7 ms·

While I don't support Dell's decisions, digital (and analog) clipping can easily damage speakers driven at what is normally their maximum power they can handle,

by ferongr 13y ago

While I don't support Dell's decisions, digital (and analog) clipping can easily damage speakers driven at what is normally their maximum power they can handle, since the resulting clipped waveform carries a lot more energy than an unclipped one.

guard-of-terra 13y ago
Nobody prevents the manufacturing to put some electronic bits before the speakers to cap the maximum power. Even better, teach the sound card to do that.
- ferongr 13y ago
  Analog limiter ICs (that I know of) limit the signal according to input voltage, something that does not catch this issue. Furthermore, limiting power at the audio codec level would require for the codec to have that feature and I don't know of audio codecs with that capability.
  - guard-of-terra 13y ago
    Then one should figure out why nobody needed this feature. I'm no expert but maybe it's actually possible to create speaker that doesn't go bad. (I could code the protection into the sound driver in half a day probably)
    - TeMPOraL 13y ago
      > I'm no expert but maybe it's actually possible to create speaker that doesn't go bad. Of course it is, but nobody wants to do that; it's not profitable. * sigh * . I sometimes wonder how difficult would it be to create, say, a brand of kettles with lifetime warranty, designed to last 50+ years instead of 50+ weeks the ones we have do. How hard would it be to sell them and to what ends would the competition go to stop you from killing their market?
      - guard-of-terra 13y ago
        I never had either failing speaker or failing* kettle in my life so it's hard for me to say anything. Maybe there just so much you can get from laptop speakers so it makes sense to cap their power hard? * Some no loger turn off automatically, tho
        amenod 13y ago
        Kettle: I had one with the same problem. That's a terminal failure to me...
        SixSigma 13y ago
        > * Some no loger turn off automatically, tho And you don't think a kettle burning its way through your kitchen is a fail !
      - fnordfnordfnord 13y ago
        Dell isn't winning much on this. The service calls and now the reputation hit cost them more than they could make repairing laptops.
      - PhantomGremlin 13y ago
        Over on this side of the pond we generally use "coffee makers" instead of "kettles". But the problem is the same. They fail after a few years. It's all cheap consumer crap made in China. We've given up trying to buy "good" coffee makers. We just buy something cheap that's on sale and throw it away when it breaks. There should be the equivalent of "Gresham's Law" for consumer goods (the bad eliminates the good). But I don't know the name of that. Perhaps "China's Law" would be appropriate.
        laumars 13y ago
        We have coffee makers as well, but you can't make a cup of tea with a coffee machine. Nor make a cup-a-soup, pot/instant noodles, pre-boil water for cooking veg, hot water bottle, etc.
  - weland 13y ago
    Cheap shit. Seriously, cheap shit. There are several ways to solve this, all of which boil down to "use more chips" or "use better speakers", thus increasing the price of already incredibly shitty speakers. Source: I'm involved in designing consumer electronic products and the only corners that don't get cut are those that, if cut, result in certifications not being granted and therefore in products you're unable to sell. Everyone, Apple included, does it. The easiest way to solve this is use stronger materials for the speaker construction. This does, yes, affect the quality of the sound, which is already somewhere between catastrophic and abysmal on a laptop. The frequency response is likely to be worse, on a set of speakers that already has a frequency response that makes them quite useless for music. You can also do it in electronics, as measuring the power delivered to speakers is not exactly rocket science. The response time wouldn't be great, but it's continuously shredding the speakers with square signals that damages them, not a couple of pulses every once in a while. You can also detect heavy slopes. You can even detect heavy slopes in software. But no, seriously, this is a problem that can be solved. The mere fact that a lot of manufacturers manage to come up with speakers that don't break should be a testimony to this. I have (granted, desktop) speakers that have gone through a decade of heavy metal, grindcore and fucking SIDs and MODs, on bad ALSA drivers that I could barely get to work for years. They're fine. This is just Dell selling cheap shit.
    - etjossem 13y ago
      This is great information - and I'd love to know which speaker manufacturers you approve of. I'd rather own an expensive set of desktop speakers that lasts ten years than deal with Dell parts breaking all the time.
      - fnordfnordfnord 13y ago
        Get a pair of these. Don't pay more than $40 for the pair. They are quite decent, especially at this (retail) price point. Dayton Audio B652 6-1/2"
      - weland 13y ago
        > I'd love to know which speaker manufacturers you approve of I have no idea how things go in the laptop market, so I can't say there's one I approve of in particular. The fact that the ones on the laptops are crap can, however, be assessed quite easily by ear :).
- astral303 13y ago
  A square wave has twice the power of a sine wave of identical amplitude. This is why distorting the amplifiers (especially digital clipping) is so much worse for speakers than overpowering the speakers. This is a very well known fact in audio circles. Playing metal at full volume is not as damaging as playing anything intensely digitally clipped at full volume. Yes, Dell is putting on shitty speakers on their laptops, (what else is new), but VLC should be amplifying the output using level-limiting (hard-limiting), which would effectively bring all the quiet parts to be just as loud as the loud parts, instead of just digitally clipping the output.
  - KaiserPro 13y ago
    Its well known if you are over pumping your speakers. You shouldn't be able to blow up your speakers if your amp, speakers and limiters are setup correctly. After all, if you're busting out £20k on a set of decent speakers, you don't want to burn them out because your idiot producer "turned it up to 11"
  - guard-of-terra 13y ago
    I actually never seen VLC clip sound that much on volumes > 100%. Maybe if your speakers are very quiet and crappy and you have to get every control to 100% (hardware, system, vlc) in order to hear anything - maybe in this case it will. Hard clipping is very noticeable for the listener so not many people will set it to the maximum if they have choice. Anyway, you can protect from this either in hardware or in the driver.
  - fnordfnordfnord 13y ago
    >A square wave has twice the power of a sine wave of identical amplitude. 1.41 x >Playing metal at full volume is not as damaging as playing anything intensely digitally clipped at full volume. The output of the amplifier really ought to be bandwidth limited either as a natural consequence of the components used or explicitly with a filter. It is silly to spend power on things that can't be heard.
    - gamache 13y ago
svantana 13y ago
True, but VLC actually applies soft limiting when set to above 100%, so that hard clipping never occurs. I use this feature alot, it's great for taking down the usually thunderous sound effects in hollywood movies a few notches, while keeping quiet scenes at hearable levels.
tmoertel 13y ago
> ... the resulting clipped waveform carries a lot more energy than an unclipped one. Is this claim actually true? My understanding is that if you take a waveform and clip it, the resulting waveform actually carries less energy (think of the corresponding integral), but more of that energy is pushed into the higher frequencies. It's this – the unexpectedly large amount of high-frequency energy – that kills speakers because their crossover networks push it into the tiny, tiny tweeters, and they are utterly unprepared for it.
- lutusp 13y ago
  >> ... the resulting clipped waveform carries a lot more energy than an unclipped one. > Is this claim actually true? Yes, it is -- but it depends on how we define "clipped". > My understanding is that if you take a waveform and clip it, the resulting waveform actually carries less energy (think of the corresponding integral) Only if the clipping reduces the peak value. If you compare a sinewave with a peak value of 1, and a square wave with a peak value of 1, the square wave has a substantially higher average level (with a ratio of pi / 2). > It's this – the unexpectedly large amount of high-frequency energy – that kills speakers because their crossover networks push it into the tiny, tiny tweeters, and they are utterly unprepared for it. Yes -- the rate at which the speaker cones are required to move is an additional factor. But for a "clipping" definition that clips by means of trying to exceed the available voltage, these two effects add. http://i.imgur.com/oE5NFZ9.png http://i.imgur.com/oE5NFZ9.png In the above linked image, the red trace is sin(x), the integral for the interval 0 < x < pi is 2. The green trace produces an integral of pi. The ratio of the two is pi/2, and the speaker power difference is (pi/2)^2 = 2.46 (because the speaker's power is the square of the applied voltage). The green trace is what you would get if you simply turned up the volume beyond any reasonable setting -- the amplifier produces a clipped version of the sine wave and the peak value is equal to the supply voltage.
  - tmoertel 13y ago
    > Only if the clipping reduces the peak value... I don't know of any widely used definition of "clipping", nor any definition that fits the context of this discussion, that allows for a signal's peak value not to be reduced. It's called clipping because the extreme values look to have been clipped away, as if by scissors. The parts that have been clipped away contain energy, don't they? So won't the clipped signal will carry less energy than the original? When you take a sine wave of peak value 1 and clip it, what you get is not a square wave with a peak value of 1. Rather, you get a peak-truncated sine wave in which values in excess of the clipping threshold V are replaced with the clipping threshold. There's less power in this clipped sine wave than in the original because min(V, |sin t|) <= |sin t| for all t.

by ferongr 13y ago

guard-of-terra 13y ago

Nobody prevents the manufacturing to put some electronic bits before the speakers to cap the maximum power. Even better, teach the sound card to do that.

ferongr 13y ago
Analog limiter ICs (that I know of) limit the signal according to input voltage, something that does not catch this issue. Furthermore, limiting power at the audio codec level would require for the codec to have that feature and I don't know of audio codecs with that capability.
- guard-of-terra 13y ago
  Then one should figure out why nobody needed this feature. I'm no expert but maybe it's actually possible to create speaker that doesn't go bad. (I could code the protection into the sound driver in half a day probably)
  - TeMPOraL 13y ago
    > I'm no expert but maybe it's actually possible to create speaker that doesn't go bad. Of course it is, but nobody wants to do that; it's not profitable. * sigh * . I sometimes wonder how difficult would it be to create, say, a brand of kettles with lifetime warranty, designed to last 50+ years instead of 50+ weeks the ones we have do. How hard would it be to sell them and to what ends would the competition go to stop you from killing their market?
    - guard-of-terra 13y ago
      I never had either failing speaker or failing* kettle in my life so it's hard for me to say anything. Maybe there just so much you can get from laptop speakers so it makes sense to cap their power hard? * Some no loger turn off automatically, tho
      - amenod 13y ago
        Kettle: I had one with the same problem. That's a terminal failure to me...
      - SixSigma 13y ago
        > * Some no loger turn off automatically, tho And you don't think a kettle burning its way through your kitchen is a fail !
    - fnordfnordfnord 13y ago
      Dell isn't winning much on this. The service calls and now the reputation hit cost them more than they could make repairing laptops.
    - PhantomGremlin 13y ago
      Over on this side of the pond we generally use "coffee makers" instead of "kettles". But the problem is the same. They fail after a few years. It's all cheap consumer crap made in China. We've given up trying to buy "good" coffee makers. We just buy something cheap that's on sale and throw it away when it breaks. There should be the equivalent of "Gresham's Law" for consumer goods (the bad eliminates the good). But I don't know the name of that. Perhaps "China's Law" would be appropriate.
      - laumars 13y ago
        We have coffee makers as well, but you can't make a cup of tea with a coffee machine. Nor make a cup-a-soup, pot/instant noodles, pre-boil water for cooking veg, hot water bottle, etc.
- weland 13y ago
  Cheap shit. Seriously, cheap shit. There are several ways to solve this, all of which boil down to "use more chips" or "use better speakers", thus increasing the price of already incredibly shitty speakers. Source: I'm involved in designing consumer electronic products and the only corners that don't get cut are those that, if cut, result in certifications not being granted and therefore in products you're unable to sell. Everyone, Apple included, does it. The easiest way to solve this is use stronger materials for the speaker construction. This does, yes, affect the quality of the sound, which is already somewhere between catastrophic and abysmal on a laptop. The frequency response is likely to be worse, on a set of speakers that already has a frequency response that makes them quite useless for music. You can also do it in electronics, as measuring the power delivered to speakers is not exactly rocket science. The response time wouldn't be great, but it's continuously shredding the speakers with square signals that damages them, not a couple of pulses every once in a while. You can also detect heavy slopes. You can even detect heavy slopes in software. But no, seriously, this is a problem that can be solved. The mere fact that a lot of manufacturers manage to come up with speakers that don't break should be a testimony to this. I have (granted, desktop) speakers that have gone through a decade of heavy metal, grindcore and fucking SIDs and MODs, on bad ALSA drivers that I could barely get to work for years. They're fine. This is just Dell selling cheap shit.
  - etjossem 13y ago
    This is great information - and I'd love to know which speaker manufacturers you approve of. I'd rather own an expensive set of desktop speakers that lasts ten years than deal with Dell parts breaking all the time.
    - fnordfnordfnord 13y ago
      Get a pair of these. Don't pay more than $40 for the pair. They are quite decent, especially at this (retail) price point. Dayton Audio B652 6-1/2"
    - weland 13y ago
      > I'd love to know which speaker manufacturers you approve of I have no idea how things go in the laptop market, so I can't say there's one I approve of in particular. The fact that the ones on the laptops are crap can, however, be assessed quite easily by ear :).
astral303 13y ago
A square wave has twice the power of a sine wave of identical amplitude. This is why distorting the amplifiers (especially digital clipping) is so much worse for speakers than overpowering the speakers. This is a very well known fact in audio circles. Playing metal at full volume is not as damaging as playing anything intensely digitally clipped at full volume. Yes, Dell is putting on shitty speakers on their laptops, (what else is new), but VLC should be amplifying the output using level-limiting (hard-limiting), which would effectively bring all the quiet parts to be just as loud as the loud parts, instead of just digitally clipping the output.
- KaiserPro 13y ago
  Its well known if you are over pumping your speakers. You shouldn't be able to blow up your speakers if your amp, speakers and limiters are setup correctly. After all, if you're busting out £20k on a set of decent speakers, you don't want to burn them out because your idiot producer "turned it up to 11"
- guard-of-terra 13y ago
  I actually never seen VLC clip sound that much on volumes > 100%. Maybe if your speakers are very quiet and crappy and you have to get every control to 100% (hardware, system, vlc) in order to hear anything - maybe in this case it will. Hard clipping is very noticeable for the listener so not many people will set it to the maximum if they have choice. Anyway, you can protect from this either in hardware or in the driver.
- fnordfnordfnord 13y ago
  >A square wave has twice the power of a sine wave of identical amplitude. 1.41 x >Playing metal at full volume is not as damaging as playing anything intensely digitally clipped at full volume. The output of the amplifier really ought to be bandwidth limited either as a natural consequence of the components used or explicitly with a filter. It is silly to spend power on things that can't be heard.
  - gamache 13y ago

svantana 13y ago

True, but VLC actually applies soft limiting when set to above 100%, so that hard clipping never occurs. I use this feature alot, it's great for taking down the usually thunderous sound effects in hollywood movies a few notches, while keeping quiet scenes at hearable levels.

tmoertel 13y ago

> ... the resulting clipped waveform carries a lot more energy than an unclipped one. Is this claim actually true? My understanding is that if you take a waveform and clip it, the resulting waveform actually carries less energy (think of the corresponding integral), but more of that energy is pushed into the higher frequencies. It's this – the unexpectedly large amount of high-frequency energy – that kills speakers because their crossover networks push it into the tiny, tiny tweeters, and they are utterly unprepared for it.

lutusp 13y ago
>> ... the resulting clipped waveform carries a lot more energy than an unclipped one. > Is this claim actually true? Yes, it is -- but it depends on how we define "clipped". > My understanding is that if you take a waveform and clip it, the resulting waveform actually carries less energy (think of the corresponding integral) Only if the clipping reduces the peak value. If you compare a sinewave with a peak value of 1, and a square wave with a peak value of 1, the square wave has a substantially higher average level (with a ratio of pi / 2). > It's this – the unexpectedly large amount of high-frequency energy – that kills speakers because their crossover networks push it into the tiny, tiny tweeters, and they are utterly unprepared for it. Yes -- the rate at which the speaker cones are required to move is an additional factor. But for a "clipping" definition that clips by means of trying to exceed the available voltage, these two effects add. http://i.imgur.com/oE5NFZ9.png http://i.imgur.com/oE5NFZ9.png In the above linked image, the red trace is sin(x), the integral for the interval 0 < x < pi is 2. The green trace produces an integral of pi. The ratio of the two is pi/2, and the speaker power difference is (pi/2)^2 = 2.46 (because the speaker's power is the square of the applied voltage). The green trace is what you would get if you simply turned up the volume beyond any reasonable setting -- the amplifier produces a clipped version of the sine wave and the peak value is equal to the supply voltage.
- tmoertel 13y ago
  > Only if the clipping reduces the peak value... I don't know of any widely used definition of "clipping", nor any definition that fits the context of this discussion, that allows for a signal's peak value not to be reduced. It's called clipping because the extreme values look to have been clipped away, as if by scissors. The parts that have been clipped away contain energy, don't they? So won't the clipped signal will carry less energy than the original? When you take a sine wave of peak value 1 and clip it, what you get is not a square wave with a peak value of 1. Rather, you get a peak-truncated sine wave in which values in excess of the clipping threshold V are replaced with the clipping threshold. There's less power in this clipped sine wave than in the original because min(V, |sin t|) <= |sin t| for all t.