Re:Can someone explain to me?
on
Tetherless Wireless
·
· Score: 4, Interesting
In this case, it has to do with spectrum and equipment. It's easy for the tower to blast out a strong signal with tons of data, and each mobile to hear it. The only limit here is spectrum allocation.
In the reverse direction, the signal from the mobile is much more tightly power-limited, so if there's too much data per unit of energy, the tower can't hear it above the noise. The solution if you can't yell, is to speak slowly.
For wireline services, it's murkier. The noise budget of a DSLAM has a few things in common with the wireless situation, but in most cases, the upload could go much faster than they sell, and yes, it's a political decision rather than a technical one. With cable, the upload is a shared channel, so they're fairly conservative in what they allocate. They should allow more upload when the network is busy, but that would take effort on their part, and only help a few percent of the customer base.
Here's what's funny: The EV-DO tower equipment is served by T1 circuits, which are symmetric. I understand using T1s for the voice stuff, since it's delay-sensitive, but they could've saved a bundle by using DSL for the data. The equipment is capable of it too, just in a nonobvious way. I bet it was never even considered.
Absolutely right. EV-DO has a 2MBPS shared channel, per sector, for all the downstream data. (I don't know whether more carriers can be added with the spectrum they have allocated.) That's fine and dandy for rural or suburban use, but once you get into dense urban environments, it's very hard to put enough bits per second per square meter to satisfy the userbase, with this technology.
The moral of the story? Don't live downtown.
Been there, done that. What about Ricochet?
on
Tetherless Wireless
·
· Score: 1
Yeah, Ricochet only advertised 128k, what's with that? From every report I've found, it was almost never below 200k.
Ricochet's indoor penetration was also second to none. With the radios blasting out a full watt at 900MHz, it would go through anything, and with 5 poletops per square mile in covered areas, you were always right under one. The modems were able to "gearshift" between modulations on the fly, to work around interference or signal fade. From the blistering speed of 64QAM to the bulletproof penetration of 2FSK, your data would get there, no matter what.
What baffles me is that, with so much Ricochet hardware already out there, and the modems going for $5 on ebay, why hasn't YDI lit the rest of the network back up?
In the meantime, there's some effort underway to reverse-engineer what we can. Check out the Ricochet wiki if you can help.
You think it's laggy, but you haven't tried Nextel's Packetstream data offering. It's been around for years, and I've used it since 2002, so it's not really a fair comparison, but oh lord! Round-trip times approaching 3000ms REALLY make SSH a pain in the ass.
EV-DO is a dream to use in comparison. I got to play with the Detroit network during installation (it's currently being fine-tuned before they open it to the public) and the 150ms lag I experienced was quite acceptable.
Also, the damn "Vortex" compression software sucks. Don't install it. The service works fine without it, and if you do any image work at all, you don't want it slaughtering the incoming pictures anyway.
Coverage is going to be spotty indoors, but they're aiming for 100% on major roads. Hint: The 5220 card has an MC jack, same as on the orinoco 802.11b cards, so you probably already own the pigtail. Tweak a 2.4GHz antenna down to 1.9, and you're sitting pretty.
Replying to myself, since I finally elucidated a detail that's been bugging me since the initial post:
While wardriving, the position readings from the GPSr are delayed when sent to the laptop. Using standard NMEA0183 settings, an update is only sent every 2 seconds. So, the derived position of the APs would lag the actual position of the vehicle by an average of 1 second.
This is easily nulled out when you drive the same area in different directions. But that's not common! One-way streets mean that you'll always see certain APs while moving south, and others while moving north, so their positions will be skewed accordingly. Even on two-way streets, the fact that we drive on the right side of the road means that observations of some points will always contain the directional lag error.
Open question: How much delay do the wifi card drivers introduce into RSSI measurements? Are they from the previous packet, previous n packets, averaged over previous second, only from beacon frames, etc?
There was a programmer named Gus Who spent all his nights in a fuss. As he lay in his bed All that went through his head Was (while !asleep()) sheep++;
But the original wardriving that put the APs into the PlaceLab database was using GPS as a reference.
I see what you're saying though, that a moving GPS on a single wardrive will have some error based on atmospheric effects, but repeated resurveys of an area on different days would tend to average these out, similar to the long-term averaging of a stationary GPS receiver.
My point was that the spokesdude in the article is either misquoted or misinformed about the accuracy of GPS, and that the neither Skyhook nor PlaceLab is likely to return better outdoor results than a consumer GPS receiver. Indoors is where this concept really shines.
Strange how the world turns, I just mentioned PlaceLab to a friend before loading Slashdot. Spooky!
PlaceLab's big advantage is the ability to use multiple sources. Wardriving data is just one potential input. If you have a GPS receiver and a wi-fi card and a CDMA phone all connected, it'll use whichever is giving the most trustworthy results. So you can move smoothly between urban, rural, and indoor environments.
What absolutely makes me giggle is this: "Morgan adds that GPS typically only locates things within a few hundred meters, whereas the Wi-Fi location system can get within 20 to 40 meters of an object."
Try Craigslist.
on
Homeless Wires?
·
· Score: 5, Interesting
eBay isn't sensible for stuff that costs more to ship than it's worth.
Post on Craigslist and get some local geeks to come pick it up.
The jump from AT to ATX motherboards was a step backwards environmentally. We used to have "Super I/O" cards with all our interfaces on them, and we could reuse that card when we switched motherboards to support a new processor.
These days, with all the ports on the mobo, we throw away an extra pound of plastic every time we change chips. This stuff seldom changes -- the ATX port cluster still includes a parallel port, PS/2 ports, and USB ports like when it was introduced. How much of this stuff is sitting in a landfill now?
I'd like to see most of that integrated with the case. Like the front-mounted USB and audio ports, why not put an actual USB hub and USB audio device, along with USB serial and parallel devices, perhaps a USB ethernet adapter, on a PCB inside the case? Let it connect to a single motherboard USB header.
This would give case designers the ultimate flexibility in putting the ports where they want, since a lot of casemodding these days seems to involve port rearrangement. It would liberate the mobo designers from having to mount and support all that plastic, which would in turn allow motherboards to be smaller for those who don't need all those ports. And, for those of us who don't care to have it integrated into the case, we could stick our port cards or port bays into whatever slot or drive mountings we chose.
There was a similar discussion on the CDex forums some time ago, about distributed ripping and encoding of CDs.
I think the final verdict was that if most encoding nodes also have ripping drives, and they only grab material from the network when they have nothing local to chew on, the problem is minimized and almost irrelevant. If you only have one drive supplying multiple encoders, things get complicated.
Don't forget the software layers on the NFS/CIFS/etc server! I'm not aware of how other OS's are optimized, but under win98, accessing a lot of small files will max out most CPUs before it saturates a 100mbit link. Large files should minimize the effect, but depending on the interface drivers and protocol stack, your server might have to worry about more than link speed.
Side note: If you're considering posting to Ask Slashdot, take a look at Ask Metafilter instead. It's made for this, and you'll suffer a lot less mocking from people who find the answer obvious.
I'd been looking for that device a while ago, but I didn't find anything promising. It still seems like vaporware, I'm unable to find anyone actually selling it. For those whose VGA ports are on the docking station, the USB2VGA would be a better way of driving a projector from a small laptop. Or driving a whole pile of monitors at once.
Being USB 2.0, I'm surprised there's a bandwidth problem. I was running MaxiVista at 10Mbps, and it was tolerable. At USB1.1's 12Mbps, I'd expect similar. But at 480Mbps? It should be capable of full-screen movies. I'm pointing my finger at the USB controller or drivers.
Anyway, most of the USB video output devices I find are made to drive NTSC composite or S-video. That could also be fun for a lot of applications.
There's a product called Buddy that's been doing this for many years. Originally, the Buddy card was a combined PCI video card and PS/2 keyboard+mouse controller, which spit all the signals out an 8-position modular jack (RJ45 for the cretins), and a little breakout box at the other end of a (long, shielded cat-5) cable accepted the monitor and input devices. The software gave two Windows95 users the impression that they were the only one on the machine, and I'm still not sure how they did that on a non-NT architecture, but it worked and worked well. Only trouble is, the video bandwidth of the cable was limited, and the RAMDAC in the video card didn't support sync rates over 60Hz, so the flicker on the slave station was pretty obnoxious.
In the years since Buddy was first released, PCI video cards have learned to play nice with their neighbors, and USB has provided a way to connect oodles of keyboards and mice to the same machine. Thus, Buddy is reincarnated as BeTwin, a software-only product that associates specific keyboards, mice, and video cards with specific sessions on the machine. (I'm not sure how it deals with sound. Multiple soundcards would seem easy.)
They say it only supports 5 users, but that sounds like an arbitrary limit and I'm sure they'd tweak a 28-user version if you felt like it.
Related links... I'm going off-topic here, but playing stupid tricks with virtual hardware is fun.
Check out MaxiVista, a "virtual video card" which Windows treats as a second monitor, allowing you to do multi-head tricks. The data for the second display goes out over the network (a la VNC) to a client machine, which simply pipes it into the video buffer. Turn that scrap laptop into a second monitor! I stuffed a 10base-T card in my old lappy and it was perfectly usable for everything except fullscreen video. At 100 or gigabit, it'd be worth a try.
Xinerama is Linux software that does the same thing, creating one large virtual X display, which then chops up the image and sends it to a number of smaller actual displays, some or all of which can obviously be located across the network.
As long as we're doing silly tricks with virtual hardware, you should be aware of Virtual Audio Cable, which enables digitally-perfect audio patching between applications' outputs and inputs, even if the apps themselves think they have exclusive control over the soundcard. (Also enables multiclient sound output under 9x, even if your card doesn't support it, because it does software mixing.)
If video is your thing, try Softcam, to feed your videoconferencing software any old source you feel like. Switch between actual cameras, use your desktop screenshot as a "camera input", add effects, etc. Their WaveMux tool is a nice complement to VAC, too.
Wow, I've been suggesting this for years, specifically to get around draconian copyright restrictions. I'd show movies on it, offer a helicopter-ferried dinner-and-theater package.:)
Anyway, as far as data service goes, send it straight! If international waters start 3 miles out, I'm sure you can name a few radio technologies that have no trouble covering more distance than that.
So you can only reach the users who live near the shore, big deal! Most of the population lives near the coasts anyway. It'll be a special perk of oceanfront property. And once you're into a shoreside connection, VPN out to wherever.
Anyway, who needs an ocean liner to run a server? I'd love to see someone pack enough processor and storage into a satellite. Launch the world's most expensive Freenet node. The trouble is, FCC regs prohibit amateurs from using encryption, so ground stations in the US would have to hit the thing with part 15 gear. I'm sure it's possible.:)
We need search on the Internet as much as we need DNS.
Agreed! I use "I'm feeling lucky" as a replacement for DNS these days. Sites could just as easily be hosted on numeric IPs, as long as they were ranked appropriately in the search results, the name of the host is essentially irrelevant.
Not the point you were making, but it's one that's stuck out at me a few times lately.
Funny, because photographers routinely store their film in lead-lined pouches to prevent inadvertent x-ray exposure if they'll be near airports or hospitals.
I wonder if the aluminized mylar antistatic bags, or potato chip bags, would work the same? When all you have is a poorly written law, everything looks like a hammer.
I did some contracting in walmart stores last summer, and rolling a cart full of my personal tools in and out would set off the alarm every time. I narrowed it down: The cordless drill/flashlight combo kit was doing it. The flashlight was doing it.
Not the battery, mind you, but the flashlight body, which consisted of nothing but a switch, a bulb, and two metal straps that connected them to the battery. For whatever reason, that must've resonated on exactly the right frequency to set off the alarm. I started leaving it in the truck and using a different light.
It occurs to me that a group of mischievous individuals could slowly identify a pile of such items, then one day, visit the store and have each member set off the alarm. The critical element will be a shotgun mic and zoom lens, perched in the parking lot and manned by a member of the media. Does your local TV news show have a "problemsolvers" feature? It's not likely that they'd run a piece on punks getting harassed by mall security, but if it were the chess club...
Ahh, more bad press for the retail establishment. Take that, you capitalist bastards!
I'd argue that if those nano-ceramic discs were made in the volume that polycarbonate discs are, they'd be just as cheap. Think about it, when CD-R media fist came out, it was ~$10/disc for a while, and that was real cost, because the manufacturing plants were expensive and the technology wasn't refined yet. Anything can be made cheaply if you have a market that'll eat up 500 million of them anually.
I'd also argue that that if long warranties were the norm, it would behoove manufacturers to fix the common failure modes.
I'd happily pay twice the price for an optical drive with a 5-year warranty. Hard drives not so much, since I don't have them fail very often. The quality is already high enough.
Re:Most popular thing people use World Wind for
on
Satellite Easter Eggs
·
· Score: 1
There are a number of threads on the Worldwind forums where people post funny things captured in the images, like airplanes in flight. Someone posted a particularly impressive shot of a plane landing, you can tell by the shadow it's about 30 feet off the runway.
What's especially fun is if an object, like a plane, is straddling a tile boundary when the image is taken. When the adjacent images are captured, the other half of the object isn't there, so you have a tailsection flying merrily along...
On the more detailed images, traffic jams and parades are especially fun to find. You can tell what state the traffic lights were in by how the cars are positioned.
I think it's because the event isn't specifically targeted at hackers, like defcon is. You could call it a "hacker con", but it's not really. It's broader than that, with a lot of attention on art and community. There are definitely some hard technical talks, and plenty of hackers will be in attendance, but that's not to the exclusion of anyone else.
Read the schedule and the speaker biographies. I think you'll get a better sense of how things will play out, and why "hacker" is too narrow a term for this event. With everything from crypto to photoshop, electronic music to telephony, there should be a much more diverse audience than at events like defcon. (There was last year, at least!) Not everyone's going to want to see every talk, and that's deliberate.
While I'm posting, I should also shamelessly plug what's sure to be one of the weirdest contests ever. Are you creative, motivated, and skilled enough to make the winning entry?
Carrying on a few traditions, like "being in the midwest" and "being fun to attend".
Notacon's definitely trying to avoid a few other rubicon traditions, like "getting broken up by the cops" and "having the attendees vandalize the hotel".
I don't think there's any real connection between the two. None of the same people are putting notacon together, and the target audience is pretty different. I think there's a strong emphasis towards leaving the kiddies at home for this one.
It seems to me the issue isn't downloading. It's uploading! The people getting sued are the ones sharing the art, not the ones leeching it.
Ask the panel to clarify their personal feelings on where the misdeed, if there is one, is committed. Is it the giver or the taker? If it's the giver, what about accidentally open shared folders? Can you have a crime without intent?
Ask the members of the panel (the ones who download music, anyway) whether they prefer to get a whole album, or just the one or two "hit singles" that're worth listening to.
I think part of the problem, the reason people feel "justified" in copying music, is that the albums suck. There's one decent song on an album, and I won't buy a $12 CD for one song. But the albums don't actually suck! If I listened to it myself, having never heard the band on the radio, I might conclude that 4 or 5 songs were enjoyable, and thus the album had been worth the money.
The music industry discourages this, however. They pick one song from an album, and pay radio stations to play it until my ears fall off. By cultivating one-hit wonders, they force the buying public to feel that the rest of the album wasn't as good, or else it would've gotten radio play, right?
By giving artists more time to work on a truly interesting record, and playing the entire thing on the radio, they'd alleviate some of the overfocused "single" problem which makes consumers feel ripped off. A satisfied audience will gladly spend money on a product they feel is worth it.
In the online distribution channels, try this: If a song goes for $1, then make any additional songs from the same album $0.10 or so. Pretty soon, the listener has a fuller picture of the artist's work, and is more likely to buy things like concert tickets. Because really, who enjoys a concert where the band spends an hour playing music nobody except the diehard fans knows, then plays their one hit song? Feeling loyal to a band sells tickets.
Parent has it right. The Advanced Hex Editor (AXE) has this functionality. Lots of fun when looking at uncompressed graphic formats like icons stored in executables.:)
Grab a few megs from the start of the disk and use sox, the sond exchange to tack audio headers onto it, and try various codec conversions, endian swaps, etc.
There's every chance that the audio chip was interfaced to the drive very simply, as you theorized, without a filesystem. I'm aware of a product which lets you access an ATA device via RS232, it's called the StampDrive. As far as I can tell, it's a PICmicro that's been taught a basic subset of the ATA spec, and it acts as a storage broker for any device that can speak async serial.
People who build their own dataloggers have lots of experience with this sort of dirt-cheap interfacing. Your audio bug is, after all, just a specialized datalogger. A few minutes with a search engine should find plenty of info on the subject.
You know that you want to keep the signal digital until it's as close to your amp as possible. Assuming your amplifier has an optical input, simply running fiber from a soundcard's optical output is the best choice. This puts the burden of clarity on the amplifier's internal DAC and power supply. Optical SPDIF seems capable of 15 meters on standard cable with normal drivers. Since the PC end is all digital, component choice is essentially irrelevant. PCI soundcards with optical outputs are common, so let reputation and support be your guide.
If your amp only accepts analog inputs, things get more complicated. A standalone SPDIF-analog converter seems obvious (and leaves a simple amplifier upgrade path in the future) but consider that such gizmos, while overpriced, usually include a heinously noisy wall-wart power supply. Ripple on the DAC's inputs translates to noise in your audio. Careful design can filter this crap, but caveat emptor. Do listening tests.
This can also be a problem with many of the USB audio devices available. Since they're powered from the USB, a bit of digital noise is inevitably coupled to the analog side. Component choice and careful design are essential here. I'd trust any of the big names to get this right. M-Audio and Edirol both make some slick little USB audio dongles with excellent analog stages. A plethora of USB and firewire audio interfaces are avilable.
If your PC is just a few meters from your stereo, then USB is probably the way to go. My first question would be about ground potential differences, between the USB signal and the amplifier's idea of analog signal ground. Feeding the whole mess from the same branch circuit is an easy way to sidestep the question, but I'm sure someone has tackled it. (Clueful? Please reply!)
If you're dealing with a longer distance, real networking may be the way to go. The idea here is to let your PC in the next room serve the files, but put enough intelligence in the hifi rack to do the decoding as well as the DAC step locally. This usually includes a display and interface of some sort, so you don't need to mess with wireless keyboards or whatever. Various network music players are available, with varying levels of software sophistication and hardware quality. I don't believe any of them include audiophile-quality components in the outputs, and power supply noise is usually an issue in these cheapie designs done by digital engineers without an analog bone in their bodies. If you can find one that supports raw WAV file input, give it a try and see if the audio quality suits you.
Most such players rely heavily on the ID3 tag info for database and display purposes, so tagless WAVs might be awkward at best. Alternately, "tune" the network player to an "internet radio station" which is really a stream running from your desktop's player software. The stream server can then stuff tag information into the stream's metadata, which will appear on the display.
Someone mentioned using the Airport Express as an output device that iTunes could throw digitized audio at. Cute, but I'd be skeptical of any analog components sitting so close to a power supply. Anyone done SNR measurements on this sucker? If it worked with software besides iTunes, it wouldn't suck so hard.
Slashdot Mirror
In this case, it has to do with spectrum and equipment. It's easy for the tower to blast out a strong signal with tons of data, and each mobile to hear it. The only limit here is spectrum allocation.
In the reverse direction, the signal from the mobile is much more tightly power-limited, so if there's too much data per unit of energy, the tower can't hear it above the noise. The solution if you can't yell, is to speak slowly.
For wireline services, it's murkier. The noise budget of a DSLAM has a few things in common with the wireless situation, but in most cases, the upload could go much faster than they sell, and yes, it's a political decision rather than a technical one. With cable, the upload is a shared channel, so they're fairly conservative in what they allocate. They should allow more upload when the network is busy, but that would take effort on their part, and only help a few percent of the customer base.
Here's what's funny: The EV-DO tower equipment is served by T1 circuits, which are symmetric. I understand using T1s for the voice stuff, since it's delay-sensitive, but they could've saved a bundle by using DSL for the data. The equipment is capable of it too, just in a nonobvious way. I bet it was never even considered.
Absolutely right. EV-DO has a 2MBPS shared channel, per sector, for all the downstream data. (I don't know whether more carriers can be added with the spectrum they have allocated.) That's fine and dandy for rural or suburban use, but once you get into dense urban environments, it's very hard to put enough bits per second per square meter to satisfy the userbase, with this technology.
The moral of the story? Don't live downtown.
Yeah, Ricochet only advertised 128k, what's with that? From every report I've found, it was almost never below 200k.
Ricochet's indoor penetration was also second to none. With the radios blasting out a full watt at 900MHz, it would go through anything, and with 5 poletops per square mile in covered areas, you were always right under one. The modems were able to "gearshift" between modulations on the fly, to work around interference or signal fade. From the blistering speed of 64QAM to the bulletproof penetration of 2FSK, your data would get there, no matter what.
What baffles me is that, with so much Ricochet hardware already out there, and the modems going for $5 on ebay, why hasn't YDI lit the rest of the network back up?
In the meantime, there's some effort underway to reverse-engineer what we can. Check out the Ricochet wiki if you can help.
You think it's laggy, but you haven't tried Nextel's Packetstream data offering. It's been around for years, and I've used it since 2002, so it's not really a fair comparison, but oh lord! Round-trip times approaching 3000ms REALLY make SSH a pain in the ass.
EV-DO is a dream to use in comparison. I got to play with the Detroit network during installation (it's currently being fine-tuned before they open it to the public) and the 150ms lag I experienced was quite acceptable.
Also, the damn "Vortex" compression software sucks. Don't install it. The service works fine without it, and if you do any image work at all, you don't want it slaughtering the incoming pictures anyway.
Coverage is going to be spotty indoors, but they're aiming for 100% on major roads. Hint: The 5220 card has an MC jack, same as on the orinoco 802.11b cards, so you probably already own the pigtail. Tweak a 2.4GHz antenna down to 1.9, and you're sitting pretty.
Replying to myself, since I finally elucidated a detail that's been bugging me since the initial post:
While wardriving, the position readings from the GPSr are delayed when sent to the laptop. Using standard NMEA0183 settings, an update is only sent every 2 seconds. So, the derived position of the APs would lag the actual position of the vehicle by an average of 1 second.
This is easily nulled out when you drive the same area in different directions. But that's not common! One-way streets mean that you'll always see certain APs while moving south, and others while moving north, so their positions will be skewed accordingly. Even on two-way streets, the fact that we drive on the right side of the road means that observations of some points will always contain the directional lag error.
Open question: How much delay do the wifi card drivers introduce into RSSI measurements? Are they from the previous packet, previous n packets, averaged over previous second, only from beacon frames, etc?
There was a programmer named Gus
Who spent all his nights in a fuss.
As he lay in his bed
All that went through his head
Was (while !asleep()) sheep++;
But the original wardriving that put the APs into the PlaceLab database was using GPS as a reference.
I see what you're saying though, that a moving GPS on a single wardrive will have some error based on atmospheric effects, but repeated resurveys of an area on different days would tend to average these out, similar to the long-term averaging of a stationary GPS receiver.
My point was that the spokesdude in the article is either misquoted or misinformed about the accuracy of GPS, and that the neither Skyhook nor PlaceLab is likely to return better outdoor results than a consumer GPS receiver. Indoors is where this concept really shines.
Strange how the world turns, I just mentioned PlaceLab to a friend before loading Slashdot. Spooky!
PlaceLab's big advantage is the ability to use multiple sources. Wardriving data is just one potential input. If you have a GPS receiver and a wi-fi card and a CDMA phone all connected, it'll use whichever is giving the most trustworthy results. So you can move smoothly between urban, rural, and indoor environments.
What absolutely makes me giggle is this: "Morgan adds that GPS typically only locates things within a few hundred meters, whereas the Wi-Fi location system can get within 20 to 40 meters of an object."
eBay isn't sensible for stuff that costs more to ship than it's worth.
Post on Craigslist and get some local geeks to come pick it up.
The jump from AT to ATX motherboards was a step backwards environmentally. We used to have "Super I/O" cards with all our interfaces on them, and we could reuse that card when we switched motherboards to support a new processor.
These days, with all the ports on the mobo, we throw away an extra pound of plastic every time we change chips. This stuff seldom changes -- the ATX port cluster still includes a parallel port, PS/2 ports, and USB ports like when it was introduced. How much of this stuff is sitting in a landfill now?
I'd like to see most of that integrated with the case. Like the front-mounted USB and audio ports, why not put an actual USB hub and USB audio device, along with USB serial and parallel devices, perhaps a USB ethernet adapter, on a PCB inside the case? Let it connect to a single motherboard USB header.
This would give case designers the ultimate flexibility in putting the ports where they want, since a lot of casemodding these days seems to involve port rearrangement. It would liberate the mobo designers from having to mount and support all that plastic, which would in turn allow motherboards to be smaller for those who don't need all those ports. And, for those of us who don't care to have it integrated into the case, we could stick our port cards or port bays into whatever slot or drive mountings we chose.
There was a similar discussion on the CDex forums some time ago, about distributed ripping and encoding of CDs.
I think the final verdict was that if most encoding nodes also have ripping drives, and they only grab material from the network when they have nothing local to chew on, the problem is minimized and almost irrelevant. If you only have one drive supplying multiple encoders, things get complicated.
Don't forget the software layers on the NFS/CIFS/etc server! I'm not aware of how other OS's are optimized, but under win98, accessing a lot of small files will max out most CPUs before it saturates a 100mbit link. Large files should minimize the effect, but depending on the interface drivers and protocol stack, your server might have to worry about more than link speed.
Side note: If you're considering posting to Ask Slashdot, take a look at Ask Metafilter instead. It's made for this, and you'll suffer a lot less mocking from people who find the answer obvious.
I'd been looking for that device a while ago, but I didn't find anything promising. It still seems like vaporware, I'm unable to find anyone actually selling it. For those whose VGA ports are on the docking station, the USB2VGA would be a better way of driving a projector from a small laptop. Or driving a whole pile of monitors at once.
Being USB 2.0, I'm surprised there's a bandwidth problem. I was running MaxiVista at 10Mbps, and it was tolerable. At USB1.1's 12Mbps, I'd expect similar. But at 480Mbps? It should be capable of full-screen movies. I'm pointing my finger at the USB controller or drivers.
Anyway, most of the USB video output devices I find are made to drive NTSC composite or S-video. That could also be fun for a lot of applications.
There's a product called Buddy that's been doing this for many years. Originally, the Buddy card was a combined PCI video card and PS/2 keyboard+mouse controller, which spit all the signals out an 8-position modular jack (RJ45 for the cretins), and a little breakout box at the other end of a (long, shielded cat-5) cable accepted the monitor and input devices. The software gave two Windows95 users the impression that they were the only one on the machine, and I'm still not sure how they did that on a non-NT architecture, but it worked and worked well. Only trouble is, the video bandwidth of the cable was limited, and the RAMDAC in the video card didn't support sync rates over 60Hz, so the flicker on the slave station was pretty obnoxious.
In the years since Buddy was first released, PCI video cards have learned to play nice with their neighbors, and USB has provided a way to connect oodles of keyboards and mice to the same machine. Thus, Buddy is reincarnated as BeTwin, a software-only product that associates specific keyboards, mice, and video cards with specific sessions on the machine. (I'm not sure how it deals with sound. Multiple soundcards would seem easy.)
They say it only supports 5 users, but that sounds like an arbitrary limit and I'm sure they'd tweak a 28-user version if you felt like it.
Related links... I'm going off-topic here, but playing stupid tricks with virtual hardware is fun.
Check out MaxiVista, a "virtual video card" which Windows treats as a second monitor, allowing you to do multi-head tricks. The data for the second display goes out over the network (a la VNC) to a client machine, which simply pipes it into the video buffer. Turn that scrap laptop into a second monitor! I stuffed a 10base-T card in my old lappy and it was perfectly usable for everything except fullscreen video. At 100 or gigabit, it'd be worth a try.
Xinerama is Linux software that does the same thing, creating one large virtual X display, which then chops up the image and sends it to a number of smaller actual displays, some or all of which can obviously be located across the network.
As long as we're doing silly tricks with virtual hardware, you should be aware of Virtual Audio Cable, which enables digitally-perfect audio patching between applications' outputs and inputs, even if the apps themselves think they have exclusive control over the soundcard. (Also enables multiclient sound output under 9x, even if your card doesn't support it, because it does software mixing.)
If video is your thing, try Softcam, to feed your videoconferencing software any old source you feel like. Switch between actual cameras, use your desktop screenshot as a "camera input", add effects, etc. Their WaveMux tool is a nice complement to VAC, too.
Wow, I've been suggesting this for years, specifically to get around draconian copyright restrictions. I'd show movies on it, offer a helicopter-ferried dinner-and-theater package. :)
:)
Anyway, as far as data service goes, send it straight! If international waters start 3 miles out, I'm sure you can name a few radio technologies that have no trouble covering more distance than that.
So you can only reach the users who live near the shore, big deal! Most of the population lives near the coasts anyway. It'll be a special perk of oceanfront property. And once you're into a shoreside connection, VPN out to wherever.
Anyway, who needs an ocean liner to run a server? I'd love to see someone pack enough processor and storage into a satellite. Launch the world's most expensive Freenet node. The trouble is, FCC regs prohibit amateurs from using encryption, so ground stations in the US would have to hit the thing with part 15 gear. I'm sure it's possible.
Not the point you were making, but it's one that's stuck out at me a few times lately.
Ask the right question, get the right answer. Parent's parent hit the nail on the head here.
Now the question is, why is Adobe spinning this the way they are, if there's no problem?
Funny, because photographers routinely store their film in lead-lined pouches to prevent inadvertent x-ray exposure if they'll be near airports or hospitals.
I wonder if the aluminized mylar antistatic bags, or potato chip bags, would work the same? When all you have is a poorly written law, everything looks like a hammer.
I did some contracting in walmart stores last summer, and rolling a cart full of my personal tools in and out would set off the alarm every time. I narrowed it down: The cordless drill/flashlight combo kit was doing it. The flashlight was doing it.
Not the battery, mind you, but the flashlight body, which consisted of nothing but a switch, a bulb, and two metal straps that connected them to the battery. For whatever reason, that must've resonated on exactly the right frequency to set off the alarm. I started leaving it in the truck and using a different light.
It occurs to me that a group of mischievous individuals could slowly identify a pile of such items, then one day, visit the store and have each member set off the alarm. The critical element will be a shotgun mic and zoom lens, perched in the parking lot and manned by a member of the media. Does your local TV news show have a "problemsolvers" feature? It's not likely that they'd run a piece on punks getting harassed by mall security, but if it were the chess club...
Ahh, more bad press for the retail establishment. Take that, you capitalist bastards!
I'd argue that if those nano-ceramic discs were made in the volume that polycarbonate discs are, they'd be just as cheap. Think about it, when CD-R media fist came out, it was ~$10/disc for a while, and that was real cost, because the manufacturing plants were expensive and the technology wasn't refined yet. Anything can be made cheaply if you have a market that'll eat up 500 million of them anually.
I'd also argue that that if long warranties were the norm, it would behoove manufacturers to fix the common failure modes.
I'd happily pay twice the price for an optical drive with a 5-year warranty. Hard drives not so much, since I don't have them fail very often. The quality is already high enough.
There are a number of threads on the Worldwind forums where people post funny things captured in the images, like airplanes in flight. Someone posted a particularly impressive shot of a plane landing, you can tell by the shadow it's about 30 feet off the runway.
What's especially fun is if an object, like a plane, is straddling a tile boundary when the image is taken. When the adjacent images are captured, the other half of the object isn't there, so you have a tailsection flying merrily along...
On the more detailed images, traffic jams and parades are especially fun to find. You can tell what state the traffic lights were in by how the cars are positioned.
I think it's because the event isn't specifically targeted at hackers, like defcon is. You could call it a "hacker con", but it's not really. It's broader than that, with a lot of attention on art and community. There are definitely some hard technical talks, and plenty of hackers will be in attendance, but that's not to the exclusion of anyone else.
Read the schedule and the speaker biographies. I think you'll get a better sense of how things will play out, and why "hacker" is too narrow a term for this event. With everything from crypto to photoshop, electronic music to telephony, there should be a much more diverse audience than at events like defcon. (There was last year, at least!) Not everyone's going to want to see every talk, and that's deliberate.
While I'm posting, I should also shamelessly plug what's sure to be one of the weirdest contests ever. Are you creative, motivated, and skilled enough to make the winning entry?
Carrying on a few traditions, like "being in the midwest" and "being fun to attend".
Notacon's definitely trying to avoid a few other rubicon traditions, like "getting broken up by the cops" and "having the attendees vandalize the hotel".
I don't think there's any real connection between the two. None of the same people are putting notacon together, and the target audience is pretty different. I think there's a strong emphasis towards leaving the kiddies at home for this one.
It seems to me the issue isn't downloading. It's uploading! The people getting sued are the ones sharing the art, not the ones leeching it.
Ask the panel to clarify their personal feelings on where the misdeed, if there is one, is committed. Is it the giver or the taker? If it's the giver, what about accidentally open shared folders? Can you have a crime without intent?
Ask the members of the panel (the ones who download music, anyway) whether they prefer to get a whole album, or just the one or two "hit singles" that're worth listening to.
I think part of the problem, the reason people feel "justified" in copying music, is that the albums suck. There's one decent song on an album, and I won't buy a $12 CD for one song. But the albums don't actually suck! If I listened to it myself, having never heard the band on the radio, I might conclude that 4 or 5 songs were enjoyable, and thus the album had been worth the money.
The music industry discourages this, however. They pick one song from an album, and pay radio stations to play it until my ears fall off. By cultivating one-hit wonders, they force the buying public to feel that the rest of the album wasn't as good, or else it would've gotten radio play, right?
By giving artists more time to work on a truly interesting record, and playing the entire thing on the radio, they'd alleviate some of the overfocused "single" problem which makes consumers feel ripped off. A satisfied audience will gladly spend money on a product they feel is worth it.
In the online distribution channels, try this: If a song goes for $1, then make any additional songs from the same album $0.10 or so. Pretty soon, the listener has a fuller picture of the artist's work, and is more likely to buy things like concert tickets. Because really, who enjoys a concert where the band spends an hour playing music nobody except the diehard fans knows, then plays their one hit song? Feeling loyal to a band sells tickets.
Parent has it right. The Advanced Hex Editor (AXE) has this functionality. Lots of fun when looking at uncompressed graphic formats like icons stored in executables. :)
:)
Grab a few megs from the start of the disk and use sox, the sond exchange to tack audio headers onto it, and try various codec conversions, endian swaps, etc.
There's every chance that the audio chip was interfaced to the drive very simply, as you theorized, without a filesystem. I'm aware of a product which lets you access an ATA device via RS232, it's called the StampDrive. As far as I can tell, it's a PICmicro that's been taught a basic subset of the ATA spec, and it acts as a storage broker for any device that can speak async serial.
People who build their own dataloggers have lots of experience with this sort of dirt-cheap interfacing. Your audio bug is, after all, just a specialized datalogger. A few minutes with a search engine should find plenty of info on the subject.
Post back with any success stories.
You know that you want to keep the signal digital until it's as close to your amp as possible. Assuming your amplifier has an optical input, simply running fiber from a soundcard's optical output is the best choice. This puts the burden of clarity on the amplifier's internal DAC and power supply. Optical SPDIF seems capable of 15 meters on standard cable with normal drivers. Since the PC end is all digital, component choice is essentially irrelevant. PCI soundcards with optical outputs are common, so let reputation and support be your guide.
If your amp only accepts analog inputs, things get more complicated. A standalone SPDIF-analog converter seems obvious (and leaves a simple amplifier upgrade path in the future) but consider that such gizmos, while overpriced, usually include a heinously noisy wall-wart power supply. Ripple on the DAC's inputs translates to noise in your audio. Careful design can filter this crap, but caveat emptor. Do listening tests.
This can also be a problem with many of the USB audio devices available. Since they're powered from the USB, a bit of digital noise is inevitably coupled to the analog side. Component choice and careful design are essential here. I'd trust any of the big names to get this right. M-Audio and Edirol both make some slick little USB audio dongles with excellent analog stages. A plethora of USB and firewire audio interfaces are avilable.
If your PC is just a few meters from your stereo, then USB is probably the way to go. My first question would be about ground potential differences, between the USB signal and the amplifier's idea of analog signal ground. Feeding the whole mess from the same branch circuit is an easy way to sidestep the question, but I'm sure someone has tackled it. (Clueful? Please reply!)
If you're dealing with a longer distance, real networking may be the way to go. The idea here is to let your PC in the next room serve the files, but put enough intelligence in the hifi rack to do the decoding as well as the DAC step locally. This usually includes a display and interface of some sort, so you don't need to mess with wireless keyboards or whatever. Various network music players are available, with varying levels of software sophistication and hardware quality. I don't believe any of them include audiophile-quality components in the outputs, and power supply noise is usually an issue in these cheapie designs done by digital engineers without an analog bone in their bodies. If you can find one that supports raw WAV file input, give it a try and see if the audio quality suits you.
Most such players rely heavily on the ID3 tag info for database and display purposes, so tagless WAVs might be awkward at best. Alternately, "tune" the network player to an "internet radio station" which is really a stream running from your desktop's player software. The stream server can then stuff tag information into the stream's metadata, which will appear on the display.
Someone mentioned using the Airport Express as an output device that iTunes could throw digitized audio at. Cute, but I'd be skeptical of any analog components sitting so close to a power supply. Anyone done SNR measurements on this sucker? If it worked with software besides iTunes, it wouldn't suck so hard.