I do systems and IC design, RF mostly. A couple of years ago I was doing work for a very large semiconductor company. They were using an excel spreadsheet to do the majority of their system level simulations. It was a very intricate piece of work that had been revised by one individual over many years, spanning a couple of different employers. The company purchased Agilent's ADS and was having trouble getting engineers to abandon their tried and true excel spreadsheets.
I must admit, in some ways, for some simulations I preferred it as well. With ADS most components were black boxes, when input to output didn't behave as expected you didn't have much info to debug with. At least with the excel spreadsheet all the equations were there to study.
It dawned on me about half way through the project that the reason most of the engineers preferred the excel based system was in large part because it was (bear with me) open source. Many times in simulation you aren't sure if the problem is your model, or the simulation package. Analysis was much easier (or trustworthy) when all the calculation methods were easily viewed and tested. (All this is also why I use Octave!)
The author of the spreadsheet is now working at another high profile semiconductor company. Word around the campfire is that they are using the spreadsheet now too.
I wonder if there is a processing gain acievable by doing this or of the motivation is purely power dissipation and space. If so, at the end of the day it seems like the power bill delta over the usable life of the computer wouldn't make the expense of the upgrade worthwhile (especially considering VT has an on campus power plant of their own). Wouldn't it make more sense to wait around for the 'next best thing' instead of the same thing in a different package? If it ain't broke, why fix it?
But I guess they want a super-computer the football team can be proud of...
Here in the mid-atlantic region, AEP has most of their power grid strung with fiber alongside the power lines. I have a friend who works in a local office and he used to amaze me with the bandwith of their network. AEP uses the fiber for their company data and voice networks as well as leasing the lines out.
To state the obvious, I beleive whoever comes up with a reformation type fuel cell that can reliably run off of gasoline wins the game.
Lets break down the numbers...
Liquid hydrogen at 20ÂK is about 265grams/gallon
Gasoline is roughly 2727grams/gallon.
Gasoline is a blend of n=5-12 hydrocarbons, so figure about n=8 for approximation purposes. That would be C8H18 hydrocarbons which would be about 15.8% Hydrogen by weight.
So in a gallon of gas that would be 2727g*15.8%=430grams of hydrogen, versus 265grams for a gallon of (very cold) hydrogen. Thus a gallon of gas has about 62.5% more hydrogen per unit volume over liquid hydrogen.
Diesel is even better.
Another reference point...
Assume it takes about 3minutes to fill the 12gallon tank in my GTI. Gasoline is roughly 45.8MegaJoules/kg=125MJ/gallon. Twelve gallons would be 1500MegaJoules in a 3 minute period of time. This is an average power output of 8.3 MegaWatts coming out of the gas pump!!
For everyone that wants electrolysis at the gas pump, that method has it's own ineffeciency problem which would up the refueling power budget. Water is an ash...not a fuel.
Physicist James Hallock, another board member, said he does not believe a missing seal alone could have created a big enough hole for the kind of heat damage experienced by Columbia. As the plumes of hot gas entered the long, narrow gap, it probably chipped or broke away at the adjoining wing panels and created an even bigger breach -- enough to lead to the ship's destruction, he said.
Isn't this kind of like saying the bullet isn't what killed him, it was the hole it left behind?
Back in the day when I was cutting my teeth at a contract manufacturer, we used a toaster oven, an I/O card and an 8086 to do reflow soldering. That was back when the majority of components were still had leads and we only did limited runs of SMT parts. We couldn't afford a real reflow oven for a while. We used and abused that oven for years then actually SOLD the thing.
A reflow oven is just a fancy pizza oven. It has a conveyor belt and different zones so you can controll the time/temperature profile of the board as the solder starts to flow (tuning this is an art). I went to Nepcon one year (a electronics manufacturing trade show) and a company had some blonde eye candy making chocolate chip cookies in their top of the line reflow oven. It was obviously virgin equipment...the oven that is.
Some of you mentioned using light to solder...this is quite common now. White light actually works better than lasers believe it or not. The broad band spectral content transfers more energy to the solder.
I believe this was originally developed in a UCLA lab a few years ago. Rumor has it the LAPD was involved in a raid of the facility. When questioned the LAPD spokesperson would only say "it fit the description".
it's not very difficult to shield against the effects of this weapon.
Just for a rough sanity check...
Decent rigid coaxial cable offers about 100dB of shielding . One-million watts = 90dBm, so that would drop it down to -10dBm interference in a shielded signal. Not enough to damage anything, but definately enough to interfere. Bluetooth and 802.11b run at a max of 20dBm and no cars crash outside when I key up the old bit blaster.
The absorbtion of the radiated power is also an issue. Different circuits absorb different frequencies better than others. If this was a fairly narrowband emission, it would wreak havoc on some things (soft tissue maybe) but not others. If it is very wide band, then you have to jack up your total power so that many different freuquencies have a potent allotment of power.
It would just be a lot easier to interfere (jam) with guidance systems and radar. And GPS is easy jam. At least that was the FCC's standpoint with respect to UWB. But that's another thread...
So when it uses up all the oxygen in my room I guess I can expect the pressure to drop rapidly as well since the resulting water vapor far,far more dense a concentration of oxygen and hydrogen.
I am so glad someone pointed this out...I coulda' damaged me ears...
enough of the smart arse stuff...
Why do people worry so much about the tank of hydrogen? A gallon of gasoline has about 60% more hydrogen than a gallon of liquid hydrogen at 20 degrees K.
To generate these sound pressure levels you most likely would use a tuned cavity and we already know that it contains a high pressure gas. If the cavity was damaged, leaked, whatever...the cavity would no longer be tuned, thus little sound.
My guess is that the frequencies are also high to keep the size of the cavity down. At least it would keep my dogs out of the kitchen.
The best way to keep it quiet is to make the system efficient. If the efficiency is high then all the acoustic energy will be turn into heat (on the hot plate).
Slashdot Mirror
Probably no where near there yet, but this is a step toward a 'no glasses required' three dimensional display.
I do systems and IC design, RF mostly. A couple of years ago I was doing work for a very large semiconductor company. They were using an excel spreadsheet to do the majority of their system level simulations. It was a very intricate piece of work that had been revised by one individual over many years, spanning a couple of different employers. The company purchased Agilent's ADS and was having trouble getting engineers to abandon their tried and true excel spreadsheets.
I must admit, in some ways, for some simulations I preferred it as well. With ADS most components were black boxes, when input to output didn't behave as expected you didn't have much info to debug with. At least with the excel spreadsheet all the equations were there to study.
It dawned on me about half way through the project that the reason most of the engineers preferred the excel based system was in large part because it was (bear with me) open source. Many times in simulation you aren't sure if the problem is your model, or the simulation package. Analysis was much easier (or trustworthy) when all the calculation methods were easily viewed and tested. (All this is also why I use Octave!)
The author of the spreadsheet is now working at another high profile semiconductor company. Word around the campfire is that they are using the spreadsheet now too.
Not a G5 expert here...
I wonder if there is a processing gain acievable by doing this or of the motivation is purely power dissipation and space. If so, at the end of the day it seems like the power bill delta over the usable life of the computer wouldn't make the expense of the upgrade worthwhile (especially considering VT has an on campus power plant of their own). Wouldn't it make more sense to wait around for the 'next best thing' instead of the same thing in a different package? If it ain't broke, why fix it?
But I guess they want a super-computer the football team can be proud of...
Here in the mid-atlantic region, AEP has most of their power grid strung with fiber alongside the power lines. I have a friend who works in a local office and he used to amaze me with the bandwith of their network. AEP uses the fiber for their company data and voice networks as well as leasing the lines out.
To state the obvious, I beleive whoever comes up with a reformation type fuel cell that can reliably run off of gasoline wins the game.
Lets break down the numbers...
Liquid hydrogen at 20ÂK is about 265grams/gallon
Gasoline is roughly 2727grams/gallon.
Gasoline is a blend of n=5-12 hydrocarbons, so figure about n=8 for approximation purposes. That would be C8H18 hydrocarbons which would be about 15.8% Hydrogen by weight.
So in a gallon of gas that would be 2727g*15.8%=430grams of hydrogen, versus 265grams for a gallon of (very cold) hydrogen. Thus a gallon of gas has about 62.5% more hydrogen per unit volume over liquid hydrogen.
Diesel is even better.
Another reference point...
Assume it takes about 3minutes to fill the 12gallon tank in my GTI. Gasoline is roughly 45.8MegaJoules/kg=125MJ/gallon. Twelve gallons would be 1500MegaJoules in a 3 minute period of time. This is an average power output of 8.3 MegaWatts coming out of the gas pump!!
For everyone that wants electrolysis at the gas pump, that method has it's own ineffeciency problem which would up the refueling power budget. Water is an ash...not a fuel.
Isn't this kind of like saying the bullet isn't what killed him, it was the hole it left behind?
Back in the day when I was cutting my teeth at a contract manufacturer, we used a toaster oven, an I/O card and an 8086 to do reflow soldering. That was back when the majority of components were still had leads and we only did limited runs of SMT parts. We couldn't afford a real reflow oven for a while. We used and abused that oven for years then actually SOLD the thing.
A reflow oven is just a fancy pizza oven. It has a conveyor belt and different zones so you can controll the time/temperature profile of the board as the solder starts to flow (tuning this is an art). I went to Nepcon one year (a electronics manufacturing trade show) and a company had some blonde eye candy making chocolate chip cookies in their top of the line reflow oven. It was obviously virgin equipment...the oven that is.
Some of you mentioned using light to solder...this is quite common now. White light actually works better than lasers believe it or not. The broad band spectral content transfers more energy to the solder.
I believe this was originally developed in a UCLA lab a few years ago. Rumor has it the LAPD was involved in a raid of the facility. When questioned the LAPD spokesperson would only say "it fit the description".
it's not very difficult to shield against the effects of this weapon.
Just for a rough sanity check...
Decent rigid coaxial cable offers about 100dB of shielding
. One-million watts = 90dBm, so that would drop it down to -10dBm interference in a shielded signal. Not enough to damage anything, but definately enough to interfere. Bluetooth and 802.11b run at a max of 20dBm and no cars crash outside when I key up the old bit blaster.
The absorbtion of the radiated power is also an issue. Different circuits absorb different frequencies better than others. If this was a fairly narrowband emission, it would wreak havoc on some things (soft tissue maybe) but not others. If it is very wide band, then you have to jack up your total power so that many different freuquencies have a potent allotment of power.
It would just be a lot easier to interfere (jam) with guidance systems and radar. And GPS is easy jam. At least that was the FCC's standpoint with respect to UWB. But that's another thread...
So when it uses up all the oxygen in my room I guess I can expect the pressure to drop rapidly as well since the resulting water vapor far,far more dense a concentration of oxygen and hydrogen.
I am so glad someone pointed this out...I coulda' damaged me ears...
enough of the smart arse stuff...
Why do people worry so much about the tank of hydrogen? A gallon of gasoline has about 60% more hydrogen than a gallon of liquid hydrogen at 20 degrees K.
To generate these sound pressure levels you most likely would use a tuned cavity and we already know that it contains a high pressure gas. If the cavity was damaged, leaked, whatever...the cavity would no longer be tuned, thus little sound.
My guess is that the frequencies are also high to keep the size of the cavity down. At least it would keep my dogs out of the kitchen.
The best way to keep it quiet is to make the system efficient. If the efficiency is high then all the acoustic energy will be turn into heat (on the hot plate).