Does that surprise you?
What answer would you give to your own question?
OK just seen your new post. I would expect to see a set of frequency components, spaced in frequency by 1/beat, centred around the drum fundamental and its overtones. There could be components at lower frequencies.
Note that to measure 'a frequency' you need it to last for an infinite time, so a real drum does not have a frequency but it does have a waveform so it also has a Fourier transform.
What answer would you give to your own question?
OK just seen your new post. I would expect to see a set of frequency components, spaced in frequency by 1/beat, centred around the drum fundamental and its overtones. There could be components at lower frequencies.
Note that to measure 'a frequency' you need it to last for an infinite time, so a real drum does not have a frequency but it does have a waveform so it also has a Fourier transform.
Last edited:
windowing in the frequency domain gives a filter in the time domain
truncating = a "brick wall" filter
dropping the ten lowest frequency components in a 20 second record's fft, then doing the inverse fft gives a 0.5 Hz brick wall high pass filter
dft allows "fast convolution" filters - for really long fir filters you can save multiplies
does a 20 Hz hi pass analog filter affect the drum beat?
truncating = a "brick wall" filter
dropping the ten lowest frequency components in a 20 second record's fft, then doing the inverse fft gives a 0.5 Hz brick wall high pass filter
dft allows "fast convolution" filters - for really long fir filters you can save multiplies
does a 20 Hz hi pass analog filter affect the drum beat?
Last edited:
Thats good stuff. HOWEVER, the comments to it are great as well... as they are here. The mistake is to be literal about left and right and what that view was actually (a 2D construct). It points out how the mind will fill-in the missing info - as we are known to do with listening.
Some people couldnt see it rotate at all or in one direction, only. What That means is up to individual exploration. But, there were many different affects indicated for the test viewing. What it means EXACTLY is not the real issue - left-right brain dominance- but that we dont all see things the same way. nor hear them the same way. As I learned from the deaf -- hearing is learned or rather interpreting what you hear is learned. Some learn more than others... some have differing spatial perspectives and abilities... even between sexes. hearing must fall in there somewhere. Thx for the extra info.-RNM
Last edited:
Though you don't have to close your eyes. The key word you used was "try", when a system is not quite there then you do have to try, the more effort required the "worse" the system, and then you get into the realm of classic listening fatigue, as in, I can only listen for half an hour before I had enough -- the ear/brain is exhausted by the amount of energy that has to go into decoding the musical message, bypassing the haze of distortion imposed by the playback system. There is a level of quality of replay where the word "try" becomes irrelevant, the body is at complete ease while listening, the realm of true "high end" sound ...
I have a nice bundle of CDs that I use for testing this capability, one particularly useful is a collection of early 30's famous swing orchestra tracks. Play this at realistic volumes, a conventional system creates a pretty awful, scratchy mess. At the next level up the sound starts to come to come together, you get soundstaging, depth, ambience, but the tonality is still extremely wearing. Finally, at the top of the tree so to speak, you can "see" the musicians in the rows of the recording space, and the tone of the brass section comes through completely cleanly, there is no difficulty "perceiving" the instruments as being "there".
This is all psychoacoustic of course, but demonstrates how powerful the mind is in unraveling a complex sound message if you give it the right material to work with ...
Frank
I respectfully disagree. In one case it is "Try to not hear the system", in another case it is "Try to hear the system", but in both cases it is "Try", because it is evaluation
However, when you speak about listening... But it is hard for a designer not to evaluate, even when listening.
Now I think you're being deliberately obtuse. Never start and stop your CD player, you could be losing something.
So, to take it a little further or to kill a dead horse --> We get a group of men and women to look at this rotating image..... and they have different descriptions of what they see. One will know exactly what it actaully is (2D etal). but the other people do not and may never see it that way or any other way except how they see it. They might look at it with one eye or side ways and get yet another view. But it isnt the actual view of the test. So, Which is the 'real' view. the actual test or the viewing of the group as expressed by the individual viewers?
The FFT for example is what it actually is and does what it does..... It doesnt correlate to what a group of people might say it sounds like. What you see on the display - harmonics - doesnt really tell us much about hearing nor listening perceptions. Not anymore than a color analyzer tells us what we percieve or experience in viewing an image. So, yes, its a tool. But not a very good one for what individuals hear. It isnt unlike the rotating figure. A test instrument isnt going to predict what You will see or think you see when observing that rotating figure.
lets see if we can find the blind spots and fill it with tests that will show a more complete picture. Or, do as I do... take all the viewers (listeners) comments and try to understand it and accept it as if real and sometimes different from others and move on. Because as F.Toole said (in a different context) " In the end the ear wins." -Thx, RNM
The FFT for example is what it actually is and does what it does..... It doesnt correlate to what a group of people might say it sounds like. What you see on the display - harmonics - doesnt really tell us much about hearing nor listening perceptions. Not anymore than a color analyzer tells us what we percieve or experience in viewing an image. So, yes, its a tool. But not a very good one for what individuals hear. It isnt unlike the rotating figure. A test instrument isnt going to predict what You will see or think you see when observing that rotating figure.
lets see if we can find the blind spots and fill it with tests that will show a more complete picture. Or, do as I do... take all the viewers (listeners) comments and try to understand it and accept it as if real and sometimes different from others and move on. Because as F.Toole said (in a different context) " In the end the ear wins." -Thx, RNM
Last edited:
Dick, I call shenanigans. A frequency spectrum correlates perfectly with how people perceive tonal balance, level, phase anomalies, and distortion. It is indeed one tool among many, and quite a powerful one. You're perfectly aware that the FT and the original time series are interconvertible, containing exactly the same information, unless you deliberately didn't read the careful explanations by Dave, Scott, and me and never bothered to study integral transform methods in undergraduate engineering (or physics) courses.
I'm sensing a new High End Irrationalist meme developing.
I'm sensing a new High End Irrationalist meme developing.
Different meanings of the word "try". Yes, if you're evaluating some performance parameter then you apply effort in a certain mental sense to focus on the particular attribute. But I'm using that word more "loosely", in the sense of having to apply largely subconscious concentration, in order to enjoy the the quality of musical reproduction ...
An analogy: a teacher of violinists listens to a master pupil playing an impressive piece on a superb instrument. Live. Everyone else is enthralled by the gloriousness of the sound, but, Terrible!, says the teacher, Technique was all over the place! Here, I'm on the side of everyone else ...
Frank
No one made those claims, the FFT is a tool for signal analysis learn its capabilities and limitations or don't it's you're choice. BTW it's also an incredibly powerful tool in optics which they used to extract data from the broken Hubble before it was fixed.
Last edited:
Dick, I call shenanigans. A frequency spectrum correlates perfectly with how people perceive tonal balance, level, phase anomalies, and distortion. It is indeed one tool among many, and quite a powerful one. You're perfectly aware that the FT and the original time series are interconvertible, containing exactly the same information,
I do understand what the FFT shows and it IS interconvertable. Does it correlate perfectly with every perception heard... really? Then we need to do a better job of explaining what people hear, is all. Thx -RNM
I am not saying FFT doesnt do what it does and that it isnt important nor that it doesnt do what it does do.
I have never said anyone claimed other-wise rather --- rather that repeated explaining FFT theory and functionality isnt effective on most lay listeners (i do get it). Do we care? I dont know. We dont need to. But, seems like we might care to get ourselves across to the average guy better- maybe with other test which is more intuitive to the lay person..... even if it isnt as elegant as math and the FFT.
Once you have a bandwidth that is Dc to light and noise and thd are below thresholds and all.... the differences are likely to be found in the interfacing and over-all data thru-put end-to-end that is affected in processing. Nothing that cant be explained by FFt or any other test equipment of suitable sensitivity. But that by ignoring those issues, we arent ever going to convince the public.
I know there is enough info in the tests we do to explain what is heard or not heard. And, FFT does it for most everyone in science. Unfortunately, that isnt changing anything on the ground. The debate rages on.
If you had to teach elementary school kids these things --- how would you present info to them. -Dick.
Dick, I call shenanigans. A frequency spectrum correlates perfectly with how people perceive tonal balance, level, phase anomalies, and distortion. It is indeed one tool among many, and quite a powerful one. You're perfectly aware that the FT and the original time series are interconvertible, containing exactly the same information, unless you deliberately didn't read the careful explanations by Dave, Scott, and me and never bothered to study integral transform methods in undergraduate engineering (or physics) courses.
I'm sensing a new High End Irrationalist meme developing.
maybe you are right. sometimes I just like to explore and play with opposite views.... new combinations. If you look at an opposite extream view from my own I can see more of eveything in between. And, then i also exagerate for effect. Its the story teller in me. Anyway - It isnt my fault because - I was raised that way (can I use that?)
.
I am leaving the door open a bit for entry for new info to come in which would make the pro-repro-duced music seem more like live, real sounds. Right now the systems do not sound like real musical instruments played in the home. far from it. but, then, a preamp isnt the worst offender in the chain.
Last edited:
What seems to be always ignored is that the performance of an ambitious audio system is constantly in flux -- anyone who has a high capability setup and sensitive hearing knows that the sound quality that he hears at any particular time varies for seemingly myriads of reasons; classic variables are the time of the day, and what electrical devices are operating in his home. The latter are due to the system not being 100% impervious to interference from the point of view of the capability of his hearing.
So any attempts to verify the performance of a system by sticking individual elements on a separate, nicely controlled test bench, and extracting a set of static measurements are always doomed to fail, certainly at the moment. Human hearing is just too sensitive, and the measurers don't try hard enough ...
Frank
So any attempts to verify the performance of a system by sticking individual elements on a separate, nicely controlled test bench, and extracting a set of static measurements are always doomed to fail, certainly at the moment. Human hearing is just too sensitive, and the measurers don't try hard enough ...
Frank
As a postmodernist I have no trouble answering that question - each person's perception is 'the real view'. Those who cling to the notion of 'objective truth' will have to rationalize on this one
- Status
- Not open for further replies.