I wasn't the AC, but here are some quotes from _Signals&Systems_. It's written by Oppenheim and Willsky. We use it in my Signals and Systems class.
"if a signal is band limited-- i.e., if its Fourier transform is zero outside a finite band of frequencies--and if the samples are taken sufficiently close together in relation to the highest frequency present in the signal, then the samples uniquely specify the signal, and we can reconstruct it perfectly." .... "The frequency 2wm (that's two omega sub m) which, under the sampling theorem (see previous paragraph), must be exceeded by the sampling frequency, is commonly referred to as the Nyquist rate." "The frequency wm (omega sub m) corresponding to one-half the Nyquist rate is often referred to as the Nyquist frequency."
What this says is basically: Any waveforms with frequencies lower than one half the sampling rate can be reconstructed perfectly (in theory).
I think you are over simplifying the operation of digital audio. If you only sampled a sine input at four times during one period of the wave, and then played "connect the dots" between those four dots, yes you would have a saw tooth wave. But it is far from this simple. I believe the foundation for digital audio comes from the fact that if you sample a sine wave twice during one period, you know _exactly_ what wave it was you were sampling. You can, in theory recreate a perfect sine wave from only two samples per cycle. CD players are more sophisticated than what you are describing.
My roommate is running the release version. He likes it pretty well for a desktop OS. I would be interested in knowing the "bug" count for windows 95,98, and NT4 were.
Win2k is a very able replacement for windows 98. I would guess that it is for NT4 also.
I am not knowledgable enough to judge W2K on server merits, but my roommate runs it as his desktop OS. Win2k does a lot of little things better than 95/98/NT4 in the interface. It also has been very stable for him afaik.
I'm not flaming. A limit on the number of AC postings? I don't get it. Do you understand the moderation/threshold system. ACs default to zero. People with accounts default to one. Set your threshold to one and *poof* the AC postings are gone, except for AC posts which have been moderated up.
I believe that the desired level of parnoia is in between the fbi-please-trample-my-rights and the twitching-holding-a-gun-in-the-corner level of paranoia.
Trusting too much can obviously cause problems. People take advantage of you, governments gain control, too much control. On the other hand, being paranoid can consume quite a bit of energy and be counter-productive.
That being said, I remember studying the US revolution in school and thinking that the colonists were sometimes excessively paranoid, however I could never fault the result. Anyway, I hope that no one here would blindly trust the fbi, without even considering that they may not be looking out for your best interests.
Remember kids: rational fear == good irrational fear == bad
"I trust them more than I trust the god damned bastards who run the local PD where I live. Believe the level of shall we say improper conduct is a lot greater at the local level."
It makes perfect sense, the more power you have, the less corrupt you are.
You have 30-40 cds but only two computers. Isn't this a 20 times scaling factor in the wrong direction? You are talking about a factor which would say basically that for every copy purchased there are 10 or 20 copies downloaded. Which is probably true. But isn't it more useful to actually count the number of computers with linux installed and running. Counting the number of times cheap bytes sells a cd doesn't make for a good count of what is actually being run.
TV's do have _some_ advantages over monitors. I believe what the original poster was referring to is aliasing. TV's don't really have aliasing problems like monitors do. Of course for computer-like work, a TV does suck.
Purdue is similar to this in the Engineering Computing Network. We have some Ultra 10s, some NT machines. A couple years ago this is all we had. Linux is creeping in, thought. We are making a lot of the NT machines dual bootable, adding Red Hat. I do have one course, a Korn Shell/Perl programming class that is running Red Hat in the lab.
On a side note, I have a newbie question. What's up with Korn Shell on linux? I know there is pdksh, but I've heard that it is not a perfect clone of ksh, and therefore not a good idea for us to use at home in this class.
Here at Purdue, all of the win95 (maybe win98 now???) machines have software that basically syncs the hard disc in your machine, with an image they have of 'how the drive is supposed to look.'
In addition, they also have software that controls your login (you can only log in with your username and password), and automatically mounts your unix shell account onto your own desktop, so you can save your files securely and retrieve them anywhere.
I wouldn't say suing overclockers was an example of AMD being bastard-like. They didn't want people buying their products after being modified. They were concerned about consumers buying overclocked processors and coming to AMD for support when the warranty had already been voided. AMD has very clearly stated that they do not care what you do with your processor after you buy it.
Actually, I thought Intel used to not announce a product before release. I think the purpose was to not rob from it's current sales. i.e, people would put off buying a pentium mmx system because the pentium II was going to be released soon.
Think back to the days when we were trying to find out more about the Deschutes (I think), the first P2 chips. I think Intel was pretty quiet about P2 before release. Of course I could be on crack.
I'm by no means an expert, but I believe the differentiation between on die L1 and L2 cache comes from different latency values. I don't know what the exact values are. But L1 is as fast a memory access as that processor is going to see. When L2 is on die, the processor will still be able to access it very quickly, but it might only be one half or one fourth as fast as the L1 cache (pulling numbers out of my ass). As I understand it, the trade off is in the number of transistors. Again, pulling numbers out of my ass, but the integrated L1 cache has the highest number of transistors per bit stored, say maybe 16 transistors (not a real number). While at the same time the L2 cache may only have 4 transistors per bit, allowing a much bigger, but slower secondary level of cache.
Warning: I pulled all numbers in this post out of my ass, please do not repeat them or accept them (or my ass) as fact.
it's funny how many people like to bitch at people who post redundant links. Making lots of assumptions as to the poster's intentions. Perhaps it is more revealing of their own motivation.
Slashdot Mirror
You are confusing my Caustic Puppy!
No replies to any of my posts
I wasn't the AC, but here are some quotes from _Signals&Systems_. It's written by Oppenheim and Willsky. We use it in my Signals and Systems class.
"if a signal is band limited-- i.e., if its Fourier transform is zero outside a finite band of frequencies--and if the samples are taken sufficiently close together in relation to the highest frequency present in the signal, then the samples uniquely specify the signal, and we can reconstruct it perfectly."
....
"The frequency 2wm (that's two omega sub m) which, under the sampling theorem (see previous paragraph), must be exceeded by the sampling frequency, is commonly referred to as the Nyquist rate."
"The frequency wm (omega sub m) corresponding to one-half the Nyquist rate is often referred to as the Nyquist frequency."
What this says is basically: Any waveforms with frequencies lower than one half the sampling rate can be reconstructed perfectly (in theory).
I think you are over simplifying the operation of digital audio. If you only sampled a sine input at four times during one period of the wave, and then played "connect the dots" between those four dots, yes you would have a saw tooth wave. But it is far from this simple. I believe the foundation for digital audio comes from the fact that if you sample a sine wave twice during one period, you know _exactly_ what wave it was you were sampling. You can, in theory recreate a perfect sine wave from only two samples per cycle. CD players are more sophisticated than what you are describing.
That .1% chance of an error is per line of code executing. Let say for the sake of argument that:
one line of code = ten machine level instructions
one machine level instruction = 2 clock cycles
500 million clock cycles per second
That means that there will be about 25,000 errors during one second of operation.
Probably not the best way to judge the significance of errors.
My roommate is running the release version. He likes it pretty well for a desktop OS. I would be interested in knowing the "bug" count for windows 95,98, and NT4 were.
Win2k is a very able replacement for windows 98. I would guess that it is for NT4 also.
I am not knowledgable enough to judge W2K on server merits, but my roommate runs it as his desktop OS. Win2k does a lot of little things better than 95/98/NT4 in the interface. It also has been very stable for him afaik.
I get:
Fatal error: Cannot use [] for reading in utils.php on line 287
I'm not flaming. A limit on the number of AC postings? I don't get it. Do you understand the moderation/threshold system. ACs default to zero. People with accounts default to one. Set your threshold to one and *poof* the AC postings are gone, except for AC posts which have been moderated up.
That last remark was dumb. He had, imo, a valid point to make there. He could have made it without coming off like a jackass.
I believe that the desired level of parnoia is in between the fbi-please-trample-my-rights and the twitching-holding-a-gun-in-the-corner level of paranoia.
Trusting too much can obviously cause problems. People take advantage of you, governments gain control, too much control. On the other hand, being paranoid can consume quite a bit of energy and be counter-productive.
That being said, I remember studying the US revolution in school and thinking that the colonists were sometimes excessively paranoid, however I could never fault the result. Anyway, I hope that no one here would blindly trust the fbi, without even considering that they may not be looking out for your best interests.
Remember kids:
rational fear == good
irrational fear == bad
"I trust them more than I trust the god damned bastards who run the local PD where I live. Believe the level of shall we say improper conduct is a lot greater at the local level."
It makes perfect sense, the more power you have, the less corrupt you are.
You have 30-40 cds but only two computers. Isn't this a 20 times scaling factor in the wrong direction? You are talking about a factor which would say basically that for every copy purchased there are 10 or 20 copies downloaded. Which is probably true. But isn't it more useful to actually count the number of computers with linux installed and running. Counting the number of times cheap bytes sells a cd doesn't make for a good count of what is actually being run.
TV's do have _some_ advantages over monitors. I believe what the original poster was referring to is aliasing. TV's don't really have aliasing problems like monitors do. Of course for computer-like work, a TV does suck.
Purdue is similar to this in the Engineering Computing Network. We have some Ultra 10s, some NT machines. A couple years ago this is all we had. Linux is creeping in, thought. We are making a lot of the NT machines dual bootable, adding Red Hat. I do have one course, a Korn Shell/Perl programming class that is running Red Hat in the lab.
On a side note, I have a newbie question. What's up with Korn Shell on linux? I know there is pdksh, but I've heard that it is not a perfect clone of ksh, and therefore not a good idea for us to use at home in this class.
Here at Purdue, all of the win95 (maybe win98 now???) machines have software that basically syncs the hard disc in your machine, with an image they have of 'how the drive is supposed to look.'
In addition, they also have software that controls your login (you can only log in with your username and password), and automatically mounts your unix shell account onto your own desktop, so you can save your files securely and retrieve them anywhere.
I just accessed all of those sites from my home computer. Didn't take more than a few seconds for any of the websites to load (over adsl).
Cool, thanks for the information.
I wouldn't say suing overclockers was an example of AMD being bastard-like. They didn't want people buying their products after being modified. They were concerned about consumers buying overclocked processors and coming to AMD for support when the warranty had already been voided. AMD has very clearly stated that they do not care what you do with your processor after you buy it.
Actually, I thought Intel used to not announce a product before release. I think the purpose was to not rob from it's current sales. i.e, people would put off buying a pentium mmx system because the pentium II was going to be released soon.
Think back to the days when we were trying to find out more about the Deschutes (I think), the first P2 chips. I think Intel was pretty quiet about P2 before release. Of course I could be on crack.
I'm by no means an expert, but I believe the differentiation between on die L1 and L2 cache comes from different latency values. I don't know what the exact values are. But L1 is as fast a memory access as that processor is going to see. When L2 is on die, the processor will still be able to access it very quickly, but it might only be one half or one fourth as fast as the L1 cache (pulling numbers out of my ass). As I understand it, the trade off is in the number of transistors. Again, pulling numbers out of my ass, but the integrated L1 cache has the highest number of transistors per bit stored, say maybe 16 transistors (not a real number). While at the same time the L2 cache may only have 4 transistors per bit, allowing a much bigger, but slower secondary level of cache.
Warning: I pulled all numbers in this post out of my ass, please do not repeat them or accept them (or my ass) as fact.
Or he turns the monitor off, but leaves the computer on, thinking the whole thing is off
Am I on crack, or did the parent of this post have html tags that weren't rendered, just displayed?
I've notices this a lot lately. Maybe everyone else has flashed their brain so they can autmatically convert tags on the fly and just see the italics.
it's funny how many people like to bitch at people who post redundant links. Making lots of assumptions as to the poster's intentions. Perhaps it is more revealing of their own motivation.
A screen of static at start up????? I don't get it. Your display runs through the bios on bootup right? or does it just go from nothing to linux?
How many times has microsoft been sued by users?
Because of heat, power, the supporting infrastructure? why?
.18 micron process.
With the newer and smaller processes, I would guess that intel could make a pretty low power pentium (not P2, the original) on a