Well hardware has already improved, and the not 'static native code' applications are still memory/cpu hogs that behaves poorly even on new and powerfull machines.
The point is, no matter how good the hardware becomes, the software part can (and is) always pushed to the point of to waste every available resource. In that environment, JIT languages are displaced because they are a waste of resources, the hipotetical gain in programming is overcome by the penalty performance of the needed runtimes.
In short, we already know how to waste cpu cycles, no need of JIT languages for that, thanks.
Me too, konqueror has been my default browser for years. There are some websites that require firefox, that's true, but the integration with KDE, makes-it a little better IMHO. The test for me will be the switch to KDE 4, I don't like de file/web browsing separation idea, but i'll give it a try, at least they have provided a toggle to keep using konqueror as the default file browser.
Well, it seems that 'tautology' does apply but I am still clueless? Good logic my friend almost as good as: 'sampling an audio signal is a destructive transformation' is a false sentece.
And not a simple CD title (excuse me but I don't buy the 'there's mo way to be certain' attitude). have you ever listened to a good LP? Your renuence to declare a simple CD title looks like you are unsure about your 'audio quality' experience.
About 'fucking xxx' does not make your points stronger, at all, my dear AC. So if you like to be considered, just post under a name, and say to the audience: Hey! The 'insert CD title here' CD sound quality is better than any LP I've ever listened, is not that hard, also:).
So not a single example of a CD/DVD better than a good LP?. Well, that matches my experience, exactly. And yes, I do have been listenning digital studio equipment for some years now.
The link you provide seems to be is a single person test made in 1984. Interesting? Of course, but anecdotal without a proper sample size/methodology.
Now, about sampling I do know something about-it, and I am sorry, but the 'exact' reproduction of that 'continuos' wave you are comenting, is a tautonomy, that is, what Nyquist theorem says is that if you are build a wave with sinusoidal components below 22Khz you can reproduce-it exactly with a 44000 samples/sec setup. BUT that also implies that if the wave have sinusoidals components above 22Khz (as any not limited acoustic wave have) sampling at 44000 samples/secs will LOSE that higher than 22Khz frequencies, is not that hard.
Neil Young CD's are as good as they can, but listen to old Young LPs ('Everybody knows this is nowhere' comes to mind, but also a lot of 70-80's discs) for a better sound. Hell, Young has commented a lot of times how dificult is to make a just 'good enough' CD!.
1 Analog to digital is a by design destructive transformation. analog to analog does not have to be a destructive, it can be made destructive or not.
2 Now you says that 'exactly' equals 'good enough', well, that starts to make sense, you are not speaking accurately, just making a bunch of unproven assumptions, right. Then my answer is: if digital is BEYOND what the analog pàrts can notice, just show me a single CD that in your opinion has a quality level above good recorded vinil, I will buy-it and listen to it carefully, but after years of hearing music (I have hundreds of LPs and a no small numbers of CDs) I would be VERY HAPPY to learn that CDs can be made to sound as good as good LPs. That's not my dayly experience.
That fucking idiot, can be right or can be wrong, read the others post in this thread for a detailed explanation of my opinions (I just wont repeat posts here) but calling names, does not gives you reason, just makes you look bad dear AC.
Thanks for correcting my spelling mistakes, my engish is terrible, I know.
Now, two wrongs don't make a right.
1) It's a fallacy that we don't hear sounds above 22Khz. You can say that the average listener does not notice frequencies that high, but, and it's a big but, we 'hear' (feel the sound waves) with our whole body, especially with our sensorial organs and our brain. For example we have an strict filtering in our brains, that excludes a big bunch of bone vibrations indiced sounds, without that filtering, speaking or even eating would become a suffering experience. So accurately we must say that we don't 'hear' those frequencies, but they exist and interact (have measurable effects) with our bodies.
2) Nyquist is about discrete integration, as an approximation of a continuos (proximal to lineal between samples) function, that has nothing to do with continuos (no sampling based) processing.
So I am sorry but discrete systems destroys information by design. They can be made to be good enough for a given task, and of course analog system can be made worse than digital ones, but in essence they don't erase information, they just modify it.
In simple words, back to the origin of the thread (vinil encoding), if you are sampling at 44000 samples/sec and encoding at 24bps, you are (approx) encoding a vinil valley shaped surface 5 microns long and 50 microns wide in 16 millions possible values. Quite different from encoding an 'V' shaped line 50 microns wide in 16 million possible values as the original poster said.
1 by definition, sampling means to losse information.
2 true, analog can be as bad (or as good) as digital, the difference is that digital systems have an objective maximal information level, something that's not properly defined in analog systems.
3 Measurement does not equal sampling. Sampling (audio signals) is measuring the signal in fixed time slices. Physical laws doesn't work in discrete time slices, that's why analog processing can have a higher maximal information value.
You seems to be under the impession that because in principle we can model anything mathematically, the model is a replacement for (can be made indistinguishable from) reality, I am sorry but that's faith.
If you think that digital can be made as good as analog, great, then make a purely digital, I repeat purely digital musical instrument, my friend.
Still wrong, math is never against logic. Nyquist theorem just states the amplitude of the range of frequencies that can be encoded in a given sample rate.
1 That has nothing to do about analog signal processing.
2 Sampling by definition means to loose information, it shouldn't be that difficult.
Dots on a plastic surface (CD/DVD), can be read as 0/1, but the dots themselves are far more complex than that, they are macroscopic objects 'dumbed down' to only two possibilities 1/0. That's the digital approach, and that means a loss of information.
Vinil cannot hold the same amount of information as a CD/DVD if you treat the vinil as it were a CD/DVD, sure, but the whole point of vinil/needle vs CD/DVD/laser is that vinil does not convert the analog source information (music) to a dumbed down digitally sampled version.
I was pointing that comparing bit resolution to groove width does not cut, as there are other factors involved, for example: playing time is not discrete on vinil.
Don't be fooled by technological hype, digital means loss, that simple, and that's no 'mistery' is just a fact. What seems to be 'faith' is the unreasoned belief than digital is somehow magically equal to analog.
1 The physical needle movements are being ignored (the digital-analog converter is just applying smooth function, too simple to be representative of the behaviour of a physical object).
2 The continuos nature of a vinil playback, wibrations at time t add to vibrations at time t-1 and it is the physical integration of those variations that builds up the electic signal.
So nice try, but digital systems are based on a loss of information when they try to reproduce analog counterparts.
There's some programmer out there that doesn't think s/he is really good?
I think I am good too, of course, but after more than 30 years programming, the more lasting feeling is that we are just artisans, we do what we know, the better we can because we love what we do.
As someone said: to debug code is harder than to write it. So if you are coding at 100%, no matter how smart are you, you will always need a smarter than you guy to find out your errors. Now, if this smarter than you guy also codes at his 100%...
Tesla was the master of Magnetics and Resonance. That paper states just a 60W 40%efficient 2mts power transmission setup, sorry but for now MIT guys just look like amateurs.
The problem is that for real uses distance alone is not the sole/main factor, route quality, trafic conditions, horaries, etc are other factors to weight in if you really want to make or plan a real travel.
When I first learned C in the K&R era, I thought those greybeards were pretty clever. Then I came across strcpy() and I wondered what they were smoking that I wasn't sharing.
I also learn C is the K&R era, and honestly if you feel strcpy() is/was wrong, I thing you are not being logically coherent, if strcpy is 'bad', then memcpy should be the devil!:)
So no, C is just as bad as assembler is, and for me the real question is what were you smoking in those days?:)
There was a time where there was a war between rithm vs melody lovers. Now it seems the war is over (by now) rythm has won, let's go back to jungle, and forget those gentle sounds.
Just remember, the music you hear when kid will stay with you for all your life.
Slashdot Mirror
Well hardware has already improved, and the not 'static native code' applications are still memory/cpu hogs that behaves poorly even on new and powerfull machines.
The point is, no matter how good the hardware becomes, the software part can (and is) always pushed to the point of to waste every available resource. In that environment, JIT languages are displaced because they are a waste of resources, the hipotetical gain in programming is overcome by the penalty performance of the needed runtimes.
In short, we already know how to waste cpu cycles, no need of JIT languages for that, thanks.
A softer than expected inner layer was the base of the crustal shifting theory of Charles Hapgood in his Path of the Pole book.
The book and theories were prefaced/backed by Einstein, but it was rejected by geologists.
Maybe there was a seed of truth in Hapgood's work?
Millions of 'tasted' domains means extra processing for Google, I suppouse that's why they prefer to filter those domains, quite natural.
Wake me up when Google allows minimun CPC, that would be news.
Well, that's what happens when you put your private data on someone else's server. Nothing new here.
I guess that's why they call you ANONYMOUS COWARD. :)
Not just a 'technical nightmare' it's plain logically incoherent, methematically unsolvable. It's like programming done by politicians. :)
Me too, konqueror has been my default browser for years. There are some websites that require firefox, that's true, but the integration with KDE, makes-it a little better IMHO. The test for me will be the switch to KDE 4, I don't like de file/web browsing separation idea, but i'll give it a try, at least they have provided a toggle to keep using konqueror as the default file browser.
If it's a C/C++ project, start trying to understand the headers, after the docs/comments they are most descriptive part.
Well, it seems that 'tautology' does apply but I am still clueless? Good logic my friend almost as good as: 'sampling an audio signal is a destructive transformation' is a false sentece.
And not a simple CD title (excuse me but I don't buy the 'there's mo way to be certain' attitude). have you ever listened to a good LP? Your renuence to declare a simple CD title looks like you are unsure about your 'audio quality' experience.
About 'fucking xxx' does not make your points stronger, at all, my dear AC. So if you like to be considered, just post under a name, and say to the audience: Hey! The 'insert CD title here' CD sound quality is better than any LP I've ever listened, is not that hard, also :).
So not a single example of a CD/DVD better than a good LP?. Well, that matches my experience, exactly. And yes, I do have been listenning digital studio equipment for some years now.
The link you provide seems to be is a single person test made in 1984. Interesting? Of course, but anecdotal without a proper sample size/methodology.
Now, about sampling I do know something about-it, and I am sorry, but the 'exact' reproduction of that 'continuos' wave you are comenting, is a tautonomy, that is, what Nyquist theorem says is that if you are build a wave with sinusoidal components below 22Khz you can reproduce-it exactly with a 44000 samples/sec setup. BUT that also implies that if the wave have sinusoidals components above 22Khz (as any not limited acoustic wave have) sampling at 44000 samples/secs will LOSE that higher than 22Khz frequencies, is not that hard.
Neil Young CD's are as good as they can, but listen to old Young LPs ('Everybody knows this is nowhere' comes to mind, but also a lot of 70-80's discs) for a better sound. Hell, Young has commented a lot of times how dificult is to make a just 'good enough' CD!.
Wow, you almost get-it, but still wrong :)
1 Analog to digital is a by design destructive transformation. analog to analog does not have to be a destructive, it can be made destructive or not.
2 Now you says that 'exactly' equals 'good enough', well, that starts to make sense, you are not speaking accurately, just making a bunch of unproven assumptions, right. Then my answer is: if digital is BEYOND what the analog pàrts can notice, just show me a single CD that in your opinion has a quality level above good recorded vinil, I will buy-it and listen to it carefully, but after years of hearing music (I have hundreds of LPs and a no small numbers of CDs) I would be VERY HAPPY to learn that CDs can be made to sound as good as good LPs. That's not my dayly experience.
That fucking idiot, can be right or can be wrong, read the others post in this thread for a detailed explanation of my opinions (I just wont repeat posts here) but calling names, does not gives you reason, just makes you look bad dear AC.
Thanks for correcting my spelling mistakes, my engish is terrible, I know.
Now, two wrongs don't make a right.
1) It's a fallacy that we don't hear sounds above 22Khz. You can say that the average listener does not notice frequencies that high, but, and it's a big but, we 'hear' (feel the sound waves) with our whole body, especially with our sensorial organs and our brain. For example we have an strict filtering in our brains, that excludes a big bunch of bone vibrations indiced sounds, without that filtering, speaking or even eating would become a suffering experience. So accurately we must say that we don't 'hear' those frequencies, but they exist and interact (have measurable effects) with our bodies.
2) Nyquist is about discrete integration, as an approximation of a continuos (proximal to lineal between samples) function, that has nothing to do with continuos (no sampling based) processing.
So I am sorry but discrete systems destroys information by design. They can be made to be good enough for a given task, and of course analog system can be made worse than digital ones, but in essence they don't erase information, they just modify it.
In simple words, back to the origin of the thread (vinil encoding), if you are sampling at 44000 samples/sec and encoding at 24bps, you are (approx) encoding a vinil valley shaped surface 5 microns long and 50 microns wide in 16 millions possible values. Quite different from encoding an 'V' shaped line 50 microns wide in 16 million possible values as the original poster said.
Too many wrongs:
1 by definition, sampling means to losse information.
2 true, analog can be as bad (or as good) as digital, the difference is that digital systems have an objective maximal information level, something that's not properly defined in analog systems.
3 Measurement does not equal sampling. Sampling (audio signals) is measuring the signal in fixed time slices. Physical laws doesn't work in discrete time slices, that's why analog processing can have a higher maximal information value.
You seems to be under the impession that because in principle we can model anything mathematically, the model is a replacement for (can be made indistinguishable from) reality, I am sorry but that's faith.
If you think that digital can be made as good as analog, great, then make a purely digital, I repeat purely digital musical instrument, my friend.
Still wrong, math is never against logic. Nyquist theorem just states the amplitude of the range of frequencies that can be encoded in a given sample rate.
1 That has nothing to do about analog signal processing.
2 Sampling by definition means to loose information, it shouldn't be that difficult.
Wrong definition of 'information'.
Dots on a plastic surface (CD/DVD), can be read as 0/1, but the dots themselves are far more complex than that, they are macroscopic objects 'dumbed down' to only two possibilities 1/0. That's the digital approach, and that means a loss of information.
Vinil cannot hold the same amount of information as a CD/DVD if you treat the vinil as it were a CD/DVD, sure, but the whole point of vinil/needle vs CD/DVD/laser is that vinil does not convert the analog source information (music) to a dumbed down digitally sampled version.
I was pointing that comparing bit resolution to groove width does not cut, as there are other factors involved, for example: playing time is not discrete on vinil.
Don't be fooled by technological hype, digital means loss, that simple, and that's no 'mistery' is just a fact. What seems to be 'faith' is the unreasoned belief than digital is somehow magically equal to analog.
1 The physical needle movements are being ignored (the digital-analog converter is just applying smooth function, too simple to be representative of the behaviour of a physical object).
2 The continuos nature of a vinil playback, wibrations at time t add to vibrations at time t-1 and it is the physical integration of those variations that builds up the electic signal.
So nice try, but digital systems are based on a loss of information when they try to reproduce analog counterparts.
There's some programmer out there that doesn't think s/he is really good?
I think I am good too, of course, but after more than 30 years programming, the more lasting feeling is that we are just artisans, we do what we know, the better we can because we love what we do.
That's all
IT's already finished? or do you need 15 years more? :)
As someone said: to debug code is harder than to write it. So if you are coding at 100%, no matter how smart are you, you will always need a smarter than you guy to find out your errors. Now, if this smarter than you guy also codes at his 100%...
Unfortunatelly, experience shows that to be true.
Tesla was the master of Magnetics and Resonance. That paper states just a 60W 40%efficient 2mts power transmission setup, sorry but for now MIT guys just look like amateurs.
The problem is that for real uses distance alone is not the sole/main factor, route quality, trafic conditions, horaries, etc are other factors to weight in if you really want to make or plan a real travel.
I also learn C is the K&R era, and honestly if you feel strcpy() is/was wrong, I thing you are not being logically coherent, if strcpy is 'bad', then memcpy should be the devil! :)
So no, C is just as bad as assembler is, and for me the real question is what were you smoking in those days? :)
There was a time where there was a war between rithm vs melody lovers. Now it seems the war is over (by now) rythm has won, let's go back to jungle, and forget those gentle sounds.
Just remember, the music you hear when kid will stay with you for all your life.
Where's the probleming?