When your computer gors *crrrunch* . . . . . .

[aneeshm]

Today , in computer class , having completed my work , I thought I'd do something trivial and interesting . So I whipped together a simple program to simulate the sound of a siren ( setting the base frequency , variance , delay , and number of repetitions ) .

For the technically oriented ( if you're not interested , skip right to the end ) ( the code is old , I had to make it work on Turbo C++ ) :

Code attached to third post .

So when I ran this thing , with base frequency = 5000 , and variance 5000 , it effectively ran over the frequency range (5000 - 5000 =) 0 to 10000 (= 5000 + 5000) , I could hear a *crrunch* sound at some point .

It seems that the resonance frequency of weither the cabinets or the motherboard of the computer was in this range , and so it was going *crrrunch* when I ran this .

I'd like to ask if this could have damaged the computers seriously , and also if anyone else has such anecdotes to relate .

[aneeshm]

The code tags messed up my code ! WTF ?

[aneeshm]

Here is the attached file .

[Harovan]

Resonances can inflict mechanical damage to those parts of your computer, which are able to vibrate with a more or less constant frequency. Whether or not your computer is/was in danger is not readable from the code, as it is not exactly clear what the sound() and nosound() functions do, what device makes the sound and what is the distance from this device to the sensitive parts of the box; mind you the power of such effects tends to lessen with the square of the distance.

By the way, who in the world taught you to make each and every function parameter a reference, even if it's not necessary (and it is not necessary in any part of your code)?

[alva]

Sir Ralph sighting!!!!

[aneeshm]

sound(k) - produces a sound of frequency k

nosound() - stops sound

delay(k) - pauses for k milliseconds

The sound is produced by the system speaker - onboard , AFAIK . There is no distance between the speaker and the motherboard/cabinet . The speaker is a part of the motherboard , which is physically connected to the cabinet .

And I agree with you about the parameter part - only current should have been a reference , and even that should not have been passed at all , considering it is global . Gotta improve - remember , this is hacked up .

[Qilue]

While the best I could make of your code is it looked like a script for NWN and My expertise ended at that point. But, something did catch my eye.

So when I ran this thing , with base frequency = 5000 , and variance 5000 , it effectively ran over the frequency range (5000 - 5000 =) 0 to 10000 (= 5000 + 5000) , I could hear a *crrunch* sound at some point .

The part in bold is where the *crrunch* may be coming from. Audible sound for an average human is around 30hz-18000hz. Also, trying to make a sound chip reproduce 0hz may not be terriblely good for it. Did you try setting the variance at 4990 (10-9990). It would sound virtually the same without the sound chip trying to reproduce 0hz.

[aneeshm]

No , the part where the crunch came was from somewhere in the audible range .

Just makes me wonder . . . . .

Could you hardware-crash a computer just by making the system speaker emit the right frequency ?

[Harovan]

Originally posted by aneeshm
sound(k) - produces a sound of frequency k

nosound() - stops sound

delay(k) - pauses for k milliseconds

The sound is produced by the system speaker - onboard , AFAIK . There is no distance between the speaker and the motherboard/cabinet . The speaker is a part of the motherboard , which is physically connected to the cabinet.

The system speaker is of very low power and it is unlikely, that it is able to add up enough energy to break things, before the energy gain is neutralized by friction losses. You can, however, never be sure.

I don't know whether this is an hoax or not, but I heard of programs, that break the mechanical part of a computers harddisk just by letting the head switch tracks with an increasing frequency, to the point of resonance, where the sensible mechanics are endangered to break. I don't know if it is possible, but from a pure physical point of view I can imagine, that it is.

Oh, and hello alva!

[Qilue]

Originally posted by aneeshm
No , the part where the crunch came was from somewhere in the audible range .

Just makes me wonder . . . . .

Could you hardware-crash a computer just by making the system speaker emit the right frequency ?

A physics student would have understood.

Note this formula - v/f = Æ´

velocity / frequency = Wavelength

Thus using your data (I'm using a rough estimate of the speed of sound)

345m/s / 0hz = infinite wavelength

However you've decided it's a computer problem.

[Q Classic]

Aneeshm, computers are nothing but massive number-crunchers anyway...

You gave it some math, so...?

[aneeshm]

It seems you misunderstood my post . I AM a physics student , and yes , I know the formula velocity/frequency = wavelength . I will have to look up the velocity of sound in the appropriate metal ( I assume steel / painted iron ) , and calculate the resonance frequency based on that metal's elastic modulus and dimensions of the cabinet/motherboard ( I don't yet know which component exactly caused the sound ) . I rather dislike that remark about a physics student having understood , because I have been studying physics for more than a year now ( mechanics , thermodynamics , and waves and oscillations ) .

When I say hardware-crash , I mean cause mechanical damage capable of destroying the machine .

@ Sir Ralph

Yes , I doubt that the speaker would be powerful enough , but ideally it should not have been powerful enough to make the machine shake ( sort of ) when used .