Most people don't have energy efficiency as their #1, let alone only concern, when they're buying a fridge.
Maybe you don't, but I certainly have running costs foremost on my mind. I replaced a 20 year old refrigerator that was leaking energy like a sieve, and my electricity bills dropped by more than $100 a month. That's right, more than 50% of my electricity usage was going towards making food lukewarm.
I fully believe that there exist people whose belief in their conspiracies is so unshakeable that you could load them onto a rocket, fly their worthless ass to the moon, land them at the Apollo sites, let them see the items firsthand, and they would STILL deny that we went there 40-ish years ago.
That's when you crack their visors open. Win-win for everyone involved.
It seems like whether the Milky Way or Andromeda is bigger changes every couple years, as this paper or that paper claims a measurement showing one or the other is actually a lot bigger than we all thought.
We used to think the Milky Way was bigger (and before that, thought Andromeda was bigger for the longest time), and then recently we got some evidence that Andromeda was actually bigger after all. And then there's this piece about the Milky Way actually be bigger after all.
Me? I'm going to sit back and let the scientists figure it out for a few more decades before deciding. All we really know is that Andromeda and the Milky Way are by far the two biggest galaxies in our Local Group, and they're probably close enough in size to make figuring out which one is really bigger a bit tricky.
Interesting that you mention a $300 price point, because there was an Amazon.com deal a few months ago where they were selling the $400 PS3 at a $100 discount with some movies thrown in. Basically, it was a Blu-ray promotion, where one of the player choices was a PS3.
Not sure how many they moved then, but if you were waiting for your $300 PS3, it came (and went).
That's very true for software-based businesses. However, imagine someone with a great idea for a new type of processor that wants to compete with Intel. There's a LOT of capital required for manufacturing-based businesses in order to do proper R&D, establish factories, and so on... assuming the concept is so radically different that it can't be outsourced to existing fabrication plants.
It's highly unlikely a VC is going to bankroll that sort of idea. If it's a risky idea that requires a lot of money, the only place you're going to get that kind of money is from the government.
Some gravitational lensing configurations do, in fact, produce a ring. As you might expect, though, such perfect alignment is pretty rare, and you usually get partial arcs or smeared out blobs.
I'm not knowledgeable about the exact reason for the cross configuration is, but the unusual effects of gravitational lensing are often due to the fact that the lens (a massive galaxy, in most cases) isn't a perfect point source, so the optical effects are somewhat surprising.
He said he couldn't make a phone call himself (he's in Afghanistan, after all), but I certainly think it's a good idea if he could ask someone stateside to call up Dell and get the dispute worked out for him. (E-mail tag always seems like a poor way to get these things resolved.) Find someone he trusts to get it done (perhaps his significant other or parent/sibling), and send all the relevant information about the dispute over.
The whole capsule design is antiquated and relies on an incredibly tough heat shield for reentry, when reentry speeds themselves should be lowered (using a lifting fuselage, like the X-33 and SS1), vastly reducing reentry heating and eliminating burnup almost entirely as a failure mode (Columbia).
Actually, lifting bodies increase re-entry heating. Lifting bodies prolong the re-entry, and use a shallower flight path which maintains a higher velocity for longer. Blunt-nosed capsules are much more thermally efficient, which is why they're still used to deliver things like nuclear warheads.
The main benefit of a lifting profile (which the Space Shuttle is actually better at than a more compact lifting body) is that you can steer better. That's it. And even the Apollo capsules could do that--they were designed with a slight lift vector, so they could be marginally steered.
Getting the volume ID isn't a crack. If you'll note, that diagram in that article itself still requires input from the device keys to perform decryption. It's old news that was reported poorly.
this is why we have things like anydvd HD, etc, and why they pushed out BD+, which was ALSO cracked.
For further reference on the ease of ripping BLU-RAY for naive users, see this link
Game over, and once again, the MPAA loses.
I've been doing some digging into how AnyDVD HD actually works, and it apparently requires an Internet connection to download keys from an AnyDVD service. Hardly a crack in any meaningful terms, and I'm not exactly itching to sign up.
The way that works (correct me if I'm wrong here) is that each title has a "title key" (randomly generated exclusively for that release) that is used to encrypt the content.
Sony has created a set of "vendor keys", lets say 1000 of them, to give out to anyone that wants to make a bluray player and agrees to play by their rules.
I'll correct you, since you're wrong. Also, because this misconception constantly comes up every time someone tries to describe how AACS works.
Creating 1000 keys and encrypting the title key 1000 times is how CSS worked. Obviously, when they designed AACS, they wanted to do it a bit differently.
Recent research in cryptography has created new "broadcast encryption" algorithms. These essentially allow you to encrypt with a single key, and have billions of compatible "subset" keys that will decrypt the same ciphertext.
Now here's the clever bit.
You can actually derive a new key that covers all but one of those subset key, allowing you to revoke one out of a billion keys, all without needing to encode the title key a billion times.
The AACS scheme apparently requires a linear amount of storage for each revoked key. Revocations are rare enough that the overhead is minimal.
So yes, they can assign every individual player (not player model) its own, individually revocable key. And have.
It's evil, but really, you have to admit the mathematics behind it is also pretty clever.
As long as it's aimed so it doesn't reflect straight back there shouldn't be much of a problem.
Well, the thing about retroreflectors is that they always reflect straight back...
The target would still be dead, but I think the whole 'what if some of it bounces back' dilemma might have some merit. I imagine a few calculations would show the return bounce would be minimal, though.
PowerDVD/WinDVD also support Blu-ray (and HD-DVD, not like anyone still cares about that), so yes, the video codecs they install (which are usable system-wide) support VC-1 (WMV9), H.264, 1080p, etc.
PureVideo doesn't actually require the $20 NVIDIA DVD decoder (which I think they've deprecated anyway); the NVIDIA DVD decoder is just for people who want to use MCE, and don't already have an MPEG-2 decoder. PureVideo is just an umbrella name for NVIDIA's video acceleration features (with varying levels of support), which is then used by specific applications (like WinDVD and PowerDVD). It's already cooked into the free drivers.
Rather than having cards with a fixed amount of VRAM, which can't manufacturers just put a bunch of memory card sockets on the card and allow users to add memory when they want?
They used to, at least on some of the better ones (way back when "Windows accelerators" that drew rectangles faster were hot stuff).
You need to realize, though, that graphics cards are on the absolute bleeding edge of memory technology. It's not electrically feasible to pump the sort of bandwidth and latency a modern GPU requires (which is literally orders of magnitude faster than what your CPU gets), while at the same time enabling expandability (especially if you want to make it a standard socket).
Video card design takes a systems approach to integration a GPU with the right RAM and other components. It's one of the main reasons why GPU technology can advance much more quickly than CPUs do. How much benefit are you really going to get when your blazingly-fast GDDR3 RAM gets replaced the very next year with the latest hot new GDDR4 RAM technology? (This is at least one of the reasons why VRAM upgrades never really took off.)
And really, the cost of RAM is a pretty minor portion of the total; compare the price differential between otherwise comparable, say, 256 MB and 512 MB cards.
But claiming that they're just slapping a bunch of RAM on a card to drum up sales is just plain wrong. Hell, the blurb here on Slashdot even mentions the fact that it has 240 cores.
Umm, the GeForce GTX 280, a gamer card released last summer, also has 240 "cores" (as Nvidia counts them; actually stream processors).
This workstation card, as you might expect, is essentially the same thing as the consumer card, just tweaked towards the professional market (more RAM, different drivers). It's nothing especially innovative.
Those 'limbs' are in an exceptionally regular spiral pattern. If you fossilized an octopus, you'd expect the limbs to be all crossed over and tangled up. I'm guessing that those 'limbs' couldn't move independently, and are more like ridges in a sheet of material.
More to the point, a soft-tissued creature like an octopus almost never leaves a fossil record. If you find a fossil, it's because of some sort of skeletal structure the creature has left behind, which of course would naturally be rigid. Think of something like an eight-branched exoskeletal structure.
All of these concerns have easy solutions that avoid GOTO:
1) Put state into a structure, then pass a pointer or reference to the structure. Makes the calling convention much simpler, and having all your state variables bundled up can provide unexpected benefits later. And no, it doesn't add any overhead in combination with...
2) Use inline or file-scoped functions for inner loops. All modern compilers can optimize these easily; this is bread and butter stuff for them. (Some compilers can do global optimization, too, but it's much harder and less common.)
To be fair, C++ is pretty stupid about this, too. Not quite as stupid, but that's clearly where they're drawing the inspiration for the first two cases.
Actually, OS X maintains a cache so it doesn't have to probe the hardware on every boot. It does do a little probing, to detect external devices and just check that everything's still the same as the last time it started, but I can tell you from first hand experience that there can be a wide difference in the time it takes to start with and without the driver cache.
record everything to a 500G harddrive and have a 24h timeshift of all channels
Umm, I think you're greatly underestimating the storage requirement for keeping a 24 hour timeshift of even one channel.
Worst case scenario, ATSC requires a 19.4 Mbps bitrate (I know, not DVB, but I assume it's fairly close), which is 2.4 MB/s, 8.7 GB/hr, and 210 GB/day.
For one channel.
Now, if you transcode to a more efficient format than MPEG-2 at 19.4 Mbps and full HD resolution, you might be able to cut that down--slightly. But it's going to take a huge amount of processing power (especially if you want to do it for every single channel simultaneously), and you're sacrificing quality.
Realistically, you're looking at terabytes of storage needed for this little 24 hour, all channels buffer you want to maintain (a bit over 5 TB for 25 channels, to be conservative).
Not to mention you would need to stream about 60 MB/s through your system 24/7 to support 25 simultaneous channels.
And that's not even taking account actually watching some of this stuff at some point.
Plus, the tuner hardware doesn't exist (at a commercial level, anyway) that could record everything at once.
So I'm afraid there's no simple software solution to this. But maybe in the future. Like 5-10 years from now. If anyone thinks it's worth it just so they can instantly flip digital channels.
A more conservative system is certainly realistic in the near term, though. Especially if you could get broadcasters to tightly synchronize when they transmitted key frames, it might be possible to do it with just two tuners, with one rotating through the available frequencies in a fixed order. However, more advanced codecs than MPEG-2 make this even more problematic, since there's even more interframe dependence. You can't just capture one I-frame and hope to construct the rest on the fly.
I learned the hard way a year or two ago exactly what "flood" means in insurance terms. It includes a flooded basement due to a failed sump pump. Fortunately in my case, the only loss was some 20-year-old carpet.
So here's my little PSA: Even if you don't live somewhere that can really "flood" in a traditional sense, buy flood insurance if you have a basement. At least the minimal "get the mold out" insurance.
I think the other lesson to take from your experience is not to put the home theater in the basement, if you can avoid it.
How does those "back references" recover your data in case of a corrupted sector? Honest question, I do not know brfs.
AFAIK ZFS has no fsck because there is no failure case where it would really help.
Back references could help you reconstruct the file system tree during fsck, but if random data is getting corrupted, you're not going to get it back without redundancy (or forward error correction, I suppose, but that amounts to the same thing).
I can't think of many scenarios where the only kind of data corruption I'm worried about is corruption to file system metadata (which is incidentally all journaling is supposed to protect you from), but who knows.
Slashdot Mirror
Most people don't have energy efficiency as their #1, let alone only concern, when they're buying a fridge.
Maybe you don't, but I certainly have running costs foremost on my mind. I replaced a 20 year old refrigerator that was leaking energy like a sieve, and my electricity bills dropped by more than $100 a month. That's right, more than 50% of my electricity usage was going towards making food lukewarm.
I fully believe that there exist people whose belief in their conspiracies is so unshakeable that you could load them onto a rocket, fly their worthless ass to the moon, land them at the Apollo sites, let them see the items firsthand, and they would STILL deny that we went there 40-ish years ago.
That's when you crack their visors open. Win-win for everyone involved.
Anyone else notice that it's an Australian site? You know, .com.au? And that maybe all the $ prices are in Australian dollars, rather than U.S.?
It seems like whether the Milky Way or Andromeda is bigger changes every couple years, as this paper or that paper claims a measurement showing one or the other is actually a lot bigger than we all thought.
We used to think the Milky Way was bigger (and before that, thought Andromeda was bigger for the longest time), and then recently we got some evidence that Andromeda was actually bigger after all. And then there's this piece about the Milky Way actually be bigger after all.
Me? I'm going to sit back and let the scientists figure it out for a few more decades before deciding. All we really know is that Andromeda and the Milky Way are by far the two biggest galaxies in our Local Group, and they're probably close enough in size to make figuring out which one is really bigger a bit tricky.
Interesting that you mention a $300 price point, because there was an Amazon.com deal a few months ago where they were selling the $400 PS3 at a $100 discount with some movies thrown in. Basically, it was a Blu-ray promotion, where one of the player choices was a PS3.
Not sure how many they moved then, but if you were waiting for your $300 PS3, it came (and went).
That's very true for software-based businesses. However, imagine someone with a great idea for a new type of processor that wants to compete with Intel. There's a LOT of capital required for manufacturing-based businesses in order to do proper R&D, establish factories, and so on... assuming the concept is so radically different that it can't be outsourced to existing fabrication plants.
It's highly unlikely a VC is going to bankroll that sort of idea. If it's a risky idea that requires a lot of money, the only place you're going to get that kind of money is from the government.
Some gravitational lensing configurations do, in fact, produce a ring. As you might expect, though, such perfect alignment is pretty rare, and you usually get partial arcs or smeared out blobs.
I'm not knowledgeable about the exact reason for the cross configuration is, but the unusual effects of gravitational lensing are often due to the fact that the lens (a massive galaxy, in most cases) isn't a perfect point source, so the optical effects are somewhat surprising.
He said he couldn't make a phone call himself (he's in Afghanistan, after all), but I certainly think it's a good idea if he could ask someone stateside to call up Dell and get the dispute worked out for him. (E-mail tag always seems like a poor way to get these things resolved.) Find someone he trusts to get it done (perhaps his significant other or parent/sibling), and send all the relevant information about the dispute over.
The whole capsule design is antiquated and relies on an incredibly tough heat shield for reentry, when reentry speeds themselves should be lowered (using a lifting fuselage, like the X-33 and SS1), vastly reducing reentry heating and eliminating burnup almost entirely as a failure mode (Columbia).
Actually, lifting bodies increase re-entry heating. Lifting bodies prolong the re-entry, and use a shallower flight path which maintains a higher velocity for longer. Blunt-nosed capsules are much more thermally efficient, which is why they're still used to deliver things like nuclear warheads.
The main benefit of a lifting profile (which the Space Shuttle is actually better at than a more compact lifting body) is that you can steer better. That's it. And even the Apollo capsules could do that--they were designed with a slight lift vector, so they could be marginally steered.
a 2007 crack is irrevocable.
Getting the volume ID isn't a crack. If you'll note, that diagram in that article itself still requires input from the device keys to perform decryption. It's old news that was reported poorly.
this is why we have things like anydvd HD, etc, and why they pushed out BD+, which was ALSO cracked.
For further reference on the ease of ripping BLU-RAY for naive users, see this link
Game over, and once again, the MPAA loses.
I've been doing some digging into how AnyDVD HD actually works, and it apparently requires an Internet connection to download keys from an AnyDVD service. Hardly a crack in any meaningful terms, and I'm not exactly itching to sign up.
The way that works (correct me if I'm wrong here) is that each title has a "title key" (randomly generated exclusively for that release) that is used to encrypt the content.
Sony has created a set of "vendor keys", lets say 1000 of them, to give out to anyone that wants to make a bluray player and agrees to play by their rules.
I'll correct you, since you're wrong. Also, because this misconception constantly comes up every time someone tries to describe how AACS works.
Creating 1000 keys and encrypting the title key 1000 times is how CSS worked. Obviously, when they designed AACS, they wanted to do it a bit differently.
Recent research in cryptography has created new "broadcast encryption" algorithms. These essentially allow you to encrypt with a single key, and have billions of compatible "subset" keys that will decrypt the same ciphertext.
Now here's the clever bit.
You can actually derive a new key that covers all but one of those subset key, allowing you to revoke one out of a billion keys, all without needing to encode the title key a billion times.
The AACS scheme apparently requires a linear amount of storage for each revoked key. Revocations are rare enough that the overhead is minimal.
So yes, they can assign every individual player (not player model) its own, individually revocable key. And have.
It's evil, but really, you have to admit the mathematics behind it is also pretty clever.
As long as it's aimed so it doesn't reflect straight back there shouldn't be much of a problem.
Well, the thing about retroreflectors is that they always reflect straight back...
The target would still be dead, but I think the whole 'what if some of it bounces back' dilemma might have some merit. I imagine a few calculations would show the return bounce would be minimal, though.
PowerDVD/WinDVD also support Blu-ray (and HD-DVD, not like anyone still cares about that), so yes, the video codecs they install (which are usable system-wide) support VC-1 (WMV9), H.264, 1080p, etc.
PureVideo doesn't actually require the $20 NVIDIA DVD decoder (which I think they've deprecated anyway); the NVIDIA DVD decoder is just for people who want to use MCE, and don't already have an MPEG-2 decoder. PureVideo is just an umbrella name for NVIDIA's video acceleration features (with varying levels of support), which is then used by specific applications (like WinDVD and PowerDVD). It's already cooked into the free drivers.
I'm not 100% sure what "marque" means but I would suggest that gcc is the marque project of open source and has been for about 20 years.
I'm ROTFLMAO at the people suggesting that the GP meant "marque" as in brand. It's obviously a typo of "marquee," as in prominent.
Rather than having cards with a fixed amount of VRAM, which can't manufacturers just put a bunch of memory card sockets on the card and allow users to add memory when they want?
They used to, at least on some of the better ones (way back when "Windows accelerators" that drew rectangles faster were hot stuff).
You need to realize, though, that graphics cards are on the absolute bleeding edge of memory technology. It's not electrically feasible to pump the sort of bandwidth and latency a modern GPU requires (which is literally orders of magnitude faster than what your CPU gets), while at the same time enabling expandability (especially if you want to make it a standard socket).
Video card design takes a systems approach to integration a GPU with the right RAM and other components. It's one of the main reasons why GPU technology can advance much more quickly than CPUs do. How much benefit are you really going to get when your blazingly-fast GDDR3 RAM gets replaced the very next year with the latest hot new GDDR4 RAM technology? (This is at least one of the reasons why VRAM upgrades never really took off.)
And really, the cost of RAM is a pretty minor portion of the total; compare the price differential between otherwise comparable, say, 256 MB and 512 MB cards.
But claiming that they're just slapping a bunch of RAM on a card to drum up sales is just plain wrong. Hell, the blurb here on Slashdot even mentions the fact that it has 240 cores.
Umm, the GeForce GTX 280, a gamer card released last summer, also has 240 "cores" (as Nvidia counts them; actually stream processors).
This workstation card, as you might expect, is essentially the same thing as the consumer card, just tweaked towards the professional market (more RAM, different drivers). It's nothing especially innovative.
You can also enter a URL directly in various places (like the Run box and the address line in the Explorer), since IE is integrated with the shell.
Those 'limbs' are in an exceptionally regular spiral pattern. If you fossilized an octopus, you'd expect the limbs to be all crossed over and tangled up. I'm guessing that those 'limbs' couldn't move independently, and are more like ridges in a sheet of material.
More to the point, a soft-tissued creature like an octopus almost never leaves a fossil record. If you find a fossil, it's because of some sort of skeletal structure the creature has left behind, which of course would naturally be rigid. Think of something like an eight-branched exoskeletal structure.
All of these concerns have easy solutions that avoid GOTO:
1) Put state into a structure, then pass a pointer or reference to the structure. Makes the calling convention much simpler, and having all your state variables bundled up can provide unexpected benefits later. And no, it doesn't add any overhead in combination with...
2) Use inline or file-scoped functions for inner loops. All modern compilers can optimize these easily; this is bread and butter stuff for them. (Some compilers can do global optimization, too, but it's much harder and less common.)
PHP:
Attribute/Method access: $foo->bar
Static method access: Foo::bar
Namespace access: foo\bar\baz
To be fair, C++ is pretty stupid about this, too. Not quite as stupid, but that's clearly where they're drawing the inspiration for the first two cases.
Actually, OS X maintains a cache so it doesn't have to probe the hardware on every boot. It does do a little probing, to detect external devices and just check that everything's still the same as the last time it started, but I can tell you from first hand experience that there can be a wide difference in the time it takes to start with and without the driver cache.
record everything to a 500G harddrive and have a 24h timeshift of all channels
Umm, I think you're greatly underestimating the storage requirement for keeping a 24 hour timeshift of even one channel.
Worst case scenario, ATSC requires a 19.4 Mbps bitrate (I know, not DVB, but I assume it's fairly close), which is 2.4 MB/s, 8.7 GB/hr, and 210 GB/day.
For one channel.
Now, if you transcode to a more efficient format than MPEG-2 at 19.4 Mbps and full HD resolution, you might be able to cut that down--slightly. But it's going to take a huge amount of processing power (especially if you want to do it for every single channel simultaneously), and you're sacrificing quality.
Realistically, you're looking at terabytes of storage needed for this little 24 hour, all channels buffer you want to maintain (a bit over 5 TB for 25 channels, to be conservative).
Not to mention you would need to stream about 60 MB/s through your system 24/7 to support 25 simultaneous channels.
And that's not even taking account actually watching some of this stuff at some point.
Plus, the tuner hardware doesn't exist (at a commercial level, anyway) that could record everything at once.
So I'm afraid there's no simple software solution to this. But maybe in the future. Like 5-10 years from now. If anyone thinks it's worth it just so they can instantly flip digital channels.
A more conservative system is certainly realistic in the near term, though. Especially if you could get broadcasters to tightly synchronize when they transmitted key frames, it might be possible to do it with just two tuners, with one rotating through the available frequencies in a fixed order. However, more advanced codecs than MPEG-2 make this even more problematic, since there's even more interframe dependence. You can't just capture one I-frame and hope to construct the rest on the fly.
Emphasis mine:
I learned the hard way a year or two ago exactly what "flood" means in insurance terms. It includes a flooded basement due to a failed sump pump. Fortunately in my case, the only loss was some 20-year-old carpet.
So here's my little PSA: Even if you don't live somewhere that can really "flood" in a traditional sense, buy flood insurance if you have a basement. At least the minimal "get the mold out" insurance.
I think the other lesson to take from your experience is not to put the home theater in the basement, if you can avoid it.
How does those "back references" recover your data in case of a corrupted sector? Honest question, I do not know brfs.
AFAIK ZFS has no fsck because there is no failure case where it would really help.
Back references could help you reconstruct the file system tree during fsck, but if random data is getting corrupted, you're not going to get it back without redundancy (or forward error correction, I suppose, but that amounts to the same thing).
I can't think of many scenarios where the only kind of data corruption I'm worried about is corruption to file system metadata (which is incidentally all journaling is supposed to protect you from), but who knows.
Indeed, let us all be glad that Microsoft won the PC war instead of Apple. Jobs would have been worse.
Are those the only two choices?