You can use your cellphone. Those tryiing to actually run the emergency services would appreciate the ability to communicate outside of the area, though. And HF radio is MUCH more reliable than cell, and people can talk from different states, which is really key to a situation like Katrina.
Network communications can actually be located in TX, AL, AR, etc, or based out in the Gulf on a boat, and easily contact people with stations setup in NOLA on generator/battery power.
The way the system works in the US is that the phone manufacturers develop a phone, and work with the network to produce a phone specifically for that network. Then the carrier buys all the phones from the manufacturer, and the carrier sells the phones to the end-user, usually at some massive discount from "retail". For instance, Sprint takes $150 off the price of a new phone in exchange for a 2-year contract, Verizon does something similar.
The GSM carriers even try to "lock" the handsets to the original carrier, but that appears to be easily breakable.
In the US the carriers sorta compete, but mainly have each hewed out their own markets. Instead of competing within the same overall market with phone manufacturers and service providers, instead there are about 5 major markets, each ruled by a carrier, and then the markets compete for each other.
It's screwed up and bizzare.
(I write software that ends up on shipping phones for various different CDMA carriers)
"hollow" or "empty" would have been my psuedo-audiophile term of choice. It think it has a lot to do with the subwoofer channel usage, and how the mix is directed to the LFE channel vs. the main L/R, and possibly also the fact that most center-channel speakers don't do low frequencies well, but a lot of content is centered, so it ends up needed to be produced there.
What I want to do is sit down with my stereo (has preamp outs) and record the outputs in different modes, and then compare the frequency contents of the different recordings, and see if the sound is actually "shaped" differently, or it just changes sound due to the speakers it's routed to.
The other issue I've noticed with the DVD-A formats (or std dvds of music), is that the surround mixed audio sounds distinctly "off" compared to the straight stereo mix of the same thing. I've especially noticed this on my Metallica S&M DVD and on a friend's Blue Man Group DVD. The sound changes radically, and on my system at home, it's still going through the same 2.1 setup either way.
I haven't sat down and figured out how the sound is engineered differently for the two systems, but I know it sounds a hell of a lot better on both my system and a friends (~$10K) system when left in stereo instead of in the surround modes.
The biggest issue I see with SACD and DVD-A is that the engineers don't know how to deal with it. HiDef is easy, as you can easily see all the added detail.
However, it seems like as time goes on, audio gets more and more dynamically compressed (not size-compressed, but dynamic range compression). I know the dynamic compressors were first used for radio, to make FM sound better (knock down the spikes in volume, bring up the volume of the quiet passages).
My modest stereo at home sounds decent, until I throw in a very well put together CD (which usually means it was simply recorded with a pair of stereo mics, and left alone by the engineer), and the result is just astounding. It sounds amazingly good.
SACD would be mind-blowing, but the material on the disc has to actually have the bandwidth to take advantage of it.
What's the advantage of DVD-Audio if the source material was already filtered to 20Khz before it was encoded at 96Khz. Or 24-bit dynamic range vs 16, when the source material has been compressed down to about 8 bits of range.
And all the surround modes that I've heard are utter crap. No nice evelopment of sound around the listener, but instead this very disjoint placement of sounds into the back channels.
My CE degree (mixed HW/SW) started programming at the assembly level, and worked up from their through C into Java.
By the time I was into abstract data types, I had a very good view of how the abstract data types would be implemented at the assembly/hardware level. How your choice of platform can greatly simplify things like semaphores, mutexes, and exceptions.
It seems the standard CS track is to start high-level, and slowly work your way down towards the hardware, and you might have a token class on machine architecture and assembly. I've watched my wife learn successively lower-level programming and watched the light-bulgs that go off as something clicks into place, how low-level hardware design ripples up into how languages are developed.
Now, when I'm programming in something higher-level like Java, C++, or ObjC, I still have that mental image of the registers and the stack, and what hoops will need to be jumped through by the compiler to call those virtual functions, or to handle RTTI and dynamic casts, etc.
Mostly, it's just an awareness of what the platform and hardware are doing, but I think it gives me a big edge over someone who only knows the packages made available by the high-level language, and doesn't understand how that function that they keep calling over and over again whose result value they could have cached can dig into their performance.
As we push back into the hills (I live up in the Santa Cruz Mtns), we get more and more into their territory, and as we initially knocked back their population, and then ended hunting of their prey (for all intents and purposes), now the deer population is rising, and with it, the cougar population. If you live along the edges of the hills in the South Bay, and you see deer, you've got cougars.
I've seen sign multiple times, and in 5 years, I've seen 3 total. They're around, but they don't make their prescence well known.
Wild pigs in my area are a serious menace. 400+ lbs, afraid of nothing, and they'll happily gore a person just because they were in the way.
When I was in HS, tape easily outstripped the size of the HD in the common home computer, for realtively cheap tape drives (we're talking 32-64MB drives here).
I've priced out various backup options for the 1TB media server I want to build. $$$$$$
Cheapeast actually appears to be just dumping the files to slow, big SATA drives sequentially. And that's still nearly $500 per backup copy.
Take will run you $1200 for a drive that can even get close to that, and then $50 a tape, but you need to span across several tapes, just like the drives. But at least then at twp copies you've made up the difference.
But the array itself costs nearly that much (including controller).
Scaling back for more "normal" home users, with say 250GB or so of files, they're options are either another drive, and to do it right, an external hot-swap cage and maybe 2-3 drives on rotation (so... $500?) Or to go the tape route again. But I've had a hell of a time finding decent places to research and price out the various current tape drives/formats.
The problem is that backups used to be easy. Spend maybe $100 down at the local 'puter shop, and take home a boxed drive with software and enough tapes to easily do backup rotations. That's just not the case anymore.
Yeah, they do. They're not cheap, either. Basically, you hook up your feed from your battery bank, and your mains connection from the power company (and optionally a generator), and then it figures it all out. Runs off the panels when it's bright out, as the panels are connected to the batteries, so as long as the panels are charging the batteries, it can supply current to the house from the panels, once the panels can't supply enough current to charge the batteries, the batteries start draining to the inverter (this is all controlled by the battery manager). Then the inverter drains down the batteries, and when they fall below a threshold, it starts pulling from the A/C mains to supplement, and recharges the batteries.
The inverter keeps itself fully synchronized off the A/C mains when they're present, and when they go away (as they often do in my area), then it uses it's own internal clock to keep the waveform correct. When mains is down, it stops backfeeding the mains (for safety of everything involved), and will recharge the batteries off the generator, if needed.
When mains comes back up, it slowly adjusts it's clock to match, and then switches over to running off the mains until the batteries are fully charged.
When an engine is properly designed to advance timing and boost pressure when running high octane fuel (like the 110 octane of E85), then you can get some surprising gains:
This saab 2.0L engine gains about 20-25% more power when running E85 than straight unleaded. With no loss of milage.
Ethanol is also a rather potent fuel-systems cleaner. In a car with a lot of straight gasoline mileage, this can be a problem as the varnish deposits from the fuel will start to break loose when the ethanol starts to dissolve them, and they can clog up fuel filters and injectors. After a few thousand miles of running ethanol blend, or after running an even stronger fuel deposit solvent through the engine, that problem is solved.
I found out the above in college when I moved, and had to start using 10% blend. Mileage tanked for 2-3 months, and then bounced back to where it was before.
I've never seen an inkjet print that can come close to the quality of a Lightjet print on archival photo paper.
My Lightjet print blown up to 16x20 from moderate quality print film look fantastic. Grain at that size is about as apparent as the best inkjet prints I've seen.
Step up to something like Fujichrome Velvia, or a custom image that's been mastered at the 403ppi native resolution of the printer, and it blows all the ink-based prints out of the water.
I've found that smart-pointers give me the ease of programming of a garbage collector (especially when paired with something like STLPort's allocator), but still gives me explicit control over the destruction of objects.
Java doesn't have a concept of a destructor that runs *NOW* (or not one that I'm aware of). And an object created on the stack can't do things in it's destructor like immediately freeing all references to non-memory resources that it's using (transactions, pipes, files, etc.)
The telephone bundles are impressively strong cables, power lines are under a lot of stress, and pulled much tighter (because they can't sag, because they're usually running at 20KV or so, and need to be kept about 10' from everything).
When a tree falls over, and crushes all the lines to the ground, the power lines snap, the telephone lines get striped from the nearest pole, but are probably holding up the tree. They're fantasticly strong.
Cable is usually ran under the telephone bundles, and as it's a big coax network, a lot easier to repair if it does get damaged (one wire vs. 1000s).
I'm a volunteer FF, so I see this a lot in the winter here in the Santa Cruz Mtns. Trees fall over in the storms, we close the road and wait for PG&E, and the telephone lines, holding up the fallen tree, are fine.
I try not to call myself that, but I know I fall into the category.
But if you ask someone who claims to have a "golden ear" (I don't) if they can listen to something that has a 140+dB dynamic range, and actually be able to hear both the quietest and the loudest passages (without pain), I doubt they'd say yes.
If you set the upper limit at 130dB (threshold of pain), the lower limit would be -14dB or so, roughly 20x quieter than a barely audible whisper in a dead quiet anechoic chamber (threshold of hearing). Dogs? yes, they can probably hear it They have larger ears to catch the sound.
And I've heard of audiophiles claiming to hear into the low 20Khz range, but never of one claiming to hear above 30Khz or so. Dogs? Yes, I know that they can.
It's the sample rate, moreso to me than the bits per sample. 44.1KHz's nyqust freq (22.05Khz) is close enough into the range of human hearing that artifacting can easily occur. You need VERY good filtering on A/D inputs to have sounds above say 18KHz still present, and the filter level be inaudible at 22Khz (greather than 60dB of attenuation in about a 1/4 octave, so 240dB of attentuation per octave. That's not a simple filter).
By upping the sample rate to 96Khz, the nyquist frequency is so high (48Khz), that the filter requirements aren't nearly so high, and that any artificats that do creep in do so at frequencies well beyond our hearing.
Even on modest gear, it makes a difference.
And by modest, I'm putting the system price at say $2000 (receiver, dvd player, speakers).
Granted, the D/A stage doesn't need these expensive filters that the recording stage does, but it gives the recorders a lot more lattitude to make GOOD recordings. But mass-market CDs sound like crap. The compress (dynamic compression) the hell out of the audio. A well done recording with a bit of dynamic range sounds so much better than most pop recordings do.
16 bits gets you about 96dB of dynamic range possible. 24 bits gets you 145dB of range. Again, the 16 bit version is on the edge of hearable. The 24 bit version isn't. But 96dB is definitely not bad for most systems.
But that's where the dynamic compression comes in, as most CDs I hear have very little dynamic range, and they push it out far enough close to the max volume that the disc can record to cover up any noise in the playback system at low signal levels.
CD-Audio is adequate, but not capable of really replacing high quality analog. DVD-Audio's rates are (the audiophiles mostly agree that the data-rate is higher than they can distingusih, although they'll probably complain about warmth/veiled highs, etc, their perogative, I guess).
Most modern recordings are aimed at playback on low-grade consumer hardware, both home/mobile. The noise floor in mobile audio is enough that very dynamic recordings tend to become half unintelligible, and half earsplitting. When the same CD, in a quiet room in a house, is wonderful (drums have massive impact, but quite passages are, well, quiet). Unfortunately, this is performed at the recording studio, instead of being performed in the car stereo (compressors can be cheaply implemented in the DSP that's dealing with all the rest of the sound-shaping that low-end gear does).
So, unless the recordings really try to take advantage of the format, we'll have the audio equivalent of ball-park hotdogs served on china with silver and crystal.
I thought the FCC rules for hams was limited to the low bitrate (9200bps?) that ARPS uses.
I'd love to do some several megabit ham network playing around, but that doesn't seem to be allowed unless I file with the FCC for private spectrum or use the part15 spectrum (which I thought wasn't allowed for non-licensed (protoype) devices, anyway).
I guess the whole idea around doing something like this is the old "why not"
I think this is the group that my wife's cousin was working with when at school (physics graduate student). Now he's working a summer internship at nVidia.
But they were doing a LOT of highpower physics computations that were designed to be vector-based algorithms that could be highly parallelizable, and then using nVidia or ATI graphics cards to do the heavy lifting, as even an $800-1000 card is a lot cheaper than a supercomputer.
Not as fast as a supercomputer, either, but a lot faster than CPU for vector math, especially highly parallelizable operations.
That situation already exists in most areas with cellular 911/*911 service.
In California, you get one of two CHP call centers. Novato for norther california. And maybe 10 minutes on hold if it's rush hour. And maybe blocked completely if you're within a mile or two of an accident on the freeway, and 200 morons are all calling 911 and not stopping to help, causing the operators to block all calls out of the region due to the load.
If your area has a "local" number for 911 that you can call (ask your local police/fire dept) from your cell, then use that.
My guess is that in 75% or more of rural america, the only choice for anything over 24Kbps modems is Comcast. It's 5 miles to my CO, which is wired for DSL. They've talked in the past about a remote DSL CO about 1.5 miles from my house, but nothing has developed. Comcast is the only choice, aside from a very few wireless providers (microwave directional antennas), and then you need LOS to the towers (hard in my mountainous region)).
The safe deposit boxes should be in the vault, which sure as HELL had better be fireproof (vs the rest of the building).
And with the two-key systems for opening each individual box (the bank can't open the box without your key), the boxes would have to be forced, which takes time, which is something a bank robber doesn't have.
I'd say the local bank is safer than my firesafe in my house, rated for paper for 2 hours. I definitely need something better (both larger rated for better insulation (negatives/optical media), and longer burn times (wooden structure in a rural setting).
Slashdot Mirror
You can use your cellphone. Those tryiing to actually run the emergency services would appreciate the ability to communicate outside of the area, though. And HF radio is MUCH more reliable than cell, and people can talk from different states, which is really key to a situation like Katrina.
Network communications can actually be located in TX, AL, AR, etc, or based out in the Gulf on a boat, and easily contact people with stations setup in NOLA on generator/battery power.
hardlinks on 'nix.
Yep.
The way the system works in the US is that the phone manufacturers develop a phone, and work with the network to produce a phone specifically for that network. Then the carrier buys all the phones from the manufacturer, and the carrier sells the phones to the end-user, usually at some massive discount from "retail". For instance, Sprint takes $150 off the price of a new phone in exchange for a 2-year contract, Verizon does something similar.
The GSM carriers even try to "lock" the handsets to the original carrier, but that appears to be easily breakable.
In the US the carriers sorta compete, but mainly have each hewed out their own markets. Instead of competing within the same overall market with phone manufacturers and service providers, instead there are about 5 major markets, each ruled by a carrier, and then the markets compete for each other.
It's screwed up and bizzare.
(I write software that ends up on shipping phones for various different CDMA carriers)
"hollow" or "empty" would have been my psuedo-audiophile term of choice. It think it has a lot to do with the subwoofer channel usage, and how the mix is directed to the LFE channel vs. the main L/R, and possibly also the fact that most center-channel speakers don't do low frequencies well, but a lot of content is centered, so it ends up needed to be produced there.
What I want to do is sit down with my stereo (has preamp outs) and record the outputs in different modes, and then compare the frequency contents of the different recordings, and see if the sound is actually "shaped" differently, or it just changes sound due to the speakers it's routed to.
The other issue I've noticed with the DVD-A formats (or std dvds of music), is that the surround mixed audio sounds distinctly "off" compared to the straight stereo mix of the same thing. I've especially noticed this on my Metallica S&M DVD and on a friend's Blue Man Group DVD. The sound changes radically, and on my system at home, it's still going through the same 2.1 setup either way.
I haven't sat down and figured out how the sound is engineered differently for the two systems, but I know it sounds a hell of a lot better on both my system and a friends (~$10K) system when left in stereo instead of in the surround modes.
The biggest issue I see with SACD and DVD-A is that the engineers don't know how to deal with it. HiDef is easy, as you can easily see all the added detail.
However, it seems like as time goes on, audio gets more and more dynamically compressed (not size-compressed, but dynamic range compression). I know the dynamic compressors were first used for radio, to make FM sound better (knock down the spikes in volume, bring up the volume of the quiet passages).
My modest stereo at home sounds decent, until I throw in a very well put together CD (which usually means it was simply recorded with a pair of stereo mics, and left alone by the engineer), and the result is just astounding. It sounds amazingly good.
SACD would be mind-blowing, but the material on the disc has to actually have the bandwidth to take advantage of it.
What's the advantage of DVD-Audio if the source material was already filtered to 20Khz before it was encoded at 96Khz. Or 24-bit dynamic range vs 16, when the source material has been compressed down to about 8 bits of range.
And all the surround modes that I've heard are utter crap. No nice evelopment of sound around the listener, but instead this very disjoint placement of sounds into the back channels.
My CE degree (mixed HW/SW) started programming at the assembly level, and worked up from their through C into Java.
By the time I was into abstract data types, I had a very good view of how the abstract data types would be implemented at the assembly/hardware level. How your choice of platform can greatly simplify things like semaphores, mutexes, and exceptions.
It seems the standard CS track is to start high-level, and slowly work your way down towards the hardware, and you might have a token class on machine architecture and assembly. I've watched my wife learn successively lower-level programming and watched the light-bulgs that go off as something clicks into place, how low-level hardware design ripples up into how languages are developed.
Now, when I'm programming in something higher-level like Java, C++, or ObjC, I still have that mental image of the registers and the stack, and what hoops will need to be jumped through by the compiler to call those virtual functions, or to handle RTTI and dynamic casts, etc.
Mostly, it's just an awareness of what the platform and hardware are doing, but I think it gives me a big edge over someone who only knows the packages made available by the high-level language, and doesn't understand how that function that they keep calling over and over again whose result value they could have cached can dig into their performance.
Require that the chips not leave the casino. Since they have to back all chips in play with cash, they can just require you to pay out the chips.
Then RFID chip readers at the exits, and they can find someone trying to steal chips.
As we push back into the hills (I live up in the Santa Cruz Mtns), we get more and more into their territory, and as we initially knocked back their population, and then ended hunting of their prey (for all intents and purposes), now the deer population is rising, and with it, the cougar population. If you live along the edges of the hills in the South Bay, and you see deer, you've got cougars.
I've seen sign multiple times, and in 5 years, I've seen 3 total. They're around, but they don't make their prescence well known.
Wild pigs in my area are a serious menace. 400+ lbs, afraid of nothing, and they'll happily gore a person just because they were in the way.
how?
When I was in HS, tape easily outstripped the size of the HD in the common home computer, for realtively cheap tape drives (we're talking 32-64MB drives here).
I've priced out various backup options for the 1TB media server I want to build. $$$$$$
Cheapeast actually appears to be just dumping the files to slow, big SATA drives sequentially. And that's still nearly $500 per backup copy.
Take will run you $1200 for a drive that can even get close to that, and then $50 a tape, but you need to span across several tapes, just like the drives. But at least then at twp copies you've made up the difference.
But the array itself costs nearly that much (including controller).
Scaling back for more "normal" home users, with say 250GB or so of files, they're options are either another drive, and to do it right, an external hot-swap cage and maybe 2-3 drives on rotation (so... $500?) Or to go the tape route again. But I've had a hell of a time finding decent places to research and price out the various current tape drives/formats.
The problem is that backups used to be easy. Spend maybe $100 down at the local 'puter shop, and take home a boxed drive with software and enough tapes to easily do backup rotations. That's just not the case anymore.
Yeah, they do. They're not cheap, either. Basically, you hook up your feed from your battery bank, and your mains connection from the power company (and optionally a generator), and then it figures it all out. Runs off the panels when it's bright out, as the panels are connected to the batteries, so as long as the panels are charging the batteries, it can supply current to the house from the panels, once the panels can't supply enough current to charge the batteries, the batteries start draining to the inverter (this is all controlled by the battery manager). Then the inverter drains down the batteries, and when they fall below a threshold, it starts pulling from the A/C mains to supplement, and recharges the batteries.
The inverter keeps itself fully synchronized off the A/C mains when they're present, and when they go away (as they often do in my area), then it uses it's own internal clock to keep the waveform correct. When mains is down, it stops backfeeding the mains (for safety of everything involved), and will recharge the batteries off the generator, if needed.
When mains comes back up, it slowly adjusts it's clock to match, and then switches over to running off the mains until the batteries are fully charged.
When an engine is properly designed to advance timing and boost pressure when running high octane fuel (like the 110 octane of E85), then you can get some surprising gains:
, 20967,1069364,00.html
http://www.popsci.com/popsci/automotive/article/0
This saab 2.0L engine gains about 20-25% more power when running E85 than straight unleaded. With no loss of milage.
Ethanol is also a rather potent fuel-systems cleaner. In a car with a lot of straight gasoline mileage, this can be a problem as the varnish deposits from the fuel will start to break loose when the ethanol starts to dissolve them, and they can clog up fuel filters and injectors. After a few thousand miles of running ethanol blend, or after running an even stronger fuel deposit solvent through the engine, that problem is solved.
I found out the above in college when I moved, and had to start using 10% blend. Mileage tanked for 2-3 months, and then bounced back to where it was before.
If you're gride-tied solar, doesn't your inverter track this for you anyway?
The Trace units I've been looking at have computer connections (IIRC).
I've never seen an inkjet print that can come close to the quality of a Lightjet print on archival photo paper.
My Lightjet print blown up to 16x20 from moderate quality print film look fantastic. Grain at that size is about as apparent as the best inkjet prints I've seen.
Step up to something like Fujichrome Velvia, or a custom image that's been mastered at the 403ppi native resolution of the printer, and it blows all the ink-based prints out of the water.
I've found that smart-pointers give me the ease of programming of a garbage collector (especially when paired with something like STLPort's allocator), but still gives me explicit control over the destruction of objects.
Java doesn't have a concept of a destructor that runs *NOW* (or not one that I'm aware of). And an object created on the stack can't do things in it's destructor like immediately freeing all references to non-memory resources that it's using (transactions, pipes, files, etc.)
You have to wait for the GC to do it's scan.
Most people that live in a frequent power outage area (like me), have these nifty devices called generators, or solar arrays and batteries.
Makes weathering out a winter storm that's dropped 400 trees in the county on the power lines a lot easier to tolerate.
The telephone bundles are impressively strong cables, power lines are under a lot of stress, and pulled much tighter (because they can't sag, because they're usually running at 20KV or so, and need to be kept about 10' from everything).
When a tree falls over, and crushes all the lines to the ground, the power lines snap, the telephone lines get striped from the nearest pole, but are probably holding up the tree. They're fantasticly strong.
Cable is usually ran under the telephone bundles, and as it's a big coax network, a lot easier to repair if it does get damaged (one wire vs. 1000s).
I'm a volunteer FF, so I see this a lot in the winter here in the Santa Cruz Mtns. Trees fall over in the storms, we close the road and wait for PG&E, and the telephone lines, holding up the fallen tree, are fine.
Audiophile dogs: The new terror/moneymaker of audio shops everywehre...
I try not to call myself that, but I know I fall into the category.
But if you ask someone who claims to have a "golden ear" (I don't) if they can listen to something that has a 140+dB dynamic range, and actually be able to hear both the quietest and the loudest passages (without pain), I doubt they'd say yes.
If you set the upper limit at 130dB (threshold of pain), the lower limit would be -14dB or so, roughly 20x quieter than a barely audible whisper in a dead quiet anechoic chamber (threshold of hearing). Dogs? yes, they can probably hear it They have larger ears to catch the sound.
And I've heard of audiophiles claiming to hear into the low 20Khz range, but never of one claiming to hear above 30Khz or so. Dogs? Yes, I know that they can.
It's the sample rate, moreso to me than the bits per sample. 44.1KHz's nyqust freq (22.05Khz) is close enough into the range of human hearing that artifacting can easily occur. You need VERY good filtering on A/D inputs to have sounds above say 18KHz still present, and the filter level be inaudible at 22Khz (greather than 60dB of attenuation in about a 1/4 octave, so 240dB of attentuation per octave. That's not a simple filter).
By upping the sample rate to 96Khz, the nyquist frequency is so high (48Khz), that the filter requirements aren't nearly so high, and that any artificats that do creep in do so at frequencies well beyond our hearing.
Even on modest gear, it makes a difference.
And by modest, I'm putting the system price at say $2000 (receiver, dvd player, speakers).
Granted, the D/A stage doesn't need these expensive filters that the recording stage does, but it gives the recorders a lot more lattitude to make GOOD recordings. But mass-market CDs sound like crap. The compress (dynamic compression) the hell out of the audio. A well done recording with a bit of dynamic range sounds so much better than most pop recordings do.
16 bits gets you about 96dB of dynamic range possible. 24 bits gets you 145dB of range. Again, the 16 bit version is on the edge of hearable. The 24 bit version isn't. But 96dB is definitely not bad for most systems.
But that's where the dynamic compression comes in, as most CDs I hear have very little dynamic range, and they push it out far enough close to the max volume that the disc can record to cover up any noise in the playback system at low signal levels.
CD-Audio is adequate, but not capable of really replacing high quality analog. DVD-Audio's rates are (the audiophiles mostly agree that the data-rate is higher than they can distingusih, although they'll probably complain about warmth/veiled highs, etc, their perogative, I guess).
Most modern recordings are aimed at playback on low-grade consumer hardware, both home/mobile. The noise floor in mobile audio is enough that very dynamic recordings tend to become half unintelligible, and half earsplitting. When the same CD, in a quiet room in a house, is wonderful (drums have massive impact, but quite passages are, well, quiet). Unfortunately, this is performed at the recording studio, instead of being performed in the car stereo (compressors can be cheaply implemented in the DSP that's dealing with all the rest of the sound-shaping that low-end gear does).
So, unless the recordings really try to take advantage of the format, we'll have the audio equivalent of ball-park hotdogs served on china with silver and crystal.
I thought the FCC rules for hams was limited to the low bitrate (9200bps?) that ARPS uses.
I'd love to do some several megabit ham network playing around, but that doesn't seem to be allowed unless I file with the FCC for private spectrum or use the part15 spectrum (which I thought wasn't allowed for non-licensed (protoype) devices, anyway).
I guess the whole idea around doing something like this is the old "why not"
I think this is the group that my wife's cousin was working with when at school (physics graduate student). Now he's working a summer internship at nVidia.
But they were doing a LOT of highpower physics computations that were designed to be vector-based algorithms that could be highly parallelizable, and then using nVidia or ATI graphics cards to do the heavy lifting, as even an $800-1000 card is a lot cheaper than a supercomputer.
Not as fast as a supercomputer, either, but a lot faster than CPU for vector math, especially highly parallelizable operations.
That situation already exists in most areas with cellular 911/*911 service.
In California, you get one of two CHP call centers. Novato for norther california. And maybe 10 minutes on hold if it's rush hour. And maybe blocked completely if you're within a mile or two of an accident on the freeway, and 200 morons are all calling 911 and not stopping to help, causing the operators to block all calls out of the region due to the load.
If your area has a "local" number for 911 that you can call (ask your local police/fire dept) from your cell, then use that.
My guess is that in 75% or more of rural america, the only choice for anything over 24Kbps modems is Comcast. It's 5 miles to my CO, which is wired for DSL. They've talked in the past about a remote DSL CO about 1.5 miles from my house, but nothing has developed. Comcast is the only choice, aside from a very few wireless providers (microwave directional antennas), and then you need LOS to the towers (hard in my mountainous region)).
The safe deposit boxes should be in the vault, which sure as HELL had better be fireproof (vs the rest of the building).
And with the two-key systems for opening each individual box (the bank can't open the box without your key), the boxes would have to be forced, which takes time, which is something a bank robber doesn't have.
I'd say the local bank is safer than my firesafe in my house, rated for paper for 2 hours. I definitely need something better (both larger rated for better insulation (negatives/optical media), and longer burn times (wooden structure in a rural setting).