---So it is truly fascinating to think of a computation that can be fully reversible / adiabatic. Is there no work being performed ? If you reverse the computation, are you still "allowed" to know the answer ?
My best estimate why they arent being worked upon is that they require different equipment a drastic change in software and engineering. That would explain the force against researching this.
---Goes to show how truly different information is, compared to matter and energy.
It goes to show that the laws of thermodynamics do play here too, but under the guise of entropy. Bits keep a small amount of heat. A destruction of a bit releases a very small amount of heat. However, this happening on a 1GHz chip in which timing loops can be multiple times that, and you have a very large amount of heat: in the factor of 10's of watts/s .
---What I still do not understand it whether this can be practically built, and if so, why do not we have it already ? Global warming and all of that ?
Well, I cannot see why we dont test it now. FPGA's are simply programmable gates, right? We can adhere ourselves to a rule only to use "passthru", "nor", "not" to do an operation. Then we can implement it in the traditional way. Measuring heat from this shouldnt prove too difficult. Doing this should prove beyond a reasonable doubt (along with it be repeatable) that it exists and can be made.
---Would it require radically different software? I would think so...
Absolutely. But there's no rule to call an instruction, say "and" and have it go through a series of reversible operations instead. If I'm reading the strong implications correctly, we could emulate all of our current (chipmaker)-64 instructions via those 3 instructions with appropriate 'glue' together. But then again, this probably is the most difficult.
---That sounds very interesting. Would you mind providing a link to the literature that discusses that ? I have some trouble figuring out the thermodynamics of this. Perpetum mobile and such, you know....
Of course. It, at first, sounds too good, but here you go.
Rolf Landauer showed in 1961 that reversible logic operations could be performed by neither using energy or taking heat out. The same could not be said for irreversible logic operations.
"Irreversibility and Heat Generation in the Computing Process" IBM Journal of Research Development 17 (1973): 525-32, IBM PDF
___
In 1973, Charles Bennett proved that any computation could be derived from purely reversible computing.
Charles H. Bennett "Logical Reversibility of Computation" IBM PDF
___
Later on, Fredkin and Toffoli presented a review of the ideas of reversible computing. The essential idea is that you can save all intermediary states between an algorithm to get the answer, and then reverse the process so that no energy is used, and generated no heat. Fredkin also indicates that if we switched from irreversible to reversible computing, we would expect to lose no more than 1% efficiency.
International Journal of Theoretical Physics 21 (1982):219-53 PDF
___
And as an unsubstantiated claim, I remember hearing that due to heat/radiation sources, that volatile memory gains errors of 1 bit per billion with a time from 1 minute to 1 day ( I forget the exact time). To correct this would only require the entropy of deleting that incorrect bit. In other words, 10^8 or so magnitude heat shrinkage. But trust the stuff above.
(Many of these ideas were taken from "The Singularity is Near" by Ray Kurzweil from page 130)
Keeping in scientific fact, how much heat has to be generated for 1 MIPS?
The fact is, absolutely none. It has been shown that only the destruction of information via AND and like instructions create entropy (heat). As long as you use only 3 types of gates (pass through, not, xor), you can create a heat-free CPU. Provided we do want to check for bit errors, we could maintain a very low heat via ECC like checking. Estimates on that are 10^8 lower than present.
We could keep 98% of our efficiency of current day chips if we switched to this method.
karlandtanya: Justice is whatever society says it is, so there's no problem here.
petrus4: Atheistic relativism is fun, isn't it kids?;)
That's democracy dude. Whatever has the majority is whatever's legal, hence acceptable. Not your "application to anti-religious beliefs". Who cares about religion when we talk politics.
---I hate to say it, but Microsoft were right for once in their earlier VISTA policy of locking down the practice of hooking into the kernel.
Locking down along with no source code is simply security by obscurity. There WILL be bugs found, and those bugs will have kernel rights. Do you think that is good? Guess what, I dont.
Vista will only reassure that bug releasers should not publish bugs, but rather sit on them. BTW, how do you clean out a kernel-infected Windows machine?
---It's that feature in XP that allows malware to flourish.
Is there an executable preventer on Linux? Nosiree, there's nothing preventing a user from affecting his own dataspace. What do you think is bad: Trashing the whole system, or trashing your ~ ? A system can be reinstalled, but most people dont back up their data.
Now, why dont Linux malwares work? They do, if the user lets them. It's just that much harder to make a program run from a browser window or from bad servers on various ports. Linux machines are usually more locked down to prevent evil stuff on the outside.
Backwards, it spelled the name of two developers: Boon and Tobias.
As a note, he make his first appearance as a super-hard hidden fight. You had to fight 50 times in 2-player on Mortal Kombat 2, and then you fought the Noob.
The noob looked like Scorpion with the ninja garb, but completely black. He was just a "shadow". he also could kick your ass super-quick.
---Some portion of the system has to be decrypted in order to boot. From that point, it's a matter of stepping through the hoops of each encryption barrier to get what you need.
Too true. I can get to it, so its theoretically able for you to get to it. My goal is to make sure you cant get to it within the statute of limitations.
---OR they just raid your trash for the evidence.
Nice that they want rotten eggs, toiletries, coffee grounds and the like. I have a burn barrel that gets all sensitive stuff. My mom has one too, for the very same reason.
---OR they put a key sniffer on the system
My computer doesnt have PS/2 ports. It uses USB, and syslog logs every USB insert and remove. I remove my MP3 and USB hard storage a lot, but I never remove my keyboard. My tripwire would catch that kind of activity.
---OR they correlate your network traffic to behavior patterns.
True. However, Most of my "secure work" could be done via a ip6 encrypted tunnel. Unless you could MitM me (not likely), you wont break that one. All you could see is data from my machine to another.. IP headers with no useful data is rather... useless. If its out of the states, good luck with those warranted searches. Then again.. Think TOR.
---OR you don't get caught. That is, in fact, a reasonable possibility.
Knowing what I know about the Police, not many know about data integrity or how to analyze a non-windows machine. Most of them have windows at home, and are trained on how to deal with a Windows-based attacker. Yes, they could hire a Unix-admin type for hundreds+$$ per hour and have them figure it out, but unless it's a serious matter, you'll just get off. And if they do attack the file system/computer, you can still get them by claiming that there's no certified tools for data integrity for this specific distribution working on this particular branch of linux and its filesystem.
Well, Matey-O, are there any breaks in an AES crypto-loopback of all of/home ?
If I was being careful, I'd also crypt/etc also. Good luck getting global configs and our home directories. The only problem is that I have to be here to mount/home and/etc.
I betcha it could be done in that a login triggers a user-mount of the cryptoloopback and unmount when the user is gone for 3+ minutes.
How would you break someone who is intelligent how computers run?
---This isn't about seeing the police side of this. This is about the legitimate derivation of power within a Constitutional Republic.
Wrong. Firing a device upon a car, when there is no search warrant, and no probable cause is a failure of my rights under the Constitution, as this is an unwarranted search.
---History is filled with dire examples of why it is best for the citizenry to disallow authority for the sake of political or legal ease. At the same time there are no lighthouse examples of why a well controlled government would be a Bad Thing.
Any monarchy, dictatorship, and communism are perfect lighthouse examples why we need a "Well Controlled Government". By control, I mean we, the people should control them, not the other way around.
You also probably think that those cell phone companies GIVE you free cell phones, right?
Guess what? Many models have subsidies in which one is given a product and receives another related (as in, you need) for money.
When I get a cell phone, I must pay for cell service through that provider. Or, when I buy computer equipment, I expect, no, DEMAND, that they provide drivers for at least the populous system. Right now, that system they provide for is Windows, but they failed even that.
---I do know how to deal with Windows issues. I've been using it for ten years, after all (in various flavors, of course). Is it easy? No, maybe not. Sometimes I just need to reinstall - which sucks, I'll admit - but the fact is, I know Windows. I'm familiar with it, and it does what I want it to do.
Back in the day (Win3.1, Win95, Win98) it was easy to handle Windows. I could pack up a Windows installation so it would fit on a zip drive undoublestacked. If I fried the ~/windows , I had a backup of just that, along with fresh drivers. I could solve it without going the fdisk/reinstall method.
Windows NT was the first start of the pain. You touch it wrong, and you get those damned "Inaccessable Boot Device" garbage. It doesnt matter if you copied everything over 100%, it doesnt run. And there's no intuitive way to fix it. At least in Linux if you have a inaccessable/, you know either the partitions not there, drive's acting up, or you used a kernel without your specific file system (had that a few times with early reiserfs).
Now, with XP, you need to ask daddy Bill permission to run your system if you changed something they think is crucial. Not Cool. Now, in the latest version, they use "STRONG CRYPTO" so you cant do diddly shit without being ordained as MS Minister of Information. And if you use YOUR computer in unforseen ways, MS will attack those very methods until you have a crippled machine, whether it be via the graphics or audio engines.... if you have drivers for them.
Im sorry, but Im a paying user, and the systems are now attacking me. It'd serve me better if I just used the latest crack on Piratebay, but who does that eventually serve?
---I have nothing against Linux - as a matter of fact I have a CentOS 4 web/smb/etc. server sitting behind me that I painstakingly built and configured as a sort of learning experiment - but I'll stick to Windows for my everyday use, thanks much.
There is another way you could have went. Try Debian Stable, and apt-get install samba webmin webmin-samba . Webmin provides a nice overview of your samba config via a secure session. Anyways, Ive found that serverwise, debian stable is pretty much the best in low maintenance and great uptime.
Ok...I've put the site up so that I don't get any annoying messages from the likes of Chris_S stating "We don't use NVidia's forums to collect legal information."
This site is intended to:
* Post your screen shots / box covers, etc...where Vista compatibility is stated outright or implied.
* Collect contact information about the class (done via your user account - email addresses are fine)
The issue is currently under review by one legal firm and a response is expected within three business days. I will attempt to work with two additional firms if necessary to have the case reviewed and will post findings here. In the event it is declined across the board, the site will be disabled.
Please be very careful about what you post - this isn't a place to rant - we've done that enough. This is a place to provide useful, constructive information conducive to supporting a legal action. While the content is not strictly moderated, spot reviews will be conducted at random and any posts contrary to this will be deleted - period. If the posts begin getting out of control, again, the site will be shut down.
Any repeated posts will be considered spam and the user will be a.) deleted from the site, b.) banned by IP address/block or c.) have their network operator contacted.
We've all come here for a reason - let's be constructive in our pursuit of this.
Admin's Note - Due to some of the posts being made, registration is being required in order to view the forums. I've already followed up with the ISPs corresponding to several of the abusive posts to have them blocked on the ISP side.
Do you understand the basic procedure Linux starts up all the way up to KDM/GDM/XDM login?
It's all in the config files and well documented (at least it is on Debian). I understand how these processes work, and how to fix them.
One major component it takes to do that is simple input/simple output. The simple part cannot be "graphical", as how do you decipher human-interfaced screens? Not smart, that move. Instead you use basic text and simple data streams to build larger structures. Those larger structures can build your guis and everything, as Linux has done.
Windows has went with the "bigger is better" in which it is a big box. You cant modify it too much as we dont fully understand how it works. And now with Vista, even most developers wont have an idea.
One idea is good for expanding onto while keeping understandability, yet another is good at keeping the users away from how it works and how to fix it. Guess which one I will use..
Slashdot Mirror
I guess the "troller" has never played half-life. Wow what an idiot.
When I look at the Wiki, I think of something 'a person' in the security field says time and time again:
"Any person can invent a security system so clever that she or he can't think of how to break it."
-Bruce Schneier
Id like to sick with just RUNNING Xen, not being in there.
---So it is truly fascinating to think of a computation that can be fully reversible / adiabatic. Is there no work being performed ? If you reverse the computation, are you still "allowed" to know the answer ?
My best estimate why they arent being worked upon is that they require different equipment a drastic change in software and engineering. That would explain the force against researching this.
---Goes to show how truly different information is, compared to matter and energy.
It goes to show that the laws of thermodynamics do play here too, but under the guise of entropy. Bits keep a small amount of heat. A destruction of a bit releases a very small amount of heat. However, this happening on a 1GHz chip in which timing loops can be multiple times that, and you have a very large amount of heat: in the factor of 10's of watts/s .
---What I still do not understand it whether this can be practically built, and if so, why do not we have it already ? Global warming and all of that ?
Well, I cannot see why we dont test it now. FPGA's are simply programmable gates, right? We can adhere ourselves to a rule only to use "passthru", "nor", "not" to do an operation. Then we can implement it in the traditional way. Measuring heat from this shouldnt prove too difficult. Doing this should prove beyond a reasonable doubt (along with it be repeatable) that it exists and can be made.
---Would it require radically different software? I would think so...
Absolutely. But there's no rule to call an instruction, say "and" and have it go through a series of reversible operations instead. If I'm reading the strong implications correctly, we could emulate all of our current (chipmaker)-64 instructions via those 3 instructions with appropriate 'glue' together. But then again, this probably is the most difficult.
---That sounds very interesting. Would you mind providing a link to the literature that discusses that ? I have some trouble figuring out the thermodynamics of this. Perpetum mobile and such, you know....
Of course. It, at first, sounds too good, but here you go.
Rolf Landauer showed in 1961 that reversible logic operations could be performed by neither using energy or taking heat out. The same could not be said for irreversible logic operations.
"Irreversibility and Heat Generation in the Computing Process" IBM Journal of Research Development 17 (1973): 525-32, IBM PDF
___
In 1973, Charles Bennett proved that any computation could be derived from purely reversible computing.
Charles H. Bennett "Logical Reversibility of Computation" IBM PDF
___
Later on, Fredkin and Toffoli presented a review of the ideas of reversible computing. The essential idea is that you can save all intermediary states between an algorithm to get the answer, and then reverse the process so that no energy is used, and generated no heat. Fredkin also indicates that if we switched from irreversible to reversible computing, we would expect to lose no more than 1% efficiency.
International Journal of Theoretical Physics 21 (1982):219-53 PDF
___
And as an unsubstantiated claim, I remember hearing that due to heat/radiation sources, that volatile memory gains errors of 1 bit per billion with a time from 1 minute to 1 day ( I forget the exact time). To correct this would only require the entropy of deleting that incorrect bit. In other words, 10^8 or so magnitude heat shrinkage. But trust the stuff above.
(Many of these ideas were taken from "The Singularity is Near" by Ray Kurzweil from page 130)
Keeping in scientific fact, how much heat has to be generated for 1 MIPS?
The fact is, absolutely none. It has been shown that only the destruction of information via AND and like instructions create entropy (heat). As long as you use only 3 types of gates (pass through, not, xor), you can create a heat-free CPU. Provided we do want to check for bit errors, we could maintain a very low heat via ECC like checking. Estimates on that are 10^8 lower than present.
We could keep 98% of our efficiency of current day chips if we switched to this method.
karlandtanya: Justice is whatever society says it is, so there's no problem here.
;)
petrus4: Atheistic relativism is fun, isn't it kids?
That's democracy dude. Whatever has the majority is whatever's legal, hence acceptable. Not your "application to anti-religious beliefs". Who cares about religion when we talk politics.
Biologically ??? Hell NO!! That means you end up with some weird-ass babies in blue-green-beige skin tones ala Capt'n Kirk style.
Me, I'd just prefer ANATOMICALLY compatible.
Wow. What a crap scam.
Doenst even LOOK legit. Even spammers and crapware dealers pay a few bucks for a pretty web presence.
Thats easily remedied by running the core of your program (encryption/obfuscation mechanism) within the CPU cache.
I dont know of many debuggers that pull from that ram.
I was intentionally leaving that out, as its a huge pain in the ass to set up and maintain.
I've only tried it once, and did a pretty bad job. Windows ACL's are only moderately better.
But they didnt lock the kernel down for the benefit of us users, they instead locked it down for a purely content driven media delivery tool.
That means we have even less access to their system. This applies to tinkerers AND system trashers.
I wonder what "premium content" spyware could do?
---I hate to say it, but Microsoft were right for once in their earlier VISTA policy of locking down the practice of hooking into the kernel.
Locking down along with no source code is simply security by obscurity. There WILL be bugs found, and those bugs will have kernel rights. Do you think that is good? Guess what, I dont.
Vista will only reassure that bug releasers should not publish bugs, but rather sit on them. BTW, how do you clean out a kernel-infected Windows machine?
---It's that feature in XP that allows malware to flourish.
Is there an executable preventer on Linux? Nosiree, there's nothing preventing a user from affecting his own dataspace. What do you think is bad: Trashing the whole system, or trashing your ~ ? A system can be reinstalled, but most people dont back up their data.
Now, why dont Linux malwares work? They do, if the user lets them. It's just that much harder to make a program run from a browser window or from bad servers on various ports. Linux machines are usually more locked down to prevent evil stuff on the outside.
Actually, that was "Noob Saibot".
Backwards, it spelled the name of two developers: Boon and Tobias.
As a note, he make his first appearance as a super-hard hidden fight. You had to fight 50 times in 2-player on Mortal Kombat 2, and then you fought the Noob.
The noob looked like Scorpion with the ninja garb, but completely black. He was just a "shadow". he also could kick your ass super-quick.
---Some portion of the system has to be decrypted in order to boot. From that point, it's a matter of stepping through the hoops of each encryption barrier to get what you need.
Too true. I can get to it, so its theoretically able for you to get to it. My goal is to make sure you cant get to it within the statute of limitations.
---OR they just raid your trash for the evidence.
Nice that they want rotten eggs, toiletries, coffee grounds and the like. I have a burn barrel that gets all sensitive stuff. My mom has one too, for the very same reason.
---OR they put a key sniffer on the system
My computer doesnt have PS/2 ports. It uses USB, and syslog logs every USB insert and remove. I remove my MP3 and USB hard storage a lot, but I never remove my keyboard. My tripwire would catch that kind of activity.
---OR they correlate your network traffic to behavior patterns.
True. However, Most of my "secure work" could be done via a ip6 encrypted tunnel. Unless you could MitM me (not likely), you wont break that one. All you could see is data from my machine to another.. IP headers with no useful data is rather... useless. If its out of the states, good luck with those warranted searches. Then again.. Think TOR.
---OR you don't get caught. That is, in fact, a reasonable possibility.
Knowing what I know about the Police, not many know about data integrity or how to analyze a non-windows machine. Most of them have windows at home, and are trained on how to deal with a Windows-based attacker. Yes, they could hire a Unix-admin type for hundreds+$$ per hour and have them figure it out, but unless it's a serious matter, you'll just get off. And if they do attack the file system/computer, you can still get them by claiming that there's no certified tools for data integrity for this specific distribution working on this particular branch of linux and its filesystem.
So technically, Windows is UNIX.
Scary thought.
Well, Matey-O, are there any breaks in an AES crypto-loopback of all of /home ?
/etc also. Good luck getting global configs and our home directories. The only problem is that I have to be here to mount /home and /etc.
If I was being careful, I'd also crypt
I betcha it could be done in that a login triggers a user-mount of the cryptoloopback and unmount when the user is gone for 3+ minutes.
How would you break someone who is intelligent how computers run?
---This isn't about seeing the police side of this. This is about the legitimate derivation of power within a Constitutional Republic.
Wrong. Firing a device upon a car, when there is no search warrant, and no probable cause is a failure of my rights under the Constitution, as this is an unwarranted search.
---History is filled with dire examples of why it is best for the citizenry to disallow authority for the sake of political or legal ease. At the same time there are no lighthouse examples of why a well controlled government would be a Bad Thing.
Any monarchy, dictatorship, and communism are perfect lighthouse examples why we need a "Well Controlled Government". By control, I mean we, the people should control them, not the other way around.
Its free..... Until you use it!!
Is that you Ian?
In other words, IT DOESNT WORK. Take the piece of shit back.
My god, please say that you put it on your credit card, so you can charge-back for bad hardware/software.
You also probably think that those cell phone companies GIVE you free cell phones, right?
Guess what? Many models have subsidies in which one is given a product and receives another related (as in, you need) for money.
When I get a cell phone, I must pay for cell service through that provider. Or, when I buy computer equipment, I expect, no, DEMAND, that they provide drivers for at least the populous system. Right now, that system they provide for is Windows, but they failed even that.
---I do know how to deal with Windows issues. I've been using it for ten years, after all (in various flavors, of course). Is it easy? No, maybe not. Sometimes I just need to reinstall - which sucks, I'll admit - but the fact is, I know Windows. I'm familiar with it, and it does what I want it to do.
/, you know either the partitions not there, drive's acting up, or you used a kernel without your specific file system (had that a few times with early reiserfs).
Back in the day (Win3.1, Win95, Win98) it was easy to handle Windows. I could pack up a Windows installation so it would fit on a zip drive undoublestacked. If I fried the ~/windows , I had a backup of just that, along with fresh drivers. I could solve it without going the fdisk/reinstall method.
Windows NT was the first start of the pain. You touch it wrong, and you get those damned "Inaccessable Boot Device" garbage. It doesnt matter if you copied everything over 100%, it doesnt run. And there's no intuitive way to fix it. At least in Linux if you have a inaccessable
Now, with XP, you need to ask daddy Bill permission to run your system if you changed something they think is crucial. Not Cool. Now, in the latest version, they use "STRONG CRYPTO" so you cant do diddly shit without being ordained as MS Minister of Information. And if you use YOUR computer in unforseen ways, MS will attack those very methods until you have a crippled machine, whether it be via the graphics or audio engines.... if you have drivers for them.
Im sorry, but Im a paying user, and the systems are now attacking me. It'd serve me better if I just used the latest crack on Piratebay, but who does that eventually serve?
---I have nothing against Linux - as a matter of fact I have a CentOS 4 web/smb/etc. server sitting behind me that I painstakingly built and configured as a sort of learning experiment - but I'll stick to Windows for my everyday use, thanks much.
There is another way you could have went. Try Debian Stable, and apt-get install samba webmin webmin-samba . Webmin provides a nice overview of your samba config via a secure session. Anyways, Ive found that serverwise, debian stable is pretty much the best in low maintenance and great uptime.
Text of website, as seen through .nyud.net:8090
_______
Ok...I've put the site up so that I don't get any annoying messages from the likes of Chris_S stating "We don't use NVidia's forums to collect legal information."
This site is intended to:
* Post your screen shots / box covers, etc...where Vista compatibility is stated outright or implied.
* Collect contact information about the class (done via your user account - email addresses are fine)
The issue is currently under review by one legal firm and a response is expected within three business days. I will attempt to work with two additional firms if necessary to have the case reviewed and will post findings here. In the event it is declined across the board, the site will be disabled.
Please be very careful about what you post - this isn't a place to rant - we've done that enough. This is a place to provide useful, constructive information conducive to supporting a legal action. While the content is not strictly moderated, spot reviews will be conducted at random and any posts contrary to this will be deleted - period. If the posts begin getting out of control, again, the site will be shut down.
Any repeated posts will be considered spam and the user will be a.) deleted from the site, b.) banned by IP address/block or c.) have their network operator contacted.
We've all come here for a reason - let's be constructive in our pursuit of this.
Admin's Note - Due to some of the posts being made, registration is being required in order to view the forums. I've already followed up with the ISPs corresponding to several of the abusive posts to have them blocked on the ISP side.
Do you understand the basic procedure Linux starts up all the way up to KDM/GDM/XDM login?
It's all in the config files and well documented (at least it is on Debian). I understand how these processes work, and how to fix them.
One major component it takes to do that is simple input/simple output. The simple part cannot be "graphical", as how do you decipher human-interfaced screens? Not smart, that move. Instead you use basic text and simple data streams to build larger structures. Those larger structures can build your guis and everything, as Linux has done.
Windows has went with the "bigger is better" in which it is a big box. You cant modify it too much as we dont fully understand how it works. And now with Vista, even most developers wont have an idea.
One idea is good for expanding onto while keeping understandability, yet another is good at keeping the users away from how it works and how to fix it. Guess which one I will use..