Naked Science Forum
Life Sciences => Plant Sciences, Zoology & Evolution => Topic started by: Huw Macdonald on 27/04/2010 01:30:03
-
Huw Macdonald asked the Naked Scientists:
Hi Chris and team.
I discovered your show (http://www.thenakedscientists.com/HTML/podcasts/) through the BBC Bitesize website last year and since then I have been really enjoying the programmes (http://www.thenakedscientists.com/HTML/podcasts/).
I keep quite a few pets around my house and often I like to play music on the radio or CD player, which is sometimes within earshot of the pets. I was wondering whether animals can be excited or affected by music in the same way that humans often are, or is it meaningless to them?.
Thanks
Huw
What do you think?
-
I think they like records and tapes, but digital music seems not to affect them.
-
I think they like records and tapes, but digital music seems not to affect them.
Really? That seems completely counter intuitive. Why would animals care what format the music is stored in? It's only vibrations in air to them.
-
Some dogs certainly seem to want to join in with some music i.e. they'll start to howl, but show no [other] signs of discomfort, when certain pieces of music are played to them. In view of the fact that the these dogs don't try to get away from the music, but just start 'accompanying' it I'd have to say that they're not howling because they find the music unpleasant. Whether they're getting pleasure from it, or whether they're just responding in some sort of instinctual fashion seems open to question though.
In view of this, it seems clear that some animals are affected by music.
I can't see what difference the source of the music makes either i.e. whether it comes from a digital or analogue source, because it has to be amplified by an analogue power amplifier and played through analogue loudspeakers to be heard.
-
Digital music recorded at 16 bits (standard CD) has only 0.39% of the information of the same music recorded at 24 bits. The CD standard was arrived at empirically as good enough for human ears. I read about one farmer who had a tape setup of classical music recorded from albums that he played to his cows and got 15% more milk. When CD's came along he replaced the analogue system and his milk production returned to the original amount. I think that its just that the analogue music on albums can contain more musical information depending on the skill of the engineers and the quality of the equipment. All of the original recording is almost always done on digital tape, but those are expensive machines recording at 24 bit or better. Most of those bits are lost when they pipe it down to 16 bit CD.
I haven't even touched on the fact that the original recording is usually done at a higher frame rate (samples per second) which is piped down to the CD standard, resulting in more orders of magnitude of lost information.
-
Digital music recorded at 16 bits (standard CD) has only 0.39% of the information of the same music recorded at 24 bits.
You might want to check that figure of 0.39% - are you really suggesting that more than 99% of the info recorded at 24bit is lost when down-sampled to 16bit?
The CD standard was arrived at empirically as good enough for human ears.
Yes, the 16bit CD standard exceeds the hearing range of most people.
I read about one farmer who had a tape setup of classical music recorded from albums that he played to his cows and got 15% more milk. When CD's came along he replaced the analogue system and his milk production returned to the original amount.
This story, as presented, is far too simplistic and I'm not sure what you're trying to imply by retelling it here. Is it that are cows more discerning than people when it comes to music reproduction? As I mentioned earlier, whatever the sound source, the music signal must still be amplified by analogue amplifiers and reproduced using analogue loudspeakers and unless these items are of audiophile quality (and cost) you'll be hard pressed to hear the difference. In fact, as the music was being reproduced from tape (almost certainly compact cassette as I can't see a farmer messing around with reel-to-reel tape) then switching from tape to CD will have certainly improved the resolution and quality of the sound.
I think that its just that the analogue music on albums can contain more musical information depending on the skill of the engineers and the quality of the equipment. All of the original recording is almost always done on digital tape, but those are expensive machines recording at 24 bit or better. Most of those bits are lost when they pipe it down to 16 bit CD.
While it's true that the resolution of vinyl can exceed that of 16bit digital audio, actually achieving the feat is far from easy and requires very expensive audiophile equipment, set up in a carefully controlled environment. For example, the moving-coil cartridges required to achieve these very high levels of resolution will only operate properly within limited ranges of temperature and humidity.
Also, "Most of those bits" are clearly not lost when the 24bit data is down-sampled to 16bit. 16 > 24/2, so not even half of the bits would be 'lost', even if the down-sampling were to be simply accomplished by discarding the least significant bits.
I haven't even touched on the fact that the original recording is usually done at a higher frame rate (samples per second) which is piped down to the CD standard, resulting in more orders of magnitude of lost information.
Well, actually, you just did exactly that, so not only is there not more lost information, the information that is lost is not of orders of magnitude in size.
While your point that analogue can exceed CD digital audio in resolution and quality is true, you're doing yourself no favours by exaggerating the truth to try to make that point.
-
Well, 2^16 = 65536 versus 2^24 = 16777216 so you throw away 16711680 bits with each sample. By your math, reducing your sample size to eight bits would result in half as good sound, while in actuality, eight bit sound is not listenable. The point is that animals may have more accurate hearing than us and so 16 bit sound may sound like eight bit to them. Visually, flies see at a higher sample frequency than us, so a TV movie looks like a series of still pictures to them.
-
... I read about one farmer who had a tape setup of classical music recorded from albums that he played to his cows and got 15% more milk. When CD's came along he replaced the analogue system and his milk production returned to the original amount.
If the change in milk yield was caused by the change in audio system it could have been the hiss from the analog sound system, not present on the digital, which affected the cows, May be relevant that white noise hiss does sound like water (river/rain).
-
Well, 2^16 = 65536 versus 2^24 = 16777216 so you throw away 16711680 bits with each sample.
When audio is down-sampled from a higher rate to a lower rate it is not simply a question of discarding the least significant bits, and even if it was it would still not result in a linear reduction in the fidelity of the reproduced waveform. Instead, I understand that the original waveform is virtually recreated and then re-sampled at the lower rate. But in any case, the higher sampling rates can only increase the resolution at the very highest frequencies. Have a read of: http://en.wikipedia.org/wiki/Sampling_rate#Sampling_theorem (http://en.wikipedia.org/wiki/Sampling_rate#Sampling_theorem)
By your math, reducing your sample size to eight bits would result in half as good sound, while in actuality, eight bit sound is not listenable.
I'm not sure how you would quantify sound as being 'half as good' but I have listened to a considerable amount of 8bit audio and while I'll admit that it is not of Hi-Fi quality, I would certainly describe it as being better than 'half' as good as 16bit audio. I think you might be confusing the bit depth with the sampling rate here. The bit depth of the samples only sets a limit on the resolution of the dynamic range, but doesn't stop you from sampling at a high rate and recording correspondingly high frequencies.
The point is that animals may have more accurate hearing than us and so 16 bit sound may sound like eight bit to them.
I'm afraid that suggesting something may be true proves nothing; it's a null argument and means nothing. Unless you have some evidence to support that assertion then that argument can be countered by simply saying that it may equally be untrue.
Visually, flies see at a higher sample frequency than us, so a TV movie looks like a series of still pictures to them.
Although flies can see at a higher rate than us, they have very low resolution vision, so a TV movie actually looks like a sequence of blurs to them. This analogy is actually quite close to the 8bit sound situation, except that there's more detail in 8bit sound than there is detail in the sight of a fly.
And after all that, there's still the issue of reproducing the sound through the analogue power amplifiers and loudspeakers, both of which will have limited bandwidth (or are you suggesting that the farmer had really kitted out his cowshed with audiophile amplifiers and loudspeakers, not to mention the careful suppression of any other resonant surfaces and objects in the cowshed?).
-
Apparently, cows have the second largest eardrums in the animal kingdom, plus they hear frequencies up to 35 Khz. So it's more reasonable to expect that they hear better than us than that they don't. If we could hear frequencies that high, I think we would find our electronic gizmos were noisier than we imagined. I doubt if there are any regulations prohibiting the production of frequencies too high for us to hear. So if there are ultrasonic noises put out by amplifiers or their inputs they may be filtered out or not. Probably the cheaper brands don't bother as the customer never complains. And cows don't vote. Or they vote with their teat. So they produce less milk.
-
So if there are ultrasonic noises put out by amplifiers or their inputs they may be filtered out or not. Probably the cheaper brands don't bother as the customer never complains.
I doubt it. The cheaper brands are more likely to lack the frequency response, and, even if they don't, they still have to drive some sort of audio tranducer (loudspeaker). Even if the electronics don't filter out the ultrasonic frequencies, the speakers will.
-
The electronics may not be designed to amplify or process ultrasonic frequencies, but that doesn't mean that they can't. We're talking about noise, not signal. The signal to noise ratio may drop off at higher frequencies, but that just means that the noise is getting amplified, not the signal. Modern tweeters shouldn't have any problem with arbitrarily high frequencies, they just aren't designed to give a flat frequency response to sounds we can't hear.
-
Modern tweeters shouldn't have any problem with arbitrarily high frequencies.
I don't think so. They might get as high as an octave above the threshold of human hearing, if you are lucky.
Ultrasonic transducers are rather specialized devices.
-
http://www.sonicelectronix.com/item_13189_MB+Quart+DSH+216.htmlhttp://
There are dozens of cheap car speaker systems on this site which are rated far into the ultrasonic.
-
http://www.sonicelectronix.com/item_13189_MB+Quart+DSH+216.htmlhttp://
There are dozens of cheap car speaker systems on this site which are rated far into the ultrasonic.
That's strange. None of the ones I looked at are specified beyond 20 kHz. Anyway, even if they go up another octave to 40 kHz, they are hardly producing "arbitrarily high frequencies".
-
40Khz is plenty of room to produce noise that is heard by cows.
-
I never said it wasn't. I was objecting to your claim that audio speakers can produce "arbitrarily high" frequencies. They don't.