the 0.5 micron chip", you're mistaken, because it is well known how difficult multiprocessor systems are, one processor with twice the power is much easier to use.
Besides, most transistors in processors today are used in cache. Doubling the cache most certainly does not double performance (see any computer architecture text)
The second problem is the lack of a good thermal oxide in the GaAs material system. Silicon uses SiO2 which is an excellent insulator and more importantly has an extremely clean interface with silicon, so there are very few traps at the oxide-si interface. Because GaAs doesn't have a good oxide, MOS field-effect transistors (MOSFETS) are impossible and so digital GaAs chips use MESFETS, which are FETs without the oxide. It turns out the good oxide in silicon makes a lot of things possible that are impossible in GaAs. For example, the si oxide makes for a very high input impedance for Si transistors so they can be used to make dense RAM and very simple registers that rely on a high impedenence node. This structures are not possible in GaAs so more complicated, higher power circuits are required in GaAs to achieve the same functionality.
That is a very arrogant and uninformed attitude. So you think machinists, electricians, teachers, police officers, firefighters, craftspeople, and nurses are unskilled? Come on, many working people are much more skilled than the average office worker. If fact, most unskilled laborers aren't in unions anyway, they're the guys you see in the morning hanging around 7-11 waiting for someone to pick them up for a day's work.
Unions exist to protect workers from exploitation from their employers and to promote a more equitible split of the fruits of their labor, pure and simple. And if you think unions are bad because they let working people earn a reasonable living, well, think of the alternative where our middle class evaporates and we have a somewhat sizeable well-to-do group and a HUGE poor population. Doesn't bode too well for the future, does it?
Even if you aren't in a union you benefit from their existance because they tend to normalize the employment market and keep corporations from acting too avariciously. Do you enjoy only working 5 days a week? How about getting medical insurance and retirement support? Do you like paid vacation? Thank the union movement.
No, they aren't. Irony is when what something seems at face value is actually opposite from the truth. Here are two simple examples of irony:
1. A giant man named "Tiny".
2. The song "Little Things" by Good Charlotte, in which they describe life-changing events, such as their Mother's mental problems, their father abandoning them, and the social humiliation of being poor as "little things".
By the way, "dramatic irony" is when the audience knows something the characters don't which enables the audience to interpret events differently from the characters.
We had some yahoo give a seminar about this at my uni and he was gushing about how it had come up with a patented voltage regulator topology and how it would revolutionize analog design. Well, an analog design that depends on its environment is simply a sucky analog design. In practice, analog circuits have to work in the harshest, most variable environments. I don't think that us "experts" are useless quite yet.