I first saw HDTV on a large runco projector... they brought in a studio-quality deck to play the source material since there were no on-air broadcasts at the time (we in dc were one of the first cities to get on-air broadcasts - they had demos of this in national airport).
The HDTV picture I saw was on a large screen (8 foot by 4 foot?), and was film-quality perfect. Lots of detail, no scan lines. Kindof what you'd expect for a $30k TV. But the coolest part was a much smaller normal-definition projection TV that was in the foreground... it was perhaps 3 feet wide, and despite the much smaller picture, it looked worse than the blown-up HDTV picture. It was amazing.
Of course, take this with a grain of salt... I don't own a TV.
Argh! And I had better text to go with it, too. I should have saved a copy for just this moment. Alas, I'll just recap my concerns from my submission:
1. Because of cost, this will be probably implemented first only in the larger denomination bills. (stated in the article)
2. The security model is flawed. The authentication process encrypts the serial number, so without the algorithm you can't tell the bill's denomination. (You can track the bills by the unique encrypted number, irregardless if you know the algorithm).
3. You put these two facts together, and the mere presence of an RFID bill in your pocket means you have at least ~USD$200. If you have 10 RFID bills, you've got at least $2000. Without bypassing the encryption, you can pick off the most worthwhile people to rob.
4. Fortuantly, these are readable only at a short distance (~12 inches), but two antennas by the bathroom door will scan a whole lot of people.
----------------
Your recent submissons
Here are your recent submissions to Slashdot, and their status within the system:
2001-12-19 16:14:26 RFID in Euro Bank notes by 2005 (articles,money) (rejected)
But, besides the fact that the grafitti is on ford-owned property, the judge's analogy has another problem: the grafitti is probably poorly written and is obviously not professionally done, so the public would know that it's just grafitti and not the offical position of ford.
So, here's another one: You walk into a dealership that sells multiple brands of trucks, and you say you want a truck that would spank the crap out of any general motors product. If the dealer sold you a ford, would ford sue the dealer?
In both the linking case and the truck-buying case, the explicit language is introduced solely by the consumer. The dealer/dns service does the customer-language to physical-object translation, and neither service is under the direct control of ford.
Samsung has unveiled a gigabit flash (128MB) that, unlike this matrix part, is erasable. If that's not enough, you can buy a 2 gigabit flash stack (256 MB) from irvine sensors.
True, these are packaging techniques for more density, and aren't as cool as putting more memory on one die, but don't overlook them: they offer about the same densities. Hopefully, we can eventually combine both techniques, for even greater densities.
I think you're talking about the fabled Signetics 25120 fully encoded, 9046 x N, Random Access, write-only-memory, released in 1974. Here are page 1 and page 2 of the data sheets.
Considering that these modifications cost between US$10m and US$20m per suit above the initial cost of the suit itself (which is pretty amazing because the suits are $12m! - maybe one of these prices is off), you'd think that for this kind of money, you could pressurize your rec room. This will make games like twister much easier to play.
This doesn't go to a satellite! It goes via line-of-site to their tower, presumably on a tall building or somewhere with a wide view of the surrounding area. I guess you could move it a little bit, but if you started showing up on another antenna (maybe at the same site, but pointed in a different direction), they'd know.
I think that you could what you say with satellite-based services, such as starband, since those satellites cover 1/2 of the country (as long as you didn't switch coasts)
.. and all that comes inside is the network cable that connects directly to your NIC.
Actually, there's a power cable, too. The cell tower doesn't have that much power! From the faq:
Your equipment includes:
* A 14" square dish, which is mounted on the side of your home that best faces the Wireless Internet Tower.
* A receiver, approximately 14" x 10". This small box is mounted outside your home near the dish. This is the device that sends and receives data to and from your PC.
* A cable that runs from the receiver into your home. The cable will connect to an electrical outlet and to your computer's Network Interface Card (NIC)
I wonder how they mount the fairly big receiver box. Even though it has to be weather proofed and operate over an extended temperature range, there are far fewer mistakes that a customer can make with a CAT5 cable than an RF cable.
These satellites usually follow a sun-synchronous orbit. Basically, this is a polar orbit (meaning that the satellites fly over the north pole, the equator, the south pole, the other side of the equator, and then back over the north pole). One of these rotations happens every 101-103 minutes. Now, the beauty of the whole thing is that during the orbit doesn't fly over the same spot of the equator every time... it moves a bit. This bit is almost equal to the distance that the earth has rotated in those 103 minutes... thus, the satellite always follows the sun as works its way around the earth.
This orbit strikes a good balance between maximum sun exposure (useful for taking pictures), global coverage, and revist time. Half the time it's in dark, the other half it's taking photos at noontime.
Being in the light all the time would require a much higher orbit (near geostationary) that would make the optics work much harder. Since the satellite would be moving much slower with respect to the earth, the revisit time would also greatly suffer.
Orbital was working on the quickbird and earlybird satellites (the names got changed around as schedules, ahem, moved). At the same time I was also working on our tractor-trailer tracking system. I figured that if we could save a lot of money if just ditch our GPS/cellular tracking hardware and put giant bar codes on the trailers, and track them visually with the satellites. But, alas, we only had 1 meter resolution and even with a 53' trailer, there wasn't enough room for a suitable bar code. But, with this better resolution, my plan's now feasable!!
Fun fact: giant shipping companies lose one or two trailers a year each because they don't know where they left them.
p.s. patent pending. Ok, not really, but if anyone tries this, please let this post serve as evidence of prior art.
I've got a piece of wire I found outside after they did some repair work on the transformer for our 6-story office building. It's no ordinary piece- the stranded metal conductor is about an inch in diameter, wrapped in a strong rat-resistant rubber insulator. It's only 2 feet long, weighs about 4 pounds, an no one has been able to bend it with their hands. It would make a good weapon. It's aluminimun(sp) of course; copper is too expensive for the power company to use.
So, I took it to my favorite hifi store and claimed it was silver speaker cable and I wanted to upgrade it. They said sure!
1200 W, not 1000W.
Lifetime: 1500 Hours (~2 months)
Control interface: RS485
Output: 46 Amps @ 26 volts
Unit must be protected from weather, sand, dust, marine, and freezing conditions in product packaging (I assume coleman does this to some extent)
The Z80, in its day, was a complicated chip. Now allotting that much board space for 'only' that much functionality seems like a waste. As a result, the Z80 has being sold more and more as IP to be embedded in a larger chip than as a whole chip itself. Here aresomecores.
The list of Z80-like parts includes: Z80, Z180, 84C011 (toshiba), 84C11, 84C013, 80C13, 84C015, 84C15, 64180 (hitachi), NCS800 (national), Z181, Z182.
But, unless you've got some legacy code, your best bet is finding a microcontroller that has the correct mix of peripherials, power, speed, and price. Technology has advanced so much since then. Even if you have legacy code... it's probably still worth considering alternatives.
Re:Will heat be a problem?
on
Linux On HP Blades
·
· Score: 3, Informative
This thread has 5 replies and no one has an answer yet?
On the data sheet (there's a nice link in the article, I'm sure you can find it), you'll find the specs you're looking for:
Capable of 50 Watts per slot.
Single Pentium III 700 MHz, 512 MB ECC (PC100), 30GB IDE 2.5" HD, cPCI hot swap, dual 10/100base-T.
smart temperature monitor and failsafe circuitry
Actually, good point. It is a derivative work, I did get a bit carried away. But, from what I remember from playing Civ II, there weren't many sentences to translate -- mainly labels for things. I think that simple translations should be pretty easy to defend against (like making alternate light switch indicators labeled 'auf' and 'weg'), but I don't know how the courts see it. Especially the German courts, who have juristion here (and whoose laws are probably bound by some sort of international treaty).
And, of course, translating the manual is another thing--- I don't think that would be right; aclean-room style manual would be needed.
No, I don't think that's a good analogy. A chapter is a specific section of a book; it is just as usable as if one were to have the book and read only that chapter. Your analogy would be like releasing a fully-playble version of Doom, but limited to the first level only. (which is pretty close to what happened, but not the point...)
A better analogy would be to offer a list of new character names and objects and where in the book these names should be inserted. The purpose, of course, would be to localize the book: the book could be cajunized-- Jonny Pottieu would be chompin' on crawfish at a charivari, instead of whatever harry potter ate...
There is no copyright violation... you are replacing the text from the original program, and not distributing any byte that was in the original. The user must own the original to apply the patches to; otherwise the patches are useless.
> there should be evidence of colonization everywhere
Every single inhabitable planet we know of has already been inhabited... what more proof do you want?
Also, bringing life to a planet isn't as easy as simply flying a ship over. To make the colonies self-sufficient, you've got to design a diverse ecosystem specifically adapted to the destination planet and allow it to spread.. not an easy feat. If it's not diverse or self-sufficient, the first plague or late supply ship kills it off.
Besides commiting to travelling to another planet, you've got to have a reason. It's taken us over 4 million years to fill this planet up; At this rate per colonization, 1-10 million years is not neary enough time to expect to fill the whole galaxy.
DivX seems to work well on my 266MHz PII laptop... I don't know if my MPEG decoder card is helping. Some other things that might make a big difference: cache size, processor multimedia instructions (such as MMX), video card bandwidth (PCI vs. AGP), and DRAM memory bandwidth (My laptop does 528 MB/sec, my desktop does 2100 MB/sec)
Oh, you mean some sort of simple technology, like a stick, or, say, maybe a rock or maybe just dirt? It's the technology's fault that it gets corrupted.
> conventional wisdom says that if you want ultimate speed, use C
Not to nitpick, but the conventional wisdom says to program in the lowest language possible -- which is usually assembly. Assuming, of course, speed is your goal, not reliability (the two aren't necessarily mutually exclusive).
Of course, if you can go even lower, then all the better: microcode, vhdl, specialized computer architecture, asics. But these options aren't available for 99.9% of pc programmers, but are routinely used by embedded programmers.
...and the first two failures had my satellites on them : (
Oh well, there's always LLV. Ooops. That was one of mine, too.
It sucks making satellites that cost than the rocket; we're totally expendable. I'm not bitter, (ok just a bit); it's just a fact of the small satellitebusiness.
I'm not sure, but how would you get a Pegasus into an L-1011? Rear door? Baggage area? Sneak it through security and stash it in the overhead compartment? Hope there's a way... that would be much better.
Yeah, solid fuel is definately preferable, but still, the L-1011's lands at 120-150 knots... What are the chances that the friction from the ground is going to generate enough heat to ignite something? (I don't know).
Yep, Orbital didn't own that plane; they own the L-1011. You can both at a non-official site here.
Did anyone notice how close the rocket is to the ground on the L-1011? The body is close, but the fins are even closer! The top fin sticks into the body of the plane. I don't think I'd want to be on that plane when it lands -- the 43,457 pounds of propellent in that missle makes it too close to a bomb. Incidently, landing with the rocket is SOP - it's used to ferry the rocket to the launch and/or integration sites.
Incidently, a friend was at one of the launches. There were two fighter jet chase planes - one NASA owned, and one military (air force?). The NASA guys used the whole runway for takeoffs and landings; the AF guy used as little as possible. Interesting difference.
Slashdot Mirror
I first saw HDTV on a large runco projector... they brought in a studio-quality deck to play the source material since there were no on-air broadcasts at the time (we in dc were one of the first cities to get on-air broadcasts - they had demos of this in national airport).
The HDTV picture I saw was on a large screen (8 foot by 4 foot?), and was film-quality perfect. Lots of detail, no scan lines. Kindof what you'd expect for a $30k TV. But the coolest part was a much smaller normal-definition projection TV that was in the foreground... it was perhaps 3 feet wide, and despite the much smaller picture, it looked worse than the blown-up HDTV picture. It was amazing.
Of course, take this with a grain of salt... I don't own a TV.
Argh! And I had better text to go with it, too. I should have saved a copy for just this moment. Alas, I'll just recap my concerns from my submission:
1. Because of cost, this will be probably implemented first only in the larger denomination bills. (stated in the article)
2. The security model is flawed. The authentication process encrypts the serial number, so without the algorithm you can't tell the bill's denomination. (You can track the bills by the unique encrypted number, irregardless if you know the algorithm).
3. You put these two facts together, and the mere presence of an RFID bill in your pocket means you have at least ~USD$200. If you have 10 RFID bills, you've got at least $2000. Without bypassing the encryption, you can pick off the most worthwhile people to rob.
4. Fortuantly, these are readable only at a short distance (~12 inches), but two antennas by the bathroom door will scan a whole lot of people.
----------------
Your recent submissons
Here are your recent submissions to Slashdot, and their status within the system:
2001-12-19 16:14:26 RFID in Euro Bank notes by 2005 (articles,money) (rejected)
But, besides the fact that the grafitti is on ford-owned property, the judge's analogy has another problem: the grafitti is probably poorly written and is obviously not professionally done, so the public would know that it's just grafitti and not the offical position of ford.
So, here's another one: You walk into a dealership that sells multiple brands of trucks, and you say you want a truck that would spank the crap out of any general motors product. If the dealer sold you a ford, would ford sue the dealer?
In both the linking case and the truck-buying case, the explicit language is introduced solely by the consumer. The dealer/dns service does the customer-language to physical-object translation, and neither service is under the direct control of ford.
Comments, please?
Samsung has unveiled a gigabit flash (128MB) that, unlike this matrix part, is erasable. If that's not enough, you can buy a
2 gigabit flash stack (256 MB) from irvine sensors.
True, these are packaging techniques for more density, and aren't as cool as putting more memory on one die, but don't overlook them: they offer about the same densities. Hopefully, we can eventually combine both techniques, for even greater densities.
I think you're talking about the fabled Signetics 25120 fully encoded, 9046 x N, Random Access, write-only-memory, released in 1974. Here are page 1 and page 2 of the data sheets.
Considering that these modifications cost between US$10m and US$20m per suit above the initial cost of the suit itself (which is pretty amazing because the suits are $12m! - maybe one of these prices is off), you'd think that for this kind of money, you could pressurize your rec room. This will make games like twister much easier to play.
This doesn't go to a satellite! It goes via line-of-site to their tower, presumably on a tall building or somewhere with a wide view of the surrounding area. I guess you could move it a little bit, but if you started showing up on another antenna (maybe at the same site, but pointed in a different direction), they'd know.
I think that you could what you say with satellite-based services, such as starband, since those satellites cover 1/2 of the country (as long as you didn't switch coasts)
.. and all that comes inside is the network cable that connects directly to your NIC.
Actually, there's a power cable, too. The cell tower doesn't have that much power! From the faq:
Your equipment includes:
* A 14" square dish, which is mounted on the side of your home that best faces the Wireless Internet Tower.
* A receiver, approximately 14" x 10". This small box is mounted outside your home near the dish. This is the device that sends and receives data to and from your PC.
* A cable that runs from the receiver into your home. The cable will connect to an electrical outlet and to your computer's Network Interface Card (NIC)
I wonder how they mount the fairly big receiver box. Even though it has to be weather proofed and operate over an extended temperature range, there are far fewer mistakes that a customer can make with a CAT5 cable than an RF cable.
I was just curious and did the math...
Figures from article:
PS2 Nov 11th - Dec 8th - 962k
Xbox Nov 15th (release) - Dec 8th - 934k
NGC Nov 18th (release) - Dec 8th - 602k
PS2: 28 days = 34,357/day
XBOX: 24 days = 38,916/day
NGC: 21 days = 28,666/day
These satellites usually follow a sun-synchronous orbit. Basically, this is a polar orbit (meaning that the satellites fly over the north pole, the equator, the south pole, the other side of the equator, and then back over the north pole). One of these rotations happens every 101-103 minutes. Now, the beauty of the whole thing is that during the orbit doesn't fly over the same spot of the equator every time... it moves a bit. This bit is almost equal to the distance that the earth has rotated in those 103 minutes... thus, the satellite always follows the sun as works its way around the earth.
This orbit strikes a good balance between maximum sun exposure (useful for taking pictures), global coverage, and revist time. Half the time it's in dark, the other half it's taking photos at noontime.
Being in the light all the time would require a much higher orbit (near geostationary) that would make the optics work much harder. Since the satellite would be moving much slower with respect to the earth, the revisit time would also greatly suffer.
Orbital was working on the quickbird and earlybird satellites (the names got changed around as schedules, ahem, moved). At the same time I was also working on our tractor-trailer tracking system. I figured that if we could save a lot of money if just ditch our GPS/cellular tracking hardware and put giant bar codes on the trailers, and track them visually with the satellites. But, alas, we only had 1 meter resolution and even with a 53' trailer, there wasn't enough room for a suitable bar code. But, with this better resolution, my plan's now feasable!!
Fun fact: giant shipping companies lose one or two trailers a year each because they don't know where they left them.
p.s. patent pending. Ok, not really, but if anyone tries this, please let this post serve as evidence of prior art.
I've got a piece of wire I found outside after they did some repair work on the transformer for our 6-story office building. It's no ordinary piece- the stranded metal conductor is about an inch in diameter, wrapped in a strong rat-resistant rubber insulator. It's only 2 feet long, weighs about 4 pounds, an no one has been able to bend it with their hands. It would make a good weapon. It's aluminimun(sp) of course; copper is too expensive for the power company to use.
So, I took it to my favorite hifi store and claimed it was silver speaker cable and I wanted to upgrade it. They said sure!
Here's the spec sheet from Ballard of the Nexa module used in the coleman. Some interesting differences:
1200 W, not 1000W.
Lifetime: 1500 Hours (~2 months)
Control interface: RS485
Output: 46 Amps @ 26 volts
Unit must be protected from weather, sand, dust, marine, and freezing conditions in product packaging (I assume coleman does this to some extent)
Nope, the TI calculators are based on the 68000 from motorola.
The Z80, in its day, was a complicated chip. Now allotting that much board space for 'only' that much functionality seems like a waste. As a result, the Z80 has being sold more and more as IP to be embedded in a larger chip than as a whole chip itself. Here are some cores.
The list of Z80-like parts includes: Z80, Z180, 84C011 (toshiba), 84C11, 84C013, 80C13, 84C015, 84C15, 64180 (hitachi), NCS800 (national), Z181, Z182.
But, unless you've got some legacy code, your best bet is finding a microcontroller that has the correct mix of peripherials, power, speed, and price. Technology has advanced so much since then. Even if you have legacy code... it's probably still worth considering alternatives.
P.S. here is a good source of documentation
This thread has 5 replies and no one has an answer yet?
On the data sheet (there's a nice link in the article, I'm sure you can find it), you'll find the specs you're looking for:
Capable of 50 Watts per slot.
Single Pentium III 700 MHz, 512 MB ECC (PC100), 30GB IDE 2.5" HD, cPCI hot swap, dual 10/100base-T.
smart temperature monitor and failsafe circuitry
So, it's just good performance, not ultra-high.
malloc(-1) != NULL
(that's assuming, of course, that size_t is still 2^32)
Actually, good point. It is a derivative work, I did get a bit carried away. But, from what I remember from playing Civ II, there weren't many sentences to translate -- mainly labels for things. I think that simple translations should be pretty easy to defend against (like making alternate light switch indicators labeled 'auf' and 'weg'), but I don't know how the courts see it. Especially the German courts, who have juristion here (and whoose laws are probably bound by some sort of international treaty).
And, of course, translating the manual is another thing--- I don't think that would be right; aclean-room style manual would be needed.
No, I don't think that's a good analogy. A chapter is a specific section of a book; it is just as usable as if one were to have the book and read only that chapter. Your analogy would be like releasing a fully-playble version of Doom, but limited to the first level only. (which is pretty close to what happened, but not the point...)
A better analogy would be to offer a list of new character names and objects and where in the book these names should be inserted. The purpose, of course, would be to localize the book: the book could be cajunized-- Jonny Pottieu would be chompin' on crawfish at a charivari, instead of whatever harry potter ate...
There is no copyright violation... you are replacing the text from the original program, and not distributing any byte that was in the original. The user must own the original to apply the patches to; otherwise the patches are useless.
> there should be evidence of colonization everywhere
Every single inhabitable planet we know of has already been inhabited... what more proof do you want?
Also, bringing life to a planet isn't as easy as simply flying a ship over. To make the colonies self-sufficient, you've got to design a diverse ecosystem specifically adapted to the destination planet and allow it to spread.. not an easy feat. If it's not diverse or self-sufficient, the first plague or late supply ship kills it off.
Besides commiting to travelling to another planet, you've got to have a reason. It's taken us over 4 million years to fill this planet up; At this rate per colonization, 1-10 million years is not neary enough time to expect to fill the whole galaxy.
DivX seems to work well on my 266MHz PII laptop... I don't know if my MPEG decoder card is helping. Some other things that might make a big difference: cache size, processor multimedia instructions (such as MMX), video card bandwidth (PCI vs. AGP), and DRAM memory bandwidth (My laptop does 528 MB/sec, my desktop does 2100 MB/sec)
Oh, you mean some sort of simple technology, like a stick, or, say, maybe a rock or maybe just dirt? It's the technology's fault that it gets corrupted.
> conventional wisdom says that if you want ultimate speed, use C
Not to nitpick, but the conventional wisdom says to program in the lowest language possible -- which is usually assembly. Assuming, of course, speed is your goal, not reliability (the two aren't necessarily mutually exclusive).
Of course, if you can go even lower, then all the better: microcode, vhdl, specialized computer architecture, asics. But these options aren't available for 99.9% of pc programmers, but are routinely used by embedded programmers.
...and the first two failures had my satellites on them : (
Oh well, there's always LLV. Ooops. That was one of mine, too.
It sucks making satellites that cost than the rocket; we're totally expendable. I'm not bitter, (ok just a bit); it's just a fact of the small satellitebusiness.
I'm not sure, but how would you get a Pegasus into an L-1011? Rear door? Baggage area? Sneak it through security and stash it in the overhead compartment? Hope there's a way... that would be much better.
Yeah, solid fuel is definately preferable, but still, the L-1011's lands at 120-150 knots... What are the chances that the friction from the ground is going to generate enough heat to ignite something? (I don't know).
I found this neat picture of the rocket attached to the underbelly in flight... the site also has a pretty complete history.
Yep, Orbital didn't own that plane; they own the L-1011. You can both at a non-official site here.
Did anyone notice how close the rocket is to the ground on the L-1011? The body is close, but the fins are even closer! The top fin sticks into the body of the plane. I don't think I'd want to be on that plane when it lands -- the 43,457 pounds of propellent in that missle makes it too close to a bomb. Incidently, landing with the rocket is SOP - it's used to ferry the rocket to the launch and/or integration sites.
Incidently, a friend was at one of the launches. There were two fighter jet chase planes - one NASA owned, and one military (air force?). The NASA guys used the whole runway for takeoffs and landings; the AF guy used as little as possible. Interesting difference.
Don't miss the Pegasus User's Guide -- it's an interesting read.