Well, you're still calculating your significant digits with full precision. This is about many "significant" digits that only need to be precise in more than 99% of the calculations.
Choosing a smaller datatype is going to be less precise in 100% of the cases, inaccurate computing with a larger datatype is going to be less precise in 0.1% of the cases - AND faster. See the point?
What if good enough really IS good enough? In a fast paced game, would you trade a tenfold increase in graphics speed for having a tiny percentage of pixels being barely noticable off the correct color?
Repeating the process is not going to improve the results.
Suppose you repeat your calculation 3 times. How will you know that the result of comparing the three results with one another is correct?
What if the only answer you can obtain is "A equals B with a probability of 0.9998"? Recursively repeat this comparison and then compare the results?:)
If you bought a popular artist recently, your music is autotuned already.
Anyway, this means that less than 0.1 percent of your decoded audio samples will be less 0.1 percent off. This is probably very acceptable outside concert halls and living rooms if it delivers large bonuses in battery saving or calculation speed.
For example, we could use a much beefier compression algorithm than MP3 or current algorithms even longer on even smaller devices.
Even more so: we intentionally gather entropy to improve the pseudo random numbers. With intentionally inaccurate CPU cores, we could scrap all that and gather entropy en-passant AND be much faster anyway.
High accuracy is required for encoding music and video, though.
Maybe we could have a selective accuracy, where programmers can set their needs via registers or direct it to different CPU cores. Accurate cores for banking transactions, AV-stream encoding and 2D GUI operations while inaccurate cores are used for AV-stream decoding, and computer game 3d drawing and AI decisions.
There's a whole lot we are calculating now without the need for more than 3 significant digits - and a whole bunch where we intentionally use random numbers, sometimes even with strong hardware entropy gathering.
These are all cases where we could just scrap the accuracy for faster processing or longer battery times. No one cares about single bit errors in portable audio decoders or in high fps 3d gaming.
My thoughts exactly. Lighting embedded in the walls and in heights out of reach for the maximum security prison. TV sets and radio embedded and mounted out of reach the same way in minimum security or "bonus"-level prison for inmates who cooperate.
Well, technically, I don't use just "Windows", I use "Windows XP SP3 Build 5.1.2600" - does that matter? A bit. To end users? Not in the least...
No matter how you twist it, in the end "Linux" is the generic name for a family of operating systems, with their associated kernels, GUI models and support applications.
A sentence is semantically correct, when a large percentage of the intended audience understands it without blinking an eye.
"I am using Linux, Ubuntu Version 9.0 nightly 2 to be precise" is a sentence most high school literates would somehow understand. "Oh, he's not using Windows, not Apple, but Linux and this Ubuntu is probably one of the flavors or version of this family, much like Win ME or Win XP are of the Windows group"
This mnemonic association is permanent and trying to establish a different brand is probably not only futile but damaging to all "free POSIX-compatible OS implementations" which I reckon is the politically correct naming scheme for what everyone else just calls "Linux".
Educating everyone about the subtle differences between Ubuntu and Fedora is not only futile but could very well be dangerous to non-Windows-OS adoption rates and public acceptance. From an end user's perspective, the differences between Fedora and Ubuntu are much smaller than between the different flavors of Windows.
Even some hardcore OSS advocates talk about "The Kernel" these days and while Torvalds may be fighting tooth and nail for his trademark on this kernel, he has already lost in the same way that Hormel lost the trademark on their variant of Spiced Ham. After all, what would be the alternative in practical use? Billboard-sized compatibility lists for common hardware enumerating every flavor of Linux distros known to man?:)
No, ideally, they give the source code to a POSIX-compatible kernel and put a sticker with a Penguin and "Linux-compatible" on the box. Problem solved and everyone is happy.
Every minute you write about Linux being the kernel and not the OS, you waste one minute pointing out obvious advantages over, say, Windows OR actually writing new code and improving it.
People call the thing Linux and everyone understands what they mean, even you do, deep inside. They will understand the differences between Linux and Windows, but the would not begin to grasp what eg. FreeBSD is about.
The interface, capabilities and system structures are very similar and programs can usually be used across most Linux flavors. It is sensible to group these OS'es and Linux is a fitting and memorable name for that, now give this dead horse a rest, please.
By writing your post, you used intellectual property licensed to you by Uncle Sams elementary school. You now owe ol' Ms. Crumplebottom 10USD in licensing fees and 10000 in statutory damages.
An ant-like army of smallish, barely controllable passenger planes in the hands of a thousand civilians have nothing to do with the War on Terror, but everything with the War on Drugs.
We can barely manage the Mexican border on the GROUND. Imagine hundreds of speedy, low-flying, radar-evading vessels crossing the borders, landing on any empty stretch of road and then hiding in a plain old basement garage.
I for one welcome our controlled-substance trafficking overlords, that's for sure.
Wankel engines may not seize, but they can deform, leak oil and burn.
But as they guzzle gas like there's no tomorrow, they are unsuitable for most non life-threatening mission profiles. I mean, there's the Mazda R-8 with a very advanced Wankel engine but with at max 15mpg it eats into your wallet hard and fast.
With a favorable torque profile, extremely fast rpm changes and the mesmerizing jet engine sound, it has it's merits. But outside the race track or until a full tank of gas is 10$ again, it's just a silly boytoy.
Why would anyone use a Wankel engine in aviation? Gasoline piston engines consume enough fuel as it is...:)
The same plain old diesel your car uses. They adapted Volkswagen TDI engines for aircraft use, primarily for highly variable ambient pressure and a single gear.
Come thinking of it, I hardly know why we haven't used diesel engines before. They are well-known for highly reliable and economical operation under high torque loads.
Some truck engines can last for a million miles and some decades of continuous operation, regular professional maintenance provided. But they also last for quite some time even with incompetent drivers, irregular maintenance if any and constantly overloaded cargo compartment. Middle America and Africa has countless machines like this and they amazingly still work. So why do we have comparatively flimsy and guzzling gasoline piston engine in general aviation aircrafts, at all?
Well, the rules above ground are drafted for professionals with thousands of logged hours. We can either require that for everyone who wants to fly or we can draft simpler rules in height-confined airspace. Maybe we already have, in VFR flight levels, I don't know.
We don't talk about crossing a street in heavy traffic here. That's a pretty controlled and repeatable environment and if you still fail, we can recruit more volunteers to try again.
If someone actively messes up our climate, well, do we have a second planet to correct our errors?
Someone compared climate change theory to former theories of eugenics - which were viewed very positively on both sides of the Atlantic, until this Austrian with the ridiculous mustache came along and put the theory to enforced action.
I hope we can do better than that and recognize junk science this time before millions are dead.
So you say the news team - faked the whole scene - specifically selected footage fitting their agenda - manipulated footage to fit their agenda - specifically selected this location and event to film over other, more racially diverse acts of looting
Well, you have an inkling of proof, do you?
Assume all carjackings in your city are done by people with their hair dyed red. EVERY SINGLE INCIDENT. Would you call a statistic or a video report on this fact racist or realistic?
That's why they're searching everybody all the same right now.
Searching everyone and ignoring differential clues for not hurting feelings of intelligent and reasonable adults is pretty retarded.
Everyone has a nose. Not everyone is Muslim. Yet almost every terrorist is Muslim. So why should we ignore this strong pattern? Because of someone's sissy feelings?
"Well, here I have this delicious grilled medaillon of pork. To qualify for speedy check-in, you have to eat it."
Works for all ages and genders, except for Jews, Buddhists, Vegetarians and Vegans. Has a catch-22 side effect on Muslims: eat pork to get through security easy, but not to Heaven and the 72 virgins afterwards - or not eat pork and be security-checked and probably not get the virgins, either.
Win-win for pork eaters, hard times for innocent non-pork eaters mentioned above.
Great posts like yours are the reason why intellectuals of every kind are usually sent to re-education in socialist revolutions. You know too much and present hard facts, which ideologists avoid like the plague.
That's what I meant, exactly. If you see a clear pattern yielding clues to prevent further attacks, you have to use that pattern, wisely, of course.
You'd not even need to know the reasoning behind the connection of red hair and bombings, if it is by chance or essential part of a cultish behavior - it's just there and clearly identifiable.
If you only protect against redheads, you're vulnerable when all cult members dye their hair blue - you're protecting against the color of last week, so to say.
But having clues and doing nothing with them would be infinitely stupid. Doing nothing with these clues just for the reason to not hurt anyone's FEELINGS would be criminal negligent, though.
Dead people don't scream "racism!" but in my view life and limb trump feelings all the time.
But you'd be wasting costly resources protecting against very unlikely threats.
Anyway, I think we're all agreeing on the simple fact that it's impossible to identify the religion (or any other mindset) of all possible passengers. Orthodox Jews, Catholic Nuns, Shia Imams clearly stand out, but people in business attire could belong to any possible group and whose look would not yield any clues.
We should conclude that we don't do equal searches of all passengers to not hurt anyone's feelings but to protect against idiots from all sides. Strip searching even Buddhist monks would otherwise make no sense.
Slashdot Mirror
Well, you're still calculating your significant digits with full precision. This is about many "significant" digits that only need to be precise in more than 99% of the calculations.
Choosing a smaller datatype is going to be less precise in 100% of the cases, inaccurate computing with a larger datatype is going to be less precise in 0.1% of the cases - AND faster. See the point?
If you're a programmer: enjoy the speed boost if needed or program for the accurate-only CPU when lazy.
If you're a program user: you will not notice anything except raw speed, just like with dual or quad core systems now.
What if good enough really IS good enough? In a fast paced game, would you trade a tenfold increase in graphics speed for having a tiny percentage of pixels being barely noticable off the correct color?
http://de.wikipedia.org/wiki/GPGPU
Repeating the process is not going to improve the results.
Suppose you repeat your calculation 3 times. How will you know that the result of comparing the three results with one another is correct?
What if the only answer you can obtain is "A equals B with a probability of 0.9998"? Recursively repeat this comparison and then compare the results? :)
If you bought a popular artist recently, your music is autotuned already.
Anyway, this means that less than 0.1 percent of your decoded audio samples will be less 0.1 percent off. This is probably very acceptable outside concert halls and living rooms if it delivers large bonuses in battery saving or calculation speed.
For example, we could use a much beefier compression algorithm than MP3 or current algorithms even longer on even smaller devices.
Even more so: we intentionally gather entropy to improve the pseudo random numbers. With intentionally inaccurate CPU cores, we could scrap all that and gather entropy en-passant AND be much faster anyway.
High accuracy is required for encoding music and video, though.
Maybe we could have a selective accuracy, where programmers can set their needs via registers or direct it to different CPU cores. Accurate cores for banking transactions, AV-stream encoding and 2D GUI operations while inaccurate cores are used for AV-stream decoding, and computer game 3d drawing and AI decisions.
There's a whole lot we are calculating now without the need for more than 3 significant digits - and a whole bunch where we intentionally use random numbers, sometimes even with strong hardware entropy gathering.
These are all cases where we could just scrap the accuracy for faster processing or longer battery times. No one cares about single bit errors in portable audio decoders or in high fps 3d gaming.
My thoughts exactly. Lighting embedded in the walls and in heights out of reach for the maximum security prison. TV sets and radio embedded and mounted out of reach the same way in minimum security or "bonus"-level prison for inmates who cooperate.
No power sockets, though - no one needs them
Well, technically, I don't use just "Windows", I use "Windows XP SP3 Build 5.1.2600" - does that matter? A bit. To end users? Not in the least...
No matter how you twist it, in the end "Linux" is the generic name for a family of operating systems, with their associated kernels, GUI models and support applications.
A sentence is semantically correct, when a large percentage of the intended audience understands it without blinking an eye.
"I am using Linux, Ubuntu Version 9.0 nightly 2 to be precise" is a sentence most high school literates would somehow understand. "Oh, he's not using Windows, not Apple, but Linux and this Ubuntu is probably one of the flavors or version of this family, much like Win ME or Win XP are of the Windows group"
This mnemonic association is permanent and trying to establish a different brand is probably not only futile but damaging to all "free POSIX-compatible OS implementations" which I reckon is the politically correct naming scheme for what everyone else just calls "Linux".
Educating everyone about the subtle differences between Ubuntu and Fedora is not only futile but could very well be dangerous to non-Windows-OS adoption rates and public acceptance. From an end user's perspective, the differences between Fedora and Ubuntu are much smaller than between the different flavors of Windows.
Even some hardcore OSS advocates talk about "The Kernel" these days and while Torvalds may be fighting tooth and nail for his trademark on this kernel, he has already lost in the same way that Hormel lost the trademark on their variant of Spiced Ham. After all, what would be the alternative in practical use? Billboard-sized compatibility lists for common hardware enumerating every flavor of Linux distros known to man? :)
No, ideally, they give the source code to a POSIX-compatible kernel and put a sticker with a Penguin and "Linux-compatible" on the box. Problem solved and everyone is happy.
Every minute you write about Linux being the kernel and not the OS, you waste one minute pointing out obvious advantages over, say, Windows OR actually writing new code and improving it.
People call the thing Linux and everyone understands what they mean, even you do, deep inside. They will understand the differences between Linux and Windows, but the would not begin to grasp what eg. FreeBSD is about.
The interface, capabilities and system structures are very similar and programs can usually be used across most Linux flavors. It is sensible to group these OS'es and Linux is a fitting and memorable name for that, now give this dead horse a rest, please.
By writing your post, you used intellectual property licensed to you by Uncle Sams elementary school. You now owe ol' Ms. Crumplebottom 10USD in licensing fees and 10000 in statutory damages.
An ant-like army of smallish, barely controllable passenger planes in the hands of a thousand civilians have nothing to do with the War on Terror, but everything with the War on Drugs.
We can barely manage the Mexican border on the GROUND. Imagine hundreds of speedy, low-flying, radar-evading vessels crossing the borders, landing on any empty stretch of road and then hiding in a plain old basement garage.
I for one welcome our controlled-substance trafficking overlords, that's for sure.
Hummer H1 2007: 193 liters
Prius 2008: 45 liters
Hummer H1: 10mpg (rare max) = 4.25 km/l
Prius 2008: 45mpg (avg) = 19.1 km/l
Hummer H1 range: ~820km / 510mi
Prius 2008 range: ~860km / 535mi
Using SI units of France and Satan. Deal with it.
Wankel engines may not seize, but they can deform, leak oil and burn.
But as they guzzle gas like there's no tomorrow, they are unsuitable for most non life-threatening mission profiles. I mean, there's the Mazda R-8 with a very advanced Wankel engine but with at max 15mpg it eats into your wallet hard and fast.
With a favorable torque profile, extremely fast rpm changes and the mesmerizing jet engine sound, it has it's merits. But outside the race track or until a full tank of gas is 10$ again, it's just a silly boytoy.
Why would anyone use a Wankel engine in aviation? Gasoline piston engines consume enough fuel as it is... :)
That's why there's diesel powered piston aircraft engines now:
http://www.centurion-engines.com/
The same plain old diesel your car uses. They adapted Volkswagen TDI engines for aircraft use, primarily for highly variable ambient pressure and a single gear.
Come thinking of it, I hardly know why we haven't used diesel engines before. They are well-known for highly reliable and economical operation under high torque loads.
Some truck engines can last for a million miles and some decades of continuous operation, regular professional maintenance provided. But they also last for quite some time even with incompetent drivers, irregular maintenance if any and constantly overloaded cargo compartment. Middle America and Africa has countless machines like this and they amazingly still work. So why do we have comparatively flimsy and guzzling gasoline piston engine in general aviation aircrafts, at all?
Followed by rabid environmentalists demanding that you don't use your car for long trips or that you don't even travel that far.
They will then battle with the radical feminists claiming you subjugated a free womyn by forcing her to have kids.
Well, let's get some popcorn while we wait :)
Well, the rules above ground are drafted for professionals with thousands of logged hours. We can either require that for everyone who wants to fly or we can draft simpler rules in height-confined airspace. Maybe we already have, in VFR flight levels, I don't know.
Well, encode a 40kB keyfile then. AES-40000 should be enough for anyone :)
We don't talk about crossing a street in heavy traffic here. That's a pretty controlled and repeatable environment and if you still fail, we can recruit more volunteers to try again.
If someone actively messes up our climate, well, do we have a second planet to correct our errors?
Someone compared climate change theory to former theories of eugenics - which were viewed very positively on both sides of the Atlantic, until this Austrian with the ridiculous mustache came along and put the theory to enforced action.
I hope we can do better than that and recognize junk science this time before millions are dead.
So you say the news team
- faked the whole scene
- specifically selected footage fitting their agenda
- manipulated footage to fit their agenda
- specifically selected this location and event to film over other, more racially diverse acts of looting
Well, you have an inkling of proof, do you?
Assume all carjackings in your city are done by people with their hair dyed red. EVERY SINGLE INCIDENT. Would you call a statistic or a video report on this fact racist or realistic?
It's only slander if it's untrue, you know.
That's why they're searching everybody all the same right now.
Searching everyone and ignoring differential clues for not hurting feelings of intelligent and reasonable adults is pretty retarded.
Everyone has a nose. Not everyone is Muslim. Yet almost every terrorist is Muslim. So why should we ignore this strong pattern? Because of someone's sissy feelings?
Insightful, my ass.
"Well, here I have this delicious grilled medaillon of pork. To qualify for speedy check-in, you have to eat it."
Works for all ages and genders, except for Jews, Buddhists, Vegetarians and Vegans. Has a catch-22 side effect on Muslims: eat pork to get through security easy, but not to Heaven and the 72 virgins afterwards - or not eat pork and be security-checked and probably not get the virgins, either.
Win-win for pork eaters, hard times for innocent non-pork eaters mentioned above.
Great posts like yours are the reason why intellectuals of every kind are usually sent to re-education in socialist revolutions. You know too much and present hard facts, which ideologists avoid like the plague.
Thanks for this calculation.
That's what I meant, exactly. If you see a clear pattern yielding clues to prevent further attacks, you have to use that pattern, wisely, of course.
You'd not even need to know the reasoning behind the connection of red hair and bombings, if it is by chance or essential part of a cultish behavior - it's just there and clearly identifiable.
If you only protect against redheads, you're vulnerable when all cult members dye their hair blue - you're protecting against the color of last week, so to say.
But having clues and doing nothing with them would be infinitely stupid. Doing nothing with these clues just for the reason to not hurt anyone's FEELINGS would be criminal negligent, though.
Dead people don't scream "racism!" but in my view life and limb trump feelings all the time.
That's not a small risk, you're right.
But you'd be wasting costly resources protecting against very unlikely threats.
Anyway, I think we're all agreeing on the simple fact that it's impossible to identify the religion (or any other mindset) of all possible passengers. Orthodox Jews, Catholic Nuns, Shia Imams clearly stand out, but people in business attire could belong to any possible group and whose look would not yield any clues.
We should conclude that we don't do equal searches of all passengers to not hurt anyone's feelings but to protect against idiots from all sides. Strip searching even Buddhist monks would otherwise make no sense.