There was an article on Zigbee in Circuit Cellar that said the same thing about being an X-10 replacement.
The big advantage of Zigbee over X-10 is that it can do handshaking. Say you have a wireless light switch implemented with Zigbee - the light switch sens out a "turn-on" signal and will keep doing so until it hears an acknowledgement. It could also be set up to awake every now and then to re-send in case of power failure.
AOL and Compuserv also started as competitors to the Internet.
I would beg to differ about Compuserve being started as a competitor to the Internet - Compuserve's roots, IIRC, predate the Internet by a few months. Compuserve did have its own data network up until the great AOL-Compuserve-Worldcom eff-over took place.
Up until 1996-97, Compuserve was very friendly to all sorts of oddball computing platforms - as long as your box could speak ascii to a modem, you could log onto CIS. In the early 90's they started working on their proprietary HMI and started to make it mandatory by '97.
The forums on CIS were very much like Usenet done right, the moderation kept the flaming to a minimum. Usenet, FWIW, was distributed by UUCP for several years before nntp was developed.
Re:Alvin Fernald, Danny Dunn, and Henry Reed
on
The Big Kerplop
·
· Score: 1
My favorite, "Danny Dunn and the Homework Machine"
Man that brings back some memories. Read that book in 5th grade - that was 1965. I beleive the book was written in the early 60's, the one about the time machine was set in 1963 (and 1763).
The storyline did foreshadow word processors - the original "homework machine" in the book was a board suspended by strings and holding two pens - allowing two copies of a handwritten paper to be done by one person.
One amusing detail - Danny had a problem when his nemesis (don't remember the name - been over 38 years since I've read the book) turned down the thermostat on the computer - causing it to garble output from being too cold...
Funny to think that the idea of a computer in the house was as far out as a time machine.
First, it turns out that the cost of electric at the wall-socket is not dominated by the cost of production, but by the cost of the power grid. If the power were completely free, cost/kW-h at the home would only go down by about 50 percent.
Sounds about right - figures I've seen state that generation, transmission and distribution costs are about 40/20/40. BTW, although I am not a power engineer, I did several power systems courses at the Big U.
Hydro power, wind power and some solar power have essentially zero marginal cost of production - it is the capital costs that kill ya. Conversely, a small combustion turbine can have very low capital costs, but the marginal costs can be killers (especially if oil and natural gas prices have spiked).
If the energy is in electrical form, then essentially all fixed energy consumers would switch to electricity. The market for oil would be limited to transportation - and that may be satisfied by capturing CO2 from the air and reacting it with H2 from water (remember, hydrocarbons have a much higher energy density than hydrogen).
Most countries have this gap in the ranges because it is not a very good transmition band due to it being the resonant frequency of water.
You're off by almost a factor of 10 - the resonant frequency of water vapor is about 22 GHz NOT 2.45 GHz. The resonance absorption line for liquid water is incredibly broad - I've seen an industrial "microwave oven" running at 916 MHz with about 1MW of power (even heard of heating from 144 MHz). Note that 916 MHz and 2.45 GHz are in the middle of the respective ISM (Industrial Scientific Medical) bands.
Physics lesson - the dielectric constant of water is about 81 at low frequencies and 2 to 3 at optical frequencies. At low frequencies, the water molecules have time to follow an oscillating electric field (and thus produce a large E-field polarization). At optical frequencies, the water molecules are essentially stationary (and thus produce very little polarization). Near the resonance, the molecules mostly follow the oscillating E-field but absorb a lot of energy in the process - i.e. the heating from a microwave oven.
As far as the 106 CPU Version Cluster the XServes the same way Sun does it.
You're way off.
For starters, the Sun E15K is a system not a cluster. Any one of the 106 processors can address any byte of the maximum of 500 GB of memory in the E15K, with worst case access on the order of 100 nsec more than the typical case. Try doing that with a cluster. FWIW, SGI's Origin 3K can go to even larger memory images.
Another thing, the UltraSparc 3 used in the larger SUN boxes is capable of maintaining cache coherency with up to 1023 processors. IIRC, the Opteron is good for 8 processors if you get the version of the Opteron designed for 8 processors. Similarly, a difference between the 970 and the Power 4 is the maximum number of processors in a unified system.
Cache coherency is important for performance reasons. Simply put, cache is there for the reason of reducing time to access memory content. In a single processor system, there's only one processor writing to memory (ignoring DMA for the moment) and the cached memory content is good by default. In a multiprocessor system, processor 'B' can write to memory cached by processor 'A', so processor 'A' needs to have some way of knowing the cache is good - the slow way is re-reading that memory location (thus defeating the purpose of the cache). The fast way is to have a means to communicate when the cache lines are invalid, then and only then, would the processor have to reload the cache line.
Re:"Regular" 100-volt?
on
42-Volt Autos
·
· Score: 1
Wait! It's 110AC RMS that doesn't convert to +/= 110VDC with just a regulator, was it just the same for marketing reasons?
The Edison system used three wires per customer, one was at plus 110VDC, one was at 0VDC (the neutral) and the last was at -110VDC. Connecting to the plus and minus side gave you 220VDC. The U.S residential standard is 110/220 VAC is essentially the same scheme, the two "line" conductors are 180 degrees out of phase (i.e. +/-) and 110 V rms was chosen so that light bulbs would work correctly.
Re:"Regular" 100-volt?
on
42-Volt Autos
·
· Score: 1
What's "regular" about that? Last time I checked ~1/10 of the world (i.e. USA, and maybe some of its "allies") used it.
Edison's Pearl Street station is considered to be the first electric utility - supplied voltage was +/- 110VDC (hence the US single phase standard of 110 VAC). Pre-WWII, line frequency could be DC, 25 Hz, 50 Hz, 60 Hz or 133 Hz. Off-grid power standard was 32VDC (including RR passenger cars.
It was the "rest of the world" that was non-complying...
Re:Not a EE, car nut or ME, but...
on
42-Volt Autos
·
· Score: 2, Interesting
Wouldn't it make more sense to just install a (slightly) larger generator in the drive line, then use a powered blower to compress the incoming air? Less parts/chance of failure, less back pressure on the engine, etc, etc, etc... It also seems like it'd be more mechanically efficient as well.
W-e-l-l if you've looked into the literature on turbocharging, you'd find that back pressure isn't usually a problem (unless it gets really high). The Wright R-3350 turbo compounds got about 20% more power by coupling the exhaust turbines through a fluid clutch. The idea is that there is a lot of wasted kinetic energy is the exhaust of a non-turbo'ed piston engine.
How ? do 42V alternators have a better efficiency than 12V ones ?
They can, remember that the diode drop is less of an issue at 42V than 14V. Also alternators generally become more efficient as size increases.
I can't wait to see cars with seat heaters
They've been here for a couple of decades, e.g. SAAB and Volvo.
what are the mechanical parts that can't be replaced by 12V electrical equivalents that could be replaced by 42V ones ?
Power steering (hydraulic pump is always on in mechanical systems), Air conditioning - easier to mount the compressor (get it away from the #@&%$! engine heat) - also compressor not dependant on engine RPM. Last but not least, hybdrid drive.
Now what would really be neat is making the turbocharger part of the hybrid system - instead of using a wastegate to dump excess exhaust energy, use the turbine to run an alternator - then the compressor could be electrically driven (no turbo lag) and excess electricity can charge the battery or be fed into the drivetrain. -- Anyone remember the Curtiss Wright Turbo Compound egine used on the L-1649 Connie's, the DC-7C's and others?
We're already dropping nuclear material all over ourselves, and for the most part, you aren't going to hear about it until it's declassified.
To say nothing of the 5 tonnes of Pu that was put into the atmosphere from weapons testing. (Hint: the best bombs fission maybe 25% of their fissile inventory.)
A couple of other posters have pointed out that reactors aren't all that dirty until they've been "fired up".
Anyway the fallout due to a launch accident will be a drop in a bucket compared to what is already out there.
The issue with GSM was that Qualcomm 'sponsored' a senator to do its bidding.
Darrell Issa is a congressman, not a senator. For the more part he does seem to have a much better clue with what's going on in the world than most congresscritters.
The Goodyear blimps typically get a half dozen bullet holes in them while crossing the U.S. Whether there are fewer rednecks near York as opposed to the central U.S. is a good question.
My original point is that heat had to be taken out of the air in order to get the water vapor (humidity) to condense. The reject heat can then be used to reheat the air (and will be warmer than the ambient air temp). This is related to why you get such good convection cells going in thunderclouds.
I'll beg to disagree, having a separate dehumidifier may be more effiecient than having the air conditioner do all the work.
The issue is that dehumidification requires a fixed amount of heat to be removed from the air - with a dehumidifier, the temperature difference at which this heat is being removed is smaller than the air conditioner (using the already cooled air). Since the air conditioner is not dehumidifying, you can run more air past the evaporator coils, thus improving the efficiency of the air conditioner.
Also with drier air, you may be comfortable with a higher room air temperature.
Re:considered the father of Linux?
on
Today's SCO News
·
· Score: 1
Have you heard of the Timeline case against Microsquat? (M$ paid patent royalties for their own use, but not for M$ customers.)
The upshot is that the end users may indeed be liable for patent infringement.
Your exact same scenario would apply if the employee stole the code and put it into a closed-source program. If this was a problem then buyers would have to worry about *all* software they buy, not just open source.
This is indeed the case - witness the brouhaha over MS SQL and the patent infringement case.
The one supposed difference is that the seller of commercial software may take the liability for any violation of copyright. Whether that works out in practice is another story.
Let's say you were a coder at IBM facing a really annoying problem in the Linux kernel. You have access to the relevant SCO code and cut and paste in a bit of code so that you can finish and get home early, thinking surely no one will ever notice.
Assuming that is what happened - that coder would be well advised to get a good lawyer - he/she could potentially face some criminal charges as well as civil charges. (Flamebait reminder: I am talking about a course of action based on the assumption being correct, but I am not presuming that this is what actually happened.)
IIRC, the pringles can antenna was about 1 wavelength long (basically a disk Yagi). Figure 11 to 12 dBi.
A 2 foot diameter dish antenna will give you about 24 dBi of gain - enough to require scaling back on the power. There was a story on/. a while back about a link from San Diego to San Clemente Island that had to have the power throttled back.
My father-in-law was showing me his bound volumes of the first five years of Trains magazine - in the final issue of the first year there was a picture of the Descanso sitting at the summit of Cajon pass. Remember it sitting at the intersection of Broadway and Alpine for a number of years at OERM.
OERM shipped the two Muni PCC's (1033 and 1039) back to the Muni and got a nice chunk of change in return.
Hmm, don't think the JBL would be interested in acquiring the Tandy Trolleys.
How do they prevent people outside the company from owning stock?
Simple, all the stock had to be bought and sold through Bull, Inc., which is SAIC's stock trading subisdiary. Needless to say, Bull has a pretty good idea of who qualifies to own stock. Employess can keep the stock in their 401(k) plans after leaving the company.
There are also provisions for family members to "own" stock, but that has to go on the employee's account - kind of like stock held in trust for a minor.
but when you leave you're forced to liquidate all your holdings in the company
There are a couple of exceptions - the stock in the 401(k) plan can be held after you leave. The other exception are the people who own SAI stock (essentially pre 1982).
One other point - SAIC is an Employee owned company not a privately held company.
6/7/8/9, 7/8/9 cards... good old SCOPE, UCB worked on it and called it CAL Improved Design Of SCOPE
At Bezerkeley, The job cards were white with the UC logo (actually had job cards for the 360 at UCSF), the 6/7/8/9 cards were pink and the data cards were "IBM card tan" with a blue stripe on top. Got yelled at once for typing out a 6/7/8/9 card with a data card.
Slashdot Mirror
The big advantage of Zigbee over X-10 is that it can do handshaking. Say you have a wireless light switch implemented with Zigbee - the light switch sens out a "turn-on" signal and will keep doing so until it hears an acknowledgement. It could also be set up to awake every now and then to re-send in case of power failure.
I would beg to differ about Compuserve being started as a competitor to the Internet - Compuserve's roots, IIRC, predate the Internet by a few months. Compuserve did have its own data network up until the great AOL-Compuserve-Worldcom eff-over took place.
Up until 1996-97, Compuserve was very friendly to all sorts of oddball computing platforms - as long as your box could speak ascii to a modem, you could log onto CIS. In the early 90's they started working on their proprietary HMI and started to make it mandatory by '97.
The forums on CIS were very much like Usenet done right, the moderation kept the flaming to a minimum. Usenet, FWIW, was distributed by UUCP for several years before nntp was developed.
Man that brings back some memories. Read that book in 5th grade - that was 1965. I beleive the book was written in the early 60's, the one about the time machine was set in 1963 (and 1763).
The storyline did foreshadow word processors - the original "homework machine" in the book was a board suspended by strings and holding two pens - allowing two copies of a handwritten paper to be done by one person.
One amusing detail - Danny had a problem when his nemesis (don't remember the name - been over 38 years since I've read the book) turned down the thermostat on the computer - causing it to garble output from being too cold...
Funny to think that the idea of a computer in the house was as far out as a time machine.
Sounds about right - figures I've seen state that generation, transmission and distribution costs are about 40/20/40. BTW, although I am not a power engineer, I did several power systems courses at the Big U.
Hydro power, wind power and some solar power have essentially zero marginal cost of production - it is the capital costs that kill ya. Conversely, a small combustion turbine can have very low capital costs, but the marginal costs can be killers (especially if oil and natural gas prices have spiked).
If the energy is in electrical form, then essentially all fixed energy consumers would switch to electricity. The market for oil would be limited to transportation - and that may be satisfied by capturing CO2 from the air and reacting it with H2 from water (remember, hydrocarbons have a much higher energy density than hydrogen).
You're off by almost a factor of 10 - the resonant frequency of water vapor is about 22 GHz NOT 2.45 GHz. The resonance absorption line for liquid water is incredibly broad - I've seen an industrial "microwave oven" running at 916 MHz with about 1MW of power (even heard of heating from 144 MHz). Note that 916 MHz and 2.45 GHz are in the middle of the respective ISM (Industrial Scientific Medical) bands.
Physics lesson - the dielectric constant of water is about 81 at low frequencies and 2 to 3 at optical frequencies. At low frequencies, the water molecules have time to follow an oscillating electric field (and thus produce a large E-field polarization). At optical frequencies, the water molecules are essentially stationary (and thus produce very little polarization). Near the resonance, the molecules mostly follow the oscillating E-field but absorb a lot of energy in the process - i.e. the heating from a microwave oven.
You're way off.
For starters, the Sun E15K is a system not a cluster. Any one of the 106 processors can address any byte of the maximum of 500 GB of memory in the E15K, with worst case access on the order of 100 nsec more than the typical case. Try doing that with a cluster. FWIW, SGI's Origin 3K can go to even larger memory images.
Another thing, the UltraSparc 3 used in the larger SUN boxes is capable of maintaining cache coherency with up to 1023 processors. IIRC, the Opteron is good for 8 processors if you get the version of the Opteron designed for 8 processors. Similarly, a difference between the 970 and the Power 4 is the maximum number of processors in a unified system.
Cache coherency is important for performance reasons. Simply put, cache is there for the reason of reducing time to access memory content. In a single processor system, there's only one processor writing to memory (ignoring DMA for the moment) and the cached memory content is good by default. In a multiprocessor system, processor 'B' can write to memory cached by processor 'A', so processor 'A' needs to have some way of knowing the cache is good - the slow way is re-reading that memory location (thus defeating the purpose of the cache). The fast way is to have a means to communicate when the cache lines are invalid, then and only then, would the processor have to reload the cache line.
The Edison system used three wires per customer, one was at plus 110VDC, one was at 0VDC (the neutral) and the last was at -110VDC. Connecting to the plus and minus side gave you 220VDC. The U.S residential standard is 110/220 VAC is essentially the same scheme, the two "line" conductors are 180 degrees out of phase (i.e. +/-) and 110 V rms was chosen so that light bulbs would work correctly.
Edison's Pearl Street station is considered to be the first electric utility - supplied voltage was +/- 110VDC (hence the US single phase standard of 110 VAC). Pre-WWII, line frequency could be DC, 25 Hz, 50 Hz, 60 Hz or 133 Hz. Off-grid power standard was 32VDC (including RR passenger cars.
It was the "rest of the world" that was non-complying...
W-e-l-l if you've looked into the literature on turbocharging, you'd find that back pressure isn't usually a problem (unless it gets really high). The Wright R-3350 turbo compounds got about 20% more power by coupling the exhaust turbines through a fluid clutch. The idea is that there is a lot of wasted kinetic energy is the exhaust of a non-turbo'ed piston engine.
They can, remember that the diode drop is less of an issue at 42V than 14V. Also alternators generally become more efficient as size increases.
I can't wait to see cars with seat heaters
They've been here for a couple of decades, e.g. SAAB and Volvo.
what are the mechanical parts that can't be replaced by 12V electrical equivalents that could be replaced by 42V ones ?
Power steering (hydraulic pump is always on in mechanical systems), Air conditioning - easier to mount the compressor (get it away from the #@&%$! engine heat) - also compressor not dependant on engine RPM. Last but not least, hybdrid drive.
Now what would really be neat is making the turbocharger part of the hybrid system - instead of using a wastegate to dump excess exhaust energy, use the turbine to run an alternator - then the compressor could be electrically driven (no turbo lag) and excess electricity can charge the battery or be fed into the drivetrain. -- Anyone remember the Curtiss Wright Turbo Compound egine used on the L-1649 Connie's, the DC-7C's and others?
To say nothing of the 5 tonnes of Pu that was put into the atmosphere from weapons testing. (Hint: the best bombs fission maybe 25% of their fissile inventory.)
A couple of other posters have pointed out that reactors aren't all that dirty until they've been "fired up".
Anyway the fallout due to a launch accident will be a drop in a bucket compared to what is already out there.
Darrell Issa is a congressman, not a senator. For the more part he does seem to have a much better clue with what's going on in the world than most congresscritters.
The Goodyear blimps typically get a half dozen bullet holes in them while crossing the U.S. Whether there are fewer rednecks near York as opposed to the central U.S. is a good question.
My original point is that heat had to be taken out of the air in order to get the water vapor (humidity) to condense. The reject heat can then be used to reheat the air (and will be warmer than the ambient air temp). This is related to why you get such good convection cells going in thunderclouds.
The issue is that dehumidification requires a fixed amount of heat to be removed from the air - with a dehumidifier, the temperature difference at which this heat is being removed is smaller than the air conditioner (using the already cooled air). Since the air conditioner is not dehumidifying, you can run more air past the evaporator coils, thus improving the efficiency of the air conditioner.
Also with drier air, you may be comfortable with a higher room air temperature.
The upshot is that the end users may indeed be liable for patent infringement.
This is indeed the case - witness the brouhaha over MS SQL and the patent infringement case.
The one supposed difference is that the seller of commercial software may take the liability for any violation of copyright. Whether that works out in practice is another story.
Assuming that is what happened - that coder would be well advised to get a good lawyer - he/she could potentially face some criminal charges as well as civil charges. (Flamebait reminder: I am talking about a course of action based on the assumption being correct, but I am not presuming that this is what actually happened.)
IIRC, the pringles can antenna was about 1 wavelength long (basically a disk Yagi). Figure 11 to 12 dBi.
A 2 foot diameter dish antenna will give you about 24 dBi of gain - enough to require scaling back on the power. There was a story on /. a while back about a link from San Diego to San Clemente Island that had to have the power throttled back.
OERM shipped the two Muni PCC's (1033 and 1039) back to the Muni and got a nice chunk of change in return.
Hmm, don't think the JBL would be interested in acquiring the Tandy Trolleys.
Why do I get the impression that someone is nostalgic for the Crown tape recorders?
Simple, all the stock had to be bought and sold through Bull, Inc., which is SAIC's stock trading subisdiary. Needless to say, Bull has a pretty good idea of who qualifies to own stock. Employess can keep the stock in their 401(k) plans after leaving the company.
There are also provisions for family members to "own" stock, but that has to go on the employee's account - kind of like stock held in trust for a minor.
There are a couple of exceptions - the stock in the 401(k) plan can be held after you leave. The other exception are the people who own SAI stock (essentially pre 1982).
One other point - SAIC is an Employee owned company not a privately held company.
I got to play with a cross assembler for the 8008! (this was 30 years ago) The code was then run on an 8008 simulator running on the 6400.
At Bezerkeley, The job cards were white with the UC logo (actually had job cards for the 360 at UCSF), the 6/7/8/9 cards were pink and the data cards were "IBM card tan" with a blue stripe on top. Got yelled at once for typing out a 6/7/8/9 card with a data card.
Bet you had fun with the blue cards...