Imagine being in the GTA world with hundreds of other players and being able to take sides with one of the factions or with the local cops if you so choose. Imagine going on a mission not knowing if you're going to be facing a group of dim-witted NPC's or player characters. When you run a police baracade one of the cops may comondeer a civilian sports car to chase after you.
Of course then Jack would be able to say that you're killing real people in the game world and would start babling even louder about the evil murder simulator.
EQ2 is pretty much mouse playable as well. The 6 mouse functions that you listed are identical in EQ2. You can have up to 3 hotbars with 12 buttons each with any spells or commands you want in there and spell queing so that you can select your next spell before the current one has finished. Most NPC's and objects respond to a double click and everything that has any action you can perform on it will pop a context menu on a right click.
All in all very mouse playable. (except for chatting)
But for me it's actually more important that I can play with just the keyboard and no mouse. It's hard to accurately move the mouse pointer when you've got the phone tucked up on your shoulder, a cigarette in one hand and dinner in the other while you're trying to keep your group from wiping deep in RunnyEye.
Using what I described, the contolling electronics would be similar in price to a cheap LCD system. The biggest differences will be in whether the digital paper is easier to produce or more eye catching than a similar LCD. Plus there's the safety and liability issue of gluing either a chunk of glass or a thin flexible sheet of plastic to a ceral box marketed at kids.
Then six months later the marketing guys demand color and the whole thing turns into a giant headache.
You wouldn't need to store pixel data because you'd be doing something more akin to hard wiring the images. Think along the lines of the old LCD style handheld video games where if you pressed on the screen you'd see little pictures of game objects. In those devices you'd sometimes see a situation, let's use a fighting game for example, where the neutral stance is a picture of a the character, a kick is the character plus an extended leg and a punch is the character plus an extended arm. You'd make each image by activating the correct combination of those three regions. In this case, all that you're storing is the info for what combination of regions to activate.
If you could hard wire clumps of pixels on this digital paper in a similar fashion you could save dramatically on component count compared to a raster or RLE based system. I think the only real requisite is that the circuitry involved with the display would need to have two layers to work with.
If display area desperately needs a strobing row or column system to work then an approach more akin to a raster system would be more appropriate. But from what I read of TFA, there was no mention as to how the new technology works other than that it could be produced in a non cleanroom environment. If it does need a raster system, I like your idea of RLE compression of XOR data though.
There is another possible way of decreasing memory requirements and other manufacturing costs if your are content with a device that can only display a very small fixed selection of images.
Imagine the classic three circle Venn diagram with regions A,B and C. Now choose three images that you want displayed on your device. Find pixels exclusive to only one image and put them in the non overlapping areas of A,B and C respectively. Find the pixels shared by any two and put them in the appropriate overlaped areas. Find the pixels shared by all three and put them smack in the center of the Venn where all three regions meet and overlap. Run a lead to each walled in little sub-region to turn them on and off.
Doing something like this will mean that a 3 image device will need [(2^3)-1] bits per image for a total of 21 bits to hold all three.
Sadly the requirements scale exponentially as a device with n images would require [(2^n)-1] bits per image or n[(2^n)-1) bits total. But with just a handful of images you escape the need for the full fledged frame buffer that would be needed by a general purpose display.
With such a simplified method you could also escape the need for any sort of processing by having the regions controlled by a chain of transistor loops. I don't feel like drawing this out right now so I'm guessing a little here, and again n is the number of images, but it could probably be done with [n^2+3n] to [n^2+4n] individual transistors and a small amount of diodes or, if they spend a lot of time tailoring the circuit for each different design, [4n] to [5n] transistors and a something close to [n^2] diodes. Both figures are in addition to the timer circuit that only needs to tick at reasonably consistant intervals.
Sadly, if they do implement something like what I'm imagining, the end result will not be anything general purpose enough to be worth hacking. Sorry guys.
NBC would still have several large pipes to get the content out quickly. A torrent style P2P setup would just mean that you'd be sucking off those large pipes while every other user is squirting data at you at the same time.
And there's no reason for downloaded shows to completely dominate your hard disks. Think a 15gig limit on the buffer compared to 250gig HDs that cost just north of $120. Disable Windows restore and you've already made up for the lost space.
If I could pay a flat monthly rate and have free reign to download as many shows as I want, I'd be tempted to shell out for it. They could keep the content DRM'd or stuff it in some massive buffer ala Rhapsody.
A subscription would let me explore the content without feeling cheated out of a download fee if I stop playing a show after watching 5 minutes of it and determining that it's pure crap.
Yes DRM and proprietary media players would prevent me from viewing the content on a linux box or archiving a show on DVD so that I can dig it out and watch it after the system eventually shuts down (or I decide to stop paying). But I can live with that, I know I'm paying for a service and not buying a product.
There is a market for subscription entertainment services. Think of how many people pay $50-$100 in monthly fees for TV from Comcast or DirectTV? Ignoring TiVo, do you own the shows or are you just paying for the right to view them and the infrastructure to get them to your house?
Tomorrow's cable box may become a media-center PC.
There's already alot of people whining because 3-4 gig isn't enough.
If AMD produced a chip with 2 gig onboard they'd also include a means of accessing external memory. Since some memory is on the CPU silicon and some is sitting on the mainboard, there will be a difference in speed and access methods to get to the two areas of RAM. This would end up creating something similar to the 640K barrier that emerged in the DOS days.
We do not need another one of these little quirks to further confuse the issue of memory management on the x86 platform again.
Lets back up a couple years.
Would productivity be lost if a typewriter jammed, eating the page it was working on? Is the typewriter manufacturer liable for lost man hours and that sheet of paper? Would the typewriter manufacturer be responsible if their machine was too heavy for a glass desk and caused an injury when it broke through?
What would be the effects of misplacing a physical file folder in the legal department? Is the manufacturer of the file cabinets in that department liable for any damages encurred by losing an important notarized document? If there are file cabinets from different manufacturers in the same room, which one carries the liability, the one where the file was supposed to be or the one that the file ended up going to?
How about if the reciever switch in your telephone fails? Is the company that manufactured that liable for business lost due to predicted earnings based on calls that might have been recieved on that phone?
Software liability doesn't make sense. Machines fail, they get placed in situations that the designers wouldn't normally think of, they are faced with sometimes truely absurd user errors, and they are sometimes run by themselves in situations where redundancy would be appropriate. Why should we hold consumer software to a higher standard than any other piece of typical office equipment? A desktop application is not a bridge or a skyscraper, it's a tool. And as a tool it need only be safe enough to prevent reasonable paths of personal injury to its users. Lost time/money is not a personal injury.
We expect a typewriter to jam periodically because it has a hundred moving parts working along interfering paths. How many different discrete functions are in the source code for MS Office? How many of them are expected to work on the open document without interfering with all the other functions working on the same open document? MS Office is considerably more complex than a consumer grade typewriter. My old typewriter would jam on a whim, my student edition of office, aside from ticking me off with autoformating, has yet to eat any of the reports I've written. And I still keep those backed up (HD + thumbdrive) just in case anyway.
And in some situations there is a liability issue with software. If the control software on an elevator allowed the carriage to move while the door was open, that would be a liability issue that the manufacturer/software developer would have to answer to.
Slashdot Mirror
You just gave me an idea for the next GTA game.
Imagine being in the GTA world with hundreds of other players and being able to take sides with one of the factions or with the local cops if you so choose. Imagine going on a mission not knowing if you're going to be facing a group of dim-witted NPC's or player characters. When you run a police baracade one of the cops may comondeer a civilian sports car to chase after you.
Of course then Jack would be able to say that you're killing real people in the game world and would start babling even louder about the evil murder simulator.
EQ2 is pretty much mouse playable as well. The 6 mouse functions that you listed are identical in EQ2. You can have up to 3 hotbars with 12 buttons each with any spells or commands you want in there and spell queing so that you can select your next spell before the current one has finished. Most NPC's and objects respond to a double click and everything that has any action you can perform on it will pop a context menu on a right click.
All in all very mouse playable. (except for chatting)
But for me it's actually more important that I can play with just the keyboard and no mouse. It's hard to accurately move the mouse pointer when you've got the phone tucked up on your shoulder, a cigarette in one hand and dinner in the other while you're trying to keep your group from wiping deep in RunnyEye.
Using what I described, the contolling electronics would be similar in price to a cheap LCD system. The biggest differences will be in whether the digital paper is easier to produce or more eye catching than a similar LCD. Plus there's the safety and liability issue of gluing either a chunk of glass or a thin flexible sheet of plastic to a ceral box marketed at kids.
Then six months later the marketing guys demand color and the whole thing turns into a giant headache.
You wouldn't need to store pixel data because you'd be doing something more akin to hard wiring the images. Think along the lines of the old LCD style handheld video games where if you pressed on the screen you'd see little pictures of game objects. In those devices you'd sometimes see a situation, let's use a fighting game for example, where the neutral stance is a picture of a the character, a kick is the character plus an extended leg and a punch is the character plus an extended arm. You'd make each image by activating the correct combination of those three regions. In this case, all that you're storing is the info for what combination of regions to activate.
If you could hard wire clumps of pixels on this digital paper in a similar fashion you could save dramatically on component count compared to a raster or RLE based system. I think the only real requisite is that the circuitry involved with the display would need to have two layers to work with.
If display area desperately needs a strobing row or column system to work then an approach more akin to a raster system would be more appropriate. But from what I read of TFA, there was no mention as to how the new technology works other than that it could be produced in a non cleanroom environment. If it does need a raster system, I like your idea of RLE compression of XOR data though.
There is another possible way of decreasing memory requirements and other manufacturing costs if your are content with a device that can only display a very small fixed selection of images.
Imagine the classic three circle Venn diagram with regions A,B and C. Now choose three images that you want displayed on your device. Find pixels exclusive to only one image and put them in the non overlapping areas of A,B and C respectively. Find the pixels shared by any two and put them in the appropriate overlaped areas. Find the pixels shared by all three and put them smack in the center of the Venn where all three regions meet and overlap. Run a lead to each walled in little sub-region to turn them on and off.
Doing something like this will mean that a 3 image device will need [(2^3)-1] bits per image for a total of 21 bits to hold all three.
Sadly the requirements scale exponentially as a device with n images would require [(2^n)-1] bits per image or n[(2^n)-1) bits total. But with just a handful of images you escape the need for the full fledged frame buffer that would be needed by a general purpose display.
With such a simplified method you could also escape the need for any sort of processing by having the regions controlled by a chain of transistor loops. I don't feel like drawing this out right now so I'm guessing a little here, and again n is the number of images, but it could probably be done with [n^2+3n] to [n^2+4n] individual transistors and a small amount of diodes or, if they spend a lot of time tailoring the circuit for each different design, [4n] to [5n] transistors and a something close to [n^2] diodes. Both figures are in addition to the timer circuit that only needs to tick at reasonably consistant intervals.
Sadly, if they do implement something like what I'm imagining, the end result will not be anything general purpose enough to be worth hacking. Sorry guys.
Couldn't find Qwest on there. Although.. sometimes the automated voice recognition system seems more capable than their live employees.
Chapter 1: It's not just for bombs anymore.
// Describes non-military applications of atomic energy.
Chapter 2: Nuclear not Nuke-U-lar.
// What it is and how it works.
Chapter 3: The inanimate carbon rod.
// Safety systems and their use.
Chapter 4: A meltdown is not a type of open-face sandwich.
// How to know if you have a problem that needs to be fixed *NOW*
Chapter 5: Let's fire it up!
// Startup procedures
Chapter 6: Careful where you dump that.
// Waste managment procedures
Chapter 7: And then everyone turned on their blowdryers at the same time.
// Handling fluxuations in energy demand
Chapter 8: Badges? We don't need no stinking badges.
// Security measures and why they're important
Chapter 9: It glows when I pee.
// Health concerns for plant employees
Chapter 10: It only leaked a little.
// Detecting and cleaning rogue radioactive material
Chapter 11: Where do I dump the moth balls?
// extended shutdown procedures.
** sorry I don't have any of the cute little one frame comics for you. Maybe ed. 2 **
NBC would still have several large pipes to get the content out quickly. A torrent style P2P setup would just mean that you'd be sucking off those large pipes while every other user is squirting data at you at the same time.
And there's no reason for downloaded shows to completely dominate your hard disks. Think a 15gig limit on the buffer compared to 250gig HDs that cost just north of $120. Disable Windows restore and you've already made up for the lost space.
If I could pay a flat monthly rate and have free reign to download as many shows as I want, I'd be tempted to shell out for it. They could keep the content DRM'd or stuff it in some massive buffer ala Rhapsody.
A subscription would let me explore the content without feeling cheated out of a download fee if I stop playing a show after watching 5 minutes of it and determining that it's pure crap.
Yes DRM and proprietary media players would prevent me from viewing the content on a linux box or archiving a show on DVD so that I can dig it out and watch it after the system eventually shuts down (or I decide to stop paying). But I can live with that, I know I'm paying for a service and not buying a product.
There is a market for subscription entertainment services. Think of how many people pay $50-$100 in monthly fees for TV from Comcast or DirectTV? Ignoring TiVo, do you own the shows or are you just paying for the right to view them and the infrastructure to get them to your house?
Tomorrow's cable box may become a media-center PC.
*twitch*
There's already alot of people whining because 3-4 gig isn't enough.
If AMD produced a chip with 2 gig onboard they'd also include a means of accessing external memory. Since some memory is on the CPU silicon and some is sitting on the mainboard, there will be a difference in speed and access methods to get to the two areas of RAM. This would end up creating something similar to the 640K barrier that emerged in the DOS days.
We do not need another one of these little quirks to further confuse the issue of memory management on the x86 platform again.
Lets back up a couple years. Would productivity be lost if a typewriter jammed, eating the page it was working on? Is the typewriter manufacturer liable for lost man hours and that sheet of paper? Would the typewriter manufacturer be responsible if their machine was too heavy for a glass desk and caused an injury when it broke through? What would be the effects of misplacing a physical file folder in the legal department? Is the manufacturer of the file cabinets in that department liable for any damages encurred by losing an important notarized document? If there are file cabinets from different manufacturers in the same room, which one carries the liability, the one where the file was supposed to be or the one that the file ended up going to? How about if the reciever switch in your telephone fails? Is the company that manufactured that liable for business lost due to predicted earnings based on calls that might have been recieved on that phone? Software liability doesn't make sense. Machines fail, they get placed in situations that the designers wouldn't normally think of, they are faced with sometimes truely absurd user errors, and they are sometimes run by themselves in situations where redundancy would be appropriate. Why should we hold consumer software to a higher standard than any other piece of typical office equipment? A desktop application is not a bridge or a skyscraper, it's a tool. And as a tool it need only be safe enough to prevent reasonable paths of personal injury to its users. Lost time/money is not a personal injury. We expect a typewriter to jam periodically because it has a hundred moving parts working along interfering paths. How many different discrete functions are in the source code for MS Office? How many of them are expected to work on the open document without interfering with all the other functions working on the same open document? MS Office is considerably more complex than a consumer grade typewriter. My old typewriter would jam on a whim, my student edition of office, aside from ticking me off with autoformating, has yet to eat any of the reports I've written. And I still keep those backed up (HD + thumbdrive) just in case anyway. And in some situations there is a liability issue with software. If the control software on an elevator allowed the carriage to move while the door was open, that would be a liability issue that the manufacturer/software developer would have to answer to.