This has always struck me as a ridiculous limitation, given the ubiquity of 44.1Khz source material. I don't know the reason for this limitation, but I imagine it has something to do with the psychoacoustics being tuned by hand for the various potential source sampling rates. Still, it seems like a glaring omission.
I especially don't understand why, as most lossy codecs work on a frequency spectrogram based system, where all your data is going to manipulated in a floating point space calculated via FFT/DCT anyway, at which point, realistically, you lose most meaningful relationship between the sampling rate of the compressed/decompressed streams.
Well, the *lowest* performance embedded systems tend to have in-order execution, so there's a plus there at least. e.g. the original Atom CPUs (pre-Silvermont) were in-order, so speculative execution is at least not a problem on that front. That's the same reason a lot of embedded ARM systems are safe, etc....also the cache-access-before-protection-check problem with Meltdown requires reliable cache timing, which means they are easier to exploit on systems with large caches. I imagine this is harder to exploit on those ultra-low-cost systems with small caches... and the cache-flush mitigation strategies will have less of an impact on them since the caches weren't that big to begin with.
For instance, Germany censors media heavily when it contains Nazi imagery......does that mean it is now legal for you to access it in Germany if you acquired the access somewhere else in the EU?
Lastly I dont figure why companies insist of having one large office in the middle of the city, impractical for commuting. I'd be nice if they had a few smaller offices spread around. For example here in Seattle if they had a sub office in Tacoma, or even in Puyallup people that currently commute with everyone else from south locations could reverse/cross commute, and even get access to larger talent pools, wouldnt that be nice! Course the people that could make those choices are paid enough to live in convenient places so its not of any interest to them.
Because in most major cities it's the easiest place to get to for the most number of people. The population tends to be clustered in a circle around the central area, and the city itself has a large population of talent to draw upon.
Moreover, in most major cities in the U.S., the only decent public transit is in the core downtown area (NYC being probably the sole exception). For instance in the bay area it's trivial for both drivers and car-free people to get to downtown SF via BART/Muni/ferry. Car-free people tend to live near stations, and drivers who live in the suburbs can drive to a suburban BART station or ferry terminal to continue their trip downtown (and avoid the worst of the automobile traffic). People living in most any direction have at least some way to get from suburbs to central core quickly.
Companies that build suburban campuses wind up limiting their talent pool because driving from one far flung suburb to another means long trips in traffic unless they happen to be near each other. Try getting to that job in Cupertino if you live in Vallejo. The central city (downtown SF) is a *compromise* location that is equidistant from the far flung locations.
Furthermore, suburban offices mean you can't hire the car-free/car-lite talent pool. This is the problem companies like Google have -- and they had to go to the extreme measure of *creating their own transit system* to deal with the problem (and since the only system they can feasibly build is buses running on public roads, they are still screwed by auto traffic).
Building lots of small offices in every little town to hold hires of every type is not a very desirable option because it breaks teams up -- effectively everyone is telecommuting *but still has to commute to an office*. That's why when companies do have a large number of locations, each one tends to specialize in only certain departments.
A central location with large numbers of different types of transportation to get there (road, rail, water) just makes sense when you think of the logistics.
Previous to this, the Amiga was the emulator platform of choice -- which I guess sort of limited the audience for emulation in the U.S. Emulators on the Amiga used all sorts of trickery to get performance improvements (and was mostly focused on emulating 8/16-bit micros, rather than consoles).
Still, I remember being amazed at what emulators could do with the brute force of post-Pentium x86 systems. I guess that's what was really needed for an era of pervasive and accurate emulation of such a wide array of platforms.
We used to joke that the Amiga could never be emulated correctly. (UAE's original name was the Unusable Amiga Emulator because no hardware at the time has the power to run it -- like trying to play Crysis on a 3DFX Voodoo Banshee). Of course, time marches on, and it was an amazing day when I saw a machine running Windows able to emulate a 1985 Amiga 1000 without glitches or slowdown. It must have been in 2000 or so.
Emulation of complex platforms like the Amiga has come a long way, but some not so much. It took many years to get a great Amiga emulator, but the problem is mostly solved now. Compare that to the Sega Saturn (with its gajillion coprocessors I consider it to be of the same blood of the Amiga, hardware-design-style, only almost a decade younger.). Saturn emulation has sort of languished. =/
The truly interesting barrier to emulation for future generations though won't be due to arcane hardware reverse engineering though. The x86 arch used for Windows/MacOS/etc. is well documented, so running today's computer software should be fine in future decades (individual programs DRM aside). The PS4 and XBoxOne, however, are generic x86 boxes. It's getting past the massive walls of DRM around their entire software architecture that will determine whether today's console games will be around for our grandchildren to play.
To clarify, at one point Sierra tried to create their own online gaming network. This was *NOT* an internet-based network, but something you could connect to directly via dialup with a POTS modem. This later on became the ImagiNation network, which was purchased by AT&T.
As I understand it, the facilities originally created for this (since upgraded to support DSL service) were repurposed by the people involved into an ISP. All of this is based in the old Sierra headquarters in Oakhurst. It's funny, because what was originally "On-Line Systems", with no networking component, later became "Sierra On-Line". This became "Sierra Entertainment", which then attempted to create an on-line network, which later became an ISP. Therefore SierraTel is now more "on-line" than "On-Line Systems" ever was.
For those not in the know, this company is the heir to Sierra On-Line/Sierra Entertainment/Yosemite Entertainment in Oakhurst, CA. They created King's Quest, Space Quest, Police Quest, Leisure Suit Larry, et al. After the studio joined Codemasters they remained in Oakhurst until at some point it became an ISP. I'm not sure if any of the original folk are still there.
Typical "Oxi-clean" (hydrogen peroxide) additives often work well.
Also you can use an ozone generator and pump your closet full of ozone for a similar effect.
Ozonating your whole house can get rid of some stubborn odors.
(Be aware though that excessive use of either method will cause some rubber products to break down faster than with normal exposure. Also, if you ozonate your house, don't be in it at the time.)
Lossy codecs typically have two major stages -- the lossy parts (e.g. dct while throwing out some component frequencies, motion prediction, etc.) -- followed by lossless entropy coding (e.g. Huffman in JPEG) to further compress the resultant data.
These compression algorithms just decompress the lossless part of the process and then recompress it with a more efficient lossless algorithm. On decompression, it then recompresses with the standard algorithm. In some cases (e.g. JPEG) you can keep a copy of the Huffman table that lets you recompress the data into a bit-accurate copy of the original file (you can include a small bit of extra information to make sure any remaining metadata matches up exactly).
The MacOS compression software StuffIt did this years ago.
In my experience, Mugen makes the best extended batteries (both in size and performance).
Of course this is not useful if your phone does not have a replaceable battery (e.g. iPhones). But in general any popular phone with a replaceable battery will have extended batteries made for it. You just put the extended fat battery in then use the provided replacement back panel that includes an enlarged area to hold the new fat battery.
I *always* get this for my phones because I get sick of having to remember to charge them.
Simply put -- consumers can't be trusted to be able to deal with complex secure authentication schemes. That's why there's so many easy-to-guess "What city did you grow up in?" password-reset functions. There are so many weak links in the chain of trust, it takes a concerted effort on the individual's part to secure it.
The CEO of Cloudflare fell victim to this when someone CONVINCED AT&T TO REROUTE HIS VOICEMAIL, starting a chain of events that wound up with the interloper having complete control over Cloudflare and the myriad of sites that use CF (and therefore trust it to send legitimate data).
It's a bit exciting/fascinating to read about the chain of events, (particularly the timeline):
I'm not discussing LANDING it. The goal is to park it into a high Earth orbit, where it can be mined relatively cheaply. Only the most valuable materials would be landed, while bulk materials plentiful on earth (water, iron/nickel, etc.) would be used to build and service spacecraft.
You could capture it with a minimum of propellant fairly easily. Reorienting its orbit relative to Earth doesn't take much of a push if you do it far enough away (which is why when you do course corrections on a spacecraft, you make the big ones early on, and make small, fine-tuning ones when you get closer to your target).
Then you can get most of your delta-v by aerobraking it in Earth's upper atmosphere, aiming it just deep enough to slow it down to just barely below Earth's escape velocity. You'd save a vast amount of propellant and make an amazing light show for anyone watching. =)
Then you give it one more nudge at apogee (probably the most expensive part of the endeavor) to circularize its orbit enough that it doesn't hit the atmosphere again (which is important). After that last high-thrust burn you could then further circularize the orbit with low-thrust, high-efficiency electric thrusters.
Given enough time and a nuclear reactor, this could all be done using reaction mass acquired on-site, so you wouldn't have to actually haul the propellant to the asteroid, and only take just enough to get your reactor and fuel-manufacturing plant to it.
Part of the problem with biking culture in the US it is an evolution of racing/track/BMX bikes. These are designed for weight reduction and aerodynamics rather than comfort. Exposed chains are almost universal, necessitating having your leg cuff rolled up or rubber banded, if you try to wear normal clothes.
Meanwhile in places like The Netherlands and Denmark, bikes are built to be practical for normal people in normal clothes to ride in a comfortable position. Step-through bikes are the norm and are not considered "women's" bikes.
1) UPRIGHT POSTURE -- for comfort rather than aerodynamics 2) FULL CHAIN CASE -- So you can wear *regular clothes* without getting grease all over them or having them get caught in the gears. 3) COAT GUARD OVER REAR WHEEL -- If you wear loose, long clothes like coats, jackets, or skirts (or a tux), it will not get caught in the rear spokes. 4) LARGE FENDERS -- Also to keep your clothes clean if the ground is wet or dirty!
These things add weight to the bike or add wind resistance. Sports bikes in the US shun all these things. Unfortunately, sports bike design has affected even "city" bikes in the US, which means that people barely remember what a full chain case or coat guard are anymore.
In the Netherlands, people go out clubbing on their bikes wearing their sexy outfits. Members of parliament bike to work wearing their suit and tie.
If we want people to switch to bikes in the US, we need features like these so people don't have the inconvenience of having to change clothes or roll up their pant leg (and still risk grease or nicks on their calves). These are all obvious solutions that are just not as obvious to American bicyclists because they never see them now.
Looks like it's Oystein, (a.k.a. edison) in charge of the operation. He's one sharp cookie and has been in the Norweigan scene for decades. I remember how fun we he was 20 yrs ago. xD I'm sure it will be solved soon.
This building is completely not built on a human scale. It places offices and services far from eachother. It's seemed DESIGNED to make people drive.
Take the giant ring and compress it into a 20-40 story dome. Not only would it result in better interconnection between offices, cafeterias, and such, but it would bemore energy efficient (a dome has the least amount of surface area to exchange heat with the outside).
It would use less land, leaving more space for parkland, a farm, solar plant, whatever you want to use it for.
Instead of building a huge fucking parking garage you could place it next to a Caltrain station, and encourage people to use Caltrain to get to work instead of driving.
Hell they could build it in Santa Clara by the Caltrain station there (there's a ton of poorly used space on the north side of it). This is a stop for not only Caltrain (San Jose San Francisco), but also Capitol Corridor (San Jose Oakland Sacramento), and ACE (San Jose Livermore Stockton).
Except driving REALLY costs over fifty cents a mile when you figure in all the money you're actually spending on your car, not just the gasoline (gasoline in the US is very cheap, and is actually NOT the majority cost of driving).
Moreover, even at 50 cents a mile, you're still assuming that you're getting free labor from an unpaid worker -- the driver (who is probably you). During this whole time when you could be working on a paper, talking to friends, drinking vodka, playing angry birds, or whatever.... instead you're focusing on not killing yourself or others while you operate a large heavy machine that could veer out of control at any second if your concentration lapses.
Slashdot Mirror
CEHT? o.o;
This has always struck me as a ridiculous limitation, given the ubiquity of 44.1Khz source material. I don't know the reason for this limitation, but I imagine it has something to do with the psychoacoustics being tuned by hand for the various potential source sampling rates. Still, it seems like a glaring omission.
I especially don't understand why, as most lossy codecs work on a frequency spectrogram based system, where all your data is going to manipulated in a floating point space calculated via FFT/DCT anyway, at which point, realistically, you lose most meaningful relationship between the sampling rate of the compressed/decompressed streams.
Well, the *lowest* performance embedded systems tend to have in-order execution, so there's a plus there at least. e.g. the original Atom CPUs (pre-Silvermont) were in-order, so speculative execution is at least not a problem on that front. That's the same reason a lot of embedded ARM systems are safe, etc. ...also the cache-access-before-protection-check problem with Meltdown requires reliable cache timing, which means they are easier to exploit on systems with large caches. I imagine this is harder to exploit on those ultra-low-cost systems with small caches... and the cache-flush mitigation strategies will have less of an impact on them since the caches weren't that big to begin with.
For instance, Germany censors media heavily when it contains Nazi imagery... ...does that mean it is now legal for you to access it in Germany if you acquired the access somewhere else in the EU?
Lastly I dont figure why companies insist of having one large office in the middle of the city, impractical for commuting. I'd be nice if they had a few smaller offices spread around. For example here in Seattle if they had a sub office in Tacoma, or even in Puyallup people that currently commute with everyone else from south locations could reverse/cross commute, and even get access to larger talent pools, wouldnt that be nice! Course the people that could make those choices are paid enough to live in convenient places so its not of any interest to them.
Because in most major cities it's the easiest place to get to for the most number of people. The population tends to be clustered in a circle around the central area, and the city itself has a large population of talent to draw upon.
Moreover, in most major cities in the U.S., the only decent public transit is in the core downtown area (NYC being probably the sole exception). For instance in the bay area it's trivial for both drivers and car-free people to get to downtown SF via BART/Muni/ferry. Car-free people tend to live near stations, and drivers who live in the suburbs can drive to a suburban BART station or ferry terminal to continue their trip downtown (and avoid the worst of the automobile traffic). People living in most any direction have at least some way to get from suburbs to central core quickly.
Companies that build suburban campuses wind up limiting their talent pool because driving from one far flung suburb to another means long trips in traffic unless they happen to be near each other. Try getting to that job in Cupertino if you live in Vallejo. The central city (downtown SF) is a *compromise* location that is equidistant from the far flung locations.
Furthermore, suburban offices mean you can't hire the car-free/car-lite talent pool. This is the problem companies like Google have -- and they had to go to the extreme measure of *creating their own transit system* to deal with the problem (and since the only system they can feasibly build is buses running on public roads, they are still screwed by auto traffic).
Building lots of small offices in every little town to hold hires of every type is not a very desirable option because it breaks teams up -- effectively everyone is telecommuting *but still has to commute to an office*. That's why when companies do have a large number of locations, each one tends to specialize in only certain departments.
A central location with large numbers of different types of transportation to get there (road, rail, water) just makes sense when you think of the logistics.
Previous to this, the Amiga was the emulator platform of choice -- which I guess sort of limited the audience for emulation in the U.S. Emulators on the Amiga used all sorts of trickery to get performance improvements (and was mostly focused on emulating 8/16-bit micros, rather than consoles).
Still, I remember being amazed at what emulators could do with the brute force of post-Pentium x86 systems. I guess that's what was really needed for an era of pervasive and accurate emulation of such a wide array of platforms.
We used to joke that the Amiga could never be emulated correctly. (UAE's original name was the Unusable Amiga Emulator because no hardware at the time has the power to run it -- like trying to play Crysis on a 3DFX Voodoo Banshee). Of course, time marches on, and it was an amazing day when I saw a machine running Windows able to emulate a 1985 Amiga 1000 without glitches or slowdown. It must have been in 2000 or so.
Emulation of complex platforms like the Amiga has come a long way, but some not so much. It took many years to get a great Amiga emulator, but the problem is mostly solved now. Compare that to the Sega Saturn (with its gajillion coprocessors I consider it to be of the same blood of the Amiga, hardware-design-style, only almost a decade younger.). Saturn emulation has sort of languished. =/
The truly interesting barrier to emulation for future generations though won't be due to arcane hardware reverse engineering though. The x86 arch used for Windows/MacOS/etc. is well documented, so running today's computer software should be fine in future decades (individual programs DRM aside). The PS4 and XBoxOne, however, are generic x86 boxes. It's getting past the massive walls of DRM around their entire software architecture that will determine whether today's console games will be around for our grandchildren to play.
To clarify, at one point Sierra tried to create their own online gaming network. This was *NOT* an internet-based network, but something you could connect to directly via dialup with a POTS modem. This later on became the ImagiNation network, which was purchased by AT&T.
https://blog.codinghorror.com/...
As I understand it, the facilities originally created for this (since upgraded to support DSL service) were repurposed by the people involved into an ISP. All of this is based in the old Sierra headquarters in Oakhurst. It's funny, because what was originally "On-Line Systems", with no networking component, later became "Sierra On-Line". This became "Sierra Entertainment", which then attempted to create an on-line network, which later became an ISP. Therefore SierraTel is now more "on-line" than "On-Line Systems" ever was.
For those not in the know, this company is the heir to Sierra On-Line/Sierra Entertainment/Yosemite Entertainment in Oakhurst, CA. They created King's Quest, Space Quest, Police Quest, Leisure Suit Larry, et al. After the studio joined Codemasters they remained in Oakhurst until at some point it became an ISP. I'm not sure if any of the original folk are still there.
Relevant Wikipedia Entry
(The Sierra name lives on as a trademark of Activision, but in name only. The hallowed halls of that great studio are now an ISP.)
True, which is why you have to not overuse it. (Most dyes hold up pretty well though.)
Typical "Oxi-clean" (hydrogen peroxide) additives often work well.
Also you can use an ozone generator and pump your closet full of ozone for a similar effect.
Ozonating your whole house can get rid of some stubborn odors.
(Be aware though that excessive use of either method will cause some rubber products to break down faster than with normal exposure. Also, if you ozonate your house, don't be in it at the time.)
You mean this one?
https://www.youtube.com/watch?...
Contrary to what the mass media says sometimes, they do not use water vapor.
They work by boiling propylene glycol and/or vegetable glycerine. These have a much lower boiling point than water.
Directly inhaling boiling water from from an inch-long metal pipe wouldn't let water vapor cool, and would cause nasty burns.
Lossy codecs typically have two major stages -- the lossy parts (e.g. dct while throwing out some component frequencies, motion prediction, etc.) -- followed by lossless entropy coding (e.g. Huffman in JPEG) to further compress the resultant data.
These compression algorithms just decompress the lossless part of the process and then recompress it with a more efficient lossless algorithm. On decompression, it then recompresses with the standard algorithm. In some cases (e.g. JPEG) you can keep a copy of the Huffman table that lets you recompress the data into a bit-accurate copy of the original file (you can include a small bit of extra information to make sure any remaining metadata matches up exactly).
The MacOS compression software StuffIt did this years ago.
In my experience, Mugen makes the best extended batteries (both in size and performance).
Of course this is not useful if your phone does not have a replaceable battery (e.g. iPhones). But in general any popular phone with a replaceable battery will have extended batteries made for it. You just put the extended fat battery in then use the provided replacement back panel that includes an enlarged area to hold the new fat battery.
I *always* get this for my phones because I get sick of having to remember to charge them.
http://www.mugenpowerbatteries...
I think you mean Parkfield, rather than Garfield.
(Trivia, there was at one point an actual Garfield, California at one point in history -- it was in the central valley.)
Energia had TWO flights. Buran (successful) and Polyus (Energia stage successful, payload unsuccessful due to upside-down sensor).
Energia was pretty cool, and in both flights the launch vehicle itself worked as designed.
What about the 7xxM series? 780M being the flagship?
Simply put -- consumers can't be trusted to be able to deal with complex secure authentication schemes. That's why there's so many easy-to-guess "What city did you grow up in?" password-reset functions. There are so many weak links in the chain of trust, it takes a concerted effort on the individual's part to secure it.
The CEO of Cloudflare fell victim to this when someone CONVINCED AT&T TO REROUTE HIS VOICEMAIL, starting a chain of events that wound up with the interloper having complete control over Cloudflare and the myriad of sites that use CF (and therefore trust it to send legitimate data).
It's a bit exciting/fascinating to read about the chain of events, (particularly the timeline):
http://blog.cloudflare.com/the...
http://blog.cloudflare.com/pos...
I'm not discussing LANDING it. The goal is to park it into a high Earth orbit, where it can be mined relatively cheaply. Only the most valuable materials would be landed, while bulk materials plentiful on earth (water, iron/nickel, etc.) would be used to build and service spacecraft.
You could capture it with a minimum of propellant fairly easily. Reorienting its orbit relative to Earth doesn't take much of a push if you do it far enough away (which is why when you do course corrections on a spacecraft, you make the big ones early on, and make small, fine-tuning ones when you get closer to your target).
Then you can get most of your delta-v by aerobraking it in Earth's upper atmosphere, aiming it just deep enough to slow it down to just barely below Earth's escape velocity. You'd save a vast amount of propellant and make an amazing light show for anyone watching. =)
Then you give it one more nudge at apogee (probably the most expensive part of the endeavor) to circularize its orbit enough that it doesn't hit the atmosphere again (which is important). After that last high-thrust burn you could then further circularize the orbit with low-thrust, high-efficiency electric thrusters.
Given enough time and a nuclear reactor, this could all be done using reaction mass acquired on-site, so you wouldn't have to actually haul the propellant to the asteroid, and only take just enough to get your reactor and fuel-manufacturing plant to it.
Part of the problem with biking culture in the US it is an evolution of racing/track/BMX bikes. These are designed for weight reduction and aerodynamics rather than comfort. Exposed chains are almost universal, necessitating having your leg cuff rolled up or rubber banded, if you try to wear normal clothes.
Meanwhile in places like The Netherlands and Denmark, bikes are built to be practical for normal people in normal clothes to ride in a comfortable position. Step-through bikes are the norm and are not considered "women's" bikes.
The first image on this page is a Dutch-style bike. The lower pics are the closest thing America has to offer. http://clevercycles.com/blog/2007/06/26/dutchness/
Notice on the Dutch bike:
1) UPRIGHT POSTURE -- for comfort rather than aerodynamics
2) FULL CHAIN CASE -- So you can wear *regular clothes* without getting grease all over them or having them get caught in the gears.
3) COAT GUARD OVER REAR WHEEL -- If you wear loose, long clothes like coats, jackets, or skirts (or a tux), it will not get caught in the rear spokes.
4) LARGE FENDERS -- Also to keep your clothes clean if the ground is wet or dirty!
These things add weight to the bike or add wind resistance. Sports bikes in the US shun all these things. Unfortunately, sports bike design has affected even "city" bikes in the US, which means that people barely remember what a full chain case or coat guard are anymore.
In the Netherlands, people go out clubbing on their bikes wearing their sexy outfits. Members of parliament bike to work wearing their suit and tie.
If we want people to switch to bikes in the US, we need features like these so people don't have the inconvenience of having to change clothes or roll up their pant leg (and still risk grease or nicks on their calves). These are all obvious solutions that are just not as obvious to American bicyclists because they never see them now.
Products in Japan tend to come in packaging that is easy to open and close again (for durable goods anyway).
In fact, I often store some things from Japan in the original packaging because it also makes a convenient case to hold it.
(Japanese FOOD, on the other hand, tends to come hideously overpackaged in many concentric layers of paper and plastic that all goes to waste.)
Looks like it's Oystein, (a.k.a. edison) in charge of the operation. He's one sharp cookie and has been in the Norweigan scene for decades. I remember how fun we he was 20 yrs ago. xD I'm sure it will be solved soon.
This building is completely not built on a human scale. It places offices and services far from eachother. It's seemed DESIGNED to make people drive.
Take the giant ring and compress it into a 20-40 story dome. Not only would it result in better interconnection between offices, cafeterias, and such, but it would bemore energy efficient (a dome has the least amount of surface area to exchange heat with the outside).
It would use less land, leaving more space for parkland, a farm, solar plant, whatever you want to use it for.
Instead of building a huge fucking parking garage you could place it next to a Caltrain station, and encourage people to use Caltrain to get to work instead of driving.
Hell they could build it in Santa Clara by the Caltrain station there (there's a ton of poorly used space on the north side of it). This is a stop for not only Caltrain (San Jose San Francisco), but also Capitol Corridor (San Jose Oakland Sacramento), and ACE (San Jose Livermore Stockton).
Except driving REALLY costs over fifty cents a mile when you figure in all the money you're actually spending on your car, not just the gasoline (gasoline in the US is very cheap, and is actually NOT the majority cost of driving).
Moreover, even at 50 cents a mile, you're still assuming that you're getting free labor from an unpaid worker -- the driver (who is probably you). During this whole time when you could be working on a paper, talking to friends, drinking vodka, playing angry birds, or whatever.... instead you're focusing on not killing yourself or others while you operate a large heavy machine that could veer out of control at any second if your concentration lapses.
Driving is NOT cheap, we just ignore the costs.