No, the computer did what it was programmed to do.
Unfortunately the programming specs did not require the program to do the most reasonable thing when presented with bad input data.
There should have been some testing spec along the lines of "and furthermore it should be possible to replace any air data input by/dev/random without affecting safety of flight".
In the old days I'd require that any program would tolerate a simulated money at the keyboard. Sooo many programs failed that simple test!
This sounds like a scenario where lawyers are trying to act as engineers. That works about as well as you might expect.
There are these engineering things, amusingly called "Schmoo plots", that map out a chip's operating envelope of voltage versus speed versus temperature. From those an engineer can forsee how hot you can run a chip before its rise and fall time margins start to get marginal.
There is very little Intel can do to stretch thing by another 5 degrees. It's not something that can be imposed by fiat. Intel engineers have already juggled all the variables to come up with the best performance possible. SOMETHING is going to have to give. Either the chips will have to be selected and graded for speed, lowering the overall envelope for the chips everyone else gets, or they'll have to fudge some other parameters, hoping nobody will notice, or worse yet they'll tweak some variable right to the edge of raggedness, resulting in worse reliability down the road.
Lawyers and accountants generally don't know you can't have everything. let's hope the engineers educate them.
The story would be more believable if they did not get certain basic facts wrong. Mercury has a very low vapor pressure, it's not going to evaporate very quickly. That's why you don't see mercury fog inside a mercury switch or thermometer. The cost and weight of the mercury are inconsequential compared to the cost of the rocket to lift the telescope up there.
amazing what doesnt get asked
on
C# In-Depth
·
· Score: 0, Flamebait
Talk about softballs, some of us would like the answers to these questions:
(1) How much did Microsoft have to pay you to switch from the side of the angels to their side?
(2) Who decided to make C# just enough like Java to fragment and muddle the whole programming scene?
(3) How can you consider C# to be a success when Microsoft tried using it to replace large parts of Windows and ended up throwing the whole mess away?
100 kilowatt hours over one year is a constant load of about 12 watts.
Air has about 0.03% CO2. Let's round that up to 0.1% or one part per thousand.
Now comes the hard part-- how much air can a 12-watt fan force through one square meter of CO2 filter material every second? Hard to say, let's estimate ten liters for starters. That would require an efficient fan and duct and a not very restrictive filter.
Air has a density of about 1.2 g/liter, so we're processing 12g/sec of air. Of that about 12mg is CO2. Per year that's about 380kg. Not too far off a ton, but not too close either.
Now we have to figure out how much energy it took to make the filter material. To capture a ton of CO2 takes about an equal amount of caustic calcium or lithium hydroxides. Unfortunately if we include the CO2 emitted to make a ton of calcium or sodium hydroxide, that's way beyond me, but it may be more than 380kg equivalent of CO2.
You have to look at all the costs and emissions involved in the whole cycle.
looking at the PDF of their paper reveals some weird stuff.
After a lot of blabbing they admit the viscosity change is only 10%. Looking at the curves for Diesel fuel viscosity, that's equivalent to heating the fuel another 10 degrees C.
In their "tests" they used an injector pressure of 200PSI. Typical cars use 2,000 PSI and some of the newer Diesels use up to 22,000 PSI! Makes you wonder why they used such a low pressure.
Their real-world test was with a Mercedes Benz diesel engine hooked up to a dynamometer, but apparently running AT IDLE. A fuel consumption of 500 grams per hour. A power output of 1/3 horsepower or so. Does not sound like typical engine operating conditions.
I would be very wary of this device given the bizarre test conditions.
First problem, if electric fields reduced viscosity or surface tension, it would have been discovered like 100 years ago. Thousands of scientists have put stuff in electric fields. it's mighty unlikely none of them noticed this effect.
Secondly, there are totally free and reliable ways of reducing viscosity, like wrapping the fuel line around the exhaust manifold. Free, does not require any fancy electronics, and has the added benefit of actually working.
If they're paying ten cents a kilowatt-hour, that 10MW data center is paying about $9M/yr for power. Cooling systems move about 15 times the power than what they draw. So the savings for a 10MW datacenter would be around $600K. Wonder how they came up with $2.9M ?
Read a little closer. They say the graphene sheets are conductive, so they're more like part of the plates. In any case, anything atom-thick is going to be problematic, whether it's plate or dielectric.
It's golly-gee wonderful if they can make a one-atom thick graphene sheet. Give them a lollipop.
But in making a capacitor, you need other attributes than just thinness. You need a capacitor plate that can carry current, remain in place in the face of strong electrostatic fields, be compatible with dielectrics, be reliable, and be manufacturable.
A one-atom thick sheet is not going to be able to do any of those things. Capacitor makers have been depositing thin electrodes for 60 years now. They know full well what the limits are. The limits are about four powers of ten higher than these neophytes are talking about.
>"unmanned flying drones can track targets even inside of buildings."
Very, very unlikely. If the drone is using passive sensors, it would have to be sensing some emanations from the buildings. Assuming the baddies are smart enough to not use cell-phones or WiFi, that leaves infrared. Adobe and brick are darn good insulators in the infrared.
If the drones are using active means, the baddies can use lo-tech sensors (the eyeball, Fuzzbuster radar detectors, binoculars) to spot them and act casual.
Maybe the writer meant that the drones have enough endurance to loiter in the area until they come back out into the open.
The TI invention was not what we would consider an "IC". It had components, but it wasnt practical to mass produce. Not even TI followed up on its development.
The real practical IC, with photoetched traces on a planar silicon substrate was developed by Fairchild.
>I've seen micrographs of hard disc platters. The previously written bits are as clear as day.
We've all seen those pics. The pics are misleading. the typical picture is of an erased area shaped into the letters "MIT" or "IBM". You can still see traces of bits under there, yeah. Useless. A real disk head never writes a string of zeros, that violates the basic principles of group code recording and DC balance. A real head would be writing new data with uncorrelated 1's and 0's, unsynced on the bit and amplitude levels of the previous data. Try reading the old bits THEN.
Read the original posting. They were talking about wardriving to collect tons of addresses, not spending many minutes driving in circles around a block to hone in on one specific point. That is of course doable in principle, if not practical. And even so, of what use is knowing that 123 Easy Street has a router named "linksys_2367443" ?
>Wifi OTOH using radio allows anyone in range to find out its address. Thats the problem.
What's the problem with knowing a MAC address? The MAC is not a key to anything except sending a packet to the router. Which is the whole point of having a WiFi router.
You don't need malware or anything else to get a router's MAC address, it's in every packet the router sends out.
And you can't easily get an exact street address from wardriving. All you know is somewhere along the antenna's main lobe there is a router. Could be 10 feet away, could be 500.
And knowing the MAC address is of no earthly use. Well, in the old days you could map it to a ethernet chip manufacturer, but now most routers have changeable MAC addresses.
You can't map MAC address to email addresses either, as the summary claims. Sheesh.
Eh, any idea how they'd cool the thing? It's fine to split atoms to make heat, but on the Moon you need to have a closed-loop cooling system. So you have to cool off the turbine exhaust so you can feed it back into the reactor. Problem-- no atmosphere and no lakes or rivers to carry away the heat. No groundwater either. Many many many meters of loose insulating moon-dust and rock fragments before you get down to bedrock, which in itself is not all that great at conducting away heat.
Methinks the Moon is not a great place to be running a reactor or power plant of the heat-cycle variety. Maybe solar cells.
Why would you need 24 contiguous hours of battery life? Most of us sleep at least 8 hours out of every 24.
uh, no, that's not the reason
on
DIY Hybrid Car Kit
·
· Score: 5, Informative
Uh, building a hybrid at home probably makes more pollution than making it in a factory.
The reason they sell it as a kit is to avoid all the federal vehicle rules. By passing on assembly to the end-user, it becomes THEIR problem to get the car licensed.
Also I don't quite get the "zero dollars to manufacture". Lots of the steps involve lots of time, welding, painting, trips to the hardware store. That all costs many $$$.
Slashdot Mirror
No, the computer did what it was programmed to do.
Unfortunately the programming specs did not require the program to do the most reasonable thing when presented with bad input data.
There should have been some testing spec along the lines of "and furthermore it should be possible to replace any air data input by /dev/random without affecting safety of flight".
In the old days I'd require that any program would tolerate a simulated money at the keyboard. Sooo many programs failed that simple test!
This sounds like a scenario where lawyers are trying to act as engineers. That works about as well as you might expect.
There are these engineering things, amusingly called "Schmoo plots", that map out a chip's operating envelope of voltage versus speed versus temperature. From those an engineer can forsee how hot you can run a chip before its rise and fall time margins start to get marginal.
There is very little Intel can do to stretch thing by another 5 degrees. It's not something that can be imposed by fiat. Intel engineers have already juggled all the variables to come up with the best performance possible. SOMETHING is going to have to give. Either the chips will have to be selected and graded for speed, lowering the overall envelope for the chips everyone else gets, or they'll have to fudge some other parameters, hoping nobody will notice, or worse yet they'll tweak some variable right to the edge of raggedness, resulting in worse reliability down the road.
Lawyers and accountants generally don't know you can't have everything. let's hope the engineers educate them.
The story would be more believable if they did not get certain basic facts wrong. Mercury has a very low vapor pressure, it's not going to evaporate very quickly. That's why you don't see mercury fog inside a mercury switch or thermometer. The cost and weight of the mercury are inconsequential compared to the cost of the rocket to lift the telescope up there.
Talk about softballs, some of us would like the answers to these questions:
(1) How much did Microsoft have to pay you to switch from the side of the angels to their side?
(2) Who decided to make C# just enough like Java to fragment and muddle the whole programming scene?
(3) How can you consider C# to be a success when Microsoft tried using it to replace large parts of Windows and ended up throwing the whole mess away?
The numbers don't make any sense.
100 kilowatt hours over one year is a constant load of about 12 watts.
Air has about 0.03% CO2. Let's round that up to 0.1% or one part per thousand.
Now comes the hard part-- how much air can a 12-watt fan force through one square meter of CO2 filter material every second? Hard to say, let's estimate ten liters for starters. That would require an efficient fan and duct and a not very restrictive filter.
Air has a density of about 1.2 g/liter, so we're processing 12g/sec of air. Of that about 12mg is CO2. Per year that's about 380kg. Not too far off a ton, but not too close either.
Now we have to figure out how much energy it took to make the filter material. To capture a ton of CO2 takes about an equal amount of caustic calcium or lithium hydroxides. Unfortunately if we include the CO2 emitted to make a ton of calcium or sodium hydroxide, that's way beyond me, but it may be more than 380kg equivalent of CO2.
You have to look at all the costs and emissions involved in the whole cycle.
See, Ted was right. The tubes to DC clogged up somewhere in Tennessee. They had to hire moonshiner to swab them out.
looking at the PDF of their paper reveals some weird stuff.
After a lot of blabbing they admit the viscosity change is only 10%. Looking at the curves for Diesel fuel viscosity, that's equivalent to heating the fuel another 10 degrees C.
In their "tests" they used an injector pressure of 200PSI. Typical cars use 2,000 PSI and some of the newer Diesels use up to 22,000 PSI! Makes you wonder why they used such a low pressure.
Their real-world test was with a Mercedes Benz diesel engine hooked up to a dynamometer, but apparently running AT IDLE. A fuel consumption of 500 grams per hour. A power output of 1/3 horsepower or so. Does not sound like typical engine operating conditions.
I would be very wary of this device given the bizarre test conditions.
First problem, if electric fields reduced viscosity or surface tension, it would have been discovered like 100 years ago. Thousands of scientists have put stuff in electric fields. it's mighty unlikely none of them noticed this effect.
Secondly, there are totally free and reliable ways of reducing viscosity, like wrapping the fuel line around the exhaust manifold. Free, does not require any fancy electronics, and has the added benefit of actually working.
As a sometime seller,this isn't significant.
In about 1,200 sales, I've had exactly one bad check, and that was about six years ago.
I've had 99+ buyers paying by PayPal for the last four years or so, with no push from me.
So this change is not going to change things for most buyers or sellers. It does simplifiy things a bit.
It would be nice to have some answers to some very simple questions:
(1) How much will this thing weigh?
(2) How you gonna pay for the launch costs? At $5K a pound, that's a lot of bucks.
(3) What do you do about lightning strikes and winds?
(4) What are the chances you'll melt the ribbon with the lasers?
(5) What insurance company will insure this thing for liability and collision? You think a Ferrari is expensive to insure!
It's fun to think about a space elevator, for the first 45 seconds, then you get to the sticky bits.
oops, I forgot about the BTU to watts conversion. My bad. The numbers do kinda make sense then.
If they're paying ten cents a kilowatt-hour, that 10MW data center is paying about $9M/yr for power.
Cooling systems move about 15 times the power than what they draw. So the savings for a 10MW datacenter would be around $600K. Wonder how they came up with $2.9M ?
Read a little closer. They say the graphene sheets are conductive, so they're more like part of the plates. In any case, anything atom-thick is going to be problematic, whether it's plate or dielectric.
It's golly-gee wonderful if they can make a one-atom thick graphene sheet. Give them a lollipop.
But in making a capacitor, you need other attributes than just thinness. You need a capacitor plate that can carry current, remain in place in the face of strong electrostatic fields, be compatible with dielectrics, be reliable, and be manufacturable.
A one-atom thick sheet is not going to be able to do any of those things. Capacitor makers have been depositing thin electrodes for 60 years now. They know full well what the limits are. The limits are about four powers of ten higher than these neophytes are talking about.
>"unmanned flying drones can track targets even inside of buildings."
Very, very unlikely. If the drone is using passive sensors, it would have to be sensing some emanations from the buildings. Assuming the baddies are smart enough to not use cell-phones or WiFi, that leaves infrared. Adobe and brick are darn good insulators in the infrared.
If the drones are using active means, the baddies can use lo-tech sensors (the eyeball, Fuzzbuster radar detectors, binoculars) to spot them and act casual.
Maybe the writer meant that the drones have enough endurance to loiter in the area until they come back out into the open.
The TI invention was not what we would consider an "IC". It had components, but it wasnt practical to mass produce. Not even TI followed up on its development.
The real practical IC, with photoetched traces on a planar silicon substrate was developed by Fairchild.
transistor: 1948
Miniskirt: 1968
I know it was a while ago, but 20 yrs is still kinda large time span.
>I've seen micrographs of hard disc platters. The previously written bits are as clear as day.
We've all seen those pics. The pics are misleading.
the typical picture is of an erased area shaped into the letters "MIT" or "IBM". You can still see traces of bits under there, yeah. Useless. A real disk head never writes a string of zeros, that violates the basic principles of group code recording and DC balance. A real head would be writing new data with uncorrelated 1's and 0's, unsynced on the bit and amplitude levels of the previous data. Try reading the old bits THEN.
Read the original posting. They were talking about wardriving to collect tons of addresses, not spending many minutes driving in circles around a block to hone in on one specific point. That is of course doable in principle, if not practical. And even so, of what use is knowing that 123 Easy Street has a router named "linksys_2367443" ?
>Wifi OTOH using radio allows anyone in range to find out its address. Thats the problem.
What's the problem with knowing a MAC address?
The MAC is not a key to anything except sending a packet to the router. Which is the whole point of having a WiFi router.
Microsoft did not connect billions. They did not create TCP/IP, SMTP, the Web, or much of anything else.
They have ridden the wave with mediocre email apps and web browsers, but that's not much to crow about.
( And you would not have to crawl under a car to diagnose a blown head gasket, so there )
You don't need malware or anything else to get a router's MAC address, it's in every packet the router sends out.
And you can't easily get an exact street address from wardriving. All you know is somewhere along the antenna's main lobe there is a router. Could be 10 feet away, could be 500.
And knowing the MAC address is of no earthly use. Well, in the old days you could map it to a ethernet chip manufacturer, but now most routers have changeable MAC addresses.
You can't map MAC address to email addresses either, as the summary claims. Sheesh.
Eh, any idea how they'd cool the thing? It's fine to split atoms to make heat, but on the Moon you need to have a closed-loop cooling system. So you have to cool off the turbine exhaust so you can feed it back into the reactor. Problem-- no atmosphere and no lakes or rivers to carry away the heat. No groundwater either. Many many many meters of loose insulating moon-dust and rock fragments before you get down to bedrock, which in itself is not all that great at conducting away heat.
Methinks the Moon is not a great place to be running a reactor or power plant of the heat-cycle variety. Maybe solar cells.
Why would you need 24 contiguous hours of battery life?
Most of us sleep at least 8 hours out of every 24.
Uh, building a hybrid at home probably makes more pollution than making it in a factory.
The reason they sell it as a kit is to avoid all the federal vehicle rules. By passing on assembly to the end-user, it becomes THEIR problem to get the car licensed.
Also I don't quite get the "zero dollars to manufacture". Lots of the steps involve lots of time, welding, painting, trips to the hardware store. That all costs many $$$.