"new standard" of 1080p phones? I know exactly two models with that resolution and a screen that big or larger. The vast majority of brand new smartphones are still less than 5" and are at most 1280x768; most are actually 1280x720. Quad-core is similarly still uncommon, and is mostly a benfit on Android with its love af battery-killing background tasks.
At the time of its release, IE10 was faster than Chrome on both rendering and JS. IE10 on the phone is faster, hardware to hardware, than all but the latest browsers for Android, and the difference there is meaningless. It's got standards support, performance, and (for a phone browser) a decent feature set (I admit I'd like the synched bookmarks of Chrome, but there are apps for that).
Stop living in the past, grandpa. IE6 is dead and gone (speaking as a former web dev: thank $DEITY)
Half the reasons on that list were obsolete before WP8 even came out (and WP8 is more different under the covers from WP7 than WP7 was from WinMo6.5), and about an eighth of them were never true. Most of the rest have been obsoleted since. That list has been trotted out so many times it's *probably* got more links to it than the number of point-by-point refutations, some of which have been posted right here on Slashdot.
At this point, it's the equivalent of claiming that Windows 8 is still based on DOS because it includes cmd.exe (yes, I've seen people make that claim, both on here and elsewhere). It does nothing but label you as an idiot who badmouths things with no concept of how they work and likely no actual experience with them either.
Your comment could be taken far more seriously if you had given any indication that you even knew there was a difference between Windows and Windows Phone. Aside from both using NT-family kernels (definitely not the same kernel though, even accounting for CPU architecture) and having live tiles and sharing some APIs, the only similarity is their name. iOS and Android are more alike than Win8 and WP8.
MalachiK is using a WP7 device, which is decidedly lacking in features compared to WP8 or current Android versions. However, several of the features that Billly Gates mentions also exist on the Lumia 920, which runs WP8: data sense and warning, WiFi-based Internet sharing, voice activation, brightness sensors (not that those are anythign new) as do several other cool features that he didn't, like wireless charging and a low-light camera.
Excitement about features is cool. I totally understand. But don't confuse "Windows the workstation OS" with "Windows Phone the consumer smartphone platform". The reasons to buy the one have very little do with the reasons to buy the other, and there are plenty of people who are enthusiastic about the Lumia 92x and 1020 devices (no, I don't have one).
Oh dear, did I speak less than praisefully of somebody's favorite language? Haha, wow, so *sensitive*. I stated facts. Argue them if you want, but disagreeing with modpoints is just pure cowardice of somebody who can't back up their position.
For what it's worth, I code in both Java and C++ (and also C# and Python, plus Javascript and occasional other languages as the job requires). I'm not claiming that checked exceptions are either the Achilles' heel of Java performance, or that Java's performance is inadequate for real world work. However, I am claiming that checked exceptions are unneccessary, a waste of both developer and CPU time.
I'll go one further: they can be a source of hidden bugs. When you have excessive numbers of checked exceptions that a function might throw, and you either don't care about them or you know they won't happen, it's easiest to just wrap them in a try-swallow block that silently discards any exception. Then, when something unexpected happens and some other exception occurs, that gets swallowed too (instead of, say, crashing with an error message that can be acted on to find and fix the problem. The program keeps running, now in a corrupted state. It will probably crash at some point, but you will have a hard time tracking that back to the source of the problem. It *might* not screw up any I/O before then, too.
Personally, I *like* exceptions. They tell where something went wrong. By only ever writing try blocks around code that I actually think has a meaningful chance of generating an exception, plus usually a single global exception handler that logs the error report and exits as gracefully as possible, I don't have to worry that an unexpected but caught exception will silently put my program into an unknown state.
The ENIGMA was cool. I didn't even look at their pamphlets or anything, though. If the US found itself engaged in a legitimate war, I might consider provding my technical skills to the US government. Until then, I'll stay private sector, thank you very much.
There are rules on attendance (for example, no undercover reporters; media folks have special badges that immediately and visibly identify them, and the last one to be caught trying to pass got thoroughly humiliated and then kicked out and banned). They don't collect logs of who attends (you can't pay by credit card, because that leaves an auditable trail, for example) but they do exert some control over admission.
How big is this ring? Is there a loaded crossbow on that rack? Or even a normal bow, already strung? A chu ko nu, if loaded, would be ideal. Never bring a sword to an archery duel.
There's non-trivial overhead in exception handling. Even leaving aside the waste of developer time, it's more efficient to just test whether a given condition is going to occur than to have to wrap everything (often several layers deep, especially if you don't like automatic bubbling) in try blocks. It's one (of several) reasons that even JIT-compiled Java still performs notably worse than C++.
Whoa... I've seen good posts from you in the past, but this is way off the mark.
Claim: A publisher can whip up millions of ebooks with minimal effort and cost just as easily as easily as a pirate can. Fact: The cost to a publisher for a book (all of an ebook, or merely much of a dead-tree one; I'll focus on ebooks here) is in recouping the investment paid to contract the author, edit the book, market the book, and pay royalties (the details on contract vs. royalties vary, but there's always a major cost in one or the other for any commercial publication of a copyrighted work). The cost to a pirate is paying down the investment in a single copy of the book, which they might have actually just borrowed from a friend. If a pirate gives an ebook away for free, then it costs them basically nothing but publishers take a loss and authors (or at least ebook authors) become a bad investment. If a pirate sells the book for just under market cost, they undercut the publisher (while getting a free ride on their marketing and such, mind you) and still make a higher profit per sale. Just because there's no cost to make the copy itself doesn't mean that publishers can afford to not get paid for them! Before anybody starts whining about the evils of the publishing industry and making comparisons to the music / video industries, let me get in there first: publishing is not a giant corporate conglomerate like the RIAA or MPAA. There are tons of small, independent, and frequently struggling publishing houses out there. Few of them make anything close to MAFIAA levels of profit.
Just because it was possible to pirate copyrighted works before doesn't mean it was worth the slim profit margins and legal risks to invest the cost of doing so. Digital distribution has greatly reduced the risks, while digital copying has eliminated the investment and greatly increased the profit margin (or simply made it economical to give them away for free; after all, somebody else already footed the bill). It's true that ebooks offer the opportunity for increased profit margins and/or lower costs to legitimate publishers as well, of course, but it definitely disproportionately favors the pirates.
Now, with all that said, I don't think DRM is the solution. I don't buy DRMed media (I'm willing to rent it, so long as I know I can play it or worst-case get my money back, but I won't settle for purchasing a revocable and restrictive license). There have been experiments in the industry with DRM-free ebooks (Baen, for example, packaged dozens of freely redistributable, though not legally resalable, ebooks with their hardbound novels for a while as a sort of "try these other works by this author and similar ones for free" marketing tactic; that practice has largely died off with them so it was probably not very profitable, but the archives of their CDs are still available online at sites like baencd.thefifthimperium.com, plus they offer a few titles as free, DRM-free downloads on the Baen website at any given time). There are authors who have been successful while using copyright license which allow free redistribution. However, those success stories are few and far between.
I don't know what the solution is, but claiming that there is no problem, that ebooks are at least as beneficial to those an the legal side of copyright as those who are not, has no basis in reality.
High-tech regions tend to be high cost of living, too. In Silicon Valley, 100K USD may well be better than an entry-level salary for a dev with a 4-year degree. The cost of living is so high that this is less impressive than it sounds, though. It's a little less bad up around Seattle ("Silicon Forest") where starting salaries are more commonly in the 70-90k range, but people break six digits very quickly. I haven't job-hunted anywhere else, but at least on the west coast, a 100k estimated average might actually be low. People with more than 5 years in the industry can probably pull at least 50% more than that if they're any good. Also, that's just for a BS; get a MS or a PhD and you can definitely start at or above 100k (and yes, in this field there are high-paying jobs in industry for folks with doctorates).
A lot of motivation for insiders to disclose the "sensitive" information would go away.
(emphasis mine) That is completely irrelevant! The question is about information exfiltration, not about information publication. Industrial espionage, international intelligence, market speculation, all manner of (other) criminal operations can provide reasons to extract data from an organization. Whistleblowing is such a miniscule part of the pie that it's not even worth worrying about. It's a poorly-written question, but getting modded to +5 for not even answering it is lamer still.
That's idealistic and foolish. You're focusing on the NSA and Snowden. What you should be focusing on is any company, country, or organization which has an enemy or competitor, and any employee who is in they pay of said competition. Industrial espionage, international intelligence, market manipulation, smear campaigns, retaliation for a fully justified firing, whatever.
The vast majority of data exfiltration is not for any noble purpose of whistleblowing. Most of the time, the person in question isn't even particularly disaffected, just greedy, in a tight spot financially, has always had other loyalties, or is acting under duress (blackmail, threats, whatever). There's no amount of "feel like they're doing something good for the world" that will make up for those!
The question was asked very poorly, though. Snowden is exciting right now because of the nature of what he did and the large amount of media attention, but the question had nothing to do with whistleblowing; it's all about information control in general.
If by "lurk in the shadows" you mean "write a little Python script for a Raspberry Pi tucked behind a chair at your local Starbucks", and by "with a complete set of compromised copies of everything in the Google PlayStore" you mean "have the script be able to inject malicious code dynamically into an any APK in a few milliseconds", and by "sniffing in hopes of intercepting someone's traffic" you mean "running a persistent ARP spoofing attack that routes all external traffic across the network through said RaPi 24/7", and by "quickly to insert a compromised copy of something midstream with a man-in-the-middle attack" you mean "implement basic automated intercepting proxy functionality such as is common in dozens of existing tools"... then yes.
I don't think you realize how easy this kind of thing would be. Computers are tiny, silent, wireless, innocuous, and cheap these days. They are more than capable of modifying a typical APK in flight without introducing a human-noteworthy amount of latency. They can gain a MitM position easily an hold it for as long as the network stays up.
Yeah, those who stick to their cellular networks for app downloading are better off (unless there's a femtocell on that network and the attacker has access to it...) but for a few hours of hacking and less than $50 per location counting WiFi adaptor, you could catch a lot of people using WiFi on their phones.
Many Windows Phones (nearly all WP7 devices, and a growing handful of WP8 devices) have been "hacked" to enable custom ROMs. It's actually the preferred way to "root" most Windows Phone models, as it lets you overwrite some stuff that Microsoft locks when the OS boots up so you can't edit it regardless of permissions (technically this could be overwritten but it would require modifying kernel memory).
The XDA-Developers and WPCentral forums are full of custom ROMs and bootloader unlocks and so forth, at least for most of the older modules
Frankly, that's a very impressive flight sim for the time. Single joystick + "fire" button may make for really basic controls - notably, no throttle and probably no simulation of the acceleration on a dive or deceleration and possible stall on a climb (I suspect it limited your rate of climb/descent pretty severly, in fact) - but for a game with game with a 3D environment featuring enemies who maneuver, projectiles flying both directions, objects getting larger as they approach, enemies who could get behind you, all while you're moving in three dimensions? That's a very remarkable game, and one that I imagine most non-pilots would have severely struggled to control on their first play-throughs. The typical human just doesn't think in terms of moving in 3D.
According to Wikipedia (I never played the game; it came out somewhat before I was born), the game adapted its difficulty based on a running average of game length. My guess is that, if Sid Meyer's friend got a "good" score in the game, he probably exceeded the old average and thus was encountering more difficult waves. The fact that Sid Meyer could then double that score strongly suggests he had observed how the game increased its difficulty, and was simply able to keep going until he made a mistake that he knew was wrong, or human reaction time failed him. Either way, to get that far past the average, he must have had a very good understanding of how to pass each level, as they would have started becoming very difficult indeed.
For those interested: It's available, along with (finally) its expansion, on Good Old Games (gog.com), DRM free, patched for current Windows versions. The graphics are very dated, but the gameplay remains classic. The AI is also surprisingly good on the higher levels.
If the WDDM driver crashes twice within a timeout period, the kernel figures it's completely hosed and bugchecks (bluescreens). I've only ever personally seen this happen with nVidia drivers, but from late 2006 (early and pre-release Vista, under which I had no problems with ATi) to late 2012 (Win8) my primary PC used nVidia graphics (Vista and then Win7, not counting the various *nix distros) so it's possible I just missed the problem period. My Win8 machine with AMD graphics is very stable; I haven't even had user-mode graphics crashes (much less any bad enough to cause a bluescreen) in that period.
For the record, XP is perfectly capable of generating a kernel or full memory dump (I'm pretty sure 2000 was as well). You can also disable the immediate reboot after bugcheck for all NT versions since it was added (XP and later).
For Vista and later, the path is: Control Panel -> System -> Advanced system settings (left sidebar) -> Startup and Recovery (Settings button) For XP, it's similar except it's System -> Advanced (tab)
Um... no. That's ridiculous, in fact. As a MSDN subscriber, I can still download, from Microsoft, MS-DOS 6.0. That doesn't mean I have to call it a "0 year old" operating system!
If you want to get picky over the "12 years" claim, you could argue that I should count from the last time a major upgrade was released, which would be 5 years (and SP3, unlike SP1 and SP2, was hardly major).
Already commented here, but please, mod parent up. The drivers aren't gone, they just aren't going to be updated anymore.
Vista and later will download video drivers (not latest-and-greatest, but well-tested versions) from Windows Update; I've forgotten whether XP does that as well or not. I don't use MacOS 9, and I don't use XP.
Also, there's an important point here which isn't being addressed in the summary. Vista and later (all NT 6.x versions) use a new "WDDM" driver model for video drivers. Although there are various characteristics of WDDM, the really defining one is that only a tiny shim that basically wraps the direct hardware access lives in kernel mode. Everything else - the actual program logic of the video driver - lives in user mode. This is fantastic for a number of reasons: 1) All the crash-prone code is now user-mode. When a XP video driver crashes, it causes a bluescreen. When a Win7 video driver crashes, it causes a blank screen for about a second while the user-mode driver restarts. 2) Updating and rolling back video drivers no longer requires a reboot; in fact, it only takes a couple seconds. It's actually practical, if you really want to, to switch video drivers between games (for example, if the latest and "greatest" doesn't work with one of your older games, but you want to use it for everything else). 3) Developing and debugging user-mode code is a lot easier than doing the same for kernel-mode code. This change lets developers spend a greater portion of their time improving the driver logic, rather than making the driver work with the various configurations of the NT kernel.
My guess is that AMD decided the benefits of item #3 were worth more than continuing to release drivers for 12-year-old OS. By no longer maintaining the pre-WDDM version, they can focus their resources on supporting modern platforms that are also easier to develop for.
Slashdot Mirror
"new standard" of 1080p phones? I know exactly two models with that resolution and a screen that big or larger. The vast majority of brand new smartphones are still less than 5" and are at most 1280x768; most are actually 1280x720. Quad-core is similarly still uncommon, and is mostly a benfit on Android with its love af battery-killing background tasks.
At the time of its release, IE10 was faster than Chrome on both rendering and JS. IE10 on the phone is faster, hardware to hardware, than all but the latest browsers for Android, and the difference there is meaningless. It's got standards support, performance, and (for a phone browser) a decent feature set (I admit I'd like the synched bookmarks of Chrome, but there are apps for that).
Stop living in the past, grandpa. IE6 is dead and gone (speaking as a former web dev: thank $DEITY)
Half the reasons on that list were obsolete before WP8 even came out (and WP8 is more different under the covers from WP7 than WP7 was from WinMo6.5), and about an eighth of them were never true. Most of the rest have been obsoleted since. That list has been trotted out so many times it's *probably* got more links to it than the number of point-by-point refutations, some of which have been posted right here on Slashdot.
At this point, it's the equivalent of claiming that Windows 8 is still based on DOS because it includes cmd.exe (yes, I've seen people make that claim, both on here and elsewhere). It does nothing but label you as an idiot who badmouths things with no concept of how they work and likely no actual experience with them either.
Your comment could be taken far more seriously if you had given any indication that you even knew there was a difference between Windows and Windows Phone. Aside from both using NT-family kernels (definitely not the same kernel though, even accounting for CPU architecture) and having live tiles and sharing some APIs, the only similarity is their name. iOS and Android are more alike than Win8 and WP8.
MalachiK is using a WP7 device, which is decidedly lacking in features compared to WP8 or current Android versions. However, several of the features that Billly Gates mentions also exist on the Lumia 920, which runs WP8: data sense and warning, WiFi-based Internet sharing, voice activation, brightness sensors (not that those are anythign new) as do several other cool features that he didn't, like wireless charging and a low-light camera.
Excitement about features is cool. I totally understand. But don't confuse "Windows the workstation OS" with "Windows Phone the consumer smartphone platform". The reasons to buy the one have very little do with the reasons to buy the other, and there are plenty of people who are enthusiastic about the Lumia 92x and 1020 devices (no, I don't have one).
Oh dear, did I speak less than praisefully of somebody's favorite language? Haha, wow, so *sensitive*. I stated facts. Argue them if you want, but disagreeing with modpoints is just pure cowardice of somebody who can't back up their position.
For what it's worth, I code in both Java and C++ (and also C# and Python, plus Javascript and occasional other languages as the job requires). I'm not claiming that checked exceptions are either the Achilles' heel of Java performance, or that Java's performance is inadequate for real world work. However, I am claiming that checked exceptions are unneccessary, a waste of both developer and CPU time.
I'll go one further: they can be a source of hidden bugs. When you have excessive numbers of checked exceptions that a function might throw, and you either don't care about them or you know they won't happen, it's easiest to just wrap them in a try-swallow block that silently discards any exception. Then, when something unexpected happens and some other exception occurs, that gets swallowed too (instead of, say, crashing with an error message that can be acted on to find and fix the problem. The program keeps running, now in a corrupted state. It will probably crash at some point, but you will have a hard time tracking that back to the source of the problem. It *might* not screw up any I/O before then, too.
Personally, I *like* exceptions. They tell where something went wrong. By only ever writing try blocks around code that I actually think has a meaningful chance of generating an exception, plus usually a single global exception handler that logs the error report and exits as gracefully as possible, I don't have to worry that an unexpected but caught exception will silently put my program into an unknown state.
The ENIGMA was cool. I didn't even look at their pamphlets or anything, though. If the US found itself engaged in a legitimate war, I might consider provding my technical skills to the US government. Until then, I'll stay private sector, thank you very much.
There are rules on attendance (for example, no undercover reporters; media folks have special badges that immediately and visibly identify them, and the last one to be caught trying to pass got thoroughly humiliated and then kicked out and banned). They don't collect logs of who attends (you can't pay by credit card, because that leaves an auditable trail, for example) but they do exert some control over admission.
How big is this ring? Is there a loaded crossbow on that rack? Or even a normal bow, already strung? A chu ko nu, if loaded, would be ideal. Never bring a sword to an archery duel.
There's non-trivial overhead in exception handling. Even leaving aside the waste of developer time, it's more efficient to just test whether a given condition is going to occur than to have to wrap everything (often several layers deep, especially if you don't like automatic bubbling) in try blocks. It's one (of several) reasons that even JIT-compiled Java still performs notably worse than C++.
Whoa... I've seen good posts from you in the past, but this is way off the mark.
Claim: A publisher can whip up millions of ebooks with minimal effort and cost just as easily as easily as a pirate can.
Fact: The cost to a publisher for a book (all of an ebook, or merely much of a dead-tree one; I'll focus on ebooks here) is in recouping the investment paid to contract the author, edit the book, market the book, and pay royalties (the details on contract vs. royalties vary, but there's always a major cost in one or the other for any commercial publication of a copyrighted work). The cost to a pirate is paying down the investment in a single copy of the book, which they might have actually just borrowed from a friend. If a pirate gives an ebook away for free, then it costs them basically nothing but publishers take a loss and authors (or at least ebook authors) become a bad investment. If a pirate sells the book for just under market cost, they undercut the publisher (while getting a free ride on their marketing and such, mind you) and still make a higher profit per sale. Just because there's no cost to make the copy itself doesn't mean that publishers can afford to not get paid for them! Before anybody starts whining about the evils of the publishing industry and making comparisons to the music / video industries, let me get in there first: publishing is not a giant corporate conglomerate like the RIAA or MPAA. There are tons of small, independent, and frequently struggling publishing houses out there. Few of them make anything close to MAFIAA levels of profit.
Just because it was possible to pirate copyrighted works before doesn't mean it was worth the slim profit margins and legal risks to invest the cost of doing so. Digital distribution has greatly reduced the risks, while digital copying has eliminated the investment and greatly increased the profit margin (or simply made it economical to give them away for free; after all, somebody else already footed the bill). It's true that ebooks offer the opportunity for increased profit margins and/or lower costs to legitimate publishers as well, of course, but it definitely disproportionately favors the pirates.
Now, with all that said, I don't think DRM is the solution. I don't buy DRMed media (I'm willing to rent it, so long as I know I can play it or worst-case get my money back, but I won't settle for purchasing a revocable and restrictive license). There have been experiments in the industry with DRM-free ebooks (Baen, for example, packaged dozens of freely redistributable, though not legally resalable, ebooks with their hardbound novels for a while as a sort of "try these other works by this author and similar ones for free" marketing tactic; that practice has largely died off with them so it was probably not very profitable, but the archives of their CDs are still available online at sites like baencd.thefifthimperium.com, plus they offer a few titles as free, DRM-free downloads on the Baen website at any given time). There are authors who have been successful while using copyright license which allow free redistribution. However, those success stories are few and far between.
I don't know what the solution is, but claiming that there is no problem, that ebooks are at least as beneficial to those an the legal side of copyright as those who are not, has no basis in reality.
High-tech regions tend to be high cost of living, too. In Silicon Valley, 100K USD may well be better than an entry-level salary for a dev with a 4-year degree. The cost of living is so high that this is less impressive than it sounds, though. It's a little less bad up around Seattle ("Silicon Forest") where starting salaries are more commonly in the 70-90k range, but people break six digits very quickly. I haven't job-hunted anywhere else, but at least on the west coast, a 100k estimated average might actually be low. People with more than 5 years in the industry can probably pull at least 50% more than that if they're any good. Also, that's just for a BS; get a MS or a PhD and you can definitely start at or above 100k (and yes, in this field there are high-paying jobs in industry for folks with doctorates).
(emphasis mine)
That is completely irrelevant!
The question is about information exfiltration, not about information publication. Industrial espionage, international intelligence, market speculation, all manner of (other) criminal operations can provide reasons to extract data from an organization. Whistleblowing is such a miniscule part of the pie that it's not even worth worrying about. It's a poorly-written question, but getting modded to +5 for not even answering it is lamer still.
That's idealistic and foolish. You're focusing on the NSA and Snowden. What you should be focusing on is any company, country, or organization which has an enemy or competitor, and any employee who is in they pay of said competition. Industrial espionage, international intelligence, market manipulation, smear campaigns, retaliation for a fully justified firing, whatever.
The vast majority of data exfiltration is not for any noble purpose of whistleblowing. Most of the time, the person in question isn't even particularly disaffected, just greedy, in a tight spot financially, has always had other loyalties, or is acting under duress (blackmail, threats, whatever). There's no amount of "feel like they're doing something good for the world" that will make up for those!
The question was asked very poorly, though. Snowden is exciting right now because of the nature of what he did and the large amount of media attention, but the question had nothing to do with whistleblowing; it's all about information control in general.
If by "lurk in the shadows" you mean "write a little Python script for a Raspberry Pi tucked behind a chair at your local Starbucks", and by "with a complete set of compromised copies of everything in the Google PlayStore" you mean "have the script be able to inject malicious code dynamically into an any APK in a few milliseconds", and by "sniffing in hopes of intercepting someone's traffic" you mean "running a persistent ARP spoofing attack that routes all external traffic across the network through said RaPi 24/7", and by "quickly to insert a compromised copy of something midstream with a man-in-the-middle attack" you mean "implement basic automated intercepting proxy functionality such as is common in dozens of existing tools"... then yes.
I don't think you realize how easy this kind of thing would be. Computers are tiny, silent, wireless, innocuous, and cheap these days. They are more than capable of modifying a typical APK in flight without introducing a human-noteworthy amount of latency. They can gain a MitM position easily an hold it for as long as the network stays up.
Yeah, those who stick to their cellular networks for app downloading are better off (unless there's a femtocell on that network and the attacker has access to it...) but for a few hours of hacking and less than $50 per location counting WiFi adaptor, you could catch a lot of people using WiFi on their phones.
Many Windows Phones (nearly all WP7 devices, and a growing handful of WP8 devices) have been "hacked" to enable custom ROMs. It's actually the preferred way to "root" most Windows Phone models, as it lets you overwrite some stuff that Microsoft locks when the OS boots up so you can't edit it regardless of permissions (technically this could be overwritten but it would require modifying kernel memory).
The XDA-Developers and WPCentral forums are full of custom ROMs and bootloader unlocks and so forth, at least for most of the older modules
Frankly, that's a very impressive flight sim for the time. Single joystick + "fire" button may make for really basic controls - notably, no throttle and probably no simulation of the acceleration on a dive or deceleration and possible stall on a climb (I suspect it limited your rate of climb/descent pretty severly, in fact) - but for a game with game with a 3D environment featuring enemies who maneuver, projectiles flying both directions, objects getting larger as they approach, enemies who could get behind you, all while you're moving in three dimensions? That's a very remarkable game, and one that I imagine most non-pilots would have severely struggled to control on their first play-throughs. The typical human just doesn't think in terms of moving in 3D.
According to Wikipedia (I never played the game; it came out somewhat before I was born), the game adapted its difficulty based on a running average of game length. My guess is that, if Sid Meyer's friend got a "good" score in the game, he probably exceeded the old average and thus was encountering more difficult waves. The fact that Sid Meyer could then double that score strongly suggests he had observed how the game increased its difficulty, and was simply able to keep going until he made a mistake that he knew was wrong, or human reaction time failed him. Either way, to get that far past the average, he must have had a very good understanding of how to pass each level, as they would have started becoming very difficult indeed.
For those interested: It's available, along with (finally) its expansion, on Good Old Games (gog.com), DRM free, patched for current Windows versions. The graphics are very dated, but the gameplay remains classic. The AI is also surprisingly good on the higher levels.
If the WDDM driver crashes twice within a timeout period, the kernel figures it's completely hosed and bugchecks (bluescreens). I've only ever personally seen this happen with nVidia drivers, but from late 2006 (early and pre-release Vista, under which I had no problems with ATi) to late 2012 (Win8) my primary PC used nVidia graphics (Vista and then Win7, not counting the various *nix distros) so it's possible I just missed the problem period. My Win8 machine with AMD graphics is very stable; I haven't even had user-mode graphics crashes (much less any bad enough to cause a bluescreen) in that period.
For the record, XP is perfectly capable of generating a kernel or full memory dump (I'm pretty sure 2000 was as well). You can also disable the immediate reboot after bugcheck for all NT versions since it was added (XP and later).
For Vista and later, the path is:
Control Panel -> System -> Advanced system settings (left sidebar) -> Startup and Recovery (Settings button)
For XP, it's similar except it's System -> Advanced (tab)
Um... no. That's ridiculous, in fact. As a MSDN subscriber, I can still download, from Microsoft, MS-DOS 6.0. That doesn't mean I have to call it a "0 year old" operating system!
If you want to get picky over the "12 years" claim, you could argue that I should count from the last time a major upgrade was released, which would be 5 years (and SP3, unlike SP1 and SP2, was hardly major).
Already commented here, but please, mod parent up. The drivers aren't gone, they just aren't going to be updated anymore.
Vista and later will download video drivers (not latest-and-greatest, but well-tested versions) from Windows Update; I've forgotten whether XP does that as well or not. I don't use MacOS 9, and I don't use XP.
Also, there's an important point here which isn't being addressed in the summary.
Vista and later (all NT 6.x versions) use a new "WDDM" driver model for video drivers. Although there are various characteristics of WDDM, the really defining one is that only a tiny shim that basically wraps the direct hardware access lives in kernel mode. Everything else - the actual program logic of the video driver - lives in user mode. This is fantastic for a number of reasons:
1) All the crash-prone code is now user-mode. When a XP video driver crashes, it causes a bluescreen. When a Win7 video driver crashes, it causes a blank screen for about a second while the user-mode driver restarts.
2) Updating and rolling back video drivers no longer requires a reboot; in fact, it only takes a couple seconds. It's actually practical, if you really want to, to switch video drivers between games (for example, if the latest and "greatest" doesn't work with one of your older games, but you want to use it for everything else).
3) Developing and debugging user-mode code is a lot easier than doing the same for kernel-mode code. This change lets developers spend a greater portion of their time improving the driver logic, rather than making the driver work with the various configurations of the NT kernel.
My guess is that AMD decided the benefits of item #3 were worth more than continuing to release drivers for 12-year-old OS. By no longer maintaining the pre-WDDM version, they can focus their resources on supporting modern platforms that are also easier to develop for.
Well, or just disconnect the Kinect from the console.
Or the console from the Internet.
These are not difficult solutions. In fact, hey lookit that, Microsoft just announced that the cosole doesn't need a permanent connection anymore!
Last time I asked for a number the guy started swearing at me and then hung up. :-/ Haven't gotten a call since then but it wasn't long ago...