Unfortunately, that's really not in the best interests of the companies involved. They want you continuously forking over money as long as you have any interest in the game - with the threat that if you don't pay 'em, your character or characters are erased. Basically, it's a creative form of extortion.
This may have something to do with the new standard. If they make the wireless adapter 802.11g and the demand is for the 802.11n by the time the new XBox is released, they're stuck with an old standard. 10/100 Ethernet has been around forever, but wireless is still developing rapidly. 802.11n will supposedly have between 100Mbps and 320Mbps, which puts it equal to or better than standard ethernet, as well as increased reliability, or so I hear (this is a real problem in my 900MHz and 2.4GHz noisy area).
Compared with today's fastest supercomputers, it will be six times faster, consume 1/15th the power per computation and be 10 times more compact than today's fastest supercomputers.
-Team XBox (see link above)
1/15th the power consumption is a heck of a lot of heat that doesn't need to be dissipated. Supposedly can be run without CPU fans, too, but it's still very experimental and that may change when they crank the speed up (the heatsink-and-fanless one for supercomputers I saw was only 700MHz... I don't know if it requires cooling for practical use, though). It's also capable of 200 simultaneous computations per cycle, supposedly.
quick background: 1889 - a measure of a physical object distance 1893-1960 - a measure done through interferometry (wavelengths of light) on the physical object. 1960-present measurements of krypton or light (not physical object) for creation of SI units.
The problem is, the interferometry measurements were done using white light, not laser light (laser invented in 1960, incidentally), so there potentially is a fair degree of error in the interferometery measurements. I don't know if that is enough to make.0001, but I suspect it could be.
Space Ship One isn't restricted by government mandated standards for "safe" space travel, either. Triple-redundancy of all critical components and heavily tested radiation hardening (which is why many of the chips used are 3-4 years old), for instance.
It also doesn't have the contractual and budgetary quirks that give you a $900 toilet seat or $2000+ hammer. The main problem is that the government has no idea what a certain item will cost for R&D and construction and budgets a certain amount to a contractor. If the contractor spends $10000 and finds an acceptable hammer solution on a $500000 budget, the numbers get badly skewed, quickly. btw, it seems to me that the toilet seat also included the framework to hold it and had to have specific testing to pass military specifications, which may also have contributed to the expense.
NASA itself is really a different bird... the Shuttle is build as a heavy load lifter, not a passenger craft. A lot of the parts for the shuttle are contracted, not built in house, which has the advantage of getting competing designs, but the disadvantage of added expense (even if you go with the lowest bidder, the cost of evaluation probably makes up for the difference). NASA also has a huge R&D role and gets their fingers in everything, from new materials and fabrics to foodstuffs and weightlessness research.
SpaceShipOne fills a void, because NASA feels there is no need for passenger spacecraft - and from their point of view, they're correct. NASA's primary goals are military deployment and research, so only puts people in space to do work - construction, fixing satellites, research etc. They don't care about the commercial aspects like tourism because they're not a business. A lot of the research done or funded by NASA eventually trickles down into consumer goods.
I was wrong on the Sony part - they appear to be using a new processor of their own design. The article I read compared Nintendo and Microsoft, and then I looked up Nintendo because I though it was Sony and Microsoft...
I had thought this might be one of the reasons MS bought VirtualPC - at least, I did when I heard that the CPU they were going to use was PowerPC. Now I'm not so sure - emulating/overriding the nVidia graphics subsystem may not be possible. I was thinking they'd port the API layer and only emulate instructions as necessary (like the WINE on PPC proposed - probably much easier to do with the original source code).
The main reason to switch to the PowerPC was the "computer on a chip" tech that makes building them MUCH cheaper. Rumor has it that Sony and Microsoft may even use the same multi-core chip (and Nintendo a similar one). Not that it matters much - the GPU(s) is going to matter more for polygon count.
I think you've missed the point, but indirectly hit it, as well.
There are games that show women as sex objects like Vice City and Donkey Kong (for lack of a better example), but there are also lots of strong female character games, as well, like Beyond Good and Evil. What differentiates the two is that many of the women in the male character games are sex objects or victims, which is not the case for the female character leads, in which men are, at worst, equals. If anything, a game with a female PC and male NPC, the female character will want romance, not just using the guy and crushing his fragile little heart under 3" stilletos.
It'd be interesting to see a game with a female lead like that, but probably a bit twisted, as well. I imagine something like that would start in a RPG, where you start down a romance plot, then squash the guy's aspirations for someone else (I'm sorry I have to dump you, Gareth, you're so sweet and caring, but I met this guy, Conan, and, well, just LOOK at him!).
As an odd footnote, I've personally known more professional female game programmers than male game programmers, but that has a lot to do with knowing several female edu-ware game programmers ('cause I got to know the cute girl in AI class who worked for one - it was a swing and a miss, but fun while it lasted). Both male game programmers I knew worked for Loki at one time, oddly enough... not sure where they are now.
When I first saw this, I started wondering what tools are available for recording 5.1 surround. Google didn't find any in my searches (though I may have chosen poor search words).
Although I didn't find any free tools, I did find several commercial tools, which makes me wonder if you need a license to record and/or distribute 5.1 audio (sorta like mp3 - I'd assume this would apply to licensed technologies like EAX and Dolby). There also doesn't appear to be a good open format for 5.1 or 7.1 audio, meaning you're probably stuck licensing for it. Since most people only have stereo speakers, studios probably figure it's not worth the extra expense (especially if there's a per-CD license, though that fee often is only to display the logo, like THX, which is often a selling point).
That's just the problem - few companies do moderate real play because of deadlines.
Until code cutoff, new features are always going into the code, any of which can cause bugs (alpha). At some cutoff point, only bug fixes are allowed in (beta - but cutoff is often missed for that great feature that has to go in). Most of the time spent here is play balancing and feature verification. That last week or two before release to manufacturer is probably the only real testing done on the final product ('rc' or release candidate testing). After that, the testers get moved on to planning for the next project.
I have a feeling the real problem was the complete game didn't get thoroughly tested before the publisher deadline for an E3 release. Lack of thorough rc testing is an industry-wide problem, usually pushed by publishers, who are, in turn, pushed by investors.
Actually, you have two planes of angles to work with from the top, and two from the side.
The stealth aircraft have little odd angled "mirror rooms" (for lack of a better words - think a house of horrors hall of mirrors) that temporarily absorb signals, bounce them around a bit and let them out at various angles at various times, which is why they have a signature, but it is a lot like a flock of birds, not an airplane. A ship would probably reflect radar coming from the side into nearby waves and use them for the scattering effect and try to redirect deck waves in a direction other than straight back (thus the non-90 degree angle).
The Xeon is being tested - the Dell Precision 650 is a Xeon processor (as is the HP on a linked page). The problem is, the benchmark is playing to the strengths of the PPC and the weaknesses of this particular Xeon. Apple vs Precision 650: Memory spec: DDR400 vs DDR266 FSB: 900-1250MHz vs 533MHz drive (wild card): SATA drive vs U320 SCSI drive.
I say wild card on drive because, in general current U320 drives are faster than SATA (so setup plays a factor - Raid, # drives, cache, etc), but also since all these drives transfer at more than 120MB/sec (1000Mbit) this is probably not a bottleneck.
Also, neither machine lists CAS or any other latency, and this is one of those cases where it would be important. Since Apple added RAM, there's no telling what they put in. It could be a completely unfair 400/2/2/2/5 vs 266/3/4/4/8. Did they use DDR slot configuration? C'mon Apple, there are extreme geeks that want to know:) In Apple's defense, 99.9% of people probably don't know what the heck I just said, so it doesn't make sense to mention it in market documents. Still, I'd like to see a detailed comparison. Sigh.
The test listed is primarily a memory movement test (disk to memory, process, then back to disk), since the process time is probably minimal, the bus and memory speeds are more than likely the difference. Many of the benchmarks Apple shows are bus and memory intensive... yeah, marketing! If you're into live video or audio editing or streaming, it's an awesome system. I suspect games won't fare quite as well, but that depends on work offloaded to the GPU.
The 8080 architecture is almost a non-issue; the SIMD instructions have long ago been divided into inline RISC instructions. Longer pipelines aren't always better, it depends heavily on how good your branch prediction is and how often you have pipeline flushes (compiler is important!). Context switches between integer, SIMD and float (one running at a time) are also a thing of the past.
I'm not sure if scaling the numbers for Power4 would work, as you propose, as it really depends on what benchmark was run and how. Many are designed to hit a single processor and avoid cache, so multiple cores (basically multiple processors on a single dye) and huge cache would be ineffective. I would suspect that the CPU is actually quite a bit slower.
I was going to say basically the same - add a house with a 200k in mortgage debt and/or breed and you can easily knock yourself down an income bracket (student loans help, too). One year I actually dropped down two brackets, but that was because I had to sell mutual funds at a heavy loss to pay off student and car loans just to get the loan to buy my house. It ended up to be a good choice - the mutual funds have been stagnant in the last few years, the car has depreciated as expected, and my house value has nearly tripled. Not to mention I was able to refinance to a 15 year fixed mortgage for just $100 more than my 30 year Variable.
It's also a rather tricky bug that may not have been caught in their testing process.
I can entirely see how their Q/A team would miss the bug - Q/A would need to be playing on hard, probably with no cheats on (otherwise, why save and reload?). Save and reload probably got some visible verification (inventory there, start point correct, etc), so that was probably checked off without further testing. Most testing probably never used save and reload - mainly because that is one of the last completed parts of the game completed. Portions of save/restore may work, but until item placement and inventory items are complete, why test it?
I admit, what they need is to have some people run through the "finished" product once or twice at every difficulty to verify there are no outstanding showstoppers, but that's not always possible (time demands), so maybe they settled on a runthrough only at moderate difficulty. Due to the limited lifespan of games (about 3 months) they probably didn't want to have an open beta (no sense leaking the code to pirates any earlier than possible).
Part of the problem is that the BBC article seems to be missing some history behind the story that makes a non-island city a possibility. Plato pretty much lifted the Atlantis (city) story from the Egyptian tale of Keftiu (as well as embellished on it), a city that supposedly existed past the Pillars of Hercules (Straight of Gibraltar today, which separates Spain and Morocco). Keftiu is rooted in the Egyptian word for Pillar and was believed to be the end of the earth where the sky was held up. Atlantis means isle of Atlas - recognize the similarity? Atlas held up the world in Greek mythology. Keftiu also wasn't necessarily an island - it can either mean the Isle of Keft or the People of Keft. So, possibly due to a simple translation error, an island was born.
This could very easily be Atlantis. Minoan Crete never made sense (it never sunk) - Santorini island made more sense as most of it blew up (flooding Minoan Crete). It seems to me, though, that it was described as "west of Egypt" and that island's really NW.
That doesn't explain how it was lost, although speculation of a great city that sunk deep into a lake after an earthquake (or something like that) does exist in Mexico.
There is, however, some speculative evidence that supports your theory. German scientists claim to have found nicotine, cocaine, and THC in mummies, although mainstream archeologists say the claims are false. If cocaine was truely found in mummies, that would mean trade routes of some kind to South America (or a native crop before Egypt became desert?). How or when such a trade route started or ended is a matter of speculation. The main problem here is timeline - Atlantis supposedly sunk about 10000 years ago, meaning the 4000 year old mummies aren't old enough.
Silly, but implemented and put on some CDs already (especially in Europe). Look up CDS-100 and CDS-200 encoding (stands for Cactus Data Shield and is now owned by Macrovision).
My understanding is they add "spikes" to the wave at mixed intervals that are supposedly not perceptable when the music is played. When the music is compressed and/or converted into a different format, these spikes join together or are somehow elongated into the audible range and distort the sound.
Yeah, it's true that MS gaming division isn't making them money, but their bread-and-butter, MS-Office and MS-Windows make so much dough that they pay the losses of every losing division and still rake in over a billion dollars without even mentioning their other profitable divisions. I don't think MS has anything to worry about, except maybe the PC drying up as a gaming market. I have my doubts about that, as well, because graphics technology continues to be pushed on the PC side, which eventually influences consoles. Consoles have the plus of standardized hardware, so it's easier to write for them, but you don't get "cutting edge" graphics, except maybe on the console's release.
MS has hurt PC gaming, by buying many PC developers and moving them exclusively to XBox dev (then porting to PC with a different developer later). The FASA (Mechwarrior) and Bungie (Halo) teams are prime examples. Basically, they've moved the PC game to play second fiddle to the XBox game, but adding all the missing features later. In many ways this works out good for MS, because the XBox is basically a standardized PC, which means many less configuration problems and simplified debugging (meaning shorter release schedule and thus less investment), so they can shovel the cost of PC hardware debugging to a third party developer. Unfortunately, it also means late releases on the PC and controls that either don't work, or are so dumbed down the game is either too easy or too hard:(
I build PCs, and for $600, I've build some pretty nice machines that are perfectly capable of playing even the latest games - maybe not with everything on, but with a decent feature set at 1024x768. For $600, I have to try to keep the CPU to about $70, HardDisk to $60, Memory to $100, Windows to $100, Video to $150, mobo to $50 (sound and ethernet onboard), and floppy, mouse, keyboard, and case to the other $70. I usually have some give-and-take by scouring pricewatch and Ebay, or saving $20 by using slower memory (one of the lower impacts to games). The sweet-spot is probably closer to $850-$1000 on a new machine, though (spending more on processor, video and hard-disk; for $1000, get a better mobo and case/PSU). After that, you're splurging on stuff you can buy for significantly less in just a few months. Not splurging, however, puts you in the eternal upgrade spiral, which I've been in for about 2 years;)
Slashdot Mirror
Continuum
Brian Wilson (one of the authors) even posted the source
One of my old faves, as well.
Unfortunately, that's really not in the best interests of the companies involved. They want you continuously forking over money as long as you have any interest in the game - with the threat that if you don't pay 'em, your character or characters are erased. Basically, it's a creative form of extortion.
This may have something to do with the new standard. If they make the wireless adapter 802.11g and the demand is for the 802.11n by the time the new XBox is released, they're stuck with an old standard. 10/100 Ethernet has been around forever, but wireless is still developing rapidly. 802.11n will supposedly have between 100Mbps and 320Mbps, which puts it equal to or better than standard ethernet, as well as increased reliability, or so I hear (this is a real problem in my 900MHz and 2.4GHz noisy area).
By using the new Blue Gene processors.
To quote:
Compared with today's fastest supercomputers, it will be six times faster, consume 1/15th the power per computation and be 10 times more compact than today's fastest supercomputers.
-Team XBox (see link above)
1/15th the power consumption is a heck of a lot of heat that doesn't need to be dissipated. Supposedly can be run without CPU fans, too, but it's still very experimental and that may change when they crank the speed up (the heatsink-and-fanless one for supercomputers I saw was only 700MHz... I don't know if it requires cooling for practical use, though). It's also capable of 200 simultaneous computations per cycle, supposedly.
Actually, it does sort of make sense
.0001, but I suspect it could be.
quick background:
1889 - a measure of a physical object distance
1893-1960 - a measure done through interferometry (wavelengths of light) on the physical object.
1960-present measurements of krypton or light (not physical object) for creation of SI units.
The problem is, the interferometry measurements were done using white light, not laser light (laser invented in 1960, incidentally), so there potentially is a fair degree of error in the interferometery measurements. I don't know if that is enough to make
Space Ship One isn't restricted by government mandated standards for "safe" space travel, either. Triple-redundancy of all critical components and heavily tested radiation hardening (which is why many of the chips used are 3-4 years old), for instance.
It also doesn't have the contractual and budgetary quirks that give you a $900 toilet seat or $2000+ hammer. The main problem is that the government has no idea what a certain item will cost for R&D and construction and budgets a certain amount to a contractor. If the contractor spends $10000 and finds an acceptable hammer solution on a $500000 budget, the numbers get badly skewed, quickly. btw, it seems to me that the toilet seat also included the framework to hold it and had to have specific testing to pass military specifications, which may also have contributed to the expense.
NASA itself is really a different bird... the Shuttle is build as a heavy load lifter, not a passenger craft. A lot of the parts for the shuttle are contracted, not built in house, which has the advantage of getting competing designs, but the disadvantage of added expense (even if you go with the lowest bidder, the cost of evaluation probably makes up for the difference). NASA also has a huge R&D role and gets their fingers in everything, from new materials and fabrics to foodstuffs and weightlessness research.
SpaceShipOne fills a void, because NASA feels there is no need for passenger spacecraft - and from their point of view, they're correct. NASA's primary goals are military deployment and research, so only puts people in space to do work - construction, fixing satellites, research etc. They don't care about the commercial aspects like tourism because they're not a business. A lot of the research done or funded by NASA eventually trickles down into consumer goods.
I was wrong on the Sony part - they appear to be using a new processor of their own design. The article I read compared Nintendo and Microsoft, and then I looked up Nintendo because I though it was Sony and Microsoft...
:)
Brain fart
I had heard the same thing, which makes me suspect that it either didn't happen, or he's keeping a tighter lid on the ideas this time.
:)
I certainly didn't learn anything from the article
I had thought this might be one of the reasons MS bought VirtualPC - at least, I did when I heard that the CPU they were going to use was PowerPC. Now I'm not so sure - emulating/overriding the nVidia graphics subsystem may not be possible. I was thinking they'd port the API layer and only emulate instructions as necessary (like the WINE on PPC proposed - probably much easier to do with the original source code).
The main reason to switch to the PowerPC was the "computer on a chip" tech that makes building them MUCH cheaper. Rumor has it that Sony and Microsoft may even use the same multi-core chip (and Nintendo a similar one). Not that it matters much - the GPU(s) is going to matter more for polygon count.
I think you've missed the point, but indirectly hit it, as well.
There are games that show women as sex objects like Vice City and Donkey Kong (for lack of a better example), but there are also lots of strong female character games, as well, like Beyond Good and Evil. What differentiates the two is that many of the women in the male character games are sex objects or victims, which is not the case for the female character leads, in which men are, at worst, equals. If anything, a game with a female PC and male NPC, the female character will want romance, not just using the guy and crushing his fragile little heart under 3" stilletos.
It'd be interesting to see a game with a female lead like that, but probably a bit twisted, as well. I imagine something like that would start in a RPG, where you start down a romance plot, then squash the guy's aspirations for someone else (I'm sorry I have to dump you, Gareth, you're so sweet and caring, but I met this guy, Conan, and, well, just LOOK at him!).
As an odd footnote, I've personally known more professional female game programmers than male game programmers, but that has a lot to do with knowing several female edu-ware game programmers ('cause I got to know the cute girl in AI class who worked for one - it was a swing and a miss, but fun while it lasted). Both male game programmers I knew worked for Loki at one time, oddly enough... not sure where they are now.
I was half expecting to find this thing was water cooled :P
Intel has a history of supporting weird uses of their chips, like these
When I first saw this, I started wondering what tools are available for recording 5.1 surround. Google didn't find any in my searches (though I may have chosen poor search words).
Although I didn't find any free tools, I did find several commercial tools, which makes me wonder if you need a license to record and/or distribute 5.1 audio (sorta like mp3 - I'd assume this would apply to licensed technologies like EAX and Dolby). There also doesn't appear to be a good open format for 5.1 or 7.1 audio, meaning you're probably stuck licensing for it. Since most people only have stereo speakers, studios probably figure it's not worth the extra expense (especially if there's a per-CD license, though that fee often is only to display the logo, like THX, which is often a selling point).
dammit, why didn't you tell me earlier! I got hitched 3 years ago already...
maybe I'd even still be having sex...
and since I read (in the can) that 70% of Americans do most of their reading in the can, this is probably a bigger issue than they realize.
Maybe it'll work in France...
That's just the problem - few companies do moderate real play because of deadlines.
Until code cutoff, new features are always going into the code, any of which can cause bugs (alpha). At some cutoff point, only bug fixes are allowed in (beta - but cutoff is often missed for that great feature that has to go in). Most of the time spent here is play balancing and feature verification. That last week or two before release to manufacturer is probably the only real testing done on the final product ('rc' or release candidate testing). After that, the testers get moved on to planning for the next project.
I have a feeling the real problem was the complete game didn't get thoroughly tested before the publisher deadline for an E3 release. Lack of thorough rc testing is an industry-wide problem, usually pushed by publishers, who are, in turn, pushed by investors.
Actually, you have two planes of angles to work with from the top, and two from the side.
The stealth aircraft have little odd angled "mirror rooms" (for lack of a better words - think a house of horrors hall of mirrors) that temporarily absorb signals, bounce them around a bit and let them out at various angles at various times, which is why they have a signature, but it is a lot like a flock of birds, not an airplane. A ship would probably reflect radar coming from the side into nearby waves and use them for the scattering effect and try to redirect deck waves in a direction other than straight back (thus the non-90 degree angle).
Sounds like the fire on the USS Forrestal. I believe both the first and second groups trying to control the fire were wiped out by explosions.
There's some pretty shocking film footage of the whole thing which has been aired a couple of times on the History Channel.
The Xeon is being tested - the Dell Precision 650 is a Xeon processor (as is the HP on a linked page). The problem is, the benchmark is playing to the strengths of the PPC and the weaknesses of this particular Xeon.
:) In Apple's defense, 99.9% of people probably don't know what the heck I just said, so it doesn't make sense to mention it in market documents. Still, I'd like to see a detailed comparison. Sigh.
Apple vs Precision 650:
Memory spec: DDR400 vs DDR266
FSB: 900-1250MHz vs 533MHz
drive (wild card): SATA drive vs U320 SCSI drive.
I say wild card on drive because, in general current U320 drives are faster than SATA (so setup plays a factor - Raid, # drives, cache, etc), but also since all these drives transfer at more than 120MB/sec (1000Mbit) this is probably not a bottleneck.
Also, neither machine lists CAS or any other latency, and this is one of those cases where it would be important. Since Apple added RAM, there's no telling what they put in. It could be a completely unfair 400/2/2/2/5 vs 266/3/4/4/8. Did they use DDR slot configuration? C'mon Apple, there are extreme geeks that want to know
The test listed is primarily a memory movement test (disk to memory, process, then back to disk), since the process time is probably minimal, the bus and memory speeds are more than likely the difference. Many of the benchmarks Apple shows are bus and memory intensive... yeah, marketing! If you're into live video or audio editing or streaming, it's an awesome system. I suspect games won't fare quite as well, but that depends on work offloaded to the GPU.
The 8080 architecture is almost a non-issue; the SIMD instructions have long ago been divided into inline RISC instructions. Longer pipelines aren't always better, it depends heavily on how good your branch prediction is and how often you have pipeline flushes (compiler is important!). Context switches between integer, SIMD and float (one running at a time) are also a thing of the past.
I'm not sure if scaling the numbers for Power4 would work, as you propose, as it really depends on what benchmark was run and how. Many are designed to hit a single processor and avoid cache, so multiple cores (basically multiple processors on a single dye) and huge cache would be ineffective. I would suspect that the CPU is actually quite a bit slower.
I was going to say basically the same - add a house with a 200k in mortgage debt and/or breed and you can easily knock yourself down an income bracket (student loans help, too). One year I actually dropped down two brackets, but that was because I had to sell mutual funds at a heavy loss to pay off student and car loans just to get the loan to buy my house. It ended up to be a good choice - the mutual funds have been stagnant in the last few years, the car has depreciated as expected, and my house value has nearly tripled. Not to mention I was able to refinance to a 15 year fixed mortgage for just $100 more than my 30 year Variable.
It's also a rather tricky bug that may not have been caught in their testing process.
I can entirely see how their Q/A team would miss the bug - Q/A would need to be playing on hard, probably with no cheats on (otherwise, why save and reload?). Save and reload probably got some visible verification (inventory there, start point correct, etc), so that was probably checked off without further testing. Most testing probably never used save and reload - mainly because that is one of the last completed parts of the game completed. Portions of save/restore may work, but until item placement and inventory items are complete, why test it?
I admit, what they need is to have some people run through the "finished" product once or twice at every difficulty to verify there are no outstanding showstoppers, but that's not always possible (time demands), so maybe they settled on a runthrough only at moderate difficulty. Due to the limited lifespan of games (about 3 months) they probably didn't want to have an open beta (no sense leaking the code to pirates any earlier than possible).
Part of the problem is that the BBC article seems to be missing some history behind the story that makes a non-island city a possibility. Plato pretty much lifted the Atlantis (city) story from the Egyptian tale of Keftiu (as well as embellished on it), a city that supposedly existed past the Pillars of Hercules (Straight of Gibraltar today, which separates Spain and Morocco). Keftiu is rooted in the Egyptian word for Pillar and was believed to be the end of the earth where the sky was held up. Atlantis means isle of Atlas - recognize the similarity? Atlas held up the world in Greek mythology. Keftiu also wasn't necessarily an island - it can either mean the Isle of Keft or the People of Keft. So, possibly due to a simple translation error, an island was born.
This could very easily be Atlantis. Minoan Crete never made sense (it never sunk) - Santorini island made more sense as most of it blew up (flooding Minoan Crete). It seems to me, though, that it was described as "west of Egypt" and that island's really NW.
That doesn't explain how it was lost, although speculation of a great city that sunk deep into a lake after an earthquake (or something like that) does exist in Mexico.
There is, however, some speculative evidence that supports your theory. German scientists claim to have found nicotine, cocaine, and THC in mummies, although mainstream archeologists say the claims are false. If cocaine was truely found in mummies, that would mean trade routes of some kind to South America (or a native crop before Egypt became desert?). How or when such a trade route started or ended is a matter of speculation. The main problem here is timeline - Atlantis supposedly sunk about 10000 years ago, meaning the 4000 year old mummies aren't old enough.
Silly, but implemented and put on some CDs already (especially in Europe). Look up CDS-100 and CDS-200 encoding (stands for Cactus Data Shield and is now owned by Macrovision).
My understanding is they add "spikes" to the wave at mixed intervals that are supposedly not perceptable when the music is played. When the music is compressed and/or converted into a different format, these spikes join together or are somehow elongated into the audible range and distort the sound.
Yeah, it's true that MS gaming division isn't making them money, but their bread-and-butter, MS-Office and MS-Windows make so much dough that they pay the losses of every losing division and still rake in over a billion dollars without even mentioning their other profitable divisions. I don't think MS has anything to worry about, except maybe the PC drying up as a gaming market. I have my doubts about that, as well, because graphics technology continues to be pushed on the PC side, which eventually influences consoles. Consoles have the plus of standardized hardware, so it's easier to write for them, but you don't get "cutting edge" graphics, except maybe on the console's release.
:(
MS has hurt PC gaming, by buying many PC developers and moving them exclusively to XBox dev (then porting to PC with a different developer later). The FASA (Mechwarrior) and Bungie (Halo) teams are prime examples. Basically, they've moved the PC game to play second fiddle to the XBox game, but adding all the missing features later. In many ways this works out good for MS, because the XBox is basically a standardized PC, which means many less configuration problems and simplified debugging (meaning shorter release schedule and thus less investment), so they can shovel the cost of PC hardware debugging to a third party developer. Unfortunately, it also means late releases on the PC and controls that either don't work, or are so dumbed down the game is either too easy or too hard
I build PCs, and for $600, I've build some pretty nice machines that are perfectly capable of playing even the latest games - maybe not with everything on, but with a decent feature set at 1024x768. For $600, I have to try to keep the CPU to about $70, HardDisk to $60, Memory to $100, Windows to $100, Video to $150, mobo to $50 (sound and ethernet onboard), and floppy, mouse, keyboard, and case to the other $70. I usually have some give-and-take by scouring pricewatch and Ebay, or saving $20 by using slower memory (one of the lower impacts to games). The sweet-spot is probably closer to $850-$1000 on a new machine, though (spending more on processor, video and hard-disk; for $1000, get a better mobo and case/PSU). After that, you're splurging on stuff you can buy for significantly less in just a few months. Not splurging, however, puts you in the eternal upgrade spiral, which I've been in for about 2 years ;)