Reminds me of the infrastructure diagrams of corporate LAN's and WAN's back in the 1990's. They would have a diagram of the local network of each site with servers, workstations, routers and firewalls. Then each firewall would be connected to an X.25 cloud (which looked exactly like a big puffy cloud). If it was an internal ID department diagram, then someone would usually add four or more legs and a face or some lightning flashes (then it became an X.25 spider, an X.25 sheep, or an X.25 packet storm).
The ferromagnetic liquid will always try to achieve an equilibrium point between gravity, surface tension and the surrounding magnetic field. Gravity and surface tension will make it try and remain flat. As a magnetic field is continuous, it should be possible to have a large number of small but powerful magnets to make the liquid adopt whatever position is desired.
The Atari 400/800/1200(XL) computers had stereo track tape recorders. The first track was used for saving/loading binary data. The second track was used for playing background music while loading. In theory it might have been possible for someone to build some electronics to convert the background music track for use as binary data as well, and thus booost the save/load speed.
The human eye has about 100 million rods and cones. You need a 100 megapixel framebuffer (around 10,000 by 10,000 pixels to achieve this.
There was an article in the Independent newspaper about Virtual Reality a long time ago. In the article, one of the researchers stated that photorealistic quality was defined as 80 million textured triangles/second.
There's a guy in our work who is working on advanced lighting models (BRDF, BTF). He grades graphics cards by the amount of texture memory and the number of texture units available divided by the cost of the card.
Not really. The graphics card has 800 stream processors all running off cached texture memory to cached framebuffer memory (deferred rendering). Instead of simply fetching pixel data direct from texture memory and writing it directly out into framebuffer memory, the graphics card will maintain a texture cache (the current textures being used), and a framebuffer cache (the current area of the framebuffer being rendered). Then when there is no more pixels to be written, the framebuffer cache is written back into the framebuffer.
There is no upper limit on how much texture memory needed. For many applications like volume rendering with floating point data, the only limit is the amount of texture memory available.
You would have to have all your programs to take advantage of multi-threading ie. pthreads
With multi-threading, there are different ways of improving the performance of applications. The simple way is to have separate threads for user input, background tasks, and display.
The more complex way is to replace each single function that would run on a CPU (ie. color adjusting an image) to run on multiple cores/threads. For any function, the application creates a whole series of tasks which are retrieved and processed by separate threads. For image processing, this would involve splitting the processing of the image into many subimages - each thread processes a particular chunk of the main image.
For these kind of supercomputers, the users are going to have massive 3D grid volumes (eg. 2048^3 or greater) with each core/thread processing a particular sub-volume. Calculations like CFD can require a good many passes to run a single time-step of the simulation. And both a current generation grid and a next generation grid need to be stored. All this information needs to be processed so that it can be rendered in real-time).
Possibly MAC addresses can be reprogrammed. If you boot into the BIOS of any system, you will see the serial number there as well.
Some workstation manufacturers would actually embed a serial number into every component - which led to some interesting discoveries when a service technician came to upgrade a system in some developing world country.
Listings and printouts can be faked. Having any serial number of firmware address registered with a legally acceptable third party such as an insurance company, the manufacturer or retailer would be necessary.
Financially, it can be done. But the political cost of displacing low-income people will be a lot more, especially as people now realize that demolishing homes to make way for wider roads and car-parks only makes the problem worse in the long term.
SF does have underground car parks built as part of large office blocks. But the problem is where there is a mix of residential and business parking. They parking spaces may be free for residents between 6.00pm and 8.00am, but used for business during week days. It isn't practical to demolish a block of residential housing just to build a new car park.
I go to a university which is in the top-10 for research funding. As far as the undergraduate courses go, the recommended textbooks are the same as any university (eg. Deitel for Operating Systems, Watt for 3D Computer Graphics, Gamma for C++), and students are encouraged to read IEEE/ACM/BCS papers, along with applying for student membership. Students accomodation has broadband Internet access with institution wide access to research journals.
Unfortunately, it is really up to the motivation of the students as to whether they take advantage of all the resources available. The main problem is that they all have part-time jobs as bar tenders and waiters in order to pay their way through three years of education.
Consequently, they don't really have much time for doing coursework assignments. For exams, questions used to be based on "Explain how a.... works". Now, they are "Explain how you would implement.... using [Programming language/API]".
I gave up reading on USENET around 7 years ago - many of the technical discussion groups became spammed by junk mail and overloaded by students looking for quick solutions to their coursework assignments.
There was some mystique in dialing up your ISP, hearing than modem connect and see your newsgroups download. Then you could spend an hour or so just reading the world technical news and humour.
Standards are a totally different issue. Today, the GPUs are not even documented, let alone standardization of the programming.
An open standard allowed hardware vendors to hide their custom techniques behind a standard interface (memory mapped register sets/texture framebuffers.
Third party board developers do get access to the specifications. I know a few companies who have built embedded systems using these chips.
Exactly!! But isn't that punishable? Isn't that anti-competitive to dictate people's choice of OS even when the GPU's behaviour has nothing to do with which OS it is running on??
Absolutely. That's what I dislike about Microsoft - they sit between the hardware vendors, and the application developers, and insist on particular API's being used, which will either change rapidly, or be built on top of legacy 8086 programming standards.
VGA was probably the most open video standard for hardware programming. Once you know what all the different registers were for, you could do all sorts of fun things, like having paged framebuffers, one super big 256-color framebuffer larger than the actual screen size, or reprogram the hardware video font.
That's probably what they fear - having lots of people trying out different ideas, rather than having one company (Microsoft) deciding their future for them.
Survivalist Suppliers announces - buy a lifetimes supply of Viagra for your mountain forest bunker today. What are you going to do when the black gold runs out - you won't be able to drive to the pharmacy then. One large box lasts 5 years and only costs $99.99. Buy a lifetimes supply and get an extra 50% absolutely free. Cash payments only.
What about cryogenic storage - if people could go into hibernation for decades if not hundreds or thousands of years, what would the culture shock be like. People releaseed from jail after 10 or 20 years have considerable problems adapting to the subtle changes in technology (cellphones, PDA's, disappearance of post offices/telephone boxes).
There was a story I once read about the last survivor on Earth, who went into cryogenic storage for an incurable disease, only to be revived once a cure had been found. The only problem was that he was the only human left. The whole world had been replaced by machines as the human race died off due to disease. After spending millions of years in and out of hibernation, robot search teams eventually find a planet with primitive plant life. Eventually after a few million years, this planet develops intelligent life, and the hero is able to restart his life. I think it was a Nova collection book (hardcover book with an orange front page with some red stripes).
Slashdot Mirror
I thought Wall.E really caught the appearance of an 1970's sci-fi. A cross between V.I.N.CENT from "The Black Hole", E.T. , and the two robots from Logan's Run, and Short Circuit (not really 1970's though
Reminds me of the infrastructure diagrams of corporate LAN's and WAN's back in the 1990's. They would have a diagram of the local network of each site with servers, workstations, routers and firewalls. Then each firewall would be connected to an X.25 cloud (which looked exactly like a big puffy cloud). If it was an internal ID department diagram, then someone would usually add four or more legs and a face or some lightning flashes (then it became an X.25 spider, an X.25 sheep, or an X.25 packet storm).
The ferromagnetic liquid will always try to achieve an equilibrium point between gravity, surface tension and the surrounding magnetic field. Gravity and surface tension will make it try and remain flat. As a magnetic field is continuous, it should be possible to have a large number of small but powerful magnets to make the liquid adopt whatever position is desired.
There are a good few videos on youtube: Magnetic sculpture
The "Rambus vs. Infineon, Hynix, Micron Technologies, Siemens AG, Nvidia, anyone-else-who-has-a-memory-chip-in-their-hardware" lawsuit is another one which seems to be just going on and on...
The Atari 400/800/1200(XL) computers had stereo track tape recorders. The first track was used for saving/loading binary data. The second track was used for playing background music while loading. In theory it might have been possible for someone to build some electronics to convert the background music track for use as binary data as well, and thus booost the save/load speed.
The human eye has about 100 million rods and cones. You need a 100 megapixel framebuffer (around 10,000 by 10,000 pixels to achieve this.
There was an article in the Independent newspaper about Virtual Reality a long time ago. In the article, one of the researchers stated that photorealistic quality was defined as 80 million textured triangles/second.
There's a guy in our work who is working on advanced lighting models (BRDF, BTF). He grades graphics cards by the amount of texture memory and the number of texture units available divided by the cost of the card.
Not really. The graphics card has 800 stream processors all running off cached texture memory to cached framebuffer memory (deferred rendering). Instead of simply fetching pixel data direct from texture memory and writing it directly out into framebuffer memory, the graphics card will maintain a texture cache (the current textures being used), and a framebuffer cache (the current area of the framebuffer being rendered). Then when there is no more pixels to be written, the framebuffer cache is written back into the framebuffer.
There is no upper limit on how much texture memory needed. For many applications like volume rendering with floating point data, the only limit is the amount of texture memory available.
You would have to have all your programs to take advantage of multi-threading ie. pthreads
With multi-threading, there are different ways of improving the performance of applications. The simple way is to have separate threads for user input, background tasks, and display.
The more complex way is to replace each single function that would run on a CPU (ie. color adjusting an image) to run on multiple cores/threads. For any function, the application creates a whole series of tasks which are retrieved and processed by separate threads. For image processing, this would involve splitting the processing of the image into many subimages - each thread processes a particular chunk of the main image.
For these kind of supercomputers, the users are going to have massive 3D grid volumes (eg. 2048^3 or greater) with each core/thread processing a particular sub-volume. Calculations like CFD can require a good many passes to run a single time-step of the simulation. And both a current generation grid and a next generation grid need to be stored. All this information needs to be processed so that it can be rendered in real-time).
Possibly MAC addresses can be reprogrammed. If you boot into the BIOS of any system, you will see the serial number there as well.
Some workstation manufacturers would actually embed a serial number into every component - which led to some interesting discoveries when a service technician came to upgrade a system in some developing world country.
Listings and printouts can be faked. Having any serial number of firmware address registered with a legally acceptable third party such as an insurance company, the manufacturer or retailer would be necessary.
Standard marketing psychology used for discount TV channel marketing;
" buy an nvidia GX280 graphics card 160 stream processors within the next hour, and we'll throw in another 80 stream processors absolutely free!!!"
"Order our 3 Gigabyte memory ultimate gaming rig and we'll give you an extra 1 Gigabyte of memory at no extra charge ..."
Supermarkets used to do this all the time, but then the consumer advice columns starting giving out warnings.
Financially, it can be done. But the political cost of displacing low-income people will be a lot more, especially as people now realize that demolishing homes to make way for wider roads and car-parks only makes the problem worse in the long term.
SF does have underground car parks built as part of large office blocks. But the problem is where there is a mix of residential and business parking. They parking spaces may be free for residents between 6.00pm and 8.00am, but used for business during week days. It isn't practical to demolish a block of residential housing just to build a new car park.
I go to a university which is in the top-10 for research funding. As far as the undergraduate courses go, the recommended textbooks are the same as any university (eg. Deitel for Operating Systems, Watt for 3D Computer Graphics, Gamma for C++), and students are encouraged to read IEEE/ACM/BCS papers, along with applying for student membership. Students accomodation has broadband Internet access with institution wide access to research journals.
Unfortunately, it is really up to the motivation of the students as to whether they take advantage of all the resources available. The main problem is that they all have part-time jobs as bar tenders and waiters in order to pay their way through three years of education.
Consequently, they don't really have much time for doing coursework assignments. For exams, questions used to be based on "Explain how a .... works". Now, they are "Explain how you would implement .... using [Programming language/API]".
Eternal September?
I gave up reading on USENET around 7 years ago - many of the technical discussion groups became spammed by junk mail and overloaded by students looking for quick solutions to their coursework assignments.
There was some mystique in dialing up your ISP, hearing than modem connect and see your newsgroups download. Then you could spend an hour or so just reading the world technical news and humour.
Standards are a totally different issue. Today, the GPUs are not even documented, let alone standardization of the programming.
An open standard allowed hardware vendors to hide their custom techniques behind a standard interface (memory mapped register sets/texture framebuffers.
Third party board developers do get access to the specifications. I know a few companies who have built embedded systems using these chips.
Exactly!! But isn't that punishable? Isn't that anti-competitive to dictate people's choice of OS even when the GPU's behaviour has nothing to do with which OS it is running on??
Absolutely. That's what I dislike about Microsoft - they sit between the hardware vendors, and the application developers, and insist on particular API's being used, which will either change rapidly, or be built on top of legacy 8086 programming standards.
VGA was probably the most open video standard for hardware programming. Once you know what all the different registers were for, you could do all sorts of fun things, like having paged framebuffers, one super big 256-color framebuffer larger than the actual screen size, or reprogram the hardware video font.
That's probably what they fear - having lots of people trying out different ideas, rather than having one company (Microsoft) deciding their future for them.
Survivalist Suppliers announces - buy a lifetimes supply of Viagra for your mountain forest bunker today. What are you going to do when the black gold runs out - you won't be able to drive to the pharmacy then. One large box lasts 5 years and only costs $99.99. Buy a lifetimes supply and get an extra 50% absolutely free. Cash payments only.
The last I heard, all radioactive elements heavier than Iron, were created by neutron bombardment within a star that went supernova.
Some trees can live for thousands of years. Turtles and whales have been known to live for 200+ years.
Sounds just like the plot from Zardoz.
What about cryogenic storage - if people could go into hibernation for decades if not hundreds or thousands of years, what would the culture shock be like. People releaseed from jail after 10 or 20 years have considerable problems adapting to the subtle changes in technology (cellphones, PDA's, disappearance of post offices/telephone boxes).
There was a story I once read about the last survivor on Earth, who went into cryogenic storage for an incurable disease, only to be revived once a cure had been found. The only problem was that he was the only human left. The whole world had been replaced by machines as the human race died off due to disease. After spending millions of years in and out of hibernation, robot search teams eventually find a planet with primitive plant life. Eventually after a few million years, this planet develops intelligent life, and the hero is able to restart his life. I think it was a Nova collection book (hardcover book with an orange front page with some red stripes).
It depends - suppose people choose to have children when they are 300 years old rather than 30 years, that wouldn't be too much of a problem.
But then if they chose to have a new family every 40 years, that would be a problem,
The people who take their car to go to the gym thats 10 minutes away (come on...you're being lazy on your way to go exercise, the hell?)
I've known people like that - there was a guy in our office who did that. He drove his car two blocks to the fitness centre and then back again.
Reminds me of this fitness centre