Chip costs won't rise. They'll continue to fall, just as they always have. Building a fab is indeed a large investment, but if you have the money to invest then it's one that'll pay for itself in a very short amount of time.
Uh, this assumes you have good products in high demand and can keep the fab running continuously at or near full capacity. A fab running below half capacity can bleed red ink pretty fast! Unfortunately, there's quite a bit of overcapacity in the semiconductor industry at the moment (mostly due to rapid expansion by foundries in Taiwan and elsewhere in Asia). This is one reason why semiconductor stocks have been in the toilet for the last year or so. IBM's Fab will only make this worse. Although IBM's advanced processing technology definitely gives them an advantage, so it may be their competitors rather than IBM that feels the pinch.
Equipment is no big deal -- the building itself is a huge deal. Getting all the tolerances tight enough for 65 nm work costs a LOT of money.
Think again, equipment prices are HUGE, especially when you're talking state of the art 300mm tools! They account for the greater portion of that $2.5B price tag. Lithography tools alone run $15-25M each and a big production fab like this probably has 20-30, so you're already at $0.5B with just one step of the process. Now add in Ion implanters, Plasma etch systems, CVD equipment, diffusion furnaces, Sputtering systems, chemical mechanical polish tools, electroplating equipment, and wet clean hoods, not to mention all the analytical equipment (SEMs, Elipsometers, particle counters, Quantox systems, CV plotters etc...) needed to ensure everything is functioning properly.
I have worked in a semiconductor manufacturing environment and to me the bigger concern would be interference between the 802.11 transmitters and the sensitive electronic equipment used in some stages of manufacturing. I'm not an EM guy, so I really don't know which equipment would be most susceptible to interference, but I would imagine some plasma tools, ion implanters, and especially testing equipment (which measures very small electrical currents in the completed devices) could be effected. I do know that due to interference concerns, cell phones were not allowed in the Fabs where I worked.
Of course 802.11 is much lower power than a cell phone (especially the old analog variety), so perhaps it is not a concern. Also, certainly IBM would be aware of potential interference problems and would have taken steps to avoid them, so it is possible that sensitive equipment is located in a part of the fab that is isolated from the wireless network.
Believe me, I tried very hard to understand the instructors and I am in general quite patient when working with non-native English speakers. It's one thing if the professor has a mild accent, or is willing to acknowledge his language struggles and work extra hard to help the students to understand by providing extra office hours or inviting students to meet one-on-one in an environment where language barriers can be overcome. However, if the instructor is defensive about his language difficulties and blames the students for not understanding him even becoming hostile, it is very hard to get by. The fact that over 50% of the class got D's or worse should have been a clue to the University that this professor (the Japanese instructor) wasn't suited to teaching. Certainly, he was a very competent mathmatician, but being good at math isn't the only qualification for teaching it.
A *huge* part of which is "better" depends entirely on the instructor. I've seen fantastic University professors, and fantastic college Instructors.
I completely agree. The instructor can make all the difference. Unfortunately, my University math experience was quite poor (I was a Math minor). My calculus professor was good, but after that I got stuck with two Chinese grad students and a Japanese teacher, none of whose English was really acceptable. In my multi-variable calculus class it took several class sessions before I realized what my teacher meant by the "Wee-wector" (v-vector). Actually, the second Chinese grad student spoke reasonably well, but after 2 semesters of incomprehensible math teachers, I was so out of the habit of paying attention that I didn't really get much from him either. I did still manage to get A's and B's, but my understanding and enthusiasm were seriously hampered.
If you want basic multivariable calculus, maybe a little bit of algebra.. yes, college is they way to go. If you are serious about a deep study of mathematics... you simply cannot beat training with people who are ACTUALLY ACTIVELY DOING IT. University professors, as part of their jobs, are required to engage in active research in their field of study. The same is not generally true of college instructors.
I would disagree a little bit. I think that Community Colleges are great for the subjects that they cover. So, I would recommend taking the basic courses there, since they are cheaper and the instruction is likely to be just as good. However, few Community Colleges offer much beyond simple Calculus, so you will eventually end up at a University if you want to go into mutli-variable calculus, differential equations, or any advanced math topics.
As others have mentioned, self-study with a good book can be a cheaper alternative. The drawbacks are: 1) no one is pushing you to complete assignments, so you will have to be quite self-motivated, and 2) if you get stumped, no one is available to help you out. One solution to this second issue is to post questions on an appropriate Usenet newsgroup. I know of several math related groups including:
In general I would first recommend taking a community college or University course (my experience is that either can be acceptable, although community colleges do not offer advanced math courses). However, for a cheaper alternative, get a good math textbook; perhaps, the one being used in a nearby college math course. Then, work through the book and if you get stumped, ask questions on Usenet.
You might also want to work through some supplemental problems. There are several math books in the "Schaums" series that have lots of pre-worked example problems for you to practice on.
It seems to me that the major limitations will be alignment and contamination issues. As you say, the alignment issues can be overcome by engineering, but this will add significantly to the cost. It will probably still be cheaper than conventional photolithography, since you don't have to deal with the expensive optics, but it won't be that much cheaper.
Even more critical is the contamination issue. At various stages in wafer processing, the wafer surface is coated with materials that are incompatible with other stages of the process. Since nanoimprint lithography involves mechanical contact between parts of the equipment and the surface of the wafer, there is a significant increase in the chance of contaminating the equipment. This is not an insurmountable problem, but it may increase the need for carefully monitoring cross-contamination between wafers, which will drive up the cost of using this technique.
Do you perhaps mean electromigration? Electromigration is a process which causes failure of metal interconnects. Essentially, momentum transfer bewteen moving electrons and metal atoms causes displacement of the atoms from their lattice positions, resulting in voids that break the electronic circuit. Electromigration only affects metals and will not affect the the transistor's themselves, just the interconnects between them.
While it's true that electromigration becomes more significant as metal line sizes decrease, there are methods for reducing electromigration. Primarily these involve altering the microstructure of the metal through the use of dopants or by changing the metal deposition conditions.
Actually, I don't think it's so much manufacturability as cost. I'm fairly certain people are making GaAs devices commercially(and more exotic variants InGaAs, AlGaAs, etc...), but these are quite expensive compared to Si. SiGe on the other hand is only marginally more expensive than Si and essentially uses the same manufacturing techniques. One other advantage is that SiGe devices can be put on the same chip as conventional Si logic devices. Thus SiGe allows for a 1-chip solution, where GaAs may require a 2-chip solution.
Actually, the zdnet article is a bit misleading, since it tries to spin the IBM announcement about a communications chip into a competition between Intel and IBM over semiconductor processing superiority. However, the IBM chip is based on heterojunction bipolar transistors (HBTs), rather than MOSFETs. Intel doesn't make HBTs and probably never will, since their strength is in logic circuitry (ie. MOSFETs).
HBTs are primarily used for communications applications as low-noise, high-frequency amplifiers. Intel isn't in this market and as far as I know has no plans to enter it. Intel is trying to make inroads into the DSP portion of the communications market, but that's a different topic altogether. So, basically the zdnet article takes a bunch of unrelated information and tries to spin it together into single story and does a rather poor job.
I'm a ChemE, not an EE, but I do work in SiGe technology. I'm fairly certain the mentioned IBM circuit is based on HBTs, since Bernard Meyer (mentioned in the press release) is IBM's big SiGe HBT guy. Also, the article itself doesn't mention CPU applications, but only focusses on wireless and networking applications.
Just reading from the two articles, it seems like they have only demonstrated a ring oscillator type circuit, although they plan to release a chip that runs at 110GHz by the end of the year. They currently seem to have an 80 GHz chip. Of course these are for networking and wireless applications and are not CPUs.
As for why people are talking about Pentiums and Athlons, a little knowledge is a dangerous thing.
Finally, my question is whether this is really the fastest semiconductor circuit ever demonstrated? My understanding is that GaAs can provide faster circuits than SiGe. The main advantage of SiGe is that it can be integrated into a conventional Si production line and so is much less expensive than GaAs (also it can be integrated with Si logic circuits). Since GaAs should be faster than SiGe, I'm skeptical about the title of the article. Though I might be persuaded to believe it is the fastest Si-based circuit.
Actually, one of the advantages of SiGe is that it is compatible with conventional Si-based technology. All that is required is the addition of a CVD reactor for depositing epitaxial SiGe films. However, as I state in a later post, at the moment SiGe can't be used for making MOSFETs, so it's not really suitable for CPUs. However, it can be used for making low-noise high-frequency amplifiers, such as are needed for 100 Gigabit optical networks. The fact that it can be integrated with conventional silicon processing gives it an edge over GaAs or other more exotic materials.
Call me stupid, but why can't they use the same material in PCs to increase the chip speed? Are there some limitations/incompatibilities other than the comparitively slow speeds of memory and I/O (I guess we can all see why I never got very far in that EE major...)
First of all, the IBM transistors are not MOSFETs, the tiny switches used in CPU's and other logic-based circuitry. They are instead heterojunction bipolar transistors (HBTs). HBTs are lightning fast and can be used as low-noise amplifiers for high frequency signals, which makes them great for wireless and Gigabit optical communication applications, but they are relatively large compared to MOSFETs and so are not really suitable for making CPU's. (Notice that the IBM press release never mentions CPU applications, but instead focuses on 100 Gigabit optical communications networks).
Now, you may wonder why SiGe can't be used to make super-fast MOSFETs. The main problem is that MOSFETs require a dielectric, such as SiO2 to act as an insulating layer between the "gate" and the channel. However, attempting to grow a layer of SiO2 on SiGe results in separation of the Ge from the Si, ultimately causing device failure.
Currently, people are trying to find ways to deposit new dielectrics with higher dielectric constants, such as ZrO2, to replace the SiO2. Once this is acheived it may be possible to put such a material onto SiGe to allow creation of a MOSFET using this technology. However, development of such high-k dielectric technology is probably 3-4 years away and adaptation of this to SiGe will be a few more years beyond that, so don't expect SiGe-based CPU's anytime soon.
One last thing. I don't understand why IBM gets all the press. Motorola announced 110 GHz HBTs last October. IBM is really not as far ahead of the curve as they would like you to believe.
"Terahertz" chip. It seems both IBM and AMD had developed this technology and Intel snubbed it, citing that it was to expensive to implement. There is nothing breakthrough about "fast switching" electrons, just the fact that INTEL released a press story about it makes it interesting. Ho hum.
Just a small correction. The technology was developed by IBM and Motorola. AMD licensed the technology from Motorola.
It's interesting that this is the second Noble prize in 4 years (since 1997) received by NIST scientists working on low-temperature applications. Obviously these guys are well-connected.
I hate to break it to you and mess up all of these posts about gas, gasoline, etc..., but the ZDNet article is screwed up. The Motorola Fuel Cell uses Methanol, not methane!
This is really old news. Motorola has already demonstrated working transistors using these types of materials, see this article from 1999.
Motorola is the #1 supplier of chips to the the communications industry, makes processors for Macs as well as Palm Pilots and licenses a lot of its process technology, such as copper interconnects and SOI, to AMD, so don't worry this technology is not likely to get lost in the shuffle.
Also, come on Jon, stop rambling. The bill doesn't affect non-violent games, so what's your point in saying:
Gaming isn't merely hunt-and-kill challenges for adolescents -- it includes everything from urban-planning, trivia, gambling, bridge and chess puzzles to complex, sophisticated journeys into the imagination. And it?s making a ton of money.
The bill didn't ban gaming, just certain violent games, so the above argument is completely irrelevant.
Your article comes off as almost as bad of an emotional knee-jerk reaction to the bill, as the bill was to Columbine.
I sometimes wonder how the risk of a catastrophic nuclear event killing thousands to millions compares to the risk of a localized catastrophic natural disaster killing thousands to millions (eg. Earth Quake in California or Japan or Volcanic eruption of Mt. Ranier near Seattle Washington).
I would guess the risks are greater for the natural disaster scenario, though people still choose to live in "high-risk" areas. Maybe people aren't always as rational as they think.
You make some good points, but I think your missing the point with your statement:
The problem with Chernobyl is partly that the level was NOT low, but mostly because the level was not CONTINUOUS. A short-term exposure to a high level of radiation is NOT equivalent to the same amount of radiation spread over years for a number of reasons, and a very big one is that the damage takes place before the levels of protecting chemicals can be raised.
The article seems to imply that there was some evidence of internal radiation exposure. This means that the damage was probably not so much due to short-term high-level radiation exposure, but to long-term low-level exposure to ingested radioactive materials.
It seems most likely that the subjects ingested some amount of radioactive isotopes, either by eating contaminated food or by inhaling contaminated material. The material then concentrated in or near the reproductive organs. This means that even after leaving Chernobyl, the parents were probably still being exposed to low-levels of ionizing radiation.
The article mentions that the probability for mutations decreased with increasing time away from Chernobyl, which is consistant with the gradual removal of the radioactive materials from the parents by natural processes.
Furthermore, this would explain the difference between Chernobyl and Hiroshima, since in Hiroshima most of the radiation exposure was to external gamma-radiation. Levels of radioactive isotopes that were subsequently ingested were likely much higher in Chernobyl, since the reactor would presumably contain more radioactive material than a bomb and it would have been spread over a smaller area.
We need not look for exotic explanations for rising health problems like cancer, because the most probably explanation is the constant exposure to EM radiation in unprecedented amounts. And as for your assertion that the chances of health risks are so small, that's just stupid, since no studies have been done or could be done since the modern world is pervaded with these high levels of low-energy EM particles. Think before you write.
O.K. Let's think a bit. According to the American Cancer Society (www.cancer.org) safety standards are usually set conservatively due to the difficulting in extrapolating from high dose experiments in animals to the low-doses typically encountered by humans. "For cancer safety standards, only increased risks of one case or less per million persons over a lifetime are usually acceptable."
At the same time, non-ionizing radiation has been studied extensively, yet from the same website:
"Electromagnetic radiation at frequencies below ionizing and ultraviolet levels has not been shown to cause cancer. While some epidemiologic studies suggest associations with cancer, others do not, and experimental studies have not yielded reproducible evidence of carcinogenic mechanisms."
The absense of any repeatible evidence that non-ionizing radiation causes cancer, suggests that the probability of cancer due to non-ionizing radiation is quite low. Probably much less than the 1 case in a million guideline stated above. Therefore, my statement that your odds of getting cancer from non-ionizing radiation are in the 1 in 10-100 million range is probably quite reasonable if not an overestimate of your chances.
By contrast, clear links have been established between cancer and a large number of environmental factors including: Tobacco and alcohol, various chemicals (eg. benzene, asbestos, vinyl chloride, arsenic), obesity, ionizing radiation from x-rays or radon gas, UV radiation from the sun, Viral infections (HIV, Hepatitus B, etc.) and many others.
It is not reasonable to assume that non-ionizing radiation is responsible for the increased incidence of cancer, just because EM levels have increased during the same time period. There are many other factors which show a stronger relation to the increased incidence of cancer and which have been shown to be contributors to cancer. Therefore, the probability of getting cancer from non-ionizing radiation is most likely quite low.
We're talking entire family lines affected for every future generation, by genetic mutations.
I think you'rr reading far too much into the dangers of genetic mutations. While it is true that genetic mutations can cause serious defects, most such defects do not result in a viable fetus. This means that the more severe defects cannot be passed on to a person's offspring and will not affect the family for generations. If two mutants breed, they would be even less likely to produce a viable fetus. Also, some mutations are harmless and will have no effect.
I do not mean to imply that there is no cause for concern and that there are no harmful effects of genetic mutations, but that these effects will likely not be as severe as you seem to imply. In short, yes it may be raining on our parade, but the sky is not falling.
Slashdot Mirror
Chip costs won't rise. They'll continue to fall, just as they always have. Building a fab is indeed a large investment, but if you have the money to invest then it's one that'll pay for itself in a very short amount of time.
Uh, this assumes you have good products in high demand and can keep the fab running continuously at or near full capacity. A fab running below half capacity can bleed red ink pretty fast! Unfortunately, there's quite a bit of overcapacity in the semiconductor industry at the moment (mostly due to rapid expansion by foundries in Taiwan and elsewhere in Asia). This is one reason why semiconductor stocks have been in the toilet for the last year or so. IBM's Fab will only make this worse. Although IBM's advanced processing technology definitely gives them an advantage, so it may be their competitors rather than IBM that feels the pinch.
Equipment is no big deal -- the building itself is a huge deal. Getting all the tolerances tight enough for 65 nm work costs a LOT of money.
Think again, equipment prices are HUGE, especially when you're talking state of the art 300mm tools! They account for the greater portion of that $2.5B price tag. Lithography tools alone run $15-25M each and a big production fab like this probably has 20-30, so you're already at $0.5B with just one step of the process. Now add in Ion implanters, Plasma etch systems, CVD equipment, diffusion furnaces, Sputtering systems, chemical mechanical polish tools, electroplating equipment, and wet clean hoods, not to mention all the analytical equipment (SEMs, Elipsometers, particle counters, Quantox systems, CV plotters etc...) needed to ensure everything is functioning properly.
I have worked in a semiconductor manufacturing environment and to me the bigger concern would be interference between the 802.11 transmitters and the sensitive electronic equipment used in some stages of manufacturing. I'm not an EM guy, so I really don't know which equipment would be most susceptible to interference, but I would imagine some plasma tools, ion implanters, and especially testing equipment (which measures very small electrical currents in the completed devices) could be effected. I do know that due to interference concerns, cell phones were not allowed in the Fabs where I worked.
Of course 802.11 is much lower power than a cell phone (especially the old analog variety), so perhaps it is not a concern. Also, certainly IBM would be aware of potential interference problems and would have taken steps to avoid them, so it is possible that sensitive equipment is located in a part of the fab that is isolated from the wireless network.
Believe me, I tried very hard to understand the instructors and I am in general quite patient when working with non-native English speakers. It's one thing if the professor has a mild accent, or is willing to acknowledge his language struggles and work extra hard to help the students to understand by providing extra office hours or inviting students to meet one-on-one in an environment where language barriers can be overcome. However, if the instructor is defensive about his language difficulties and blames the students for not understanding him even becoming hostile, it is very hard to get by. The fact that over 50% of the class got D's or worse should have been a clue to the University that this professor (the Japanese instructor) wasn't suited to teaching. Certainly, he was a very competent mathmatician, but being good at math isn't the only qualification for teaching it.
A *huge* part of which is "better" depends entirely on the instructor. I've seen fantastic University professors, and fantastic college Instructors.
I completely agree. The instructor can make all the difference. Unfortunately, my University math experience was quite poor (I was a Math minor). My calculus professor was good, but after that I got stuck with two Chinese grad students and a Japanese teacher, none of whose English was really acceptable. In my multi-variable calculus class it took several class sessions before I realized what my teacher meant by the "Wee-wector" (v-vector). Actually, the second Chinese grad student spoke reasonably well, but after 2 semesters of incomprehensible math teachers, I was so out of the habit of paying attention that I didn't really get much from him either. I did still manage to get A's and B's, but my understanding and enthusiasm were seriously hampered.
If you want basic multivariable calculus, maybe a little bit of algebra.. yes, college is they way to go. If you are serious about a deep study of mathematics... you simply cannot beat training with people who are ACTUALLY ACTIVELY DOING IT. University professors, as part of their jobs, are required to engage in active research in their field of study. The same is not generally true of college instructors.
I would disagree a little bit. I think that Community Colleges are great for the subjects that they cover. So, I would recommend taking the basic courses there, since they are cheaper and the instruction is likely to be just as good. However, few Community Colleges offer much beyond simple Calculus, so you will eventually end up at a University if you want to go into mutli-variable calculus, differential equations, or any advanced math topics.
As others have mentioned, self-study with a good book can be a cheaper alternative. The drawbacks are: 1) no one is pushing you to complete assignments, so you will have to be quite self-motivated, and 2) if you get stumped, no one is available to help you out. One solution to this second issue is to post questions on an appropriate Usenet newsgroup. I know of several math related groups including:
alt.algebra
alt.algebra.help
sci.math.*
I'm surprised I haven't seen any reference to Usenet news groups. There are several math related Usenet groups including:
sci.mathh .
alt.algebra
alt.algebra.help
sci.mathematics
sci.math.num-analysis
sci.math.researc
sci.math.symbolic
In general I would first recommend taking a community college or University course (my experience is that either can be acceptable, although community colleges do not offer advanced math courses). However, for a cheaper alternative, get a good math textbook; perhaps, the one being used in a nearby college math course. Then, work through the book and if you get stumped, ask questions on Usenet.
You might also want to work through some supplemental problems. There are several math books in the "Schaums" series that have lots of pre-worked example problems for you to practice on.
Good Luck.
It seems to me that the major limitations will be alignment and contamination issues. As you say, the alignment issues can be overcome by engineering, but this will add significantly to the cost. It will probably still be cheaper than conventional photolithography, since you don't have to deal with the expensive optics, but it won't be that much cheaper.
Even more critical is the contamination issue. At various stages in wafer processing, the wafer surface is coated with materials that are incompatible with other stages of the process. Since nanoimprint lithography involves mechanical contact between parts of the equipment and the surface of the wafer, there is a significant increase in the chance of contaminating the equipment. This is not an insurmountable problem, but it may increase the need for carefully monitoring cross-contamination between wafers, which will drive up the cost of using this technique.
Do you perhaps mean electromigration? Electromigration is a process which causes failure of metal interconnects. Essentially, momentum transfer bewteen moving electrons and metal atoms causes displacement of the atoms from their lattice positions, resulting in voids that break the electronic circuit. Electromigration only affects metals and will not affect the the transistor's themselves, just the interconnects between them.
While it's true that electromigration becomes more significant as metal line sizes decrease, there are methods for reducing electromigration. Primarily these involve altering the microstructure of the metal through the use of dopants or by changing the metal deposition conditions.
Actually, I don't think it's so much manufacturability as cost. I'm fairly certain people are making GaAs devices commercially(and more exotic variants InGaAs, AlGaAs, etc...), but these are quite expensive compared to Si. SiGe on the other hand is only marginally more expensive than Si and essentially uses the same manufacturing techniques. One other advantage is that SiGe devices can be put on the same chip as conventional Si logic devices. Thus SiGe allows for a 1-chip solution, where GaAs may require a 2-chip solution.
Actually, the zdnet article is a bit misleading, since it tries to spin the IBM announcement about a communications chip into a competition between Intel and IBM over semiconductor processing superiority. However, the IBM chip is based on heterojunction bipolar transistors (HBTs), rather than MOSFETs. Intel doesn't make HBTs and probably never will, since their strength is in logic circuitry (ie. MOSFETs).
HBTs are primarily used for communications applications as low-noise, high-frequency amplifiers. Intel isn't in this market and as far as I know has no plans to enter it. Intel is trying to make inroads into the DSP portion of the communications market, but that's a different topic altogether. So, basically the zdnet article takes a bunch of unrelated information and tries to spin it together into single story and does a rather poor job.
Correction: Replace Bernard Meyer with Bernard Meyerson in the previous post (I guess my fingers were slower than my brain).
I'm a ChemE, not an EE, but I do work in SiGe technology. I'm fairly certain the mentioned IBM circuit is based on HBTs, since Bernard Meyer (mentioned in the press release) is IBM's big SiGe HBT guy. Also, the article itself doesn't mention CPU applications, but only focusses on wireless and networking applications.
Just reading from the two articles, it seems like they have only demonstrated a ring oscillator type circuit, although they plan to release a chip that runs at 110GHz by the end of the year. They currently seem to have an 80 GHz chip. Of course these are for networking and wireless applications and are not CPUs.
As for why people are talking about Pentiums and Athlons, a little knowledge is a dangerous thing.
Finally, my question is whether this is really the fastest semiconductor circuit ever demonstrated? My understanding is that GaAs can provide faster circuits than SiGe. The main advantage of SiGe is that it can be integrated into a conventional Si production line and so is much less expensive than GaAs (also it can be integrated with Si logic circuits). Since GaAs should be faster than SiGe, I'm skeptical about the title of the article. Though I might be persuaded to believe it is the fastest Si-based circuit.
Actually, one of the advantages of SiGe is that it is compatible with conventional Si-based technology. All that is required is the addition of a CVD reactor for depositing epitaxial SiGe films. However, as I state in a later post, at the moment SiGe can't be used for making MOSFETs, so it's not really suitable for CPUs. However, it can be used for making low-noise high-frequency amplifiers, such as are needed for 100 Gigabit optical networks. The fact that it can be integrated with conventional silicon processing gives it an edge over GaAs or other more exotic materials.
Call me stupid, but why can't they use the same material in PCs to increase the chip speed? Are there some limitations/incompatibilities other than the comparitively slow speeds of memory and I/O (I guess we can all see why I never got very far in that EE major...)
First of all, the IBM transistors are not MOSFETs, the tiny switches used in CPU's and other logic-based circuitry. They are instead heterojunction bipolar transistors (HBTs). HBTs are lightning fast and can be used as low-noise amplifiers for high frequency signals, which makes them great for wireless and Gigabit optical communication applications, but they are relatively large compared to MOSFETs and so are not really suitable for making CPU's. (Notice that the IBM press release never mentions CPU applications, but instead focuses on 100 Gigabit optical communications networks).
Now, you may wonder why SiGe can't be used to make super-fast MOSFETs. The main problem is that MOSFETs require a dielectric, such as SiO2 to act as an insulating layer between the "gate" and the channel. However, attempting to grow a layer of SiO2 on SiGe results in separation of the Ge from the Si, ultimately causing device failure. Currently, people are trying to find ways to deposit new dielectrics with higher dielectric constants, such as ZrO2, to replace the SiO2. Once this is acheived it may be possible to put such a material onto SiGe to allow creation of a MOSFET using this technology. However, development of such high-k dielectric technology is probably 3-4 years away and adaptation of this to SiGe will be a few more years beyond that, so don't expect SiGe-based CPU's anytime soon.
One last thing. I don't understand why IBM gets all the press. Motorola announced 110 GHz HBTs last October. IBM is really not as far ahead of the curve as they would like you to believe.
Just a small correction. The technology was developed by IBM and Motorola. AMD licensed the technology from Motorola.
It's interesting that this is the second Noble prize in 4 years (since 1997) received by NIST scientists working on low-temperature applications. Obviously these guys are well-connected.
Don't give CmdrTaco too hard of a time, even ZDNet got it wrong. The Motorola cell phone uses Methanol (a liquid), not methane (a gas).
see:
http://www.cellular.co.za/battery_technology.htm
and
here
I hate to break it to you and mess up all of these posts about gas, gasoline, etc..., but the ZDNet article is screwed up. The Motorola Fuel Cell uses Methanol, not methane!
2 000/upi_fuelcell_31950.asp
See: http://www.cellular.co.za/battery_technology.htm
and http://www.enn.com/news/wire-stories/2000/09/0927
This is really old news. Motorola has already demonstrated working transistors using these types of materials, see this article from 1999.
Motorola is the #1 supplier of chips to the the communications industry, makes processors for Macs as well as Palm Pilots and licenses a lot of its process technology, such as copper interconnects and SOI, to AMD, so don't worry this technology is not likely to get lost in the shuffle.
Since you sound like you know something about this work, I'd like to ask a few questions.
1. How does this perovskovite material differ from that used previously by Motorola (see reference here)?
2. What will be the primary limitation for bringing this technology into a manufacturable process and how far off in the future is this?
Thanks.
Yeah!
Also, come on Jon, stop rambling. The bill doesn't affect non-violent games, so what's your point in saying:
Gaming isn't merely hunt-and-kill challenges for adolescents -- it includes everything from urban-planning, trivia, gambling, bridge and chess puzzles to complex, sophisticated journeys into the imagination. And it?s making a ton of money.
The bill didn't ban gaming, just certain violent games, so the above argument is completely irrelevant.
Your article comes off as almost as bad of an emotional knee-jerk reaction to the bill, as the bill was to Columbine.
I sometimes wonder how the risk of a catastrophic nuclear event killing thousands to millions compares to the risk of a localized catastrophic natural disaster killing thousands to millions (eg. Earth Quake in California or Japan or Volcanic eruption of Mt. Ranier near Seattle Washington).
I would guess the risks are greater for the natural disaster scenario, though people still choose to live in "high-risk" areas. Maybe people aren't always as rational as they think.
His point is that up to a point continuous exposure to low-level ionizing radiation can actually reduce the chances of gene mutations.
You make some good points, but I think your missing the point with your statement:
The problem with Chernobyl is partly that the level was NOT low, but mostly because the level was not CONTINUOUS. A short-term exposure to a high level of radiation is NOT equivalent to the same amount of radiation spread over years for a number of reasons, and a very big one is that the damage takes place before the levels of protecting chemicals can be raised.
The article seems to imply that there was some evidence of internal radiation exposure. This means that the damage was probably not so much due to short-term high-level radiation exposure, but to long-term low-level exposure to ingested radioactive materials.
It seems most likely that the subjects ingested some amount of radioactive isotopes, either by eating contaminated food or by inhaling contaminated material. The material then concentrated in or near the reproductive organs. This means that even after leaving Chernobyl, the parents were probably still being exposed to low-levels of ionizing radiation.
The article mentions that the probability for mutations decreased with increasing time away from Chernobyl, which is consistant with the gradual removal of the radioactive materials from the parents by natural processes.
Furthermore, this would explain the difference between Chernobyl and Hiroshima, since in Hiroshima most of the radiation exposure was to external gamma-radiation. Levels of radioactive isotopes that were subsequently ingested were likely much higher in Chernobyl, since the reactor would presumably contain more radioactive material than a bomb and it would have been spread over a smaller area.
We need not look for exotic explanations for rising health problems like cancer, because the most probably explanation is the constant exposure to EM radiation in unprecedented amounts. And as for your assertion that the chances of health risks are so small, that's just stupid, since no studies have been done or could be done since the modern world is pervaded with these high levels of low-energy EM particles. Think before you write.
O.K. Let's think a bit. According to the American Cancer Society (www.cancer.org) safety standards are usually set conservatively due to the difficulting in extrapolating from high dose experiments in animals to the low-doses typically encountered by humans. "For cancer safety standards, only increased risks of one case or less per million persons over a lifetime are usually acceptable."
At the same time, non-ionizing radiation has been studied extensively, yet from the same website:
"Electromagnetic radiation at frequencies below ionizing and ultraviolet levels has not been shown to cause cancer. While some epidemiologic studies suggest associations with cancer, others do not, and experimental studies have not yielded reproducible evidence of carcinogenic mechanisms."
The absense of any repeatible evidence that non-ionizing radiation causes cancer, suggests that the probability of cancer due to non-ionizing radiation is quite low. Probably much less than the 1 case in a million guideline stated above. Therefore, my statement that your odds of getting cancer from non-ionizing radiation are in the 1 in 10-100 million range is probably quite reasonable if not an overestimate of your chances.
By contrast, clear links have been established between cancer and a large number of environmental factors including: Tobacco and alcohol, various chemicals (eg. benzene, asbestos, vinyl chloride, arsenic), obesity, ionizing radiation from x-rays or radon gas, UV radiation from the sun, Viral infections (HIV, Hepatitus B, etc.) and many others.
It is not reasonable to assume that non-ionizing radiation is responsible for the increased incidence of cancer, just because EM levels have increased during the same time period. There are many other factors which show a stronger relation to the increased incidence of cancer and which have been shown to be contributors to cancer. Therefore, the probability of getting cancer from non-ionizing radiation is most likely quite low.
We're talking entire family lines affected for every future generation, by genetic mutations.
I think you'rr reading far too much into the dangers of genetic mutations. While it is true that genetic mutations can cause serious defects, most such defects do not result in a viable fetus. This means that the more severe defects cannot be passed on to a person's offspring and will not affect the family for generations. If two mutants breed, they would be even less likely to produce a viable fetus. Also, some mutations are harmless and will have no effect.
I do not mean to imply that there is no cause for concern and that there are no harmful effects of genetic mutations, but that these effects will likely not be as severe as you seem to imply. In short, yes it may be raining on our parade, but the sky is not falling.