Has anyone out there actually had their system bottlenecked by lack of random numbers? I had thought that the bottleneck in serving large amounts of SSL content was processing the asymmetric part of the cyrpto -- hence the need for SSL accelerator cards. It's a nice invention and a creative application of physical process, but I really want to see just one case where this would be lead to a substantial benefit.
As an aside, computer simulations always use pseudoRNGs like the Mersenne Twister[1]. For a reasonable exponent (I use 19937 in my simulations), this results in a period > 10^6000 and virtually no correlations between adjacent calls. The notion of a computational physicist using a real physical RNG is laughable.
You seem to be choosing the most mission-critical life-or-death jobs like police, firefighters and EMTs and then using their situation to make a generally applicable point. This doesn't work and is a good example of confirmation bias. The vast, vast majority of jobs are not life-or-death and would not constitute "shutting down all of civilization" if those folks had more time off.
But that's exactly the point -- society has a general mechanism for deciding on what should be open and closed according to the priorities of the populace. We decide some things need to be open, while others need not.
I'll emphasize that what I am really commenting on are our priorities. [snip] [Consumer goods] are so much more valuable than quality time with people you love that whenever there is a schedule conflict, quality time is sacrificed? Do you believe that joyous, grateful, harmonious, fulfilled lives are built on this premise?
I believe very strongly in letting each individual determine her priorities according to whatever criteria best suit her. A corollary is that each individual should negotiate her own employment contract that best reflects her particular preferences.
It's the sort of thing that you can't really use facts and logic to prove. I can't write an equation that will rigorously demonstrate for you that one value system is superior to another. For this reason, if you disagree with me, then I do not believe that any amount of argument is going to result in agreement. I just wanted you to better understand what you are disagreeing with, as it is something more significant than the rather trivial objection you raise.
It's not a matter of computing whether one value system is better than another (that is, as you claim impossible). Rather, what I claim to do is compute, given the normative rankings that individuals do have, whether or not a particular service is worth keeping open or not. For instance, suppose the workers enjoy their Christmas so much that they demand 10x wages in order to come to work Dec 25th. Management can then compute whether, for that wage (and other costs), it is favorable to stay open.
What I am disagreeing with, essentially, is the sort of "objective" normative system where we do not have absolute deference to individuals' ranking of their priorities. I'm actually inclined to say that the best thing for an individual is defined by his normative preferences completely irrespective of any objective criteria that you might come up with. A man is the only person that can determine for himself what will make him happy/joyful/meaningful.
A holiday off? We can't do that, it might interefere with someone making money. This is the USA goddammit, we can't start placing quality time or family members above making money, we've got our priorities!
Who said anything about making money? Most of the fine people celebrating at home have a pretty reasonable expectation that they will have power, heat, emergency rooms, police, fire, EMT, ATC, gas stations and their internet pr0n. Just because some baby was born in a manger does not mean we have to shut down all of civilization.
The normal thing to do here is for the business/service to decide on a minimum level of service (in the case of the police/fire/ER, hopefully not too minimal) and pay their staff enough to want to show up. Part of the pay that police, ER doctors and IT professionals receive includes being on-call for the unexpected times when the shit hits the fan. That should be spelled out in your contract, including whatever level of bonus pay you expect for such work.
I've been using the same two copies of XP that I got when it came out (one MSDN from a friend, one retail). Of course, I don't feel the need to have a million different computers running at the same time -- one laptop and one desktop is plenty.
You are factually incorrect as to what the law requires (leaving aside for the moment the question of whether the law is being properly enforced).
As a US citizen, they cannot tap your phone without warrant. They can, however, tap the phone line of a suspect outside the US so that if you call that "tagged" line, they can listen in. That doesn't make you tapped because your next call to Joe's Pizza cannot be intercepted. Basically, they can monitor the phone line of the foreign person and gather your phone to him but they can't monitor your line.
Surely having records of being in the hospital unconscious while the act of terrorism takes place will never be considered an airtight alibi. Also, why wouldn't you fill out a police report while there, detailing your loss?
Since you need an individualized warrant from a Federal court to wiretap an American citizen inside the US, (yes, even with the new FISA, IIAL and I have actually read the statute), I don't see how registering the SIM to an identity has anything to do with privacy or free speech. Unless it's your position that (1) citizens do not have to submit to the lawful surveillance under the terms approved by Congress and monitored by a court of competent jurisdiction, (2) the courts are not effecting Congress' intent properly or (3) the terms, as written by Congress, are entirely unconscionable.
If I were writing the law, I would likely include more protection for privacy than is currently included. That political preference, however, does not mean that I think the law outside the bounds of reasonableness. There is room to disagree without hyperbole.
* Oh yeah, the telecom immunity part of the law was bullshit because the law was clearly being violated from 2002-2006. That said, here's an analysis of the law going forward: http://volokh.com/posts/1215699055.shtml.
(1) This isn't a legal proceeding, it's an administrative one. The board does have some limited powers to sanction offenders but it's not a court. There is therefore no notion of standing -- the board can hear anyone whose testimony it finds probative. The enforcement mechanism is laid out in 74C17 and requires the board to initiate legal proceedings in a state court against violators.
(2) Again, any attempt by him to lie or misdirect should be answered by having the board find his testimony not-credible or refuse to let him change the subject. These risks should emphatically not be dealt with by refusing him the right to speak at all.
(3) The board has broad powers to subpoena documents and compel witnesses [ 74C5(5)]. That makes them ideally suited to get to the bottom of who did what prior to trial. Just don't forget -- they are an investigative, not judicial, body.
The only thing that he's asked for is the (entirely reasonable, IMHO) chance to give specific testimony because he believes that the complaint is factually incorrect. Given his history of prevarication, I don't believe his claims but I cannot fathom why anyone would deny him the right to go in front of the board and say his piece. It's not like the board will somehow be in a worse position to sanction MediaSentry/SafeNet for whatever violations they have committed.
To me this is basically a non-story. Aside from the involvement of the RIAA, you could reduce it to: Corporation asks to give testimony in regulatory hearing that directly impacts its business. Truly shocking.
No, you don't. A dedicated 10 Mbps link would run thousands of dollars a month (a T1, which is 1.5 each way, is ~$300 in the US) and most consumers can't swing that sort of dedicated line. Instead, you paid for a connection that is 10 Mbps maximum and you knew damn well that you would be sharing it with others in your neighborhood. How else could you rationalize paying only $60/month for faster-than-T1 level service?
Pretending that you don't understand the difference between a dedicated line and a shared line is utterly unconvincing to me.
Or you could just not drive so fast you would get a ticket. I know, I am totally out of touch with reality and my ideas are correspondingly crazy. But I'll happily take a few minutes extra travel time and have a relaxed ride, because I don't have to worry about law enforcement and other drivers slowing me down.
I fully acknowledge your preference for that particular driving style. Now, would you acknowledge my preference to get where I'm going at a speed that I find comfortable without being hassled all the time? Or do you expect me to agree that your preferences are acceptable but mine, alas, are not?
Criminal fines (of the kind contemplated in the 8A) are not the same as civil judgments. The 8A does not limit the latter, only the former. Interestingly, the SCOTUS just ruled against large punitive (as opposed to compensatory) civil damages in the Exxon Valdez case. http://fortunelegalpad.files.wordpress.com/2008/06/07-219.pdf
Most of the [gaming] benchmarks show that Vista is just slower than XP.
This was true when Vista was RTMed but is just no longer the case. For any reviews after Vista SP1 (MS always takes 1 SP to get things right), the performance is about equal. Please cite any benchmarks you find to the contrary though -- I'd love to read them.
If you were expecting a huge drop in performance as your eyes scanned from the XP to the Vista results, well, surprise! As many a tech analyst predicted, Windows Vista's gaming performance conundrum has largely been solved, and it was mainly due to early graphics drivers.
We have a difference in language here. When I say 40% efficiency is the absolute maximum, I mean we can only extract usable work equal to 40% of the energy density of gasoline by burning it in internal combustion. The reason is that the system is irreversible and therefore must have some form of energy waste (or else it wouldn't add entropy to the system and that would mean it was reversible). I can prove (or at least I did once) that the Atkinson cycle is the absolute limit allowed by thermodynamics, you just can't do any better without violating the laws of physics as we currently understand them. See http://en.wikipedia.org/wiki/Thermal_efficiency for more details.
Now, you could say that a real gasoline engine that gets 39.5% of the usable work out of gasoline is really 99% efficient because it is operating at 99% of the theoretical maximum, but that doesn't appear to be what you are talking about. The sad fact is, there's no way to get 100% of the energy out of gasoline -- this is why chemists talk about the free energy. I would quote the free energy content of gasoline, but the exact number depends on the process used to extract it. http://en.wikipedia.org/wiki/Thermodynamic_free_energy
Re:Indeed, the record is an order of magnitude...
on
1000-mph Car Planned
·
· Score: 1
You have to figure on the utility value of our time too -- if I drive to work at 35mph instead of my usual 70mph (well, now I live 1 mile from my job -- it's really awesome) then I will spend twice as much time. If my time is worth something, then it might be worth it to spend more gas for the utility of being at the office sooner.
Driving slower is not the answer, IMO, unless you want people trying to cram more and more utility out of their car-time. It's bad enough with people yapping on their phones and eating their breakfast.
Yup. To some extent this becomes semantics -- what constitutes a "car". I know one when I see one and the SAE/Shell vehicles were just not cars.
Quibble: my calculation wasn't for the fuel efficiency of an actual car. It was for the maximum attainable efficiency given aerodynamic resistance. There are many things (rolling friction, imperfect regenerative breaking, internal resistance of the components) that make the number much lower. So a CD=.04 has a theoretical maximum mpg of 1800. Like I said, just a quibble.
However, I wonder if there are hidden assumptions that you are not factoring. I say this, knowing that the maximum theoretical speed of a modem is 9600 baud, but higher bps are possible, and knowing that the death of Moore's Law has been predicted for the last several years - both of which predictions are based in seemingly sound physics calculations.
Well, I have detailed my assumptions pretty clearly, I thought. If you can reduce the drag coefficient or the area of the car, that will help (although, the extent to which you can call the SAE entrants "cars" at all is highly dubious -- they are wonderful feats, just not cars). As another commenter pointed out, if you couple the aerodynamics of the cars (drafting), you can throw my figure out the door.
As to the BPS/Moore's law arguments, those were based on "limits" that are not exactly the same as thermodynamic limits. Moreover, 56k modems do operate incredibly close to the Shannon Limit, which is as absolute as the 2nd law of thermodynamics (from which the maximum efficiency of combustion engines is derived). The death of Moore's law is complicated by our incomplete master of quantum physics (part of my area of research, I'm not afraid to admit we have a long way to go). At the moment, the limits are in the photolithography which is optics -- an area in which clever tricks go a long way. There is a saying posted on the wall in our department "The diffraction limit is for undergrads" -- if you understand what that means you'll go a long way towards understanding the difference between "limit" used in the absolute sense and "limit" used in the operational sense.
Bugatti's car breaks the "rules" by using air at much higher pressure -- the friendly turbocharger on every WRX and riced-out-Civic in your neighborhood. It's a long-known trick for beating the gas/air density mixture problem.
Finally, I don't know what "potential energy capture mechanisms" are. The 40% efficiency figure comes from an engine that, consistent with the 2nd law of thermodynamics (which is absolutely absolute), uses all the energy in the gasoline and emits only enough waste heat to satisfy entropy. That is to say, it uses all the free energy (which is a technical term, wikipedia it for a quick intro) available. The rest of the energy is not "free" (to be used to do useful work) but must end up as waste heat.
Reading this discussion, it seems to me that the 1000mpg problem requires much more subtlety than the 1000mph problem, and is, to me, the far more interesting one.
Absolutely. MPG is orders of magnitude more difficult because of diminishing returns -- at least level that you remove waste, smaller and smaller effects (wave drag, rolling friction, turning friction) become more pronounced.
How many birds do you know that have a cross-sectional area of (1m)^2 and travel at 60MPH?
Furthermore, birds are masterful at using thermal drafts and to extract energy from the atmosphere -- they ride a thermal draft upwards and use the potential energy to continue moving. So, indeed a humongous bird zipping at 60MPH does in fact use ~1 KW, just not all of that is provided internally.
40% is the absolute limit for the Atkinson Cycle, which is itself the absolutely most efficient method of extracting energy from combustion. These limits are derived from the basic laws of thermodynamics -- if you violate them, you win a nobel prize (or crash the computer simulating the universe, if you think that's how it works).
If you want to quibble about parameters, my figure of 0.1 for the drag coefficient and (1m)^2 for the orthogonal area are much better targets. I got the former from the most aerodynamic car ever made and the latter as a guess for the smallest cross-sectional area for what would still be called a car (as opposed to a motorcycle).
(1) Power plant efficiency cannot be compared to internal combustion (Otto Cycle) efficiency. On a theoretical level, they are very different thermodynamic systems. On a practical level, power plants use combined-cycle where the output from each level is fed into the input for the next level (some have 5-6 turbines in descending order of energetics).
(2) You are correct that I computed the efficiency for a car who is aerodynamically uncoupled from all the other cars (or driving down the lonesome road, as a fellow once said). Drafting cars behind one another would reduce drag quite a bit and, in fact, I do enjoy drafting behind large trucks -- not for the mpg difference but because in a manual car with a tight throttle and a loud engine, you can viscerally hear and feel the difference as you move into the vacuum.
Finally, who the heck wants to drive controlled by a computer?! I like driving.:-)
Re:Indeed, the record is an order of magnitude...
on
1000-mph Car Planned
·
· Score: 1
Yes, those are impressive feats of engineering.
I think they beat my estimates because they are (a) much smaller than my 1m^2 cross section, (b) go significantly slower (recall that drag goes up at the cube of velocity and (c) use a fuel-cell that is much more efficient (90%) than internal combustion. They are a bit dishonest in reporting an "equivalent mpg" for use of hydrogen -- are they computing based on the total energy content of gasoline/hydrogen or the available energy after efficiency is computed. Moreover, are they figuring in that an internal-combustion engine that burns gasoline weighs much more, has much more complex emissions control (the fact that we do not allow cars to pollute the air actually adds significantly to their inefficiency) and would not allow their novel steering mechanism.
Like I said, they are very impressive. They do not, however, answer the same question that I answered.
Slashdot Mirror
Has anyone out there actually had their system bottlenecked by lack of random numbers? I had thought that the bottleneck in serving large amounts of SSL content was processing the asymmetric part of the cyrpto -- hence the need for SSL accelerator cards. It's a nice invention and a creative application of physical process, but I really want to see just one case where this would be lead to a substantial benefit.
As an aside, computer simulations always use pseudoRNGs like the Mersenne Twister[1]. For a reasonable exponent (I use 19937 in my simulations), this results in a period > 10^6000 and virtually no correlations between adjacent calls. The notion of a computational physicist using a real physical RNG is laughable.
[1] http://en.wikipedia.org/wiki/Mersenne_twister
You seem to be choosing the most mission-critical life-or-death jobs like police, firefighters and EMTs and then using their situation to make a generally applicable point. This doesn't work and is a good example of confirmation bias. The vast, vast majority of jobs are not life-or-death and would not constitute "shutting down all of civilization" if those folks had more time off.
But that's exactly the point -- society has a general mechanism for deciding on what should be open and closed according to the priorities of the populace. We decide some things need to be open, while others need not.
I'll emphasize that what I am really commenting on are our priorities. [snip] [Consumer goods] are so much more valuable than quality time with people you love that whenever there is a schedule conflict, quality time is sacrificed? Do you believe that joyous, grateful, harmonious, fulfilled lives are built on this premise?
I believe very strongly in letting each individual determine her priorities according to whatever criteria best suit her. A corollary is that each individual should negotiate her own employment contract that best reflects her particular preferences.
It's the sort of thing that you can't really use facts and logic to prove. I can't write an equation that will rigorously demonstrate for you that one value system is superior to another. For this reason, if you disagree with me, then I do not believe that any amount of argument is going to result in agreement. I just wanted you to better understand what you are disagreeing with, as it is something more significant than the rather trivial objection you raise.
It's not a matter of computing whether one value system is better than another (that is, as you claim impossible). Rather, what I claim to do is compute, given the normative rankings that individuals do have, whether or not a particular service is worth keeping open or not. For instance, suppose the workers enjoy their Christmas so much that they demand 10x wages in order to come to work Dec 25th. Management can then compute whether, for that wage (and other costs), it is favorable to stay open.
What I am disagreeing with, essentially, is the sort of "objective" normative system where we do not have absolute deference to individuals' ranking of their priorities. I'm actually inclined to say that the best thing for an individual is defined by his normative preferences completely irrespective of any objective criteria that you might come up with. A man is the only person that can determine for himself what will make him happy/joyful/meaningful.
A holiday off? We can't do that, it might interefere with someone making money. This is the USA goddammit, we can't start placing quality time or family members above making money, we've got our priorities!
Who said anything about making money? Most of the fine people celebrating at home have a pretty reasonable expectation that they will have power, heat, emergency rooms, police, fire, EMT, ATC, gas stations and their internet pr0n. Just because some baby was born in a manger does not mean we have to shut down all of civilization.
The normal thing to do here is for the business/service to decide on a minimum level of service (in the case of the police/fire/ER, hopefully not too minimal) and pay their staff enough to want to show up. Part of the pay that police, ER doctors and IT professionals receive includes being on-call for the unexpected times when the shit hits the fan. That should be spelled out in your contract, including whatever level of bonus pay you expect for such work.
Newegg is the most reputable store in the internet and they are selling Windows XP Home for $90.
$350 for a copy of Windows is insanity. Fortunately, a copy of XP will set you back less than $100. About 30 seconds on froogle got me to http://www.newegg.com/Product/Product.aspx?Item=N82E16832116511&nm_mc=OTC-Froogle&cm_mmc=OTC-Froogle-_-Software+-+Operating+Systems-_-Microsoft-_-32116511.
I've been using the same two copies of XP that I got when it came out (one MSDN from a friend, one retail). Of course, I don't feel the need to have a million different computers running at the same time -- one laptop and one desktop is plenty.
You are factually incorrect as to what the law requires (leaving aside for the moment the question of whether the law is being properly enforced).
As a US citizen, they cannot tap your phone without warrant. They can, however, tap the phone line of a suspect outside the US so that if you call that "tagged" line, they can listen in. That doesn't make you tapped because your next call to Joe's Pizza cannot be intercepted. Basically, they can monitor the phone line of the foreign person and gather your phone to him but they can't monitor your line.
Please read the link in my OP.
Surely having records of being in the hospital unconscious while the act of terrorism takes place will never be considered an airtight alibi. Also, why wouldn't you fill out a police report while there, detailing your loss?
Since you need an individualized warrant from a Federal court to wiretap an American citizen inside the US, (yes, even with the new FISA, IIAL and I have actually read the statute), I don't see how registering the SIM to an identity has anything to do with privacy or free speech. Unless it's your position that (1) citizens do not have to submit to the lawful surveillance under the terms approved by Congress and monitored by a court of competent jurisdiction, (2) the courts are not effecting Congress' intent properly or (3) the terms, as written by Congress, are entirely unconscionable.
If I were writing the law, I would likely include more protection for privacy than is currently included. That political preference, however, does not mean that I think the law outside the bounds of reasonableness. There is room to disagree without hyperbole.
* Oh yeah, the telecom immunity part of the law was bullshit because the law was clearly being violated from 2002-2006. That said, here's an analysis of the law going forward: http://volokh.com/posts/1215699055.shtml.
(1) This isn't a legal proceeding, it's an administrative one. The board does have some limited powers to sanction offenders but it's not a court. There is therefore no notion of standing -- the board can hear anyone whose testimony it finds probative. The enforcement mechanism is laid out in 74C17 and requires the board to initiate legal proceedings in a state court against violators.
(2) Again, any attempt by him to lie or misdirect should be answered by having the board find his testimony not-credible or refuse to let him change the subject. These risks should emphatically not be dealt with by refusing him the right to speak at all.
(3) The board has broad powers to subpoena documents and compel witnesses [ 74C5(5)]. That makes them ideally suited to get to the bottom of who did what prior to trial. Just don't forget -- they are an investigative, not judicial, body.
http://www.ncga.state.nc.us/EnactedLegislation/Statutes/HTML/ByChapter/Chapter_74C.html
The only thing that he's asked for is the (entirely reasonable, IMHO) chance to give specific testimony because he believes that the complaint is factually incorrect. Given his history of prevarication, I don't believe his claims but I cannot fathom why anyone would deny him the right to go in front of the board and say his piece. It's not like the board will somehow be in a worse position to sanction MediaSentry/SafeNet for whatever violations they have committed.
To me this is basically a non-story. Aside from the involvement of the RIAA, you could reduce it to:
Corporation asks to give testimony in regulatory hearing that directly impacts its business. Truly shocking.
If I pay for 10Mbps download speed
No, you don't. A dedicated 10 Mbps link would run thousands of dollars a month (a T1, which is 1.5 each way, is ~$300 in the US) and most consumers can't swing that sort of dedicated line. Instead, you paid for a connection that is 10 Mbps maximum and you knew damn well that you would be sharing it with others in your neighborhood. How else could you rationalize paying only $60/month for faster-than-T1 level service?
Pretending that you don't understand the difference between a dedicated line and a shared line is utterly unconvincing to me.
Or you could just not drive so fast you would get a ticket. I know, I am totally out of touch with reality and my ideas are correspondingly crazy. But I'll happily take a few minutes extra travel time and have a relaxed ride, because I don't have to worry about law enforcement and other drivers slowing me down.
I fully acknowledge your preference for that particular driving style. Now, would you acknowledge my preference to get where I'm going at a speed that I find comfortable without being hassled all the time? Or do you expect me to agree that your preferences are acceptable but mine, alas, are not?
The confrontation clause applies to criminal trials but parking tickets are a civil, not criminal violation (as such, they cannot include jail time).
Criminal fines (of the kind contemplated in the 8A) are not the same as civil judgments. The 8A does not limit the latter, only the former. Interestingly, the SCOTUS just ruled against large punitive (as opposed to compensatory) civil damages in the Exxon Valdez case. http://fortunelegalpad.files.wordpress.com/2008/06/07-219.pdf
Most of the [gaming] benchmarks show that Vista is just slower than XP.
This was true when Vista was RTMed but is just no longer the case. For any reviews after Vista SP1 (MS always takes 1 SP to get things right), the performance is about equal. Please cite any benchmarks you find to the contrary though -- I'd love to read them.
http://www.bit-tech.net/bits/2008/03/25/windows_vista_sp1_gaming_performance/10
http://www.extremetech.com/article2/0,2845,2302527,00.asp
http://www.gamespot.com/features/6188289/index.html?sid=6188289&cpage=8
If you were expecting a huge drop in performance as your eyes scanned from the XP to the Vista results, well, surprise! As many a tech analyst predicted, Windows Vista's gaming performance conundrum has largely been solved, and it was mainly due to early graphics drivers.
Unless you are charging the batteries from an outlet, all the energy has to have come from gasoline.
We have a difference in language here. When I say 40% efficiency is the absolute maximum, I mean we can only extract usable work equal to 40% of the energy density of gasoline by burning it in internal combustion. The reason is that the system is irreversible and therefore must have some form of energy waste (or else it wouldn't add entropy to the system and that would mean it was reversible). I can prove (or at least I did once) that the Atkinson cycle is the absolute limit allowed by thermodynamics, you just can't do any better without violating the laws of physics as we currently understand them. See http://en.wikipedia.org/wiki/Thermal_efficiency for more details.
Now, you could say that a real gasoline engine that gets 39.5% of the usable work out of gasoline is really 99% efficient because it is operating at 99% of the theoretical maximum, but that doesn't appear to be what you are talking about. The sad fact is, there's no way to get 100% of the energy out of gasoline -- this is why chemists talk about the free energy. I would quote the free energy content of gasoline, but the exact number depends on the process used to extract it. http://en.wikipedia.org/wiki/Thermodynamic_free_energy
You have to figure on the utility value of our time too -- if I drive to work at 35mph instead of my usual 70mph (well, now I live 1 mile from my job -- it's really awesome) then I will spend twice as much time. If my time is worth something, then it might be worth it to spend more gas for the utility of being at the office sooner.
Driving slower is not the answer, IMO, unless you want people trying to cram more and more utility out of their car-time. It's bad enough with people yapping on their phones and eating their breakfast.
Yup. To some extent this becomes semantics -- what constitutes a "car". I know one when I see one and the SAE/Shell vehicles were just not cars.
Quibble: my calculation wasn't for the fuel efficiency of an actual car. It was for the maximum attainable efficiency given aerodynamic resistance. There are many things (rolling friction, imperfect regenerative breaking, internal resistance of the components) that make the number much lower. So a CD=.04 has a theoretical maximum mpg of 1800. Like I said, just a quibble.
However, I wonder if there are hidden assumptions that you are not factoring. I say this, knowing that the maximum theoretical speed of a modem is 9600 baud, but higher bps are possible, and knowing that the death of Moore's Law has been predicted for the last several years - both of which predictions are based in seemingly sound physics calculations.
Well, I have detailed my assumptions pretty clearly, I thought. If you can reduce the drag coefficient or the area of the car, that will help (although, the extent to which you can call the SAE entrants "cars" at all is highly dubious -- they are wonderful feats, just not cars). As another commenter pointed out, if you couple the aerodynamics of the cars (drafting), you can throw my figure out the door.
As to the BPS/Moore's law arguments, those were based on "limits" that are not exactly the same as thermodynamic limits. Moreover, 56k modems do operate incredibly close to the Shannon Limit, which is as absolute as the 2nd law of thermodynamics (from which the maximum efficiency of combustion engines is derived). The death of Moore's law is complicated by our incomplete master of quantum physics (part of my area of research, I'm not afraid to admit we have a long way to go). At the moment, the limits are in the photolithography which is optics -- an area in which clever tricks go a long way. There is a saying posted on the wall in our department "The diffraction limit is for undergrads" -- if you understand what that means you'll go a long way towards understanding the difference between "limit" used in the absolute sense and "limit" used in the operational sense.
Bugatti's car breaks the "rules" by using air at much higher pressure -- the friendly turbocharger on every WRX and riced-out-Civic in your neighborhood. It's a long-known trick for beating the gas/air density mixture problem.
Finally, I don't know what "potential energy capture mechanisms" are. The 40% efficiency figure comes from an engine that, consistent with the 2nd law of thermodynamics (which is absolutely absolute), uses all the energy in the gasoline and emits only enough waste heat to satisfy entropy. That is to say, it uses all the free energy (which is a technical term, wikipedia it for a quick intro) available. The rest of the energy is not "free" (to be used to do useful work) but must end up as waste heat.
Reading this discussion, it seems to me that the 1000mpg problem requires much more subtlety than the 1000mph problem, and is, to me, the far more interesting one.
Absolutely. MPG is orders of magnitude more difficult because of diminishing returns -- at least level that you remove waste, smaller and smaller effects (wave drag, rolling friction, turning friction) become more pronounced.
How many birds do you know that have a cross-sectional area of (1m)^2 and travel at 60MPH?
Furthermore, birds are masterful at using thermal drafts and to extract energy from the atmosphere -- they ride a thermal draft upwards and use the potential energy to continue moving. So, indeed a humongous bird zipping at 60MPH does in fact use ~1 KW, just not all of that is provided internally.
40% is the absolute limit for the Atkinson Cycle, which is itself the absolutely most efficient method of extracting energy from combustion. These limits are derived from the basic laws of thermodynamics -- if you violate them, you win a nobel prize (or crash the computer simulating the universe, if you think that's how it works).
If you want to quibble about parameters, my figure of 0.1 for the drag coefficient and (1m)^2 for the orthogonal area are much better targets. I got the former from the most aerodynamic car ever made and the latter as a guess for the smallest cross-sectional area for what would still be called a car (as opposed to a motorcycle).
Very good points. A few quibbles:
(1) Power plant efficiency cannot be compared to internal combustion (Otto Cycle) efficiency. On a theoretical level, they are very different thermodynamic systems. On a practical level, power plants use combined-cycle where the output from each level is fed into the input for the next level (some have 5-6 turbines in descending order of energetics).
(2) You are correct that I computed the efficiency for a car who is aerodynamically uncoupled from all the other cars (or driving down the lonesome road, as a fellow once said). Drafting cars behind one another would reduce drag quite a bit and, in fact, I do enjoy drafting behind large trucks -- not for the mpg difference but because in a manual car with a tight throttle and a loud engine, you can viscerally hear and feel the difference as you move into the vacuum.
Finally, who the heck wants to drive controlled by a computer?! I like driving. :-)
Yes, those are impressive feats of engineering.
I think they beat my estimates because they are (a) much smaller than my 1m^2 cross section, (b) go significantly slower (recall that drag goes up at the cube of velocity and (c) use a fuel-cell that is much more efficient (90%) than internal combustion. They are a bit dishonest in reporting an "equivalent mpg" for use of hydrogen -- are they computing based on the total energy content of gasoline/hydrogen or the available energy after efficiency is computed. Moreover, are they figuring in that an internal-combustion engine that burns gasoline weighs much more, has much more complex emissions control (the fact that we do not allow cars to pollute the air actually adds significantly to their inefficiency) and would not allow their novel steering mechanism.
Like I said, they are very impressive. They do not, however, answer the same question that I answered.
Of course, one can always ignore your calculations by doing things differently than how you expect.
Of course. I thought I made it clear that I was talking about the consumption of gasoline by internal combustion, nothing more.