Material science, computer modelling, automated assembly, and engine efficiency have, however. There are also some common-sense things, like the fact that you'd have gotten 2% better fuel economy on Concorde just by towing the damned thing to the runway instead of having it taxi under its own power. Today you might just have some auxiliary system for that, such as an electric motor in the landing gear wheel to handle taxiing or something. There are a whole bunch of vectors on which we could do better today. The biggest is probably weight. We could build a similar aircraft that was substantially lighter today. The aerodynamics can also be better, since we can model this better today. Heck, even the Tu-144 was more aerodynamic and it was from the same era.
Do you honestly think that they got everything absolutely perfect the first time around when they designed this thing in the late 50s, and that there is no room for improvement whatsoever?
Of course it was rejected. Their proposal travels at Mach 0.98. A regular aircraft (using the 787 as an example) travels at Mach 0.85. That's a really tiny difference; a flight that would have taken 6 hours would instead take 5 hours and 12 minutes. Yeah, it's an improvement, but not enough to justify the extra expense as compared to more efficient aircraft.
On the other hand, if you created an aircraft that travelled at the same speed as a Concorde but with much greater efficiency, you could do your 6 hour flight in 2 hours and 30 minutes. That's some substantial savings.
The Concorde was designed in the late 1950s. We have made rather substantial improvements in technology in the past half century that would allow an aircraft designed today to achieve substantially better fuel efficiency, not to mention the additional efficiencies we can gain via higher altitudes. The stigma of its failure will probably prevent anybody from trying again any time soon, but just because an aircraft designed in the 1950s wasn't cost effective doesn't mean an aircraft designed in the 2010s couldn't be.
Exactly. Which is why I would much rather have one that is well calibrated out of the box, even if accuracy isn't as good as some other model that could be user-calibrated better.
OEMs respond to market demands. If it's clear to an OEM that Windows 7 is going to sell more machines than Windows 8, they're going to put Windows 7 on the machines. It might take a while for this to come to pass, but if Win8 is as big a disaster as I think it will be, it'll happen.
Remember the Vista downgrade stuff? Customers hated Vista so much that OEMs would ship a machine with Vista and a promise to the customer that they could downgrade to WinXP for free? Well, if Win8 is rejected by the buying public even more strongly, we might just see them skip the downgrade step and ship the machines with the older OS to begin with.
I suspect a lot of the work is already done for games that have been previously ported to OS X. After all, that already entails the switch to OpenGL. Obviously there's a lot more to it than that, but having your game already running under OpenGL on a POSIX platform is a big head start. Especially if you started out on Windows, since that means you've already had to abstract a lot of the platform-specific stuff out to get it running on the second platform.
To summarize what people are saying and linking, the answer is basically "Yes, you do it in hardware, but you buffer entire frames before processing them, and that can happen multiple times between the monitor getting the image and displaying the image, since it's easier than doing it on the fly."
At 60Hz, even buffering a frame once immediately adds ~17ms of latency. Not so bad in isolation, but combine that with other buffers inside the monitor, and the amount of time it takes an LCD display to actually change the pixels, and it starts to add up.
Some IPS panels, like Dell monitors a few years ago, were notorious for having latency issues. The Dell 2408WFP had, IIRC, roughly 50ms of latency, which meant whatever you saw on the monitor was 3 full frames behind.
I'll agree with you on point 9. I own a Dell U2711. I love it, but it's not perfect. My two (but not only) biggest beefs with it are probably the glittery anti-reflective coating, and the crappy build quality.
It's still a great display, though. Phenomenal colour accuracy out of the box (since it's factory calibrated, and comes with a benchmark report), for example. That's kind of important... Sites like anandtech will point out that even though it's one of if not the best calibrated out-of-the-box display, there are other displays that can do better when you calibrate it yourself. The flaw here is the common assumption reviewers make that home users are going to calibrate their LCD monitors. In the real world, almost nobody does. How many people do you know who own a USB colourimeter or whatever hardware is required to do this? In practice, a monitor that is well calibrated out of the box is far more valuable to the average person than one that is poorly calibrated out of the box but CAN be perfectly calibrated if you spend the extra money and effort to do so.
There are currently no players that can do bluray menus. I would say that it will come eventually, but it took ages to get decent DVD menu support (only within the past few years did opensource players get good support for them).
There's no need to spend much more than $750 on a 1440p display, since that's what the Dell U2711 costs (in Canada, anyhow). It's the same panel as Apple's 27" Cinema Display (and possible Thunderbolt Display) use, although the build quality isn't nearly as nice. It's a great display, whose greatest weakness is that it uses a CCFL backlight rather than an LED backlight, which means it pumps out a lot of heat. It does have an obscenely huge number of inputs, though, and can act as a digital audio decoder (audio comes in via HDMI and is outputted via a 3.5mm jack), which is both rare and useful.
And Dell is currently selling the U2412M, a 24" IPS display with 1920x1200 resolution, for $250... What's your point? A 23" 1080p ASUS IPS for $270 isn't a good deal.
I've had dead pixels on big name monitors... it's not like LG or or Samsung or Dell won't tolerate them. Even when they do have a dead pixel policy, the policy usually only specifies how many of what kind of dead pixels they'll tolerate before they replace it.
In some respects, skipping all the enhancement hardware should result in lower latency, which can be a good thing.
It should be noted that one of the reasons why it's made awful (and one reason why the OP's "not a death sentence" may be wrong) is because Amazon likes to melt their employees by saving money. For many years they refused to air condition their warehouses, and refused to open garage doors to get some cross-breeze going citing "theft concerns". The result? 110 degrees Fahrenheit working environment, with 15 employees in one warehouse fainting of heatstroke in a single day.
It was only after there was a big thing in the media about it that Amazon agreed to install industrial air conditioners.
I'm not an electrical engineer, and most articles claiming Apple devices don't adhere to the spec either provide no sources are use the same ladyada.com article as a source. That said, it would appear that I was wrong.
The "proprietary charger resistor trick" was made part of the USB standard in 2007 (USB Battery Charging Specification), three years before the article you linked to purporting to have discovered "secret resistors" that enable Apple to "artificially restricts iPhone chargers"...
Apple's no saint, but if you're going to call them out on something, maybe try to stick to stuff they actually did wrong instead of making stuff up. The headphone recess thing might be one, although I'd argue that that was just a dumb design decision rather than an attempt to introduce a proprietary standard; it was still a standard 3.5mm jack, just rendered mostly useless.
HBO Canada: watered down by cancon requirements and unavailable without paying for TMN's whole lineup, which is bloody expensive ($16.22/mth to get one channel). Food Network: watered down by cancon requirements, not available in HD on all providers (like mine). Nickelodeon: Watered down by cancon requirements, not available in HD in Canada. SyFy: As you said, not available in Canada, although ironically they shoot all their stuff in Canada. Space: Only JUST became available in HD to non-Bell customers, watered down by cancon requirements, although most US scifi shows are produced in Canada, so it actually works out.
I should mention that we've had a-la-carte selection in Montreal for years. Not completely, and there are all the cancon restrictions the government puts on it, but you can get basic cable and then pick most other stuff a-la-carte except for a handful of specialties like HBO Canada, but that's a restriction from the owner of HBO Canada that forces the cable companies to sell it as a bundle (even if cable carriers wanted to they couldn't sell it unbundled)
I can't speak for "everywhere else". My cable bill is something like $60-70, but I've got a rather large number of channels, I'm not getting any bundle discounts, I'm virtually locked in via a bunch of silly rules in my lease, and the non-incumbent options haven't been around long enough to have any impact on price. Some of them do have rather large potential savings.
Right, because there are obviously no available choices for television service in Canada. Why, a person in Montreal can choose between only six different services (Illico digital cable from Videotron, FibeTV IPTV from Bell, BellTV satellite from Bell, ShawDirect satellite from Shaw, Colbanet IPTV from Colbanet, Zazeen IPTV from Acanac) from five distinct companies over three different transport mediums... No competition whatsoever!
The only real upside is that, while you may get the same number of channels for the same price, you might be able to get more channels that you want by replacing the ones you don't want.
In my apartment, in the heart of downtown of the second largest city in Canada, or the sixth largest city in the US/Canada, I get a grand total of zero digital channels. Before the digital switchover, I got three or four analog, all but one with very poor quality (double/triple images, lots of snow, etc), and that one that came in well was only viewable if I stood in a certain place in the room, less the image degrade.
At the same time, satellite dishes are forbidden, and IPTV from the phone company requires you to have an internet connection with them (which would have me paying twice as much for a third the monthly cap).
So, yeah, cable television is kind of a must. It's the only viable option for me. My only other hope is that the phone company loosens their mandatory bundling requirement for IPTV. Such bundling is illegal anyhow, although nobody seems to care.
Amusingly, most of the channels mentioned (HBO, SyFy, Food Network, Nickelodeon) are not available directly in Canada. AMC is, assuming it's not a watered down Canadian version, but most Viacom channels (like Comedy Central) are not available.
The CRTC has "cancon" (Canadian Content) regulations that require:
1) Canadian channels to show Canadian productions for a certain percentage of their airtime 2) Cable providers to have Canadian channels as a certain percentage of channels offered 3) Consumers to subscribe to a certain percentage of Canadian channels
For cable companies in Canada that already have a-la-carte offerings (my provider, Videotron, will sell you basic cable and you can a-la-carte the rest) require that your a-la-carte selections adhere to the cancon restrictions.
That assumes that the addresses are not predictable. I believe that one proposal is to use the MAC address of the machine as part of the address. The OUI is fairly predictable based on market share (for example, realtek sells a rather lot of NIC controllers), so if you assume that the target is using a realtek NIC (or if you know what manufacturer they use), that knocks off 24 bits right there. That gets you down to 40 bits, and require only 1TB of bandwidth to scan. That is fairly cheap to do with a botnet or cloud service.
Let's assume you want to scan for that Realtek auto-configured NIC using Amazon EC2. We'll assume a target timespan of one hour, since I believe that's the minimum time slice. We want to pump out 1TB of bandwidth total, and let's say you don't want to push more than 100 megabits per second to any instance. That would require roughly 24 instances, which gives us $120 in bandwidth, and ~ $0.17 in instance time (spot instance micro price).
Say you want to do it faster, in one minute: you're still at only about 1400 instances, still $120 in bandwidth, and roughly $10 in instance time...
Of course, if you know nothing about the topology, scanning 2^64 addresses would likely exceed the capacity of Amazon's entire cloud, not to mention your wallet;)
Slashdot Mirror
Material science, computer modelling, automated assembly, and engine efficiency have, however. There are also some common-sense things, like the fact that you'd have gotten 2% better fuel economy on Concorde just by towing the damned thing to the runway instead of having it taxi under its own power. Today you might just have some auxiliary system for that, such as an electric motor in the landing gear wheel to handle taxiing or something. There are a whole bunch of vectors on which we could do better today. The biggest is probably weight. We could build a similar aircraft that was substantially lighter today. The aerodynamics can also be better, since we can model this better today. Heck, even the Tu-144 was more aerodynamic and it was from the same era.
Do you honestly think that they got everything absolutely perfect the first time around when they designed this thing in the late 50s, and that there is no room for improvement whatsoever?
Of course it was rejected. Their proposal travels at Mach 0.98. A regular aircraft (using the 787 as an example) travels at Mach 0.85. That's a really tiny difference; a flight that would have taken 6 hours would instead take 5 hours and 12 minutes. Yeah, it's an improvement, but not enough to justify the extra expense as compared to more efficient aircraft.
On the other hand, if you created an aircraft that travelled at the same speed as a Concorde but with much greater efficiency, you could do your 6 hour flight in 2 hours and 30 minutes. That's some substantial savings.
The Concorde was designed in the late 1950s. We have made rather substantial improvements in technology in the past half century that would allow an aircraft designed today to achieve substantially better fuel efficiency, not to mention the additional efficiencies we can gain via higher altitudes. The stigma of its failure will probably prevent anybody from trying again any time soon, but just because an aircraft designed in the 1950s wasn't cost effective doesn't mean an aircraft designed in the 2010s couldn't be.
Exactly. Which is why I would much rather have one that is well calibrated out of the box, even if accuracy isn't as good as some other model that could be user-calibrated better.
OEMs respond to market demands. If it's clear to an OEM that Windows 7 is going to sell more machines than Windows 8, they're going to put Windows 7 on the machines. It might take a while for this to come to pass, but if Win8 is as big a disaster as I think it will be, it'll happen.
Remember the Vista downgrade stuff? Customers hated Vista so much that OEMs would ship a machine with Vista and a promise to the customer that they could downgrade to WinXP for free? Well, if Win8 is rejected by the buying public even more strongly, we might just see them skip the downgrade step and ship the machines with the older OS to begin with.
I suspect a lot of the work is already done for games that have been previously ported to OS X. After all, that already entails the switch to OpenGL. Obviously there's a lot more to it than that, but having your game already running under OpenGL on a POSIX platform is a big head start. Especially if you started out on Windows, since that means you've already had to abstract a lot of the platform-specific stuff out to get it running on the second platform.
To summarize what people are saying and linking, the answer is basically "Yes, you do it in hardware, but you buffer entire frames before processing them, and that can happen multiple times between the monitor getting the image and displaying the image, since it's easier than doing it on the fly."
At 60Hz, even buffering a frame once immediately adds ~17ms of latency. Not so bad in isolation, but combine that with other buffers inside the monitor, and the amount of time it takes an LCD display to actually change the pixels, and it starts to add up.
Some IPS panels, like Dell monitors a few years ago, were notorious for having latency issues. The Dell 2408WFP had, IIRC, roughly 50ms of latency, which meant whatever you saw on the monitor was 3 full frames behind.
After; the coating would be applied long before the monitor was even assembled, let alone tested.
I'll agree with you on point 9. I own a Dell U2711. I love it, but it's not perfect. My two (but not only) biggest beefs with it are probably the glittery anti-reflective coating, and the crappy build quality.
It's still a great display, though. Phenomenal colour accuracy out of the box (since it's factory calibrated, and comes with a benchmark report), for example. That's kind of important... Sites like anandtech will point out that even though it's one of if not the best calibrated out-of-the-box display, there are other displays that can do better when you calibrate it yourself. The flaw here is the common assumption reviewers make that home users are going to calibrate their LCD monitors. In the real world, almost nobody does. How many people do you know who own a USB colourimeter or whatever hardware is required to do this? In practice, a monitor that is well calibrated out of the box is far more valuable to the average person than one that is poorly calibrated out of the box but CAN be perfectly calibrated if you spend the extra money and effort to do so.
There are currently no players that can do bluray menus. I would say that it will come eventually, but it took ages to get decent DVD menu support (only within the past few years did opensource players get good support for them).
There's no need to spend much more than $750 on a 1440p display, since that's what the Dell U2711 costs (in Canada, anyhow). It's the same panel as Apple's 27" Cinema Display (and possible Thunderbolt Display) use, although the build quality isn't nearly as nice. It's a great display, whose greatest weakness is that it uses a CCFL backlight rather than an LED backlight, which means it pumps out a lot of heat. It does have an obscenely huge number of inputs, though, and can act as a digital audio decoder (audio comes in via HDMI and is outputted via a 3.5mm jack), which is both rare and useful.
And Dell is currently selling the U2412M, a 24" IPS display with 1920x1200 resolution, for $250... What's your point? A 23" 1080p ASUS IPS for $270 isn't a good deal.
I've had dead pixels on big name monitors... it's not like LG or or Samsung or Dell won't tolerate them. Even when they do have a dead pixel policy, the policy usually only specifies how many of what kind of dead pixels they'll tolerate before they replace it.
In some respects, skipping all the enhancement hardware should result in lower latency, which can be a good thing.
It should be noted that one of the reasons why it's made awful (and one reason why the OP's "not a death sentence" may be wrong) is because Amazon likes to melt their employees by saving money. For many years they refused to air condition their warehouses, and refused to open garage doors to get some cross-breeze going citing "theft concerns". The result? 110 degrees Fahrenheit working environment, with 15 employees in one warehouse fainting of heatstroke in a single day.
It was only after there was a big thing in the media about it that Amazon agreed to install industrial air conditioners.
I don't have the original 2007 spec (v1.1), but the current 2010 version of the spec (v1.2) is available here:
http://www.usb.org/developers/devclass_docs/BCv1.2_011912.zip
I'm not an electrical engineer, and most articles claiming Apple devices don't adhere to the spec either provide no sources are use the same ladyada.com article as a source. That said, it would appear that I was wrong.
The "proprietary charger resistor trick" was made part of the USB standard in 2007 (USB Battery Charging Specification), three years before the article you linked to purporting to have discovered "secret resistors" that enable Apple to "artificially restricts iPhone chargers"...
Apple's no saint, but if you're going to call them out on something, maybe try to stick to stuff they actually did wrong instead of making stuff up. The headphone recess thing might be one, although I'd argue that that was just a dumb design decision rather than an attempt to introduce a proprietary standard; it was still a standard 3.5mm jack, just rendered mostly useless.
HBO Canada: watered down by cancon requirements and unavailable without paying for TMN's whole lineup, which is bloody expensive ($16.22/mth to get one channel). Food Network: watered down by cancon requirements, not available in HD on all providers (like mine). Nickelodeon: Watered down by cancon requirements, not available in HD in Canada. SyFy: As you said, not available in Canada, although ironically they shoot all their stuff in Canada. Space: Only JUST became available in HD to non-Bell customers, watered down by cancon requirements, although most US scifi shows are produced in Canada, so it actually works out.
I should mention that we've had a-la-carte selection in Montreal for years. Not completely, and there are all the cancon restrictions the government puts on it, but you can get basic cable and then pick most other stuff a-la-carte except for a handful of specialties like HBO Canada, but that's a restriction from the owner of HBO Canada that forces the cable companies to sell it as a bundle (even if cable carriers wanted to they couldn't sell it unbundled)
I can't speak for "everywhere else". My cable bill is something like $60-70, but I've got a rather large number of channels, I'm not getting any bundle discounts, I'm virtually locked in via a bunch of silly rules in my lease, and the non-incumbent options haven't been around long enough to have any impact on price. Some of them do have rather large potential savings.
Right, because there are obviously no available choices for television service in Canada. Why, a person in Montreal can choose between only six different services (Illico digital cable from Videotron, FibeTV IPTV from Bell, BellTV satellite from Bell, ShawDirect satellite from Shaw, Colbanet IPTV from Colbanet, Zazeen IPTV from Acanac) from five distinct companies over three different transport mediums... No competition whatsoever!
The only real upside is that, while you may get the same number of channels for the same price, you might be able to get more channels that you want by replacing the ones you don't want.
In my apartment, in the heart of downtown of the second largest city in Canada, or the sixth largest city in the US/Canada, I get a grand total of zero digital channels. Before the digital switchover, I got three or four analog, all but one with very poor quality (double/triple images, lots of snow, etc), and that one that came in well was only viewable if I stood in a certain place in the room, less the image degrade.
At the same time, satellite dishes are forbidden, and IPTV from the phone company requires you to have an internet connection with them (which would have me paying twice as much for a third the monthly cap).
So, yeah, cable television is kind of a must. It's the only viable option for me. My only other hope is that the phone company loosens their mandatory bundling requirement for IPTV. Such bundling is illegal anyhow, although nobody seems to care.
Amusingly, most of the channels mentioned (HBO, SyFy, Food Network, Nickelodeon) are not available directly in Canada. AMC is, assuming it's not a watered down Canadian version, but most Viacom channels (like Comedy Central) are not available.
The CRTC has "cancon" (Canadian Content) regulations that require:
1) Canadian channels to show Canadian productions for a certain percentage of their airtime
2) Cable providers to have Canadian channels as a certain percentage of channels offered
3) Consumers to subscribe to a certain percentage of Canadian channels
For cable companies in Canada that already have a-la-carte offerings (my provider, Videotron, will sell you basic cable and you can a-la-carte the rest) require that your a-la-carte selections adhere to the cancon restrictions.
It wasn't: not everybody will use the privacy extensions, and there may be a flawed implementation that causes problems.
That assumes that the addresses are not predictable. I believe that one proposal is to use the MAC address of the machine as part of the address. The OUI is fairly predictable based on market share (for example, realtek sells a rather lot of NIC controllers), so if you assume that the target is using a realtek NIC (or if you know what manufacturer they use), that knocks off 24 bits right there. That gets you down to 40 bits, and require only 1TB of bandwidth to scan. That is fairly cheap to do with a botnet or cloud service.
Let's assume you want to scan for that Realtek auto-configured NIC using Amazon EC2. We'll assume a target timespan of one hour, since I believe that's the minimum time slice. We want to pump out 1TB of bandwidth total, and let's say you don't want to push more than 100 megabits per second to any instance. That would require roughly 24 instances, which gives us $120 in bandwidth, and ~ $0.17 in instance time (spot instance micro price).
Say you want to do it faster, in one minute: you're still at only about 1400 instances, still $120 in bandwidth, and roughly $10 in instance time...
Of course, if you know nothing about the topology, scanning 2^64 addresses would likely exceed the capacity of Amazon's entire cloud, not to mention your wallet ;)