I can't tag because I'm on Links, but I think it should be tagged slownewsday. While knowing how many people use a given browser is only mildly interesting, except for people who have a website (but who might yet prefer to rely on their own browser usage statistics since they're more relevant of what their audience uses), there is little we can deduce from this 400M figure.
I think we should use that study to deny the right of vote to stupid republicans so that only us smart right-thinking liberals who know everything better than anyone else could decide.
In fact, I have talked to several people that knew people that had tumors for many, many years and never had any trouble, but after their doctors talked them into removing the tumors and doing radiation/chemo treatment, they were dead within a year.
You forgot your lessons.
Rule #1 : Correlation does not imply causation Rule #2 : The plural of 'anecdote' is not 'data'
Maybe we are using different definitions for that natural selection is. I am saying that natural selection is the process by which forms of life blah blah blah...
Well my definition for natural selection is "Natural selection is the process by which favorable traits that are heritable become more common in successive generations of a population of reproducing organisms, and unfavorable traits that are heritable become less common." (from Wikipedia). Therefore, the extinction of an entire species is not natural selection, in other words none of what we talked about here has anything to do with natural selection.
So you're saying that us killing elephants and such one by one until there's hardly any left is natural selection, but that killing Jews doesn't make it natural selection because, according to you, Jews are humans? (heh, I love the ambiguity of this question)
Hey! It's called natural selection, and it's been going on for millions of years.
Natural selection, damn right, more like mass extinction. Calling it natural selection would be like saying that the dinosaurs died due to natural selection. Plus, how is that natural selection when elephants are getting killed for their ivory? If that's natural selection than I guess genocides are natural selection too and so maybe jews and darfurians are unfit to live on Earth..
Or alternatively, find a job during which you can find the time to do anything you want on a computer, and once you're home and fed up with doing everything you like to do on a computer all day, spend all your time taking care of your girlfriend.
Or find a job that takes you to remote places so that the bulk of the interactions with your girlfriend are made online, with a good balance so that you get to meet her only enough for you to do anything you want with her and not too much so that you don't get bored of her.
Wait, are you trying to say that interacting with someone online and in real-life produces different experiences!? NO WAI! Does it also mean I must put clothes on, look presentable and not pick my nose when I'd hypothetically interact with people in the real world?
Depends. As I have noticed, online relationships' realness depends on how well they pass the test of time, and how well the relationship survives the shit it goes through.
Now that I come to think about it, it's the exact same thing in real-life relationships. Real-life one night stands or relationships that live no longer than a couple of weeks have little credibility.
All I have to say is, look at the open source successes in the realm of handheld consoles, namely with the GP32 and the GP2X. The thought that such a success could be possible with non-handheld consoles is what prompted me to write my previous comment.
I think these news might have different implications than we might suspect. While we may think "that's cool, although so few gamers are running Linux", I think this move might have other repercussions than just help the Linux PC game market.
In this day and age, we've got Open Source Anything, handheld consoles, cell phones, toasters, anything. Now if we imagine that some people somewhere decide to make a gaming console to rivalize with the Xbox 360 and the Wii, an Open Source Console, running Linux, or even some Open Source AppleTV-like box, which GPU will the makers choose? Obviously the most FOSS/Unix friendly, and that would be AMD/ATI.
They might be feeling that a large market might open up soon, and that's why I think they chose to do this move, while they can easily become the first ones there.
Very nice, but their flight sim is very basic, I think they should deal with someone who's done good in the flight sim business. My first thought is that guy who made X-Plane, I think they should hire some of his services to make Google Earth Flightsim a fully fledged flightsim that would compete with MS's flightsims
The USA haven't been into "extreme capitalism" all along either. From the 1940's to the early 1970's, the Keynesian model was the rule, thanks to FDR, and it worked.
How do you know the attack, release, and countless other settings involved?
Well like I said before I have never been interested in compression algorithms before and I wasn't even aware that such parameters were needed for compression, but let me tell you, it doesn't matter that much. We're not trying to make things be exactly the same as they were before evil record companies compressed the hell out of music, but only trying to get something that sounds better, and all it takes is, as I mentionned quite a few times earlier, to detect the average level in a track, and to set it to an average of our choice which would offer us a better dynamic range, by reverting a fairly generic compression algorithm.
You can see this as utilities that attempt to diminish artifacts on JPEG images, it surely doesn't restore the images to their pre-compression state, but it helps.
I really wasn't going to reply to this, but given the condescending tone and the fact that you got modded up...
I didn't get modded up, that's Karma Bonus you see there, you must be new here;-). You might understand where the condescending tone is coming from if you check out my sig, hehe.
To say you can undo the excessive compression/limiting used in the studio well by processing 16 bit 44.1 KHz audio is just wrong.
Far fetched at worst, but by no means plain wrong, well except for limiting, but at no point was it the topic, nor does the difference between 24-bit and 16-bit audio have anything to do with it. As for the loss of quality, I maintain that it shall be minimal, it's all about rounding "errors".
And again, even if it could be done in theory, what's the point? No listener has any way to do so using their existing equipment.
Unless we create a piece of software/equipment (I'd rather go for the software) that allows you to do this, which would be fairly trivial to make, algorithmically.
Absolutely...once you've crushed that peak to average level there's no getting it back.
In case you're talking about hard limiting, you're off topic, since it's not what compression is about (except for extreme compression, but no one actually does that for it's called amplification). Otherwise, you're just wrong, I explained why in a few other of my posts:-)
As for the computational requirements for real-time compression, they're fairly light and I think it could be done on an iPod in real-time. If I got the idea of how compression works correctly on an algorithmical level, it works on a per-sample basis, and, as the (B&W) iPod has two 80 MHz ARM7 CPUs, I think we could easily make such an algorithm work in less than 30 cycles per sample per CPU, which means I think it could easily run while taking barely 1% of the iPod's CPU time. That's how trivial it is.
As for automatic de-compression (also called expansion), you could easily have iTunes to profile each song and analyze how compressed it is (very simple, get the absolute value for every sample in a song, calculate the average, then calculate how much it is in dBFS), and then expand it in real time so that songs now have the same compression level, and in real-time, it's the exact same super-trivial algorithm you can use on the iPod.
So yes, we could very well have such a compression/decompression knob/software setting.
How do you propose to tell the difference between a particular sample level that got that way as a result of dynamic-range compression, versus one at the same level that accurately reflects the recorded source?
Well, I guess you mean, how would you in practice determine where the threshold level is? Well, I have never really looked into compression/expansion in sound, but provided that sounds are actually being compressed the way I described, then you could do an histogram of the sample values in the entire track and when you would see holes (values that samples in the track systematically avoid) then I guess you'd have spotted your threshold level. But I have little knowledge of how it's actually done in practice by mastering engineers, so I'm the wrong person to ask to.
As for the loss of precision, you don't lose a lot of precision by compressing a sound because compression is information-friendly, in a way. What I mean is that, the highest the sample value, the more tolerant it is to imprecision. That's the reason why, in early numeric telephone standards, sample values were not indexed linearly (constantly spaced) by rather logarithmically, so it would have more precision for soft sounds than for loud sounds, which is comparable to compression.
Anyways, as I proposed in another post, it doesn't matter so much to know exactly how it's been done, here's something that would give a satisfying result : As indicated in the article, the level of compression is indicated by its average level with respect to the peak level, expressed in dBFS. So you could just analyze a track for its average level, and say we found a value of -6 dBFS (which means its very compressed), we could expand it to more reasonable levels around -15 dBFS, and we could have an audio player that would make this systematical, for example.. Of course it wouldn't give you a result matching bit-wise to the original uncompressed track, but it would suit your needs in that it'd stop the sound from sounding like some compressed crap.:-)
Slashdot Mirror
I can't tag because I'm on Links, but I think it should be tagged slownewsday. While knowing how many people use a given browser is only mildly interesting, except for people who have a website (but who might yet prefer to rely on their own browser usage statistics since they're more relevant of what their audience uses), there is little we can deduce from this 400M figure.
Therefore, it's hardly newsworthy.
Wait crap, I posted as AC by mistake. I'm not really a coward :-/
Haha, mod that post insightful already.
I think we should use that study to deny the right of vote to stupid republicans so that only us smart right-thinking liberals who know everything better than anyone else could decide.
Which was the millionth article then? Not that it really matters, just being curious, cause I'm like, bored..
Dude, get a clue. And by a clue a mean a sense of humour. You're only making yourself sound like an idiot by taking such things seriously.
In fact, I have talked to several people that knew people that had tumors for many, many years and never had any trouble, but after their doctors talked them into removing the tumors and doing radiation/chemo treatment, they were dead within a year.
You forgot your lessons.
Rule #1 : Correlation does not imply causation
Rule #2 : The plural of 'anecdote' is not 'data'
Maybe we are using different definitions for that natural selection is. I am saying that natural selection is the process by which forms of life blah blah blah...
Well my definition for natural selection is "Natural selection is the process by which favorable traits that are heritable become more common in successive generations of a population of reproducing organisms, and unfavorable traits that are heritable become less common." (from Wikipedia). Therefore, the extinction of an entire species is not natural selection, in other words none of what we talked about here has anything to do with natural selection.
Can you explain to me where the ambiguity is.
"according to you, Jews are humans?"
So you're saying that us killing elephants and such one by one until there's hardly any left is natural selection, but that killing Jews doesn't make it natural selection because, according to you, Jews are humans? (heh, I love the ambiguity of this question)
Hey! It's called natural selection, and it's been going on for millions of years.
Natural selection, damn right, more like mass extinction. Calling it natural selection would be like saying that the dinosaurs died due to natural selection. Plus, how is that natural selection when elephants are getting killed for their ivory? If that's natural selection than I guess genocides are natural selection too and so maybe jews and darfurians are unfit to live on Earth..
Life moves on indeed!
shallow flings are WAY more fun in real life
Haha, yeah, true, well, I guess.. All of this is still purely theorical to me.
Or alternatively, find a job during which you can find the time to do anything you want on a computer, and once you're home and fed up with doing everything you like to do on a computer all day, spend all your time taking care of your girlfriend.
Or find a job that takes you to remote places so that the bulk of the interactions with your girlfriend are made online, with a good balance so that you get to meet her only enough for you to do anything you want with her and not too much so that you don't get bored of her.
Wait, are you trying to say that interacting with someone online and in real-life produces different experiences!? NO WAI! Does it also mean I must put clothes on, look presentable and not pick my nose when I'd hypothetically interact with people in the real world?
Mind boggling!
They are real alright.
Depends. As I have noticed, online relationships' realness depends on how well they pass the test of time, and how well the relationship survives the shit it goes through.
Now that I come to think about it, it's the exact same thing in real-life relationships. Real-life one night stands or relationships that live no longer than a couple of weeks have little credibility.
All I have to say is, look at the open source successes in the realm of handheld consoles, namely with the GP32 and the GP2X. The thought that such a success could be possible with non-handheld consoles is what prompted me to write my previous comment.
I think these news might have different implications than we might suspect. While we may think "that's cool, although so few gamers are running Linux", I think this move might have other repercussions than just help the Linux PC game market.
In this day and age, we've got Open Source Anything, handheld consoles, cell phones, toasters, anything. Now if we imagine that some people somewhere decide to make a gaming console to rivalize with the Xbox 360 and the Wii, an Open Source Console, running Linux, or even some Open Source AppleTV-like box, which GPU will the makers choose? Obviously the most FOSS/Unix friendly, and that would be AMD/ATI.
They might be feeling that a large market might open up soon, and that's why I think they chose to do this move, while they can easily become the first ones there.
Very nice, but their flight sim is very basic, I think they should deal with someone who's done good in the flight sim business. My first thought is that guy who made X-Plane, I think they should hire some of his services to make Google Earth Flightsim a fully fledged flightsim that would compete with MS's flightsims
I'm a private pilot, and the other simulators disappoint.
Even X-Plane?
The USA haven't been into "extreme capitalism" all along either. From the 1940's to the early 1970's, the Keynesian model was the rule, thanks to FDR, and it worked.
How do you know the attack, release, and countless other settings involved?
Well like I said before I have never been interested in compression algorithms before and I wasn't even aware that such parameters were needed for compression, but let me tell you, it doesn't matter that much. We're not trying to make things be exactly the same as they were before evil record companies compressed the hell out of music, but only trying to get something that sounds better, and all it takes is, as I mentionned quite a few times earlier, to detect the average level in a track, and to set it to an average of our choice which would offer us a better dynamic range, by reverting a fairly generic compression algorithm.
You can see this as utilities that attempt to diminish artifacts on JPEG images, it surely doesn't restore the images to their pre-compression state, but it helps.
I really wasn't going to reply to this, but given the condescending tone and the fact that you got modded up...
I didn't get modded up, that's Karma Bonus you see there, you must be new here ;-). You might understand where the condescending tone is coming from if you check out my sig, hehe.
To say you can undo the excessive compression/limiting used in the studio well by processing 16 bit 44.1 KHz audio is just wrong.
Far fetched at worst, but by no means plain wrong, well except for limiting, but at no point was it the topic, nor does the difference between 24-bit and 16-bit audio have anything to do with it. As for the loss of quality, I maintain that it shall be minimal, it's all about rounding "errors".
And again, even if it could be done in theory, what's the point? No listener has any way to do so using their existing equipment.
Unless we create a piece of software/equipment (I'd rather go for the software) that allows you to do this, which would be fairly trivial to make, algorithmically.
Without that, you're never undoing anything without serious loss of quality.
Sure, if by serious you mean unsignifyingly low.
Absolutely...once you've crushed that peak to average level there's no getting it back.
In case you're talking about hard limiting, you're off topic, since it's not what compression is about (except for extreme compression, but no one actually does that for it's called amplification). Otherwise, you're just wrong, I explained why in a few other of my posts :-)
As for the computational requirements for real-time compression, they're fairly light and I think it could be done on an iPod in real-time. If I got the idea of how compression works correctly on an algorithmical level, it works on a per-sample basis, and, as the (B&W) iPod has two 80 MHz ARM7 CPUs, I think we could easily make such an algorithm work in less than 30 cycles per sample per CPU, which means I think it could easily run while taking barely 1% of the iPod's CPU time. That's how trivial it is.
As for automatic de-compression (also called expansion), you could easily have iTunes to profile each song and analyze how compressed it is (very simple, get the absolute value for every sample in a song, calculate the average, then calculate how much it is in dBFS), and then expand it in real time so that songs now have the same compression level, and in real-time, it's the exact same super-trivial algorithm you can use on the iPod.
So yes, we could very well have such a compression/decompression knob/software setting.
How do you propose to tell the difference between a particular sample level that got that way as a result of dynamic-range compression, versus one at the same level that accurately reflects the recorded source?
Well, I guess you mean, how would you in practice determine where the threshold level is? Well, I have never really looked into compression/expansion in sound, but provided that sounds are actually being compressed the way I described, then you could do an histogram of the sample values in the entire track and when you would see holes (values that samples in the track systematically avoid) then I guess you'd have spotted your threshold level. But I have little knowledge of how it's actually done in practice by mastering engineers, so I'm the wrong person to ask to.
As for the loss of precision, you don't lose a lot of precision by compressing a sound because compression is information-friendly, in a way. What I mean is that, the highest the sample value, the more tolerant it is to imprecision. That's the reason why, in early numeric telephone standards, sample values were not indexed linearly (constantly spaced) by rather logarithmically, so it would have more precision for soft sounds than for loud sounds, which is comparable to compression.
Anyways, as I proposed in another post, it doesn't matter so much to know exactly how it's been done, here's something that would give a satisfying result : As indicated in the article, the level of compression is indicated by its average level with respect to the peak level, expressed in dBFS. So you could just analyze a track for its average level, and say we found a value of -6 dBFS (which means its very compressed), we could expand it to more reasonable levels around -15 dBFS, and we could have an audio player that would make this systematical, for example.. Of course it wouldn't give you a result matching bit-wise to the original uncompressed track, but it would suit your needs in that it'd stop the sound from sounding like some compressed crap. :-)
By changing every sample above that same threshold level to twice its level above the threshold level
I meant four times, of course.