As a first pass you can estimate adders as 10 gates pr. bit, state as 20 gates pr. bit and multipliers as 10x bits squared (Unless it is by a power of two in which case it is free) If you need to to division in your algorithm you should redesign it. If you use floating point, everything gets huge (Try not to use floating point, remember in hardware you do not need to deal with arbitrary word size restrictions, just scale word sizes to suit the requirements)
Now, figuring out exactly what resources you need, this is where you will get into trouble. Normally you will reuse some (lots) of your arithmetic, but exactly how much depend on what performance/power/gate count target you need to hit. More reuse means less gates but faster clocks (Which can drive you to more gates if you get into trouble on timing closure). The extreme case is software which just reuse a very limited set of ALUs, the other extreme is an unrolled design where algorithmic operation have dedicated hardware, so one iteration takes one clock.
Depending on performance targets the same algorithm can have a factor 1000 difference in gate count.
Try High Frontier by Phil Eklund (http://sierra-madre-games.eu/index_high_frontier_2nd_edition.html) A game about industrializing space made by an actual rocket scientist.
The board is beautiful with spaces representing stable orbits and movements represented by delta-v needed to change orbits.
Rocket stacks are built by reasonably realistic technologies and fueling and mass adhere to the Tsiolkovsky rocket equation (Placed in an elegant table array for simple use)
Some legislatures assert that the person whose name is associated with the cars license is ultimately responsible for illegal operation of the vehicle. Such legislatures usually cancel the fine if you have reported the car stolen or will reassign it if you provide information on who actually drove the car.
The largest issue I have with the hyperloop proposal is its rather pitiful capacity. At the highest rate proposed, with once cart every 30 seconds it still only transpoprts ~3600 PAX/hr, which is about on par with a 3 lane highway and that is before mixing in the car carriers.
Bog standard high-speed train lines do 30000 PAX/hr routinely, and while the hyperloop carts might be able to scale some, based on how they do the air bearing and that I think linked carts likely will not work, I doubt they can scale much other than by building multiple tubes (which adds upp the most expensive component in the system)
The biggest value add by public transport is to be able to free up the excessive area consumption an automobile based society incurs, but to do this the public transport in question better beat the automobile in land use with a wide margin.
Looking at speed alone is a bit of a red herring. faced with increased transport speeds people have always responded by traveling further which just escalates the problem of increased land use and increased energy use for transport. Throughput pr. unit land is likely a better metric for sustainable travel solutions than raw speed.
A good chunk of tesla sales go to norway, making norway the first country outside US to get Tesla superchargers. The reason is simple: EVs have enoromus tax breaks and road access privelidges:
No VAT (Equivalent to US sales tax) Normal VAT is 25% No taxes. Most cars get added taxes to the tune of $4000 and up. Cars in the same performance bracket as tesla usually have taxes in the $20000-$50000 range. (Taxes are based on emissions and weight) Access to bus lanes (Which makes most EVs in norway largely concentrated in a single county just outside Oslo where the highways are notoriously congested)
These benefits will not last.If a lot of teslas outside-US buisness is based on such tax breaks I foresee a significant political risk for teslas buisness
A wifi module is unlikely to draw more than 50-80mW while active And it can likely idle a lot so the average should be able to hit 1mW. Replace the radio with something more power efficient like zigbee lighting (based on 802.15.4) and you can divide these numbers by a factor 10.
A LED module for domestic lighting would likely draw 1-10W while active, so unless it is on on a very short duty cycle the LED would dominate.
"He who sows roads, will harvest traffic" - H.J. Vogel
Building more highways only encourages more driving, it is the congestion that will be invariant, not the traffic. It may take a a couple of years, and the congestion may move a bit, but the congestion will remain. Fundamentally building more roads is a lousy solution to the transport problem. See also "The fundamental law of road congestion: Evidence from US cities" By Duranton & Turner
Radon (And its decay products) can most certainly get inside your body through your respiratory system. Exposure to radon is associated with lung cancer. A quick google search finds the following article: http://www.cancer.gov/cancertopics/factsheet/Risk/radon
While radon itself has a very short biological half life since it is an inert gas, it also have a rather short physical half life decaying into various heavy atoms (polonium, lead, bismuth, thallium and mercury). This turns into airborne dust that can get lodged in your lungs. Of the decay products pb-210 might be of most interest having a half life of 22 years.
Air in planes are dehumidified to minimize the amount of ice and condensation that forms on the inside of the airplanes pressure skin. If this weren't done you might literally get a shower during landing.
It was late to market as the #3 contender. * It is unclear to the user what WP7 offers over the 'safe' bets. * Why should application developers support WP7? It gives users the worst of both worlds compared to iOS/Android: * Multiple device vendors so updates and support can be variable and fractioned * Extremely limited device support effectively locking device vendors out from doing any practical differentiation. (I.e. all devices must have same resolution, same CPU family same peripherals), effectively reducing the user choice to the same device (only cosmetic differences)
In the long term WP7 and iOS will have severe growing pains as they have both tried to leverage a hardware 'monoculture' to optimize end user experience. As technology developes Apple and MS will suffer severe pains as new hardware breaks or renders obsolete existing development work, and subsequently will need to be delayed to the end user. Apple have allready been there having to do a resolution x4 jump on the display, they likely can't do that once more, and while Apple is foremost a hardware company with experience in driving hardware development ahead of the curve, microsoft will be limited by the general availability of new components.
Android have a benefit in the long term in that it was made with support for diverse hardware in mind. This caused confusion and criticism initially, but is likely starting to pay off now.
Looking at computer history we can draw the parallel to the development of the x86 wintel platform vs. a handful of vertical oriented competitors (Amiga, Atari ST, Mac). The Wintel platform was not the prettiest, nor the friendliest, but it was the only one placed to leverage the enormous momentum in ever improving hardware. Amiga and Atari died, mac survived after a fashion, but not without scrapping everything and starting over. (Kudos to apple for pulling that off, but they very nearly broke their backs doing it)
I have no problems with Google policing the android market, in fact it might be good if they did more of that to prevent rouge apps from appearing there. The key difference between android and iOS is that you can get your apps from other sources than android market so the devs of PSX4Droid can just host the app at some other server and they are in business again.
Superior how. Do you fit your code in a smaller memory footprint, can you you make it run on a slower clock, using less energy? If not I am not interested in your so called 'superior code'.
There are other SW development jobs out there that absolutely do not require OO. In fact I would be very reluctant to hire anyone who have done OO and framework based stuff most of the time. To clarify: Our SW devs are firmware developers who work with embedded devices that get all of 2-8KB of ram and 16-128KB of flash to play with. Devs that rely on libraries and frameworks will never be able to get the job done.
On top of that I will still require your code to be understandable and reusable.
"The most lucrative market in bicycles isn't cheap commodity bikes like Schwinns" Do you have references for this. In most markets it is the low end that contributes the most to the bottom line simply because the volumes are magnitudes higher than the high-end.
"Besides, if adopted, economy of scale would drop price dramatically" Of course, economies of scale will in practical terms mean replacing the printed parts with injection molded parts. Economies of scale isn't magic. In practical terms it means you can afford to pay larger up-front costs to get to more rational manufacturing. Consider SRAMs XX casette. This part is milled through a time consuming process while other casettes are stamped. SRAM will not be able to simply push a button to manufacture more XX casettes and then get them cheaper, it is the more rational manufacturing that leads to economies of scale, not the other way around.
A lot may change yet until the election. What if the republicans end up winning despite this? Then these guys might just have caused the worst candidate (in their view) to win. Sure indications right now is that this improves odds for Obama, but it isn't a win.
I would have preferred an approach where both candidates are reasonable. I don't care much for gaming the democratic system, I think if this actually have any impact it would need to drive electoral reform.
P.S: I am european and I do not have much personal opinion on any of the candidates.
telecom was deregulated in norway in the eighties. It was done quite ineptly leaving the resulting telecom company 'telenor' in control of the distribiution infrastructure as well as the last mile granting them monopoly on the existing tax funded highly costly infrastructure.
Nevertheless an immidiate positive effect occurred in that you could get a phone installed in days instead of waiting months. After a couple of years competing distribution companies turned up (as a consequence of regulation that partly undid that they did not split the last mile from distribution, forcing the ex gov. Teleco to offer last mile access at fair rates) this turned significant pressure on pricing.
Now that cell phones have made the last mile irrelevant we have very healthy competition in telecom with a lot lower prices that before deregulation.
Considering that wifi can use both the 2.4GHz and 5GHz ISM bands, how do you figure there will be more interference issues with wifi than BT (That only use 2.4GHz as of right now)
HTC et.al will likely go to microsoft with one of their android phones and say "here is the hardware, now port the software, design a new chassis and give us money to market it and we'll sell your stuff"
It's not like they will spend much green on WM7 until it is proven in the market,
"And no one gets the idea of synchronizing the charghing cycles of your car to the output changes of a windpark 100km away... so shortsighted."
The idea have been had a while ago. It take time to implement. and of course right now the load from electrical vehicles isn't very significant, so it doesn't matter that it isn't implemented yet.
The communication protocols must be defined and the infrastructure to distribute commands from the utilities must be built.
Some keywords for you to google: Smart Grid Smart meter Zigbee smart energy 2.0
Slashdot Mirror
As a first pass you can estimate adders as 10 gates pr. bit, state as 20 gates pr. bit and multipliers as 10x bits squared (Unless it is by a power of two in which case it is free) If you need to to division in your algorithm you should redesign it. If you use floating point, everything gets huge (Try not to use floating point, remember in hardware you do not need to deal with arbitrary word size restrictions, just scale word sizes to suit the requirements)
Now, figuring out exactly what resources you need, this is where you will get into trouble. Normally you will reuse some (lots) of your arithmetic, but exactly how much depend on what performance/power/gate count target you need to hit. More reuse means less gates but faster clocks (Which can drive you to more gates if you get into trouble on timing closure). The extreme case is software which just reuse a very limited set of ALUs, the other extreme is an unrolled design where algorithmic operation have dedicated hardware, so one iteration takes one clock.
Depending on performance targets the same algorithm can have a factor 1000 difference in gate count.
A useful way of leaking a document to the public while maintaining plausible deniability? The author may be sympathetic to ACLU.
Try High Frontier by Phil Eklund (http://sierra-madre-games.eu/index_high_frontier_2nd_edition.html) A game about industrializing space made by an actual rocket scientist.
The board is beautiful with spaces representing stable orbits and movements represented by delta-v needed to change orbits.
Rocket stacks are built by reasonably realistic technologies and fueling and mass adhere to the Tsiolkovsky rocket equation (Placed in an elegant table array for simple use)
Some legislatures assert that the person whose name is associated with the cars license is ultimately responsible for illegal operation of the vehicle. Such legislatures usually cancel the fine if you have reported the car stolen or will reassign it if you provide information on who actually drove the car.
The largest issue I have with the hyperloop proposal is its rather pitiful capacity. At the highest rate proposed, with once cart every 30 seconds it still only transpoprts ~3600 PAX/hr, which is about on par with a 3 lane highway and that is before mixing in the car carriers.
Bog standard high-speed train lines do 30000 PAX/hr routinely, and while the hyperloop carts might be able to scale some, based on how they do the air bearing and that I think linked carts likely will not work, I doubt they can scale much other than by building multiple tubes (which adds upp the most expensive component in the system)
The biggest value add by public transport is to be able to free up the excessive area consumption an automobile based society incurs, but to do this the public transport in question better beat the automobile in land use with a wide margin.
Looking at speed alone is a bit of a red herring. faced with increased transport speeds people have always responded by traveling further which just escalates the problem of increased land use and increased energy use for transport. Throughput pr. unit land is likely a better metric for sustainable travel solutions than raw speed.
A good chunk of tesla sales go to norway, making norway the first country outside US to get Tesla superchargers. The reason is simple: EVs have enoromus tax breaks and road access privelidges:
No VAT (Equivalent to US sales tax) Normal VAT is 25%
No taxes. Most cars get added taxes to the tune of $4000 and up. Cars in the same performance bracket as tesla usually have taxes in the $20000-$50000 range. (Taxes are based on emissions and weight)
Access to bus lanes (Which makes most EVs in norway largely concentrated in a single county just outside Oslo where the highways are notoriously congested)
These benefits will not last .If a lot of teslas outside-US buisness is based on such tax breaks I foresee a significant political risk for teslas buisness
A wifi module is unlikely to draw more than 50-80mW while active And it can likely idle a lot so the average should be able to hit 1mW. Replace the radio with something more power efficient like zigbee lighting (based on 802.15.4) and you can divide these numbers by a factor 10.
A LED module for domestic lighting would likely draw 1-10W while active, so unless it is on on a very short duty cycle the LED would dominate.
"He who sows roads, will harvest traffic" - H.J. Vogel
Building more highways only encourages more driving, it is the congestion that will be invariant, not the traffic. It may take a a couple of years, and the congestion may move a bit, but the congestion will remain. Fundamentally building more roads is a lousy solution to the transport problem. See also "The fundamental law of road congestion: Evidence from US cities" By Duranton & Turner
Prices in fabs are normallly on a pr. wafer basis. Also normally prices go down, not up...
Radon (And its decay products) can most certainly get inside your body through your respiratory system. Exposure to radon is associated with lung cancer. A quick google search finds the following article: http://www.cancer.gov/cancertopics/factsheet/Risk/radon
While radon itself has a very short biological half life since it is an inert gas, it also have a rather short physical half life decaying into various heavy atoms (polonium, lead, bismuth, thallium and mercury). This turns into airborne dust that can get lodged in your lungs. Of the decay products pb-210 might be of most interest having a half life of 22 years.
Air in planes are dehumidified to minimize the amount of ice and condensation that forms on the inside of the airplanes pressure skin. If this weren't done you might literally get a shower during landing.
It was late to market as the #3 contender.
* It is unclear to the user what WP7 offers over the 'safe' bets.
* Why should application developers support WP7?
It gives users the worst of both worlds compared to iOS/Android:
* Multiple device vendors so updates and support can be variable and fractioned
* Extremely limited device support effectively locking device vendors out from doing any practical differentiation. (I.e. all devices must have same resolution, same CPU family same peripherals), effectively reducing the user choice to the same device (only cosmetic differences)
In the long term WP7 and iOS will have severe growing pains as they have both tried to leverage a hardware 'monoculture' to optimize end user experience. As technology developes Apple and MS will suffer severe pains as new hardware breaks or renders obsolete existing development work, and subsequently will need to be delayed to the end user. Apple have allready been there having to do a resolution x4 jump on the display, they likely can't do that once more, and while Apple is foremost a hardware company with experience in driving hardware development ahead of the curve, microsoft will be limited by the general availability of new components.
Android have a benefit in the long term in that it was made with support for diverse hardware in mind. This caused confusion and criticism initially, but is likely starting to pay off now.
Looking at computer history we can draw the parallel to the development of the x86 wintel platform vs. a handful of vertical oriented competitors (Amiga, Atari ST, Mac). The Wintel platform was not the prettiest, nor the friendliest, but it was the only one placed to leverage the enormous momentum in ever improving hardware. Amiga and Atari died, mac survived after a fashion, but not without scrapping everything and starting over. (Kudos to apple for pulling that off, but they very nearly broke their backs doing it)
Soooo this looks like a thermocouple or peltier element. What's new?
I have no problems with Google policing the android market, in fact it might be good if they did more of that to prevent rouge apps from appearing there. The key difference between android and iOS is that you can get your apps from other sources than android market so the devs of PSX4Droid can just host the app at some other server and they are in business again.
Superior how. Do you fit your code in a smaller memory footprint, can you you make it run on a slower clock, using less energy? If not I am not interested in your so called 'superior code'.
Bull.
There are other SW development jobs out there that absolutely do not require OO. In fact I would be very reluctant to hire anyone who have done OO and framework based stuff most of the time. To clarify: Our SW devs are firmware developers who work with embedded devices that get all of 2-8KB of ram and 16-128KB of flash to play with. Devs that rely on libraries and frameworks will never be able to get the job done.
On top of that I will still require your code to be understandable and reusable.
"The most lucrative market in bicycles isn't cheap commodity bikes like Schwinns"
Do you have references for this. In most markets it is the low end that contributes the most to the bottom line simply because the volumes are magnitudes higher than the high-end.
"Besides, if adopted, economy of scale would drop price dramatically"
Of course, economies of scale will in practical terms mean replacing the printed parts with injection molded parts. Economies of scale isn't magic. In practical terms it means you can afford to pay larger up-front costs to get to more rational manufacturing. Consider SRAMs XX casette. This part is milled through a time consuming process while other casettes are stamped. SRAM will not be able to simply push a button to manufacture more XX casettes and then get them cheaper, it is the more rational manufacturing that leads to economies of scale, not the other way around.
A lot may change yet until the election. What if the republicans end up winning despite this? Then these guys might just have caused the worst candidate (in their view) to win. Sure indications right now is that this improves odds for Obama, but it isn't a win.
I would have preferred an approach where both candidates are reasonable. I don't care much for gaming the democratic system, I think if this actually have any impact it would need to drive electoral reform.
P.S: I am european and I do not have much personal opinion on any of the candidates.
telecom was deregulated in norway in the eighties. It was done quite ineptly leaving the resulting telecom company 'telenor' in control of the distribiution infrastructure as well as the last mile granting them monopoly on the existing tax funded highly costly infrastructure.
Nevertheless an immidiate positive effect occurred in that you could get a phone installed in days instead of waiting months. After a couple of years competing distribution companies turned up (as a consequence of regulation that partly undid that they did not split the last mile from distribution, forcing the ex gov. Teleco to offer last mile access at fair rates) this turned significant pressure on pricing.
Now that cell phones have made the last mile irrelevant we have very healthy competition in telecom with a lot lower prices that before deregulation.
Considering that wifi can use both the 2.4GHz and 5GHz ISM bands, how do you figure there will be more interference issues with wifi than BT (That only use 2.4GHz as of right now)
HTC et.al will likely go to microsoft with one of their android phones and say "here is the hardware, now port the software, design a new chassis and give us money to market it and we'll sell your stuff"
It's not like they will spend much green on WM7 until it is proven in the market,
So i can buy the one with the most radiated energy. That way i get more reliable connections.
I herby revoke your geek license.
There are no patching capabilities in any ARM core I am familiar with. (OK they have ROM patch capabilities, but that was not what you was suggesting)
"And no one gets the idea of synchronizing the charghing cycles of your car to the output changes of a windpark 100km away... so shortsighted."
The idea have been had a while ago. It take time to implement. and of course right now the load from electrical vehicles isn't very significant, so it doesn't matter that it isn't implemented yet.
The communication protocols must be defined and the infrastructure to distribute commands from the utilities must be built.
Some keywords for you to google:
Smart Grid
Smart meter
Zigbee smart energy 2.0