What type of sound is the most complex to reproduce faithfully?
Jul 13, 2010 at 6:49 PM Post #17 of 35
Good point.  I think digital sources really have the most trouble with this, whereas true analog, like vinyl and reel-to-reel tapes are better at it.
 
Quote:
traditional chinese music has all those very clear/percussive and bright sounds that are very hard to reproduce 



 
Jul 13, 2010 at 10:06 PM Post #18 of 35
Not exactly a musical instrument, but: a few years ago I started an engineering project that aimed to use acoustic analysis to detect operational abnormalities in small jet engines. It was mostly for "fun" but I learned a lot about sound and its behaviour - the way air molecules act when they're being slapped around by bits of metal spinning at 10 000rpm is pretty damn fascinating. There was a boatload of interaction and harmonic activity.
 
Reproducing these sounds accurately proved to be extraordinarily difficult to say the least, but it was an enlightening experience.
 
Jul 13, 2010 at 10:15 PM Post #19 of 35


Quote:
Not exactly a musical instrument, but: a few years ago I started an engineering project that aimed to use acoustic analysis to detect operational abnormalities in small jet engines. It was mostly for "fun" but I learned a lot about sound and its behaviour - the way air molecules act when they're being slapped around by bits of metal spinning at 10 000rpm is pretty damn fascinating. There was a boatload of interaction and harmonic activity.
 
Reproducing these sounds accurately proved to be extraordinarily difficult to say the least, but it was an enlightening experience.

 
Now thats a good one.  Didn't think about mechanical noises.  F1 cars rev to 18-20k rpm.  An M1A1 tank might be good too. Jet turbine, trackpads clomping, metal squeeling, hydraulics pumping, weapons firing all at the same time.  That's a pretty complex sound signature.
 
 
Jul 26, 2010 at 12:54 AM Post #20 of 35
We can't even accurately recreate the sound of an 18 wheeler let alone a Raptor jet.  As an auto technician, listening to an engine throughout its rev range is a key part of diagnosis of possible problems - Let me say accurately recreating the sound of a high performance V12 is next to impossible.  That is why I say - there is a lot more bass in reality than what most headphones will have you believe.
 
Jul 26, 2010 at 10:10 PM Post #22 of 35
The most difficult sound to reproduce faithfully is pink noise
k701smile.gif
.
But wait, I'm only half joking, a deviation from a flat frequency response is the most easily perceptible in pink noise track, since very few transducers have a flat response, pink noise is indeed very difficult to reproduce.
 
 
Jul 27, 2010 at 7:15 AM Post #23 of 35


Quote:
pink noise
k701smile.gif

 


 
khaos974, I think the OP meant natural, voice, and instrument sounds!
 
 
If talking about loudspeakers, I think the answer would depend on the speaker system being used. Surely it is hard to stop and reverse a large mass speaker driver when it is accelerated fast towards it's specified maximum excursion? I think that it's inertia is going to want to keep it going until it runs out of suspension with a distorting bounce and shudder. So my vote goes for "any very, very large amplitude signal at the top of the frequency range of the largest mass speaker in your loudspeaker box".
 
As for single driver systems, such as headphones, at any point in time it can never be in more than one place at once, but it has to get there relative to where it was at the previous point in time, so it would have to be sounds with very large contrasts of frequency and amplitude which require drastic dynamic changes of position. Heavy bass drum combined with a hard hit cymbal? That would require large low frequency displacement combined with tracking small harmonics up to 20+ kHz.
 
I'm just guessing.
 
J.
 
 
 
 

 
 
Jul 27, 2010 at 8:02 AM Post #24 of 35
Devin Townsend's, power metal/prog wall of sound.
 
Jul 27, 2010 at 12:34 PM Post #25 of 35
Even though it´s probably technically piano... I have not heard a headphone coming close to the KeeS modded Pro 900 in that regard. Though I wonder if not voice is a better measuring stick. We hear the human voice every day in all kind of variants. We don´t hear piano every day or to the same degree. So it should be a better reference that way even though it doesn´t include all ranges. If you are mid range focused anyway.
 
Jul 28, 2010 at 1:16 AM Post #26 of 35
Simple speech is a good test to gauge midrange tonal balance and phasing anomolies.  Hard transients are more able to stress test driver capability to it's upper limit.
 
 
Jul 28, 2010 at 12:23 PM Post #28 of 35
In terms of recording and playback, I find that a distorted electric guitar is one of the hardest things to faithfully reproduce. The range of harmonics in vocals and a purely acoustical instrument are rather small in comparison. The distorted electric guitar has all of the mid frequencies that the non-distorted version has and then it also has all of the distortion harmonics added in! Whenever I compare audio equipment, distorted guitar is always a reference for me because of how drastically the sound of the distortion changes.
 
Jul 28, 2010 at 7:46 PM Post #29 of 35


Quote:
In terms of recording and playback, I find that a distorted electric guitar is one of the hardest things to faithfully reproduce. The range of harmonics in vocals and a purely acoustical instrument are rather small in comparison. The distorted electric guitar has all of the mid frequencies that the non-distorted version has and then it also has all of the distortion harmonics added in! Whenever I compare audio equipment, distorted guitar is always a reference for me because of how drastically the sound of the distortion changes.


Nice one.
 
Aug 4, 2010 at 7:29 AM Post #30 of 35
I have heard it from many, many recording engineers and mastering engineers that the most difficult sound to get right is the piano. I would agree.
 

Users who are viewing this thread

Back
Top