Actually, I do (well, I personally don't, but folks on my team do). My gizmo does RF from DC to 3GHz, down to -120dBm, so noise is a concern. I'm guessing you do image processing, and are looking at a CCD, so you can at least bandwidth limit your signal. Wish I had that luxury.
Could be worse. My brother worked on CCD gear that would be fried by an ESD of 35 volts. LHe2 cooled, fun stuff.
First of all:
Sometimes you have to run other people's code, and you may not have control over what they targeted.
Second of all:
The cost of a ready built X86 CPU board is a fraction of the cost of a ready build board with the other chips.
Third of all:
For a low-volume house, the cost of building your own CPU board is rediculous.
Forth of all:
Most of your "designed for embedded" CPUs don't have an FPU. The embedded PowerPC, the StrongArm, and (IIRC) the SuperH and embedded MIPS don't have FPUs. If you need number crunching power, you need to go elsewhere.
Unfortunately, some of us work under all those constraints (also, time to market, availability of second sources, and such).
I suggest that if you are going to tell somebody with decades of professional embedded experience how to do his job, you should try to have a little yourself.
I've been eagerly awaiting one of the Single Board Computer (SBC) venders to make a device with a Crusoe on it. I work for a company that makes embedded devices, and currently we are using a Pentium/MMX based system at 200 MHz. Even that CPU board sucks down about 25 to 50 watts, which is not good when you are trying to make a system that needs to run from a lighter socket (ca. 120 watts maximum, more like 80 under real conditions.) Plus, you have to get rid of that heat, which adds to the cooling burden, and the @#()$& CPU fans keep crapping out on us. This is on an industrial SBC costing about US$700.
I'd love to go to a Crusoe based system, pick up a few MIPS, and cut a few watts out of the power budget. Add to this the fact that the Crusoe has the North Bridge built-in, which reduces the size of the board, and you have a great win all around for us embedded types.
However, unlike the laptop market, x86 compatibility isn't as great a deal for us embedded weenies, and therefor the StrongArm XScale is an attractive option too. Transmeta had better make this a very compelling option.
I find it interesting that the W3C is saying "Why didn't we publisize this? We did, you just didn't hear it!"
I think this DOES show the power of public outcry - had/. not publisized this, it would have quietly slipped away like the DMCA did, and they would have said "but nobody complained at the time!" Instead, they got hammered and have had to back-pedal a bit.
Just remember, don't just bitch on/. - bitch to the people that matter.
Of course, the gesture command for dismissing annoying pop-up ads should be obvious, depending upon whether you are using LOCALE=en_US or LOCALE=en_UK.
You keep bitching about not having any bandwidth. Get VA to terminate a T1 at your house, with the other end terminated on the inside of the firewall at the/. server cage. You have 1.5Mb bandwidth, you can admin/. without fear, and you and VA can write it off as a business expense.
These devices work off the MOVING of heat from place to place, not absorbing the heat (that would be a violation of thermodynamics - you have to reject some heat to the cold side of a device).
Actually, if you want to do this, just buy a Peltier module (or buy a Coleman electric cooler and rip the module out of it). Make one side hot, and one side cold, and it will make electricity.
Some people have said, "Put this on top of your Atlon|Pentium and you can make electricity." True, but a bad idea - the module will act as a thermal resistance, preventing maximum heat from your CPU. Result - one cooked chip.
The amount of electricity generated by these things is pathetic - you'd be better off buying a World Radio, ripping the crank generator out of it, and using that to power your toys.
If you do embedded software development, I urge you to fight this! Here's the comment I sent in:
As an embedded software developer, I find this proposed shift in the policies of the W3C on patent-encumbered standards quite disturbing. Some of the projects I work on have embedded HTML browsers, and due to the environment in which they are used I cannot always turn to a commercially provided solution for what I need to do. If I have to patent check every feature I try to place in my equipment, then I will simply not be able to place any new features in my equipment. And as the Web moves from standards-based documents (HTML) to non-standard documents (PDF, MS-DOC file, etc.), the utility of the Web as a platform to be built upon will be diminished.
Please, reconsider this peril-laden action. Require ALL W3C standards to be either completely patent unencumbered, or require all patents to be released for use by anybody without royalty for the purposes of implementing a W3C compliant app.
I see this crap every day of work: I cannot turn around without stumbling over a patent license that prevents me from supporting a standard. I work in the land mobile radio market. Look at the big player in this market: Motorola. They own most of the patents on the standards, and so they can pretty much prevent any competitor from gaining a foothold in the market. If you want to have a system that works correctly, you buy Motorola repeaters, Motorola consoles, Motorola mobiles, and Motorola test equipment. Try to integrate a Racal, E.F. Johnson, or Kenwood system, and all the places Big M violated the published standard break everything.
Open standards aren't just a good idea, they are the ONLY WAY to make a system that everybody can play in.
Thanks for the insight, that explains much. Of course, with folks like me the effect is quite the opposite of what is desired ( I change the station to get away from the DJ's patter).
You forget that big Mediawants us to be vapid, shallow, and stupid. If we are constantly reminded of the past, that the past even exists, then we cannot be as shallow as they wish us to be.
Remember in 1984, Animal Farm, and Fahrenheit 451, how utterly the Powers That Be controlled the past, and ruthlessly erased anything that might cause thought on the part of the masses? That is what is going on right now: if we remove the WTC from all history books, then we can forget this ever happened, and go back to being happy little sheeple. After all, the chocolate ration was just raised to 5g/week...
A question for the audience: are DJ's around the world as bad are they are here (Wichita, KS)? Specifically, these morons seem to think we tune in to hear their lips flap - they talk over the intro to the song, right up to (and often past) the point where the lyrics start, and then start yapping at the end of the song.
Is this just stupidity, vanity, or some vague and low-tech form of copy protection? Any DJ's out there that can answer?
It is so bad here that I no longer listen to the radio for any length of time (just in the mornings, for about 2 songs, on my clock radio. That way, the annoyance factor of the DJ's helps me wake up). I listen entirely to my own music collection the rest of the time.
These idiots seem to forget that if we aren't listening, we aren't pumping up their ad revenues. Of course, this also makes it hard to hear new music.
You see, in the universe I am from, we too have organizations called ZDNET and Microsoft. However, in my universe, ZDNET has always been anxious to please Microsoft any way they could, to the point of fawning over any MS release.
Reading the link from the article, I see things are different in this universe - I cannot imagine stronger commentary against Microsoft. If anybody can give me pointers on how to get back home....
On second thought, my credit cards work, my Slashdot account works, and this universe seems to be just a bit better. Maybe I'll stay.
By that logic, metal detectors are a lousy system. Anecdotally, at least 50% of the passengers trip off the metal detector. Note that it's not there to detect metal, but weapons. If 1 in 1000 people are carrying weapons, then the metal detectors are giving 500 false positives per 1000 people.
However, if you trip the metal detector, it is a simple matter of asking you to back up, place any metal items in the bowl, and step through again. The cost of rechecking a false positive is very low.
Constrast that to the cost of verifying a false trip of a biometric system: You have to detain the individual, and at a minimum fingerprint them, and run the prints against the individual the system thought they were.
Now, it's not that hard to make your fingerprints NOT match those on file: sandpaper them off, cut your fingertips, smudge the card/sensor. In those cases, it becomes very easy to convert a true positive into a "false false-positive" - a terrorist being fingered, then BSing his way out of it.
I've seen several comments that "If the system gives a false positive only 1 in 1000 times, then it must be pretty good!". This demonstrates that many people have no clue about how to properly apply probability - what is called Baysian math.
You have to start out with two probabilities that are based on the system: probability of a false positive (Pp) and probability of a false negative (Pn).
A false positive is mis-identifying a non-terrorist as a terrorist. Let us say that a collection of 1 million non-terrorists are run through a system, and it fingers one of them as a terrorist. That system has a Pp of 1 in a million, or 1E-6.
A false negative is mis-identifying a terrorist as not being a terrorist. Let us say that we run a thousand known terrorists through the system, and let us say that only one is not detected. Then this system has a Pf of 1 in a thousand, or 1E-3.
Now, that is ALL that you can say about a system. You cannot state the actual number of false positives vs. the number of false negatives in real use without an additional piece of data, the probability of any given person in a crowd being a terrorist, Pt. Let us say that in any given crowd, one in ten thousand people are terrorists (Pt = 1E-4). This may seem very high, but the lower Pt, the worse the system will perform, and I am heavily weighting this in favor of the face scanner.
Now, let's run a million random people through the system, and see what happens.
First, out of that million people, 1E6 * Pt = 1E6 * 1E-4 = 1E2 = 100 of them are terrorists. We would expect that of that 100 terrorists, 100 * Pf = 100 * 1E-3 =.1 terrorist will be mis-identified. So we will assume that all 100 of the terrorists trip the alarm.
Now, out of the remaining 999,900 people, we would expect the system to finger 999,900 * Pn = 99,900 * 1E-6 =.9999, so we will assume that one innocent person gets fingered as a terrorist.
Now, we had 101 trips, of which 1 was false, so the odds that you aren't a terrorist given that you were fingered are just under a percent. That's given the assumption that the system mis-identifies innocent people only one in a million times, and assuming that one person in ten thousand is a terrorist. Increase the false positive rate by a factor of ten (one in one hundred thousand innocents gets fingered), and decrease the terrorist population to a tenth of what we assumed (one terrorist in one hundred thousand) and you now have roughly fifty-fifty odds that a person fingered by the system is innocent.
And that, people, is why systems like this don't work.
Remember, to a Vulcan, any Terran normal environment would be COLD. Personally, I find it refreshing that the show's writer's paid such meticulous attention to detail in this matter.
There is a simple out the writers could use, that would not only resolve the speed issues, but allow them to introduce the "Stardate" system in a very sensible fashion. (care to lay odds that they actually figure this out?)
Just state that there is still some residual relativistic effects even in warp. Have Cmdr. Tucker making a comment along these lines:
Computer, begin log entry:
Cmdr. Tucker, 9 Sept. 2153 (ship time)
Computer, pause log.
Y'know, this is weird. It'd Sept. 9th on the ship, but it is already New Years back home. We're going to have to have some way to have a calendar that is consistent between warp ships and home.
Computer, resume log.
(Question: why does the person recording a log have to give the date: doesn't the log entry get its own timestamp? What, the ship's computer isn't running NTP?)
This would allow them to say that it took eighty hours to get to Khronos ship time, while having it be several weeks Earth time. That way, anytime the writers forget how really, really, amazingly, mind boggling huge space is, they can use this explaination to correct their mistake.
Actually, it wasn't the handoffs that killed the system. When the first site failed, it killed all of the sites that were part of that network via the trunk lines connecting the sites. The trunk lines are how the sites communicate hand-off messages, look up subscriber data, etc.
There was a bug in the site firmware, and so when it overloaded, it started sending garbage over the trunks.
There was the experiment that was done in the early days of cellular telephony here in the US. A guy filled a football stadium-sized parking lot with people, loaned them all cell phones, had them dial, and then everybody pressed SEND at the same time. The point he was making was that the cellular system wouldn't handle such an event gracefully.
It didn't. It crashed not only the local cell site, but several others as well.
Of course, this is in a microcosmn what happens when there is a big "event" such as an earthquake or bombing - everybody tries to call Aunt Phillis and the cell network goes down. That's when we hams get busy....
Slashdot Mirror
Actually, I do (well, I personally don't, but folks on my team do). My gizmo does RF from DC to 3GHz, down to -120dBm, so noise is a concern. I'm guessing you do image processing, and are looking at a CCD, so you can at least bandwidth limit your signal. Wish I had that luxury.
Could be worse. My brother worked on CCD gear that would be fried by an ESD of 35 volts. LHe2 cooled, fun stuff.
I was referring to the CPU board being second sourced, not the CPU itself. Unfortunately, it seems the days of the second sourced CPU are over.
Well, I get a little tired of the average /.er expounding upon items they have little comprehension of.
But, the whole thing that makes being an engineer "challenging" is the fact that the definition of "better" is a multidimensional equation.
You are correct: I have 2 YIG oscillators burning 15W each, so the CPU is just one of my problems.
First of all:
Sometimes you have to run other people's code, and you may not have control over what they targeted.
Second of all:
The cost of a ready built X86 CPU board is a fraction of the cost of a ready build board with the other chips.
Third of all:
For a low-volume house, the cost of building your own CPU board is rediculous.
Forth of all:
Most of your "designed for embedded" CPUs don't have an FPU. The embedded PowerPC, the StrongArm, and (IIRC) the SuperH and embedded MIPS don't have FPUs. If you need number crunching power, you need to go elsewhere.
Unfortunately, some of us work under all those constraints (also, time to market, availability of second sources, and such).
I suggest that if you are going to tell somebody with decades of professional embedded experience how to do his job, you should try to have a little yourself.
And this is why I can justify reading /. at work. Thanks for the link.
I've been eagerly awaiting one of the Single Board Computer (SBC) venders to make a device with a Crusoe on it. I work for a company that makes embedded devices, and currently we are using a Pentium/MMX based system at 200 MHz. Even that CPU board sucks down about 25 to 50 watts, which is not good when you are trying to make a system that needs to run from a lighter socket (ca. 120 watts maximum, more like 80 under real conditions.) Plus, you have to get rid of that heat, which adds to the cooling burden, and the @#()$& CPU fans keep crapping out on us. This is on an industrial SBC costing about US$700.
I'd love to go to a Crusoe based system, pick up a few MIPS, and cut a few watts out of the power budget. Add to this the fact that the Crusoe has the North Bridge built-in, which reduces the size of the board, and you have a great win all around for us embedded types.
However, unlike the laptop market, x86 compatibility isn't as great a deal for us embedded weenies, and therefor the StrongArm XScale is an attractive option too. Transmeta had better make this a very compelling option.
I find it interesting that the W3C is saying "Why didn't we publisize this? We did, you just didn't hear it!"
/. not publisized this, it would have quietly slipped away like the DMCA did, and they would have said "but nobody complained at the time!" Instead, they got hammered and have had to back-pedal a bit.
/. - bitch to the people that matter.
I think this DOES show the power of public outcry - had
Just remember, don't just bitch on
Of course, the gesture command for dismissing annoying pop-up ads should be obvious, depending upon whether you are using LOCALE=en_US or LOCALE=en_UK.
You keep bitching about not having any bandwidth. Get VA to terminate a T1 at your house, with the other end terminated on the inside of the firewall at the /. server cage. You have 1.5Mb bandwidth, you can admin /. without fear, and you and VA can write it off as a business expense.
These devices work off the MOVING of heat from place to place, not absorbing the heat (that would be a violation of thermodynamics - you have to reject some heat to the cold side of a device).
Actually, if you want to do this, just buy a Peltier module (or buy a Coleman electric cooler and rip the module out of it). Make one side hot, and one side cold, and it will make electricity.
Some people have said, "Put this on top of your Atlon|Pentium and you can make electricity." True, but a bad idea - the module will act as a thermal resistance, preventing maximum heat from your CPU. Result - one cooked chip.
The amount of electricity generated by these things is pathetic - you'd be better off buying a World Radio, ripping the crank generator out of it, and using that to power your toys.
w.r.t. the pile of CDs in the car, do what I did and buy an MP3 player, like this.
It's a whole lot better than a pile of CDs in the trunk...
I see this crap every day of work: I cannot turn around without stumbling over a patent license that prevents me from supporting a standard. I work in the land mobile radio market. Look at the big player in this market: Motorola. They own most of the patents on the standards, and so they can pretty much prevent any competitor from gaining a foothold in the market. If you want to have a system that works correctly, you buy Motorola repeaters, Motorola consoles, Motorola mobiles, and Motorola test equipment. Try to integrate a Racal, E.F. Johnson, or Kenwood system, and all the places Big M violated the published standard break everything.
Open standards aren't just a good idea, they are the ONLY WAY to make a system that everybody can play in.
Thanks for the insight, that explains much. Of course, with folks like me the effect is quite the opposite of what is desired ( I change the station to get away from the DJ's patter).
You forget that big Media wants us to be vapid, shallow, and stupid. If we are constantly reminded of the past, that the past even exists, then we cannot be as shallow as they wish us to be.
Remember in 1984, Animal Farm, and Fahrenheit 451, how utterly the Powers That Be controlled the past, and ruthlessly erased anything that might cause thought on the part of the masses? That is what is going on right now: if we remove the WTC from all history books, then we can forget this ever happened, and go back to being happy little sheeple. After all, the chocolate ration was just raised to 5g/week...
A question for the audience: are DJ's around the world as bad are they are here (Wichita, KS)? Specifically, these morons seem to think we tune in to hear their lips flap - they talk over the intro to the song, right up to (and often past) the point where the lyrics start, and then start yapping at the end of the song.
Is this just stupidity, vanity, or some vague and low-tech form of copy protection? Any DJ's out there that can answer?
It is so bad here that I no longer listen to the radio for any length of time (just in the mornings, for about 2 songs, on my clock radio. That way, the annoyance factor of the DJ's helps me wake up). I listen entirely to my own music collection the rest of the time.
These idiots seem to forget that if we aren't listening, we aren't pumping up their ad revenues. Of course, this also makes it hard to hear new music.
You see, in the universe I am from, we too have organizations called ZDNET and Microsoft. However, in my universe, ZDNET has always been anxious to please Microsoft any way they could, to the point of fawning over any MS release.
Reading the link from the article, I see things are different in this universe - I cannot imagine stronger commentary against Microsoft. If anybody can give me pointers on how to get back home....
On second thought, my credit cards work, my Slashdot account works, and this universe seems to be just a bit better. Maybe I'll stay.
However, if you trip the metal detector, it is a simple matter of asking you to back up, place any metal items in the bowl, and step through again. The cost of rechecking a false positive is very low.
Constrast that to the cost of verifying a false trip of a biometric system: You have to detain the individual, and at a minimum fingerprint them, and run the prints against the individual the system thought they were.
Now, it's not that hard to make your fingerprints NOT match those on file: sandpaper them off, cut your fingertips, smudge the card/sensor. In those cases, it becomes very easy to convert a true positive into a "false false-positive" - a terrorist being fingered, then BSing his way out of it.
I've seen several comments that "If the system gives a false positive only 1 in 1000 times, then it must be pretty good!". This demonstrates that many people have no clue about how to properly apply probability - what is called Baysian math.
.1 terrorist will be mis-identified. So we will assume that all 100 of the terrorists trip the alarm.
.9999, so we will assume that one innocent person gets fingered as a terrorist.
You have to start out with two probabilities that are based on the system: probability of a false positive (Pp) and probability of a false negative (Pn).
A false positive is mis-identifying a non-terrorist as a terrorist. Let us say that a collection of 1 million non-terrorists are run through a system, and it fingers one of them as a terrorist. That system has a Pp of 1 in a million, or 1E-6.
A false negative is mis-identifying a terrorist as not being a terrorist. Let us say that we run a thousand known terrorists through the system, and let us say that only one is not detected. Then this system has a Pf of 1 in a thousand, or 1E-3.
Now, that is ALL that you can say about a system. You cannot state the actual number of false positives vs. the number of false negatives in real use without an additional piece of data, the probability of any given person in a crowd being a terrorist, Pt. Let us say that in any given crowd, one in ten thousand people are terrorists (Pt = 1E-4). This may seem very high, but the lower Pt, the worse the system will perform, and I am heavily weighting this in favor of the face scanner.
Now, let's run a million random people through the system, and see what happens.
First, out of that million people, 1E6 * Pt = 1E6 * 1E-4 = 1E2 = 100 of them are terrorists. We would expect that of that 100 terrorists, 100 * Pf = 100 * 1E-3 =
Now, out of the remaining 999,900 people, we would expect the system to finger 999,900 * Pn = 99,900 * 1E-6 =
Now, we had 101 trips, of which 1 was false, so the odds that you aren't a terrorist given that you were fingered are just under a percent. That's given the assumption that the system mis-identifies innocent people only one in a million times, and assuming that one person in ten thousand is a terrorist. Increase the false positive rate by a factor of ten (one in one hundred thousand innocents gets fingered), and decrease the terrorist population to a tenth of what we assumed (one terrorist in one hundred thousand) and you now have roughly fifty-fifty odds that a person fingered by the system is innocent.
And that, people, is why systems like this don't work.
Remember, to a Vulcan, any Terran normal environment would be COLD. Personally, I find it refreshing that the show's writer's paid such meticulous attention to detail in this matter.
Just state that there is still some residual relativistic effects even in warp. Have Cmdr. Tucker making a comment along these lines:
(Question: why does the person recording a log have to give the date: doesn't the log entry get its own timestamp? What, the ship's computer isn't running NTP?)
This would allow them to say that it took eighty hours to get to Khronos ship time, while having it be several weeks Earth time. That way, anytime the writers forget how really, really, amazingly, mind boggling huge space is, they can use this explaination to correct their mistake.
Does anybody else find it odd that we've heard nothing from Senator Clinton on this?
The mouse mentioned in the article is called "the RenMouse".
Does anybody else feel the need to protect his new mouse with a rubber walrus protector?
Of course, when it stops working you can always shout "YOU BLOATED SACK OF PROTOPLASM!" at it.
I also suppose they will be bringing out the StimpyPedals as an accessory.
Actually, it wasn't the handoffs that killed the system. When the first site failed, it killed all of the sites that were part of that network via the trunk lines connecting the sites. The trunk lines are how the sites communicate hand-off messages, look up subscriber data, etc.
There was a bug in the site firmware, and so when it overloaded, it started sending garbage over the trunks.
Exactly what do they think they mean by 75,000 tons of energy? Do they mean
<sarcasm>
Gosh, I do so love reporters and editors who are right on top of sceince <sic>
</sarcasm>
There was the experiment that was done in the early days of cellular telephony here in the US. A guy filled a football stadium-sized parking lot with people, loaned them all cell phones, had them dial, and then everybody pressed SEND at the same time. The point he was making was that the cellular system wouldn't handle such an event gracefully.
It didn't. It crashed not only the local cell site, but several others as well.
Of course, this is in a microcosmn what happens when there is a big "event" such as an earthquake or bombing - everybody tries to call Aunt Phillis and the cell network goes down. That's when we hams get busy....