Single-player - I don't know. I haven't really enjoyed any single-player FPS that much. I always buy them for multiplayer.
Multiplayer - Classic Quakeworld is still my favorite. God I miss the Canal Zone QWTF map.
Q3F and now ETF are some of my current favorites, although they're not quite as fun as classic Q1.
The thing I miss most about Q1 - Explosion sounds. Explosions in every Quake game (and also the UTs) since then have been WEAK. Q1 rocket/grenade explosions would shake half my freshman year dorm building with the speakers I was using back then.:) And simultaneously detonating 6-7 pipebombs as a demoman in QWTF... droool... An incredible BOOM followed by watching your victim fly halfway across the map.:)
"As far as strong containment buildings, was there not a fission product release at TMI?"
A small amount of gas was released, stuff that was barely radioactive at all. The nastiest radioactive elements are all solids, which tend to be easier to contain.
There are coal fired power plants in Utah that release more radioactive material into the air PER DAY due to trace amounts of such material in the coal they burn that TMI released during the entire incident.
If you're doing a lot of business travel, then you probably want to go with a multipurpose device. Either choose a high-end PDA (there are some damn nice emulators available for the high-res OS5 Palm devices, I'm sure some great emulators are available for PPC too), or buy a laptop. Depends on if you'll actually be flying "business class" which is the only place you'll find DC power outlets to power your laptop inflight, and/or the length of your flight. If all of your flights are relatively short (1-3 hours), then a laptop can make it on battery only.
You could get a PalmOne Treo 650 and a large SD card or two, and you'd get the following all in one: Mobile phone Damned nice PDA Emulator for a large variety of games (Admittedly, the Palm emulation scene has issues right now. There are a bunch of commercial single-system emulators which all have various issues, and then there's Little John, which emulates 5-6 different systems with amazing accuracy but is still in very early beta) Movie/music player. There are a number of media players available for PalmOS: PocketTunes (Probably the best music player currently available) MMPlayer (Formerly the best video player, TCPMP is on par with it already though.) The Core Pocket Media Player (aka TCPMP) - Open source media player, the UI still needs lots of work (but is improving rapidly... I'm talking major steps forward in the past 2 months alone) but it's the fastest and most versatile media player already.
There's also the Tapwave Zodiac, which is a PalmOS device with added graphics acceleration. While its game library is larger than "normal" PalmOS systems, it isn't as big as other gaming systems, but it's a fully functional PalmOS PDA too. The aforementioned Little John emulator works very well on the Zodiac from what I've heard, and in fact was originally a Zodiac-only program.
If your quote were, "Change games 100,000 times and it's time for a new stick" you'd be correct. As another poster pointed out, the limitation is 100,000 writes PER CELL.
Thus you need to write to a single memory location 100,000+ times to exceed its lifetime. As the other poster mentioned, this lifetime is closer to 1 million write cycles nowadays.
Octave is 99% compatible with Matlab, although the nicest Octave features are not Matlab-compatible. (For example, I've never been able to figure out how to make a Matlab program automatically save a plot to a file, and DEFINATELY not in a format that's easy to include in other documents. In Octave, it's pretty easy to automatically output plots in a wide variety of formats without any manual intervention.)
My typical workflow for projects in grad school is as follows:
Use Octave to run whatever simulations/analysis I need to run and output any plots in Encapsulated PostScript (EPS) format. Convert all of the EPSes to small PDFs with epstopdf (This is needed for a later step) Write up the project report in LaTeX, including the plots. Use pdflatex to generate a really nice PDF of the project report complete with cross-references and internal hyperlinks. EVERYTHING is in vector format so it prints just as nicely as it looks on screen.
GnuPlot itself is pretty powerful, but I generally find it most useful/easiest to use Octave as a frontend to it.
Chernobyl was a water-cooled graphite-moderated reactor.
There were a few bad things about this design: 1) If the reactor loses all of its coolant, it does not lose its moderator. Thus, losing coolant does not slow the reaction down. In fact, I believe that the Chernobyl reactor had a number of operating regimes where increases in temperature would increase the output power. 2) Graphite is very combustible. Highly flammable materials in an extremely high-temperature environment such as a nuclear reactor is a Bad Idea. Especially in a facility with no containment building whatsoever.
U.S. reactors are very different. Like Chernobyl, they are water-cooled, BUT they are also water-moderated. If they begin losing coolant, the reaction will begin to slow down. There are no highly combustible substances in the reactor core, and even if there were, U.S. reactors have very strong containment buildings so that if something goes horribly wrong, it will not likely ever escape containment.
Liquid-metal reactors have the disadvantage that their coolants are in some cases very reactive, but that's not much of a problem with a strong containment building, especially since some of the liquid-metal reactors are FAR more efficient as far as making use of their fuel and also produce waste that has a much shorter half-life than the waste from pressurized water reactors, making disposal much easier.
As the other poster mentioned, there's no limit whatsoever on that fee.
Note, though, that you cannot restrict anyone you sell your GPL software to from redistributing it for free. This is how the first widely available TIGER/Line Census dataset was distributed - Bruce Perens paid the (rather exporbitant) $1000-2000 "copying" fee to the Census Bureau for a set of 6-7 CDs, and then burning free copies for a select set of vendors and selling more copies for $10 IIRC. Eventually after he recouped his costs it all wound up on an FTP site. Two years or so later, the Census started putting up the TIGER data for free on an FTP site themselves.
And yes, I know the TIGER dataset isn't GPLed, but it has some similar redistribution rules to GPLed software.
Of all the stations in NYC, UPN's transmitter is second only to ABC's in term of suckiness. Given a choice between recording from CBS or NBC and crashing my PVR-350's encoder trying to record UPN, I'll take CBS or NBC. (I did download all of the Enterprise shows, but if it weren't for BitTorrent, I wouldn't have watched Enterprise at all the past year or two.)
And UPN's digital broadcasts are even worse. WWOR-DT (piggybacked off of FOX 5 WNYW's signal in a subchannel) was actually WORSE than the crappy analog reception at my house. Not only was it a 480i transmission, it was a HORRENDOUS 480i transmission with horrendous macroblocking. I haven't seen MPEG that looked that bad in years.
I believe at least one of the reasons we don't have high-speed rail here is that there are additional technical difficulties in the United States that make upgrading to high-speed rail difficult. Specifically, most of the European and Japanese high-speed rail systems have very large minimum turning radiuses for their rails. In the United States, much of our existing rail infrastructure has turns that are way too tight for the types of high-speed trains found in Europe. From what I remember, Amtrak's Acela trains have a lot of additional features that the European/Japanese high-speed trains don't have in order to combat this issue, and still are not able to go as fast. Acela has had plenty of other problems too. (The most recent being problems with premature brake failure that resulted in the entire Acela fleet being taken out of service for a few months.)
Another reply to your post already commented on this, but neutron sources can be used directly for imaging of non-radioactive materials in a manner very similar to X-ray imaging, except that the capabilities of neutraon imaging are far greater. (For example, neutron imaging can find small stress fractures in metals that X-ray imaging cannot find.)
For a few decades, Cornell University ran a low-power fission reactor (unpressurized, approx. 100-200KW output power), and neutron generation for just such imaging techniques was the primary use for the Ward reactor.
Sadly, the reactor, one of the only low-power research reactors in the country, was shut down around 2000-2002. (I can't remember exactly, but it was the last half of my time as an undergrad there.)
Looking down into the containment pool to see the Cerenkov radiation coming from the core below was one of the most amazing sights I will ever see. (Cerenkov radiation is a bluish light that is emitted when a particle travels faster than the speed of light in the medium it travels through. In this case it was neutrons passing through the water at the bottom of the containment pool.)
In the example you describe (and the one described by this article), the definition you state for "cold fusion" does not hold.
In both cases, there is significant local heating. An atomic nucleus accelerated to relativistic speeds (article's example) can be considered to have an extremely high temperature.
In the case of sonoluminescence, the contents of the oscillating bubble become superheated due to adiabatic heating (If you compress a gas without energy loss to the outside, it will heat up. In the case of sonoluminescence, the bubble oscillates rapidly and its volume changes so rapidly and so significantly that the gases inside it become superheated.)
Hmm, by any chance, might that wind up being 800 watts per array element?
From what I remember, aircraft "detection" radars (not the IFF interrogators used by most airports, which depend on a transponder response from the aircraft) used much higher transmitter powers (raw transmitter power, not EIRP, which is the raw power multiplied by the antenna gain) much higher than 800 watts.
AEGIS shipboard radar is a multi-megawatt system, the onboard aircraft radars are likely at least in the multi-kilowatt range.
Although still, solid-state drives are MUCH less likely to get damaged by temperature changes (and have a much wider operating temperature range) than HDDs.
Their shock resistance is probably an order of magnitude (maybe a few orders of magnitude.) higher than that of an HDD, too.
Personally, I've ditched the USB thumb drives. I now go with a keychain SD card reader (SanDisk MobileMate SD+) and SD cards. So far I just have a single 512M SD card, but I can always add more, and the SD cards are more durable than the USB connector. Plus they work in my Treo.:)
From what I remember, we've already passed the first 2-3 such deadlines that were set. It keeps getting pushed back.
I believe the deadlines all have a clause stating that they are dependent on market penetration of suitable receivers. I forget the exact rule/clause, but it basically says, "If x% of TV owners still have analog-only TVs, the deadline is delayed". I think x was a pretty low number like 15-20%.
The power output of that radar must've been more than 800 watts - Most consumer microwave ovens have power outputs of 1000+ watts, and since all of that goes into a shielded chamber, it all eventually goes into whatever load is inside that chamber. (Whatever you're cooking.)
Also, large directional antennas usually don't have that much near-field gain. It's only until you get far away from the antenna that the gain becomes apparent.
"At that time, my grandfather died; he lived in $RAILROAD_TOWN about 1/4 of the way between the museum and the rail office. He was a civil engineer, and one of his pet peeve was about railroaders calling themselves "engineers" because they ran the engines..."
FYI, I believe this is exactly where the term "engineer" started.
i.e. those guys were the first engineers, and since then the term has become very generalized compared to its original meaning.
PNG is 24 bits per pixel for an RGB image. That translates to 8 bits per color.
The person you replied to was talking about 16 bits PER COLOR. i.e. 48 bits per pixel. Also known as "high dynamic range" in some image processing circles. Typically used for pro photography and pro video/film.
Certain types of basic switching regulators actually have a MINIMUM load that allows them to maintain regulation.
Many ATX-type switchers use a flyback-style arrangement, where each of the voltages is obtained from a tapped transformer. Regulation is often performed by monitoring the voltage on one line, and if the loads on the other lines are within the design specifications of the supply then the other lines are guaranteed to be regulated if one is.
In any PC, it's pretty easy to guarantee that the loads are within specs, as the load ranges of a properly designed supply are quite wide. Unfortunately, it is basically impossible to maintain regulation in these designs if some of the lines have zero load, ESPECIALLY the line that is monitored.
Higher-end PC power supplies (Such as some of the top-end Thermaltake and Antec units) use independently controlled switching supplies for each output, and will run fine with zero load on some of their lines, but 90%+ of ATX supplies are not built this way because it's significantly more expensive.
If this doesn't make sense, I suggest reading up on switching power supply theory. For optimum operation, switching power supplies depend on there ALWAYS being current flowing through an inductor. Whether this happens is dependent on switching frequency, inductor size, and the value of the load resistance. The higher the load resistance is, the higher the inductor size and frequency needed to keep current always flowing in the correct direction. If an ideal (no resistive losses) power supply is operated within this region, its output voltage will always be a direct function of the input voltage and the ratio of on/off times of the switching element. If the load resistance is too high, this no longer holds true and the switching circuit is no longer regulated. In non-ideal circuits, resistive losses in the inductor and switching element will impose a maximum load in addition to minimum load.
If you add circuitry that monitors the output of the switching circuit and changes the switching waveform appropriately if it is out of tolerance, you can maintain regulation outside of the aforementioned regions of operation (specifically the low-load conditions), although efficiency will often suffer in these regimes. As I mentioned before, most PC power supplies do not implement this independently for each output simply because it is not necessary, in any situation the PS was designed for, regulation on one line will imply regulation on all lines because of the way the loads are balanced. But if you leave one line totally unloaded, regulation is no longer guaranteed.
If they do it right, Sony could exercise control over the system that would counteract the hyperinflation that various cashmaking exploits have created in EQ.
Unfortunately, the only way I can think of to do that would be for Sony to "buy back" ingame items with real-world money and have those items simply be removed from ingame circulation. This would cost Sony money.
Unfortunately, hyperinflation and/or economy crashes from market overflooding are difficult to avoid in MMOGs.
An example was EVE Online - The game had an amazing concept and an incredibly complex economic system. Unfortunately, they made some mistakes in aspects of the economic system that utterly destroyed profit margins on manufactured items within a short amount of time. Basically everything in the economy became available at cost or even below cost because people were undercutting each other.
The same is going for Dark Age of Camelot right now - The market for crafted items has all but disappeared, with the result being that crafters can't even sell their wares at material cost or even close to it. (99 quality items sell at material cost for one creation of such an item or sometimes even less. Since there's a 1 in 5 chance of getting an item with that quality, this represents a pretty hefty loss.)
I think a lot of "techies" are less likely to use GMail because they don't use any form of webmail at all.
I know a LOT of non-techies that use GMail exclusively for their mail. While I have a GMail account, I don't use it, as I have my own pretty decent webmail system set up.
I have a Yahoo! account mainly because I was on a few Yahoo! Groups mailing lists. (Oh, their mailing list system SUCKS. I wonder when Google is going to get into that business... Mail archives with Google search capability would be WONDERFUL...)
I may have an account, but I hardly ever use Yahoo! anymore. Pretty much only for leeching movie listings with Plucker, and that's only because hollywood.com's mobile site neglects to put tags into some of their pages which breaks Plucker. Hollywood's mobile interface is so much better. (Their non-mobile site is Flash/JavaScript hell though.)
It only displays a small fraction of seeders/leechers, those that the tracker has told it about.
To see the total S/L ratio, you need to go to the master tracker's status page, if available. For example, I routinely see only 4-5 seeders in the client when there are 50+ listed on the tracker page, and similarly only 10% or so of the total leechers are displayed by my client.
Considering that hardware MPEG-4 encoders are nearly nonexistent even if you ignore the issue of Linux support, it's not surprising you're having problems finding MPEG-4 support. So you have three options:
a) Use a hardware MPEG-2 encoder at maximum bitrate and set MythTV to transcode to lower-bitrate MPEG-4 in non-realtime. This gives the benefit of allowing 2-pass encoding. b) Attempt realtime software encoding. I believe NuppelVideo (which is what MythTV uses for software encoding) uses an MPEG-4 codec. c) Wait. A long time. Looking at the detailed specs for Plextor's MPEG-4 encoder hardware (supposedly Linux drivers were just released for this recently), the device only supports CBR MPEG-4, not even MPEG-4 in ABR mode. MPEG-4 CBR has almost no advantage over VBR MPEG-2, and has not much advantage in ABR mode. To get good quality from MPEG-4 you need to use 2-pass encoding or constant-quantizer (aka constant-quality) encoding. Hardware encoders simply CANNOT support 2-pass encoding, and probably very few if any will ever support constant-quantizer encoding because it makes encoder buffer design MUCH more difficult. As such, it's going to be a LONG time before you see any hardware encoders that support CQ mode. In short, if you even remotely care about quality, MPEG-2 is going to beat MPEG-4 in any hardware-encoding-only situation with any currently available hardware, and it will likely stay that way for quite some time.
Once you start wanting to stream to lightweight frontends around the house (trust me, you'll want to do this once you start with Myth), you'll regret it if you're using MPEG-4 because it instantly increases the cost of any frontends you use.
And then there's HD - In this case you'll be using a transport-stream-capture device, and Myth supports both MPEG2-TS-over-1394 capture and the HD-3000 ATSC tuner card.
MPEG-4 simply does not belong in any PVR system at the current time due to the excessive encoding complexity needed for acceptable quality.
Slashdot Mirror
Single-player - I don't know. I haven't really enjoyed any single-player FPS that much. I always buy them for multiplayer.
:) And simultaneously detonating 6-7 pipebombs as a demoman in QWTF... droool... An incredible BOOM followed by watching your victim fly halfway across the map. :)
Multiplayer - Classic Quakeworld is still my favorite. God I miss the Canal Zone QWTF map.
Q3F and now ETF are some of my current favorites, although they're not quite as fun as classic Q1.
The thing I miss most about Q1 - Explosion sounds. Explosions in every Quake game (and also the UTs) since then have been WEAK. Q1 rocket/grenade explosions would shake half my freshman year dorm building with the speakers I was using back then.
"As far as strong containment buildings, was there not a fission product release at TMI?"
A small amount of gas was released, stuff that was barely radioactive at all. The nastiest radioactive elements are all solids, which tend to be easier to contain.
There are coal fired power plants in Utah that release more radioactive material into the air PER DAY due to trace amounts of such material in the coal they burn that TMI released during the entire incident.
Qualcomm is putting Linux support into their chipsets.
It'll be up to the handset manufacturers (who buy from Qualcomm) whether to actually use such support or not.
This kind of surprises me, most of the really high-end handsets use ARM-based processors (Intel XScale or TI OMAP) instead.
Qualcomm left the handset business a few years ago, they sold their handset division to Kyocera.
That's why Kyocera's first phones seemed so similar to existing Qualcomm handset designs - Because they WERE Qualcomm handset designs.
If you're doing a lot of business travel, then you probably want to go with a multipurpose device. Either choose a high-end PDA (there are some damn nice emulators available for the high-res OS5 Palm devices, I'm sure some great emulators are available for PPC too), or buy a laptop. Depends on if you'll actually be flying "business class" which is the only place you'll find DC power outlets to power your laptop inflight, and/or the length of your flight. If all of your flights are relatively short (1-3 hours), then a laptop can make it on battery only.
You could get a PalmOne Treo 650 and a large SD card or two, and you'd get the following all in one:
Mobile phone
Damned nice PDA
Emulator for a large variety of games (Admittedly, the Palm emulation scene has issues right now. There are a bunch of commercial single-system emulators which all have various issues, and then there's Little John, which emulates 5-6 different systems with amazing accuracy but is still in very early beta)
Movie/music player. There are a number of media players available for PalmOS:
PocketTunes (Probably the best music player currently available)
MMPlayer (Formerly the best video player, TCPMP is on par with it already though.)
The Core Pocket Media Player (aka TCPMP) - Open source media player, the UI still needs lots of work (but is improving rapidly... I'm talking major steps forward in the past 2 months alone) but it's the fastest and most versatile media player already.
There's also the Tapwave Zodiac, which is a PalmOS device with added graphics acceleration. While its game library is larger than "normal" PalmOS systems, it isn't as big as other gaming systems, but it's a fully functional PalmOS PDA too. The aforementioned Little John emulator works very well on the Zodiac from what I've heard, and in fact was originally a Zodiac-only program.
Dammit, where's the "-1 WRONG" mod option?
If your quote were, "Change games 100,000 times and it's time for a new stick" you'd be correct. As another poster pointed out, the limitation is 100,000 writes PER CELL.
Thus you need to write to a single memory location 100,000+ times to exceed its lifetime. As the other poster mentioned, this lifetime is closer to 1 million write cycles nowadays.
Octave + GnuPlot is an amazing combination.
Octave is 99% compatible with Matlab, although the nicest Octave features are not Matlab-compatible. (For example, I've never been able to figure out how to make a Matlab program automatically save a plot to a file, and DEFINATELY not in a format that's easy to include in other documents. In Octave, it's pretty easy to automatically output plots in a wide variety of formats without any manual intervention.)
My typical workflow for projects in grad school is as follows:
Use Octave to run whatever simulations/analysis I need to run and output any plots in Encapsulated PostScript (EPS) format.
Convert all of the EPSes to small PDFs with epstopdf (This is needed for a later step)
Write up the project report in LaTeX, including the plots.
Use pdflatex to generate a really nice PDF of the project report complete with cross-references and internal hyperlinks. EVERYTHING is in vector format so it prints just as nicely as it looks on screen.
GnuPlot itself is pretty powerful, but I generally find it most useful/easiest to use Octave as a frontend to it.
Chernobyl was a water-cooled graphite-moderated reactor.
There were a few bad things about this design:
1) If the reactor loses all of its coolant, it does not lose its moderator. Thus, losing coolant does not slow the reaction down. In fact, I believe that the Chernobyl reactor had a number of operating regimes where increases in temperature would increase the output power.
2) Graphite is very combustible. Highly flammable materials in an extremely high-temperature environment such as a nuclear reactor is a Bad Idea. Especially in a facility with no containment building whatsoever.
U.S. reactors are very different. Like Chernobyl, they are water-cooled, BUT they are also water-moderated. If they begin losing coolant, the reaction will begin to slow down. There are no highly combustible substances in the reactor core, and even if there were, U.S. reactors have very strong containment buildings so that if something goes horribly wrong, it will not likely ever escape containment.
Liquid-metal reactors have the disadvantage that their coolants are in some cases very reactive, but that's not much of a problem with a strong containment building, especially since some of the liquid-metal reactors are FAR more efficient as far as making use of their fuel and also produce waste that has a much shorter half-life than the waste from pressurized water reactors, making disposal much easier.
As the other poster mentioned, there's no limit whatsoever on that fee.
Note, though, that you cannot restrict anyone you sell your GPL software to from redistributing it for free. This is how the first widely available TIGER/Line Census dataset was distributed - Bruce Perens paid the (rather exporbitant) $1000-2000 "copying" fee to the Census Bureau for a set of 6-7 CDs, and then burning free copies for a select set of vendors and selling more copies for $10 IIRC. Eventually after he recouped his costs it all wound up on an FTP site. Two years or so later, the Census started putting up the TIGER data for free on an FTP site themselves.
And yes, I know the TIGER dataset isn't GPLed, but it has some similar redistribution rules to GPLed software.
Of all the stations in NYC, UPN's transmitter is second only to ABC's in term of suckiness. Given a choice between recording from CBS or NBC and crashing my PVR-350's encoder trying to record UPN, I'll take CBS or NBC. (I did download all of the Enterprise shows, but if it weren't for BitTorrent, I wouldn't have watched Enterprise at all the past year or two.)
And UPN's digital broadcasts are even worse. WWOR-DT (piggybacked off of FOX 5 WNYW's signal in a subchannel) was actually WORSE than the crappy analog reception at my house. Not only was it a 480i transmission, it was a HORRENDOUS 480i transmission with horrendous macroblocking. I haven't seen MPEG that looked that bad in years.
I believe at least one of the reasons we don't have high-speed rail here is that there are additional technical difficulties in the United States that make upgrading to high-speed rail difficult. Specifically, most of the European and Japanese high-speed rail systems have very large minimum turning radiuses for their rails. In the United States, much of our existing rail infrastructure has turns that are way too tight for the types of high-speed trains found in Europe. From what I remember, Amtrak's Acela trains have a lot of additional features that the European/Japanese high-speed trains don't have in order to combat this issue, and still are not able to go as fast. Acela has had plenty of other problems too. (The most recent being problems with premature brake failure that resulted in the entire Acela fleet being taken out of service for a few months.)
Another reply to your post already commented on this, but neutron sources can be used directly for imaging of non-radioactive materials in a manner very similar to X-ray imaging, except that the capabilities of neutraon imaging are far greater. (For example, neutron imaging can find small stress fractures in metals that X-ray imaging cannot find.)
For a few decades, Cornell University ran a low-power fission reactor (unpressurized, approx. 100-200KW output power), and neutron generation for just such imaging techniques was the primary use for the Ward reactor.
Sadly, the reactor, one of the only low-power research reactors in the country, was shut down around 2000-2002. (I can't remember exactly, but it was the last half of my time as an undergrad there.)
Looking down into the containment pool to see the Cerenkov radiation coming from the core below was one of the most amazing sights I will ever see. (Cerenkov radiation is a bluish light that is emitted when a particle travels faster than the speed of light in the medium it travels through. In this case it was neutrons passing through the water at the bottom of the containment pool.)
In the example you describe (and the one described by this article), the definition you state for "cold fusion" does not hold.
In both cases, there is significant local heating. An atomic nucleus accelerated to relativistic speeds (article's example) can be considered to have an extremely high temperature.
In the case of sonoluminescence, the contents of the oscillating bubble become superheated due to adiabatic heating (If you compress a gas without energy loss to the outside, it will heat up. In the case of sonoluminescence, the bubble oscillates rapidly and its volume changes so rapidly and so significantly that the gases inside it become superheated.)
Hmm, by any chance, might that wind up being 800 watts per array element?
From what I remember, aircraft "detection" radars (not the IFF interrogators used by most airports, which depend on a transponder response from the aircraft) used much higher transmitter powers (raw transmitter power, not EIRP, which is the raw power multiplied by the antenna gain) much higher than 800 watts.
AEGIS shipboard radar is a multi-megawatt system, the onboard aircraft radars are likely at least in the multi-kilowatt range.
Although still, solid-state drives are MUCH less likely to get damaged by temperature changes (and have a much wider operating temperature range) than HDDs.
:)
Their shock resistance is probably an order of magnitude (maybe a few orders of magnitude.) higher than that of an HDD, too.
Personally, I've ditched the USB thumb drives. I now go with a keychain SD card reader (SanDisk MobileMate SD+) and SD cards. So far I just have a single 512M SD card, but I can always add more, and the SD cards are more durable than the USB connector. Plus they work in my Treo.
From what I remember, we've already passed the first 2-3 such deadlines that were set. It keeps getting pushed back.
I believe the deadlines all have a clause stating that they are dependent on market penetration of suitable receivers. I forget the exact rule/clause, but it basically says, "If x% of TV owners still have analog-only TVs, the deadline is delayed". I think x was a pretty low number like 15-20%.
The power output of that radar must've been more than 800 watts - Most consumer microwave ovens have power outputs of 1000+ watts, and since all of that goes into a shielded chamber, it all eventually goes into whatever load is inside that chamber. (Whatever you're cooking.)
Also, large directional antennas usually don't have that much near-field gain. It's only until you get far away from the antenna that the gain becomes apparent.
"At that time, my grandfather died; he lived in $RAILROAD_TOWN about 1/4 of the way between the museum and the rail office. He was a civil engineer, and one of his pet peeve was about railroaders calling themselves "engineers" because they ran the engines..."
FYI, I believe this is exactly where the term "engineer" started.
i.e. those guys were the first engineers, and since then the term has become very generalized compared to its original meaning.
PNG is 24 bits per pixel for an RGB image. That translates to 8 bits per color.
The person you replied to was talking about 16 bits PER COLOR. i.e. 48 bits per pixel. Also known as "high dynamic range" in some image processing circles. Typically used for pro photography and pro video/film.
Certain types of basic switching regulators actually have a MINIMUM load that allows them to maintain regulation.
Many ATX-type switchers use a flyback-style arrangement, where each of the voltages is obtained from a tapped transformer. Regulation is often performed by monitoring the voltage on one line, and if the loads on the other lines are within the design specifications of the supply then the other lines are guaranteed to be regulated if one is.
In any PC, it's pretty easy to guarantee that the loads are within specs, as the load ranges of a properly designed supply are quite wide. Unfortunately, it is basically impossible to maintain regulation in these designs if some of the lines have zero load, ESPECIALLY the line that is monitored.
Higher-end PC power supplies (Such as some of the top-end Thermaltake and Antec units) use independently controlled switching supplies for each output, and will run fine with zero load on some of their lines, but 90%+ of ATX supplies are not built this way because it's significantly more expensive.
If this doesn't make sense, I suggest reading up on switching power supply theory. For optimum operation, switching power supplies depend on there ALWAYS being current flowing through an inductor. Whether this happens is dependent on switching frequency, inductor size, and the value of the load resistance. The higher the load resistance is, the higher the inductor size and frequency needed to keep current always flowing in the correct direction. If an ideal (no resistive losses) power supply is operated within this region, its output voltage will always be a direct function of the input voltage and the ratio of on/off times of the switching element. If the load resistance is too high, this no longer holds true and the switching circuit is no longer regulated. In non-ideal circuits, resistive losses in the inductor and switching element will impose a maximum load in addition to minimum load.
If you add circuitry that monitors the output of the switching circuit and changes the switching waveform appropriately if it is out of tolerance, you can maintain regulation outside of the aforementioned regions of operation (specifically the low-load conditions), although efficiency will often suffer in these regimes. As I mentioned before, most PC power supplies do not implement this independently for each output simply because it is not necessary, in any situation the PS was designed for, regulation on one line will imply regulation on all lines because of the way the loads are balanced. But if you leave one line totally unloaded, regulation is no longer guaranteed.
If they do it right, Sony could exercise control over the system that would counteract the hyperinflation that various cashmaking exploits have created in EQ.
Unfortunately, the only way I can think of to do that would be for Sony to "buy back" ingame items with real-world money and have those items simply be removed from ingame circulation. This would cost Sony money.
Unfortunately, hyperinflation and/or economy crashes from market overflooding are difficult to avoid in MMOGs.
An example was EVE Online - The game had an amazing concept and an incredibly complex economic system. Unfortunately, they made some mistakes in aspects of the economic system that utterly destroyed profit margins on manufactured items within a short amount of time. Basically everything in the economy became available at cost or even below cost because people were undercutting each other.
The same is going for Dark Age of Camelot right now - The market for crafted items has all but disappeared, with the result being that crafters can't even sell their wares at material cost or even close to it. (99 quality items sell at material cost for one creation of such an item or sometimes even less. Since there's a 1 in 5 chance of getting an item with that quality, this represents a pretty hefty loss.)
I'd be inclined to disagree with you on that.
I think a lot of "techies" are less likely to use GMail because they don't use any form of webmail at all.
I know a LOT of non-techies that use GMail exclusively for their mail. While I have a GMail account, I don't use it, as I have my own pretty decent webmail system set up.
(I LOVE IMP, it rocks!)
I have a Yahoo! account mainly because I was on a few Yahoo! Groups mailing lists. (Oh, their mailing list system SUCKS. I wonder when Google is going to get into that business... Mail archives with Google search capability would be WONDERFUL...)
I may have an account, but I hardly ever use Yahoo! anymore. Pretty much only for leeching movie listings with Plucker, and that's only because hollywood.com's mobile site neglects to put tags into some of their pages which breaks Plucker. Hollywood's mobile interface is so much better. (Their non-mobile site is Flash/JavaScript hell though.)
It only displays a small fraction of seeders/leechers, those that the tracker has told it about.
:)
To see the total S/L ratio, you need to go to the master tracker's status page, if available. For example, I routinely see only 4-5 seeders in the client when there are 50+ listed on the tracker page, and similarly only 10% or so of the total leechers are displayed by my client.
Up to 225 KB/sec now.
Considering that hardware MPEG-4 encoders are nearly nonexistent even if you ignore the issue of Linux support, it's not surprising you're having problems finding MPEG-4 support. So you have three options:
a) Use a hardware MPEG-2 encoder at maximum bitrate and set MythTV to transcode to lower-bitrate MPEG-4 in non-realtime. This gives the benefit of allowing 2-pass encoding.
b) Attempt realtime software encoding. I believe NuppelVideo (which is what MythTV uses for software encoding) uses an MPEG-4 codec.
c) Wait. A long time. Looking at the detailed specs for Plextor's MPEG-4 encoder hardware (supposedly Linux drivers were just released for this recently), the device only supports CBR MPEG-4, not even MPEG-4 in ABR mode. MPEG-4 CBR has almost no advantage over VBR MPEG-2, and has not much advantage in ABR mode. To get good quality from MPEG-4 you need to use 2-pass encoding or constant-quantizer (aka constant-quality) encoding. Hardware encoders simply CANNOT support 2-pass encoding, and probably very few if any will ever support constant-quantizer encoding because it makes encoder buffer design MUCH more difficult. As such, it's going to be a LONG time before you see any hardware encoders that support CQ mode. In short, if you even remotely care about quality, MPEG-2 is going to beat MPEG-4 in any hardware-encoding-only situation with any currently available hardware, and it will likely stay that way for quite some time.
Once you start wanting to stream to lightweight frontends around the house (trust me, you'll want to do this once you start with Myth), you'll regret it if you're using MPEG-4 because it instantly increases the cost of any frontends you use.
And then there's HD - In this case you'll be using a transport-stream-capture device, and Myth supports both MPEG2-TS-over-1394 capture and the HD-3000 ATSC tuner card.
MPEG-4 simply does not belong in any PVR system at the current time due to the excessive encoding complexity needed for acceptable quality.