Pundits have been saying that at least since 90nm, you know. Then 65, and again at 45, 32, 33, and now at 14.
And those limits were dodged by some new process: SOI, copper interconnects, what have you. He's not saying we won't see 10 nm, he's just saying there's a good chance it will have to be something other than CMOS.
Moving up to a 1090T or 1100T Black Edition could be nice. 3.2 or 3.3 out of the box, but 3.8 or 3.9 is almost a given without worrying about voltage bumps and awesome cooling. My 1090 was throttling at 3.8 when running SuperPi on the stock cooler, but have no such problems with a Hyper212. I would need more cooling to run all-day-every-day at 4.0.
Granted, 2.8 to 3.8 may not make a lot of real world difference when memory bandwidth comes into play, and as you pointed out, the GPU has a lot to do with it (when it's relevant at all, which it isn't when, say, running six sessions of WinRAR at once).
AMD can hold their own in the mid-high end, though they aren't cranking out the high-margin "extreme" processors. Those are more of a status symbol anyhow, for both manufacturer and consumer, and I doubt that even Intel makes enough of them to be a big chunk of its gross income. That's why they're so damn expensive. Why make them any cheaper if you can't make them rapidly enough as it is?
Sure you can give 110%, just like an engine can deliver 110% of sustained power (which is how marine and aviation engines are rated, not instantaneous power). It just can't do that indefinitely, and it decreases reliability. Either service must be more frequent, or it's going to fail sooner, or both. Nonetheless, many marine engines are held in excess of 100% for hours or days on end.
Similarly, if you take work output divided by calendar time, you can deliver 110%... for a while. If you try to sustain this, you burn out. Just look at the game development industry for an example. They expect 110% or more, all the time, consequences be damned (so long as they fall on the hapless employees).
Say I'm a business owner, and I get burned by younger women who take the job, get knocked up, take the maximum legal maternity leave (while I pay for the medical insurance), and then decide to quit at the end and not come back. You think that's not going to dissuade me from hiring members of their class (young women likely to become mothers) when there are other applicants in front of me who can't burn me that way?
Yes, it's discrimination. No, I wouldn't tell anyone I'm doing it. I'd fix the demographics by hiring women in their 40s and beyond, when available, thus meeting both gender quotas AND age quotas in one fell swoop. Obviously, I'd have to accept younger women when they're the only ones qualified for (or possibly the only ones seeking) a certain job, but I wouldn't invest the same degree of training in them, knowing they have a propensity to abandon the working world as is convenient to them. It wouldn't be personal, and I wouldn't blame them, but it's still bad for my business to dump money into training employees who are likely to leave at the drop of a hat.
Now if I ran a business where everyone was replaceable, and nobody worked enough hours to get covered medical, I wouldn't give a shit. In fact, the propensity of those same women to leave the job would help reduce the overhead of people staying on too long and expecting ever increasing wages.
I'm not a business owner, and I have no plans to do anything of the sort, but I'd have to be a blithering idiot not to at least consider the problem. I've certainly seen this play out in the real world.
Depends on the game. For some games, the skill is all about quickly executing certain key combinations, for others it involves elements of timing. For others, that include a lot of scripting hooks just so you can make common things easier, they don't want you going overboard and writing complex interactions that do conditionally respond to stuff.
That still won't stop anyone. Write a macro that gets you to a conditional branch, then two or three or however many more you need based on the situation faced. "Oh, it's going THAT way, better use the Blue layer."
I'm not an elite gamer, and I never will be, but I do like my particular flavor of hardware. If you as an organizer were to tell me I had to use the same Razer Black Widow as you're saddling everyone else with, I'd be saying "no thanks, can I get my entry fee back?" because I don't even use QWERTY, or a staggered layout for that matter. I use a tweaked Dvorak layout in a rectilinear matrix, with my cursor pad on the left. It also happens to have 32 leftover keys for programming (41 if I want to fill every available space at the cost of comfort). I can add another 30 macros in the form of the Xkeys 16 strip I paid way too much for a few years back. Telling me I can't use these would be very similar to saying I couldn't use a mouse with twelve thumb buttons – which can also be used for macros. (I don't, I actually have just a six-button mouse and use the thumb buttons as modifiers, but I wouldn't complain about someone who did.)
You can play Limit without worrying about chip stacks, but yeah, No Limit is heavily dependent on having to decide when to go all in and when to live to fight another hand. I have to assume there are re-buys.
However, keyboards and keyboard converters are easily available which can do macros in the hardware.
Soarer's. Blue Cube.Tipro. Cherry G86 (and even some G80 and G81). Xkeys. All hardware-programmable (and that's just off the top of my head). Even if you can detect the use of "illegal macro software", what about the hardware options?
Which makes me ask, what the hell is an "illegal macro" anyhow? If something is so predictable that it can be scripted and bound to a single key, then it shouldn't really take multiple presses of a key to do it in the first place. This is not just limited to games, it extends to all software where I want to do a single, moderately tortuous task efficiently and often.
I would agree that particular scenario is unlikely. I was just making the point that there may be a conflict between what is best on a local scale and what is best on a global scale, and there may even be severally equally flawed proposals which apportion the damage differently. The ones taking the brunt of the pain aren't going to be overly sympathetic, even if it needs doing.
The two of you hopefully agree that your interests are aligned, however.
What happens if the Russian plan for reducing temperature means a return of Dust Bowl conditions in the Great Plains of North America, and they start doing it unilaterally? You don't see how that could lead to a rapid escalation with Mr. "I'll nuke before I give Crimea back" Putin?
Is it really so smart to rely on Skype, a Microsoft holding, for internal operations? I would assume they have the capability to listen in to whatever they like, and would certainly not want to use Skype to transact business that is in direct competition to another one of their divisions. This is above and beyond the fact that the Feds will be able to listen in, since there is only so much they can do to avoid that anyhow.
Version numbers are arbitrary, I never said they weren't. Yes, the decimal is a divider and not a mathematical decimal, so 5.10 comes after 5.9 where it can't in the hardness scale. I never said anything about "half the new features". In fact I'd a expect large gap between the last 5.x and 6.0 and not assume the version numbers meant anything in regards to the actual quantity of features, only about their timing. That is, if a feature is introduced in 5.2, it's probably still going to be there in 5.5 (unless it was fundamentally broken and had to be yanked).
The decimals on the Mohs scale seem arbitrary as well, but that does not make them meaningless. If you have just one substance between 7 and 8, you could justifiably call it 7.5, but what do you do if it turns out that everything else between 7 and 8 scratches it? You downgrade it to 7.1, that's what. These are like old BASIC line numbers where you assign them in the hopes that you don't have to change any later because of things you plug in, but sometimes you're wrong.
It is also comparing things to OTHER things that fall between the two points. Something is not just "between 7 and 8", it is also harder than some other materials "between 7 and 8" and softer than others. The number after the decimal allows for them to be rankedrelative to one another. What you're saying is that just because the code base of Software 5.0 and 5.1 are largely the same there is no way to distinguish between them, and that everything in the 5.x line except 5.0 itself is merely "between 5 and 6".
I hope they can have two of these ready when Roseanne Barr does lose all of her central vision. There's just no justice in the world if she's the only one on the planet who doesn't have to look at her face.
One potential problem I see (it happens already) is people being ticketed for driving solo, when in fact they have a baby in the car with them. You might not feel they should be in the HOV lane because that baby would never be out driving by itself, therefore they aren't saving any congestion, but it does meet the letter of the law. It also isn't going to be easily picked up on a camera. These people already get pulled over (and released) by cops who can't see the passenger in back. What is going to happen when an automated ticket shows up in the mail? You don't think the system is going to give them the benefit of the doubt, do you? If you do, please pass me some of what you're smoking, Pollyanna.
In reality they just eat up a lane of traffic that could otherwise be used to alleviate rush hour congestion. It might be different if they actually ADDED HOV lanes instead of taking one of the normal lanes and rebranding it.
Uh, in many places they have done just that. I don't know how common it is, but I've been to a number of places in the US where the HOV lanes are even added as completely separate lanes from the rest of traffic, and I recall when one was constructed as such -- added into what was previously the wide median area of a highway.
That's how it often is around here, although they did have to shift (and in some cases narrow) the other lanes a bit to make what used to be the shoulder wide enough for a lane. Part of the result is that you end up with two wheels constantly floating across the seams in the construction that were designed to be hidden between lanes rather than within them. They should have shifted things over a full half-lane so at least we'd be completely straddling the breaks.
Ask someone with phenylketonuria. I once went to a restaurant with a group, one of whom has this disorder. When he ordered a drink, he specifically said "NOT diet, I can't have phenylalanine". They brought him Diet Coke. He drank enough that some time (maybe twenty minutes) later, he had a freak-out and would have gotten all of us tossed out if he hadn't had enough sense to explain to us what he thought was about to happen. The restaurant quickly reversed tack to make sure they weren't going to get sued, while one of the people in the group had to drive him to a hospital to make sure he'd be OK.
Again, misleading does not necessarily mean they knew it to be untrue. They could have bought into the propaganda themselves.
Agent Smith sets up misleading guidelines setting a particularly low bar for matches on fingerprints, knowing what he's doing could net the wrong people. Agent Jones follows those guidelines and testifies on the assumption that they are sound. Both the court and Agent Jones have been misled, and Jones is not guilty of perjury. Smith did not testify, thus is not guilty of perjury either. A crime has probably been committed, it's just not perjury.
It's not perjury if you really believe it to be true. Merely being wrong is not enough. Being shown to be wrong may discredit a witness, but it does not in and of itself demonstrate perjury. Perjury requires intent.
If you were told that a match to a certain level of confidence is sufficient to make a case, and you testify that your samples matched to that level (because it turns out that's a very low bar), how is that perjury? Even if your initial assumption is wrong, you are reporting what you believe to be true.
There's nothing fundamentally different about the 88-108 MHz band compared to the 2 meter band (144-148) except that it has four times the bandwidth. Any transmission mode that works well at 2 meters will also work well in the FM band, as the propagation characteristics are substantially the same.
It could also be related to the ranking of minor league affiliates to Major League Baseball teams. A teams are typically rookies, or veterans well past their prime who are looking for a way to move into coaching or some non-playing role (but they're still good enough to show off some skills). AA teams are sort of Purgatory. If a prospect doesn't move through this level relatively quickly, statistics say he probably never will. AAA teams are where the replacement-level Major League talent is being held in reserve, or fine-tuned, or where injured players go to sharpen up after time off, etc.
What would be interesting in this grading system is that AAA is not the top, as the entire system is a feeder for The Show. What would the equivalent be in game studios? I have no idea.
One thing that needs to be taken care of is to make sure there is no copyright or any other form of so-called "intellectual property" on Mars. Not only will this save lifes by not having to worry about patents / design marks and whatever they come up with next, this also allows the Martians to have complete, full access to whatever media they want (think U.S.S. Enterprise-class storage systems with "the complete cultural accomplishments of planet Earth"), and create and share freely among themselves.
When sample-based hip-hop is only legal on Mars, only Mars will have... wait, what was the down side again?
Not any more isolation than expeditions to Antarctica in the late 19th and early 20th century.
Less, actually. There were no telegraph lines on the Antarctic expedition, and I don't know how effective radio was (not very in the late 19th century obviously). Aside from when the two planets are on opposite sides of the sun, communication will merely have high latency. We'll be back to sending podcasts and video messages, not chatting on Skype, but it's still quite a bit better than what those early explorers faced without even leaving the planet.
When the two planets are on opposite sides of the sun (which is what, a period of less than a week happening less than once a year?), a third point will have to be used to "go around", reducing bandwidth and adding to latency, but it's still better than nothing.
You'll note I specified monolingual Americans, which are the bulk of the country and thus the bulk of the target market for entertainment. I admit this logic may not hold in other countries, where English is a common second (or third or fourth) language.
This is all set up so they can rake over the richer countries without entirely locking themselves out of the less wealthy countries.
In a world with region locking: "Let's charge $50 in Burgerland and Poutineville, because they'll pay it. But we also want to make some money off their neighbors to the south, who won't pay $50. (Maybe they can't, maybe it's the burned DVDs for sale on the street for $2.) But now we have to stop the people we want $50 from, from importing the $10 copies. Region locking!"
In a world without region locking: "Let's charge $50 in Burgerland and Poutineville, because they'll pay it in order to have the content right now. We'll wait until the popularity goes down so that it's worth no more than $10 anywhere, and only then will we send it to places they'll only pay $10." By that time, those $2 burned DVD vendors have saturated the market so it's not even worth $10 there any longer.
However, it seems to me there is a form of "region locking" that follows the same general divisions. It's called a "language". Don't ship discs with all languages, just the one relevant to the buyer. Monolingual Americans are not going to watch Game of Thrones in Spanish just to get it cheaper.
Slashdot Mirror
Pundits have been saying that at least since 90nm, you know. Then 65, and again at 45, 32, 33, and now at 14.
And those limits were dodged by some new process: SOI, copper interconnects, what have you. He's not saying we won't see 10 nm, he's just saying there's a good chance it will have to be something other than CMOS.
Moving up to a 1090T or 1100T Black Edition could be nice. 3.2 or 3.3 out of the box, but 3.8 or 3.9 is almost a given without worrying about voltage bumps and awesome cooling. My 1090 was throttling at 3.8 when running SuperPi on the stock cooler, but have no such problems with a Hyper212. I would need more cooling to run all-day-every-day at 4.0.
Granted, 2.8 to 3.8 may not make a lot of real world difference when memory bandwidth comes into play, and as you pointed out, the GPU has a lot to do with it (when it's relevant at all, which it isn't when, say, running six sessions of WinRAR at once).
AMD can hold their own in the mid-high end, though they aren't cranking out the high-margin "extreme" processors. Those are more of a status symbol anyhow, for both manufacturer and consumer, and I doubt that even Intel makes enough of them to be a big chunk of its gross income. That's why they're so damn expensive. Why make them any cheaper if you can't make them rapidly enough as it is?
Sure you can give 110%, just like an engine can deliver 110% of sustained power (which is how marine and aviation engines are rated, not instantaneous power). It just can't do that indefinitely, and it decreases reliability. Either service must be more frequent, or it's going to fail sooner, or both. Nonetheless, many marine engines are held in excess of 100% for hours or days on end.
Similarly, if you take work output divided by calendar time, you can deliver 110%... for a while. If you try to sustain this, you burn out. Just look at the game development industry for an example. They expect 110% or more, all the time, consequences be damned (so long as they fall on the hapless employees).
Say I'm a business owner, and I get burned by younger women who take the job, get knocked up, take the maximum legal maternity leave (while I pay for the medical insurance), and then decide to quit at the end and not come back. You think that's not going to dissuade me from hiring members of their class (young women likely to become mothers) when there are other applicants in front of me who can't burn me that way?
Yes, it's discrimination. No, I wouldn't tell anyone I'm doing it. I'd fix the demographics by hiring women in their 40s and beyond, when available, thus meeting both gender quotas AND age quotas in one fell swoop. Obviously, I'd have to accept younger women when they're the only ones qualified for (or possibly the only ones seeking) a certain job, but I wouldn't invest the same degree of training in them, knowing they have a propensity to abandon the working world as is convenient to them. It wouldn't be personal, and I wouldn't blame them, but it's still bad for my business to dump money into training employees who are likely to leave at the drop of a hat.
Now if I ran a business where everyone was replaceable, and nobody worked enough hours to get covered medical, I wouldn't give a shit. In fact, the propensity of those same women to leave the job would help reduce the overhead of people staying on too long and expecting ever increasing wages.
I'm not a business owner, and I have no plans to do anything of the sort, but I'd have to be a blithering idiot not to at least consider the problem. I've certainly seen this play out in the real world.
Depends on the game. For some games, the skill is all about quickly executing certain key combinations, for others it involves elements of timing. For others, that include a lot of scripting hooks just so you can make common things easier, they don't want you going overboard and writing complex interactions that do conditionally respond to stuff.
That still won't stop anyone. Write a macro that gets you to a conditional branch, then two or three or however many more you need based on the situation faced. "Oh, it's going THAT way, better use the Blue layer."
I'm not an elite gamer, and I never will be, but I do like my particular flavor of hardware. If you as an organizer were to tell me I had to use the same Razer Black Widow as you're saddling everyone else with, I'd be saying "no thanks, can I get my entry fee back?" because I don't even use QWERTY, or a staggered layout for that matter. I use a tweaked Dvorak layout in a rectilinear matrix, with my cursor pad on the left. It also happens to have 32 leftover keys for programming (41 if I want to fill every available space at the cost of comfort). I can add another 30 macros in the form of the Xkeys 16 strip I paid way too much for a few years back. Telling me I can't use these would be very similar to saying I couldn't use a mouse with twelve thumb buttons – which can also be used for macros. (I don't, I actually have just a six-button mouse and use the thumb buttons as modifiers, but I wouldn't complain about someone who did.)
You can play Limit without worrying about chip stacks, but yeah, No Limit is heavily dependent on having to decide when to go all in and when to live to fight another hand. I have to assume there are re-buys.
However, keyboards and keyboard converters are easily available which can do macros in the hardware.
Soarer's. Blue Cube. Tipro. Cherry G86 (and even some G80 and G81). Xkeys. All hardware-programmable (and that's just off the top of my head). Even if you can detect the use of "illegal macro software", what about the hardware options?
Which makes me ask, what the hell is an "illegal macro" anyhow? If something is so predictable that it can be scripted and bound to a single key, then it shouldn't really take multiple presses of a key to do it in the first place. This is not just limited to games, it extends to all software where I want to do a single, moderately tortuous task efficiently and often.
I would agree that particular scenario is unlikely. I was just making the point that there may be a conflict between what is best on a local scale and what is best on a global scale, and there may even be severally equally flawed proposals which apportion the damage differently. The ones taking the brunt of the pain aren't going to be overly sympathetic, even if it needs doing.
The two of you hopefully agree that your interests are aligned, however.
What happens if the Russian plan for reducing temperature means a return of Dust Bowl conditions in the Great Plains of North America, and they start doing it unilaterally? You don't see how that could lead to a rapid escalation with Mr. "I'll nuke before I give Crimea back" Putin?
Is it really so smart to rely on Skype, a Microsoft holding, for internal operations? I would assume they have the capability to listen in to whatever they like, and would certainly not want to use Skype to transact business that is in direct competition to another one of their divisions. This is above and beyond the fact that the Feds will be able to listen in, since there is only so much they can do to avoid that anyhow.
Version numbers are arbitrary, I never said they weren't. Yes, the decimal is a divider and not a mathematical decimal, so 5.10 comes after 5.9 where it can't in the hardness scale. I never said anything about "half the new features". In fact I'd a expect large gap between the last 5.x and 6.0 and not assume the version numbers meant anything in regards to the actual quantity of features, only about their timing. That is, if a feature is introduced in 5.2, it's probably still going to be there in 5.5 (unless it was fundamentally broken and had to be yanked).
The decimals on the Mohs scale seem arbitrary as well, but that does not make them meaningless. If you have just one substance between 7 and 8, you could justifiably call it 7.5, but what do you do if it turns out that everything else between 7 and 8 scratches it? You downgrade it to 7.1, that's what. These are like old BASIC line numbers where you assign them in the hopes that you don't have to change any later because of things you plug in, but sometimes you're wrong.
It is also comparing things to OTHER things that fall between the two points. Something is not just "between 7 and 8", it is also harder than some other materials "between 7 and 8" and softer than others. The number after the decimal allows for them to be rankedrelative to one another. What you're saying is that just because the code base of Software 5.0 and 5.1 are largely the same there is no way to distinguish between them, and that everything in the 5.x line except 5.0 itself is merely "between 5 and 6".
I hope they can have two of these ready when Roseanne Barr does lose all of her central vision. There's just no justice in the world if she's the only one on the planet who doesn't have to look at her face.
One potential problem I see (it happens already) is people being ticketed for driving solo, when in fact they have a baby in the car with them. You might not feel they should be in the HOV lane because that baby would never be out driving by itself, therefore they aren't saving any congestion, but it does meet the letter of the law. It also isn't going to be easily picked up on a camera. These people already get pulled over (and released) by cops who can't see the passenger in back. What is going to happen when an automated ticket shows up in the mail? You don't think the system is going to give them the benefit of the doubt, do you? If you do, please pass me some of what you're smoking, Pollyanna.
In reality they just eat up a lane of traffic that could otherwise be used to alleviate rush hour congestion. It might be different if they actually ADDED HOV lanes instead of taking one of the normal lanes and rebranding it.
Uh, in many places they have done just that. I don't know how common it is, but I've been to a number of places in the US where the HOV lanes are even added as completely separate lanes from the rest of traffic, and I recall when one was constructed as such -- added into what was previously the wide median area of a highway.
That's how it often is around here, although they did have to shift (and in some cases narrow) the other lanes a bit to make what used to be the shoulder wide enough for a lane. Part of the result is that you end up with two wheels constantly floating across the seams in the construction that were designed to be hidden between lanes rather than within them. They should have shifted things over a full half-lane so at least we'd be completely straddling the breaks.
Since when is Sucralose better than Aspartame?
Ask someone with phenylketonuria. I once went to a restaurant with a group, one of whom has this disorder. When he ordered a drink, he specifically said "NOT diet, I can't have phenylalanine". They brought him Diet Coke. He drank enough that some time (maybe twenty minutes) later, he had a freak-out and would have gotten all of us tossed out if he hadn't had enough sense to explain to us what he thought was about to happen. The restaurant quickly reversed tack to make sure they weren't going to get sued, while one of the people in the group had to drive him to a hospital to make sure he'd be OK.
Again, misleading does not necessarily mean they knew it to be untrue. They could have bought into the propaganda themselves.
Agent Smith sets up misleading guidelines setting a particularly low bar for matches on fingerprints, knowing what he's doing could net the wrong people. Agent Jones follows those guidelines and testifies on the assumption that they are sound. Both the court and Agent Jones have been misled, and Jones is not guilty of perjury. Smith did not testify, thus is not guilty of perjury either. A crime has probably been committed, it's just not perjury.
It's not perjury if you really believe it to be true. Merely being wrong is not enough. Being shown to be wrong may discredit a witness, but it does not in and of itself demonstrate perjury. Perjury requires intent.
If you were told that a match to a certain level of confidence is sufficient to make a case, and you testify that your samples matched to that level (because it turns out that's a very low bar), how is that perjury? Even if your initial assumption is wrong, you are reporting what you believe to be true.
There's nothing fundamentally different about the 88-108 MHz band compared to the 2 meter band (144-148) except that it has four times the bandwidth. Any transmission mode that works well at 2 meters will also work well in the FM band, as the propagation characteristics are substantially the same.
It could also be related to the ranking of minor league affiliates to Major League Baseball teams. A teams are typically rookies, or veterans well past their prime who are looking for a way to move into coaching or some non-playing role (but they're still good enough to show off some skills). AA teams are sort of Purgatory. If a prospect doesn't move through this level relatively quickly, statistics say he probably never will. AAA teams are where the replacement-level Major League talent is being held in reserve, or fine-tuned, or where injured players go to sharpen up after time off, etc.
What would be interesting in this grading system is that AAA is not the top, as the entire system is a feeder for The Show. What would the equivalent be in game studios? I have no idea.
You'll never be the Purple Squirrel,
You'll never even see on.
'Cause I can tell you anyhow,
They'd rather H1B one.
One thing that needs to be taken care of is to make sure there is no copyright or any other form of so-called "intellectual property" on Mars. Not only will this save lifes by not having to worry about patents / design marks and whatever they come up with next, this also allows the Martians to have complete, full access to whatever media they want (think U.S.S. Enterprise-class storage systems with "the complete cultural accomplishments of planet Earth"), and create and share freely among themselves.
When sample-based hip-hop is only legal on Mars, only Mars will have... wait, what was the down side again?
Not any more isolation than expeditions to Antarctica in the late 19th and early 20th century.
Less, actually. There were no telegraph lines on the Antarctic expedition, and I don't know how effective radio was (not very in the late 19th century obviously). Aside from when the two planets are on opposite sides of the sun, communication will merely have high latency. We'll be back to sending podcasts and video messages, not chatting on Skype, but it's still quite a bit better than what those early explorers faced without even leaving the planet.
When the two planets are on opposite sides of the sun (which is what, a period of less than a week happening less than once a year?), a third point will have to be used to "go around", reducing bandwidth and adding to latency, but it's still better than nothing.
You'll note I specified monolingual Americans, which are the bulk of the country and thus the bulk of the target market for entertainment. I admit this logic may not hold in other countries, where English is a common second (or third or fourth) language.
This is all set up so they can rake over the richer countries without entirely locking themselves out of the less wealthy countries.
In a world with region locking: "Let's charge $50 in Burgerland and Poutineville, because they'll pay it. But we also want to make some money off their neighbors to the south, who won't pay $50. (Maybe they can't, maybe it's the burned DVDs for sale on the street for $2.) But now we have to stop the people we want $50 from, from importing the $10 copies. Region locking!"
In a world without region locking: "Let's charge $50 in Burgerland and Poutineville, because they'll pay it in order to have the content right now. We'll wait until the popularity goes down so that it's worth no more than $10 anywhere, and only then will we send it to places they'll only pay $10." By that time, those $2 burned DVD vendors have saturated the market so it's not even worth $10 there any longer.
However, it seems to me there is a form of "region locking" that follows the same general divisions. It's called a "language". Don't ship discs with all languages, just the one relevant to the buyer. Monolingual Americans are not going to watch Game of Thrones in Spanish just to get it cheaper.