Well, it's back-compatible with PCIe 1.0 and 1.1, so it aside from price there's on disadvantage to including it. Whether there's demand for it is another question, but I'm sure the graphics card makers will find something to do with it. Think back to AGP 2x/4x -- those made it onto cards and motherboards fairly quickly, iirc.
OK, fine, various weird dialects are correct then. But "correct" English (as defined by, say, your average skilled editor) is a very useful language to know. It helps a lot with communication and clarity, for preventing ambiguity and misunderstandings, and for creating the desired tone of voice (as perceived by the audience) in a piece of writing. So whether you want to call that "correct" English or "standard" English or "anal-retentive proscriptivist copy-editor" English, I don't really care -- but it's damn useful to have it and to have a relatively uniform definition of what it is and uniform education in it.
Language isn't just about conversing with those you interact with on a daily basis. If you want to communicate with people outside your immediate cultural group, it's awfully useful to have some standardization. I think this is what most people are driving at when they complain about bad English skills among native speakers. Since standardization implies some proscriptivism, someone has to do the proscribing. Perhaps the linguists should stop complaining about the proscriptivism, do some proscribing, and give the 7th grade English teachers some better rules to teach -- I'm sure any linguist will agree that your average grammar class does a fairly bad job of teaching grammar, even when compared to the language it's trying to teach the grammar of (as opposed to whatever dialect or form the people in the class normally use).
So I am forced to respectfully disagree with you and say that there is something wrong with African American Vernacular English -- not in any abstract, objective sense, and not when communicating with others who are fluent in it, but in using it to attempt to communicate with those who aren't fluent in it, or in using it in situations where it isn't socially expected. And plenty of people do exactly that. I know several people who are effectively bilingual in it and "standard" English, and quite fluidly shift between the two as appropriate to the situation. That seems entirely appropriate. But expecting other people to speak your language in a social situation where the norm is for you to speak theirs is, at best, rude -- be it ignorant American tourists abroad or illegal Mexican immigrants in the US.
Actually, I'd like it if gmail let me be prolific with my tagging. Not infrequently I find that I'm looking for an email that never actually references the obvious search terms. Basically I'd like to use labels to create collections of things I look at a lot (eg a label for each mailing list; a label for work related stuff; a label for family email, etc), but I don't want dozens and dozens of different emails. I don't think it would be hard for me to notice an email coming in, decide "this might be useful to find later," and add a few keywords to it.
I guess I really want a pair of things -- I want to be able to only have some of the labels / tags show up in the box on the left, and I want a quicker way than the drop down menu to add labels / tags to a message. (Like a text box with autocomplete like slashdot tags in addition to the drop down menu. Perhaps the menu should only show the ones in the box on the left.)
No, I'm saying you can't do the switch overnight. Not shouldn't, can't. The installed base is huge. And the installed base won't change overnight, even if everyone agreed to start using metric all at once. I think metrication is a good idea and worth doing, but it can't be done by fiat. It has to be a gradual process, and it will be a painful one no matter what.
Your thoughts from Finland are basically correct;)
Here we basically do all our calculation in SI units, but tend to convert to English at either end. So if the customer wants a 1000 lb thrust engine, we convert that to Newtons, do the design, and then end up with an engine whose dimensions are specified in inches. The other thing we do is rigorously label units in all calculations. You thought it was annoying when your physics professor counted off for not writing down units -- here we care for very practical reasons;) The other reason we do this is that growing up around English units, they're much more intuitive. Telling me a line is at 1500 psi is much more informative than telling me it's at 10MPa.
Oh, and for plumbing, you actually need to stock a triple set if you're stocking both English and Metric -- you also have to stock the English / Metric adapters, because there's no way you'll never find a need to connect the two together.
In the abstract, I think metrication is a good thing. And it's well worth doing. But it's not easy, and it'll take a while, and that's a fact worth paying attention to.
As someone working on an early prototype for one of the engines involved, I'd like to say I have mixed opinions on this.
Metric is good for all the obvious reasons -- SI units haver fewer weird things going on, conversions are easier, interoperable tools and fittings, etc etc. For all things like discussing distances, velocities, thrust levels, trajectory simulations, and more, I'm completely in favor of metric everywhere.
The one place I don't like this is when it comes to fittings, fasteners, plumbing, etc. Partly it's that metric nuts, bolts, and fittings are harder to find. You can't buy metric pipe fittings around here. Sure, you can order them, but that takes longer and costs more. The cost isn't a big issue on most things, but turnaround time is -- if you find a problem, it's really nice to be able to order a different part and have it the next day, rather than waiting a few days for something from Europe to clear customs and arrive. On some things, though, it actually makes a big difference. A lot of things like large pressure regulators, specialty valves, and more are even harder to find with metric fittings on them -- specifically, they become custom parts, with associated cost increases and weeks of lead time, which is frequently unacceptable.
And before anyone says you can buy metric parts in the US -- sure, you can, as long as they're "normal." It's the specialty parts that are hard. For example, McMaster-Carr stocks 3798 different socket cap screws in English sizes, but only 1610 in metric. If you need a weird metric screw, you may very well be out of luck.
The other major thing is subcontracts -- if I hire a consultant or send a part out to be machined, the machinist needs to have metric tools. Again, most machinists have a basic set of metric tools, but not an entire shop's worth. If the consultant or machinist has to start buying new tooling, your costs and the delivery time start going up.
I'll say it again -- having to buy parts from out of the country is not just a minor nuisance; it has a direct impact on how quickly you can revise a design and do the next test, which directly translates into how long it takes to complete the project.
I'm in favor of working toward compatibility, but it's not nearly as obvious an answer as it looks when it comes to tooling, since the installed base of English tooling and suppliers is *so* *huge* while metric is really only supported because of a few foreign-made parts.
Actually, I basically agree with you -- it's impossible to prove you have the requisite reliability. Therefore, no one will fly with ground based control. But mandating methods instead of results is always bad regulatory policy, because fundamentally the public doesn't care *how* you guarantee to fly safely enough, just *that* you do so. In this case, I can't see any arguments against an onboard pilot -- but that doesn't mean I should assume there are none. There are plenty of instances of other regulations that mandated methods, which resulted in less safe operation as soon as those methods became outdated. This policy is basically well-written.
While I basically agree with what you said, you're a little off target. These rules are really for suborbital tourism, aka the SpaceShipOne flight -- up and down, not orbit. That's much easier technically, and doesn't involve the 8 km/s sideways velocity of an orbital flight. Also, the vehicles people are talking about so far will only go to a little over 100 km altitude, which is lower than any Low Earth Orbit satellite -- even at 100 km, the air is so thick orbits decay rapidly.
So, at least for this first generation of vehicles, there's no real worry about space collisions. These rules are more aimed at dealing with things like participant awareness of the risks, and protecting the uninvolved public. Both very important things, and fortunately AST (the branch of the FAA in charge of space flight stuff) is taking a very sane and reasonable approach to most of this.
While I haven't read this version, I have read some of the drafts, and talked with people intimately involved in the industry and the regulation drafting. You're a bit off base here -- taken in context, what they're actually saying is that such links are fine, as long as the operator can prove they're safe enough. That is, the burden of proving the reliability of the system (including airframe, propulsion, controls, avionics, telemetry links, etc) is on the operator. They won't reject your plan just because you use ground-based controls, you just have to meet the same level of safety as everyone else. Seems eminently reasonable to me. (Also note that you will be required to do a detailed FMEA (Failure Modes and Effects Analysis), that addresses things like common-mode failures.) This is all as it should be -- regulating agencies shouldn't mandate techniques, they should mandate results. Rather than specify how you should control the craft, or whether you should burn LOX/kerosene or tetroxide/hydrazine, they just mandate that you fly safely enough, and demonstrate to their satisfaction that you can do so.
While I know you're not being serious, something like back Playboy issues has cultural relevance and value, even if it's not exactly "literature." The same goes for plenty of other magazines; Rolling Stone back issues, anyone? How about the comics pages of your daily newspaper? Entertainment material has a lot of cultural value to it.
Besides, the rockets in question are useless as terrorist weapons, and the chemicals in question are a poor choice as an explosive.
ANFO makes a far superior terrorist explosive. If you want a delivery system, rockets are really quite bad -- they're hard to control and have a relatively short range. Model airplanes are far superior in most respects. If you're fighting a guerilla war, then cheap rockets become interesting -- see Qassam rockets in the middle east. Which, while they bear a few superficial similarities to high power hobby rockets, are really much better suited to the task than the HPR stuff anyway.
We're not building weapons, and we police our own -- even joking around about such things would be highly frowned on at a launch. Why should this be regulated any more or any differently than model airplanes? After all, model airplanes use nitromethane, which was actually used in the Oklahoma City bombing.
Just because it's a rocket doesn't mean it's ok to regulate it out of some knee-jerk "OMG Terrorists!" reaction.
Re:Gyroscopic stabilizers
on
Rocket Men
·
· Score: 1
Oops. My bad. I thought you were repeating the complaint.
Sorry about that.
Re:Gyroscopic stabilizers
on
Rocket Men
·
· Score: 1
Stop being an ass. I have good, directly relevant, second-hand knowledge of the subject matter. The other poster has indirectly relevant second-hand (if he's not writing the control code personally, it's second-hand for this discussion) knowledge of unknown (to me) quality (could be good, bad, very good -- I don't know). My information was presented to me as fact by people who certainly *ought* to know, seeing as they were investigating the exact question at hand. Hence I repeated it as fact, but am willing to admit that it could be out of date or in error. Now what, exactly, is the problem?
Re:Gyroscopic stabilizers
on
Rocket Men
·
· Score: 1
No, the segway has a huge advantage here -- the ground. If the device isn't accelerating (measured total acceleration = 1 gravity), and there's no rotational acceleration (ie it's not falling over), then you *know* that the force being applied by the ground is directly upward. Given various accelerations and knowledge of what the wheels are doing, you can back out the reference frame in less ideal conditions. This allows you to be continually fixing your gyro errors, and only use them for rotational acceleration and short-term rotation rate, which the cheap ones are perfectly capable for.
A rocket has no such luxury -- measured acceleration comes only from the rocket thrust plus any aero loads; you can't "see" the gravity vector. Alternately, imagine the engine is supplying 1G worth of thrust. Your accelerometers "see" 1G accel away from the engine. Without the gyros, you have no way of telling if oyu're vertical and hovering, or at 45 degrees, falling, and accelerating sideways. You can make guesses based on air speed and GPS and such, but it's much harder -- hence my comment that using cheap gyros depends on your ability to include other sensor data. If, on the other hand, you have an expensive FOG, you can just ask it for your current orientation and get a meaningful answer.
Re:Gyroscopic stabilizers
on
Rocket Men
·
· Score: 2, Informative
Not entirely true. I have spoken with multiple people directly involved in the area -- gyro-based IMUs for rocket vehicles -- and the FOGs are clearly superior. AIUI, some of the recent MEMS gyros *might* be good enough for low-accuracy use, depending on the application details, and how much other sensor data is available to correct with.
I do know that at least until recently, inexpensive gyros were completely unusable. Modern ones appear better, but my sources suggest that they aren't all the way there yet. It's possible there are good gyros out there that haven't been tried in this application and that I haven't heard about; I'm not averse to admitting my knowledge could be out of date or incomplete. I am, however, quite confident that the problem is not so easy as the OP implied.
And for reference, IAARS, and I have read detailed discussions by people doing actual investigation with real hardware testing of exactly this problem -- and I'd say that makes me better qualified to comment than the vast majority of posters. But, like I said, not infallible -- so please don't jump on me for admitting my fallibility.
Re:Gyroscopic stabilizers
on
Rocket Men
·
· Score: 1
I don't have a good refernce handy, but I believe the answer is no. I've heard my original comment made by people in the industry who were well aware of the DC-XA, so I'm inclined to believe it's true, but I don't know enough about it to give an authoritative answer.
Re:Gyroscopic stabilizers
on
Rocket Men
·
· Score: 3, Insightful
Hmm. I don't know how good those gyros are; I was of the impression that there wasn't really a whole lot between the cheap sensor grade stuff and the good navigation grade fiber optic ones. Also, AIUI the differences aren't just in drift rate, but also in things like vibration sensitivity and cross-axis coupling.
I suppose you could use the inexpensive ones, as long as your goal was to change the pilot requirement from "top of the line test pilot" to "very good helicopter pilot," and not an attempt to make it flyable by anyone with a bit of simulator practice.
You might do an ok job if the gyros just tried to hold the spin *rate* to zero, and let the pilot handle leveling the vehicle; one fewer integral makes for much slower error growth.
Re:And this is different from
on
Rocket Men
·
· Score: 1
You mean aside from the fact that skateboards don't have a 1000-hp power output in the form of a hot supersonic gas stream located perhaps a foot or two away from you?
Well, in that case, it's different because skateboards don't have pressurized tanks of propellant that's dangerous to get on your skin, don't have multiple pieces of complex machinery, all of which is required to operate as designed in a fairly harsh environment in order to ensure your continued survival, and furthermore are fundamentally stable (if you get on a skateboard and start moving slowly, on relatively flat terrain, it has no natural tendency to dump you on your face from 20 feet in the air).
Oh, and don't forget that it's an awful lot easier to safely test your new skateboard design. When testing is hard, as with a rocket belt, development is slow and expensive.
And for the record, IAARS -- and rocket belts scare me. A lot. I'd be much more inclined to get in a rocket powered airplane, go cart, or suborbital rocket.
Re:Gyroscopic stabilizers
on
Rocket Men
·
· Score: 5, Informative
And while we're at it, that's a *hell* of a lot easier said than done. You can't do it on cheap gyros (read: you're probably spending $5-10k per axis), and they're not particularly light weight (a couple pounds each may not seem like much, but it eats into your fuel budget quite quickly). And you need a *good* control program, which isn't easy to write. Getting it mostly right wouldn't be too hard, but would you trust your safety to "mostly right"? To date, only one VTVL rocket vehicle has demonstrated fully autonomous takeoff, hover, and landing (John Carmack's vehicle over at Armadillo Aerospace). It ain't easy.
Also, don't forget you have to build the rocket motors and feed system and such. Most belts so far are peroxide monopropellants -- a good choice IMHO, but peroxide is hard to get and takes a lot of care to handle safely. And building any size rocket motor and ensuring it's safe enough to stand next to is a bit of work.
What I'm saying is, if you're a single amateur, or a small group, then building just the rockets is a big project unto itself. It shouldn't surprise you that no one has the time, money, and skills to do that, *plus* build and test the IMU, *plus* write fly-by-wire control software for it. If a modest sized startup company decided to pursue the matter, with a bit of financial backing, I would expect they could get it all built without too much hassle (provided they had the appropriate expertise in all areas, obviously).
Oh, and don't forget that your software has to handle a non-fixed CG if the person moves about much at all.
Very true. The difference is that with, say, a database reading from a disk, it might care about 200 bytes from that 4K sector. The disk could return something on the order of 128K or more with only a very minor performance hit in terms of IO operations/s. And unlike RAM, the odds of using those bytes, while increased, aren't that great in many apps. It's a tough problem whether in RAM or disks.
how to bring random-access storage to business computing?
Now, the question is how to best make use of the *non*-random-access storage that business computing has available? Most people think of hard disks as random access, but really they're not -- there's a huge performance penalty for random reads and writes. A disk that can do many tens of MB/s of sequential reads can only do maybe 200 4kB sector reads per second. That's a *huge* difference. So much so, that it's almost free to just read a bunch of sectors before or after the requested sector, in hopes they will get used. Companies providing high performance storage products struggle a great deal with figuring out how to avoid large numbers of small random IOs, and how to actually make use of the available bandwidth when the user is requesting small blocks at a time.
OK, I'll admit I don't understand how that cost is calculated.
Was that money spent on things it wouldn't otherwise have been spent on? IE, is it a marginal cost or just that day's "share" of a fixed cost? Did it have to get taken from some other budget (either within NASA or not)?
I'm just aware there's lots of stupid people on slashdot who don't actually think about things, and besides the real audience of a comment isn't the person you're responding to, it's the readers. Same things as televised "debates" -- perhaps better described as synchronized campaign speeches;)
Slashdot Mirror
Well, it's back-compatible with PCIe 1.0 and 1.1, so it aside from price there's on disadvantage to including it. Whether there's demand for it is another question, but I'm sure the graphics card makers will find something to do with it. Think back to AGP 2x/4x -- those made it onto cards and motherboards fairly quickly, iirc.
OK, fine, various weird dialects are correct then. But "correct" English (as defined by, say, your average skilled editor) is a very useful language to know. It helps a lot with communication and clarity, for preventing ambiguity and misunderstandings, and for creating the desired tone of voice (as perceived by the audience) in a piece of writing. So whether you want to call that "correct" English or "standard" English or "anal-retentive proscriptivist copy-editor" English, I don't really care -- but it's damn useful to have it and to have a relatively uniform definition of what it is and uniform education in it.
Language isn't just about conversing with those you interact with on a daily basis. If you want to communicate with people outside your immediate cultural group, it's awfully useful to have some standardization. I think this is what most people are driving at when they complain about bad English skills among native speakers. Since standardization implies some proscriptivism, someone has to do the proscribing. Perhaps the linguists should stop complaining about the proscriptivism, do some proscribing, and give the 7th grade English teachers some better rules to teach -- I'm sure any linguist will agree that your average grammar class does a fairly bad job of teaching grammar, even when compared to the language it's trying to teach the grammar of (as opposed to whatever dialect or form the people in the class normally use).
So I am forced to respectfully disagree with you and say that there is something wrong with African American Vernacular English -- not in any abstract, objective sense, and not when communicating with others who are fluent in it, but in using it to attempt to communicate with those who aren't fluent in it, or in using it in situations where it isn't socially expected. And plenty of people do exactly that. I know several people who are effectively bilingual in it and "standard" English, and quite fluidly shift between the two as appropriate to the situation. That seems entirely appropriate. But expecting other people to speak your language in a social situation where the norm is for you to speak theirs is, at best, rude -- be it ignorant American tourists abroad or illegal Mexican immigrants in the US.
I guess I really want a pair of things -- I want to be able to only have some of the labels / tags show up in the box on the left, and I want a quicker way than the drop down menu to add labels / tags to a message. (Like a text box with autocomplete like slashdot tags in addition to the drop down menu. Perhaps the menu should only show the ones in the box on the left.)
No, I'm saying you can't do the switch overnight. Not shouldn't, can't. The installed base is huge. And the installed base won't change overnight, even if everyone agreed to start using metric all at once. I think metrication is a good idea and worth doing, but it can't be done by fiat. It has to be a gradual process, and it will be a painful one no matter what.
Here we basically do all our calculation in SI units, but tend to convert to English at either end. So if the customer wants a 1000 lb thrust engine, we convert that to Newtons, do the design, and then end up with an engine whose dimensions are specified in inches. The other thing we do is rigorously label units in all calculations. You thought it was annoying when your physics professor counted off for not writing down units -- here we care for very practical reasons ;) The other reason we do this is that growing up around English units, they're much more intuitive. Telling me a line is at 1500 psi is much more informative than telling me it's at 10MPa.
Oh, and for plumbing, you actually need to stock a triple set if you're stocking both English and Metric -- you also have to stock the English / Metric adapters, because there's no way you'll never find a need to connect the two together.
In the abstract, I think metrication is a good thing. And it's well worth doing. But it's not easy, and it'll take a while, and that's a fact worth paying attention to.
Metric is good for all the obvious reasons -- SI units haver fewer weird things going on, conversions are easier, interoperable tools and fittings, etc etc. For all things like discussing distances, velocities, thrust levels, trajectory simulations, and more, I'm completely in favor of metric everywhere.
The one place I don't like this is when it comes to fittings, fasteners, plumbing, etc. Partly it's that metric nuts, bolts, and fittings are harder to find. You can't buy metric pipe fittings around here. Sure, you can order them, but that takes longer and costs more. The cost isn't a big issue on most things, but turnaround time is -- if you find a problem, it's really nice to be able to order a different part and have it the next day, rather than waiting a few days for something from Europe to clear customs and arrive. On some things, though, it actually makes a big difference. A lot of things like large pressure regulators, specialty valves, and more are even harder to find with metric fittings on them -- specifically, they become custom parts, with associated cost increases and weeks of lead time, which is frequently unacceptable.
And before anyone says you can buy metric parts in the US -- sure, you can, as long as they're "normal." It's the specialty parts that are hard. For example, McMaster-Carr stocks 3798 different socket cap screws in English sizes, but only 1610 in metric. If you need a weird metric screw, you may very well be out of luck.
The other major thing is subcontracts -- if I hire a consultant or send a part out to be machined, the machinist needs to have metric tools. Again, most machinists have a basic set of metric tools, but not an entire shop's worth. If the consultant or machinist has to start buying new tooling, your costs and the delivery time start going up.
I'll say it again -- having to buy parts from out of the country is not just a minor nuisance; it has a direct impact on how quickly you can revise a design and do the next test, which directly translates into how long it takes to complete the project.
I'm in favor of working toward compatibility, but it's not nearly as obvious an answer as it looks when it comes to tooling, since the installed base of English tooling and suppliers is *so* *huge* while metric is really only supported because of a few foreign-made parts.
Actually, I basically agree with you -- it's impossible to prove you have the requisite reliability. Therefore, no one will fly with ground based control. But mandating methods instead of results is always bad regulatory policy, because fundamentally the public doesn't care *how* you guarantee to fly safely enough, just *that* you do so. In this case, I can't see any arguments against an onboard pilot -- but that doesn't mean I should assume there are none. There are plenty of instances of other regulations that mandated methods, which resulted in less safe operation as soon as those methods became outdated. This policy is basically well-written.
So, at least for this first generation of vehicles, there's no real worry about space collisions. These rules are more aimed at dealing with things like participant awareness of the risks, and protecting the uninvolved public. Both very important things, and fortunately AST (the branch of the FAA in charge of space flight stuff) is taking a very sane and reasonable approach to most of this.
While I haven't read this version, I have read some of the drafts, and talked with people intimately involved in the industry and the regulation drafting. You're a bit off base here -- taken in context, what they're actually saying is that such links are fine, as long as the operator can prove they're safe enough. That is, the burden of proving the reliability of the system (including airframe, propulsion, controls, avionics, telemetry links, etc) is on the operator. They won't reject your plan just because you use ground-based controls, you just have to meet the same level of safety as everyone else. Seems eminently reasonable to me. (Also note that you will be required to do a detailed FMEA (Failure Modes and Effects Analysis), that addresses things like common-mode failures.) This is all as it should be -- regulating agencies shouldn't mandate techniques, they should mandate results. Rather than specify how you should control the craft, or whether you should burn LOX/kerosene or tetroxide/hydrazine, they just mandate that you fly safely enough, and demonstrate to their satisfaction that you can do so.
While I know you're not being serious, something like back Playboy issues has cultural relevance and value, even if it's not exactly "literature." The same goes for plenty of other magazines; Rolling Stone back issues, anyone? How about the comics pages of your daily newspaper? Entertainment material has a lot of cultural value to it.
ANFO makes a far superior terrorist explosive. If you want a delivery system, rockets are really quite bad -- they're hard to control and have a relatively short range. Model airplanes are far superior in most respects. If you're fighting a guerilla war, then cheap rockets become interesting -- see Qassam rockets in the middle east. Which, while they bear a few superficial similarities to high power hobby rockets, are really much better suited to the task than the HPR stuff anyway.
We're not building weapons, and we police our own -- even joking around about such things would be highly frowned on at a launch. Why should this be regulated any more or any differently than model airplanes? After all, model airplanes use nitromethane, which was actually used in the Oklahoma City bombing.
Just because it's a rocket doesn't mean it's ok to regulate it out of some knee-jerk "OMG Terrorists!" reaction.
Sorry about that.
Stop being an ass. I have good, directly relevant, second-hand knowledge of the subject matter. The other poster has indirectly relevant second-hand (if he's not writing the control code personally, it's second-hand for this discussion) knowledge of unknown (to me) quality (could be good, bad, very good -- I don't know). My information was presented to me as fact by people who certainly *ought* to know, seeing as they were investigating the exact question at hand. Hence I repeated it as fact, but am willing to admit that it could be out of date or in error. Now what, exactly, is the problem?
A rocket has no such luxury -- measured acceleration comes only from the rocket thrust plus any aero loads; you can't "see" the gravity vector. Alternately, imagine the engine is supplying 1G worth of thrust. Your accelerometers "see" 1G accel away from the engine. Without the gyros, you have no way of telling if oyu're vertical and hovering, or at 45 degrees, falling, and accelerating sideways. You can make guesses based on air speed and GPS and such, but it's much harder -- hence my comment that using cheap gyros depends on your ability to include other sensor data. If, on the other hand, you have an expensive FOG, you can just ask it for your current orientation and get a meaningful answer.
I do know that at least until recently, inexpensive gyros were completely unusable. Modern ones appear better, but my sources suggest that they aren't all the way there yet. It's possible there are good gyros out there that haven't been tried in this application and that I haven't heard about; I'm not averse to admitting my knowledge could be out of date or incomplete. I am, however, quite confident that the problem is not so easy as the OP implied.
And for reference, IAARS, and I have read detailed discussions by people doing actual investigation with real hardware testing of exactly this problem -- and I'd say that makes me better qualified to comment than the vast majority of posters. But, like I said, not infallible -- so please don't jump on me for admitting my fallibility.
I don't have a good refernce handy, but I believe the answer is no. I've heard my original comment made by people in the industry who were well aware of the DC-XA, so I'm inclined to believe it's true, but I don't know enough about it to give an authoritative answer.
I suppose you could use the inexpensive ones, as long as your goal was to change the pilot requirement from "top of the line test pilot" to "very good helicopter pilot," and not an attempt to make it flyable by anyone with a bit of simulator practice.
You might do an ok job if the gyros just tried to hold the spin *rate* to zero, and let the pilot handle leveling the vehicle; one fewer integral makes for much slower error growth.
Well, in that case, it's different because skateboards don't have pressurized tanks of propellant that's dangerous to get on your skin, don't have multiple pieces of complex machinery, all of which is required to operate as designed in a fairly harsh environment in order to ensure your continued survival, and furthermore are fundamentally stable (if you get on a skateboard and start moving slowly, on relatively flat terrain, it has no natural tendency to dump you on your face from 20 feet in the air).
Oh, and don't forget that it's an awful lot easier to safely test your new skateboard design. When testing is hard, as with a rocket belt, development is slow and expensive.
And for the record, IAARS -- and rocket belts scare me. A lot. I'd be much more inclined to get in a rocket powered airplane, go cart, or suborbital rocket.
Also, don't forget you have to build the rocket motors and feed system and such. Most belts so far are peroxide monopropellants -- a good choice IMHO, but peroxide is hard to get and takes a lot of care to handle safely. And building any size rocket motor and ensuring it's safe enough to stand next to is a bit of work.
What I'm saying is, if you're a single amateur, or a small group, then building just the rockets is a big project unto itself. It shouldn't surprise you that no one has the time, money, and skills to do that, *plus* build and test the IMU, *plus* write fly-by-wire control software for it. If a modest sized startup company decided to pursue the matter, with a bit of financial backing, I would expect they could get it all built without too much hassle (provided they had the appropriate expertise in all areas, obviously). Oh, and don't forget that your software has to handle a non-fixed CG if the person moves about much at all.
Very true. The difference is that with, say, a database reading from a disk, it might care about 200 bytes from that 4K sector. The disk could return something on the order of 128K or more with only a very minor performance hit in terms of IO operations/s. And unlike RAM, the odds of using those bytes, while increased, aren't that great in many apps. It's a tough problem whether in RAM or disks.
Now, the question is how to best make use of the *non*-random-access storage that business computing has available? Most people think of hard disks as random access, but really they're not -- there's a huge performance penalty for random reads and writes. A disk that can do many tens of MB/s of sequential reads can only do maybe 200 4kB sector reads per second. That's a *huge* difference. So much so, that it's almost free to just read a bunch of sectors before or after the requested sector, in hopes they will get used. Companies providing high performance storage products struggle a great deal with figuring out how to avoid large numbers of small random IOs, and how to actually make use of the available bandwidth when the user is requesting small blocks at a time.
The Mouse isn't patented. It's copyrighted. There's a difference.
I suppose it could well be difficult to pump out the LH2 and actually recover enough of it to be relevant... that stuff is *really* hard to handle.
Now, to be fair, it's not (entirely) unreasonable to ignore the LOX costs... NASA pays a few pennies a pound for it, and it's worlds easier to handle.
Was that money spent on things it wouldn't otherwise have been spent on? IE, is it a marginal cost or just that day's "share" of a fixed cost? Did it have to get taken from some other budget (either within NASA or not)?
NASA accounting always confuses me.
I'm just aware there's lots of stupid people on slashdot who don't actually think about things, and besides the real audience of a comment isn't the person you're responding to, it's the readers. Same things as televised "debates" -- perhaps better described as synchronized campaign speeches ;)