I don't see how you can say this as, for instance, the music of the Indian subcontinent uses a very different system of a) tone fractions and b) beats to Western music.
Regards,
Andy
And we have 12 bar blues.. off course variations exist.. but the 16 beats of the typical Indian subcontinental style still fits the 4/4..of the normal western-world. Point is that the development of music and rhythm is profound and not only related to culture...Its rooted deeper and attached to emotions to sex..It's a fundamental...! this talking to something deeply rooted inside us is what artists try to put on their records and what we try to decode out from the recording again right in our living room...
AIUI, Western music is 7 tones to the octave? Do ray me fa so la ti ... right (as anyone who has seen The Sound of Music" will know)?
Whereas Chinese music, I believe, uses a 12-tone scale.
This is nothing to do with "12 bar blues".
Regards,
Andy
if you take the black keys there's is 12 pr octave....But that's not really the point...it's more that music is profound and developed with similar structures regardless origin..It goes a level deeper than culture and it's linked to something deep within. If this common denominator is found we may have a clue how to make systems perform better...
Maybe my ears are extra sensible
Maybe others can not hear any difference. I know a guy that owns a Forcell and can not hear any differences between using the stock very special arm cable and an ordinary supermarket wire.
Another guy started recently a discussion on Linkedin stating that he works in a pro studio an can not hear any differences between cables.
I also know a local amp builder that can not hear any difference between chinese low cost carbon resistors and naked Vishays.
Well, if you think that resistor type would make a change in sound, then you will agree with my point labeled "Point #4": If there is indeed a change in sound, perhaps it's not due to the theoretical 40pF change in capacitance but to something else related with one component (capacitor) vs the other, or some additional change at the electronic (signal) level.
yes!
in general no one really set the frequency noise, so she changed Russian, Czech and German scientists believe it = 7.83 Hz. American, Japanese, English = 7.82, while the Chinese are generally = 7.5 Hz. (I have the consistency of the Chinese electronics but!)
I relied on the Russian standard. That's the way real-time noise frequency in the air. Its fluctuations determine the conditions of the recording studio. Little nuances but ... it has a terrific sense of presence (or lack of) live music!
Antisemitic post deleted. Mr. Kunashirsky will be unable to answer your comments for the next week.I think you are saying that these Countries adopted a different recording standard. I'm OK with that.
All that needs is for the "standard" to be printed on the recording so that we listeners can switch the replay curve to match the recording curve.
But that is not what I am asking.
I am asking: why adopt 7.82 or 7.83 or 7.5 or another other fixed "standard"?
If the replay curve exactly matches the record curve, then does it make a difference to the final sound whether the "standard" adopts one value over another?
Another way to decode the riaa-emphasis is to use software filters. I have such a system running since a few month, actually using the biquad coefficients provided by Scott. Yes, it is ADC -> software filter engine -> DAC during playback. The result is amazing.
It can't be that difficult to compute the coefficients for those mentioned other curves. Perhaps somone knows how to obtain these by the aid of TI's coefficient calculator?
I seem to miss some understandig how to adjust the parameters to get along...
Or is it possible to compute those coefficents out of a Laplace transform?
Rüdiger
If the replay system can be "focused" to the curve used during the vinyl pressing, I wonder what difference that could make. Your question opened my mind to another one... Why are there so many variants ? Maybe people where searching for the "perfect tuned" curve and adopted what should seem the best considering manufacturing costs/systems availability at the time, in several regions of earth. I understand now that there are four or five determinant curves coming from Japan, Germany, Russia and USA.... What should be picked as the "best one" ?
Anyway, admiting that the majority of the pressings I am working with are relatively new (30 years) and it´s origine is USA I can deduct these use Riaa standard. After building severall phono preamps using passive filters between amplifying stages, I know there are some time constants we must try to follow very close in order to have the correct "instrument timbre".
I do not know if this aplies to designs using active filters (feedback) but Rüdiger´s post seems to translate what I mean. It is much simpler to design a digital filter to follow whatever curve we want so these must be very acurate. Any difference in the time constants used (even small) makes a big and audible difference.
The curves can be computed by useing Laplace...I do this in LT-spice to evaluate the filterings I design...Then I can get verification on how close they are to theoretical values and how tolerant they are to componentvalue tolerances..
Yes, me as well. But software de-empahsis using biquad coefficients (that form IIR-Filters with some soundmangling software) is something else. I'm almost sure it is not possible to compute coefficients with Laplace, but somehow it must be possible...
Rüdiger
Correct, IIR coefficient computation is not a closed form problem. If you search under Bob Orban's name and RIAA you will find the proceedure (though I can't seem to now). There are plenty of bad approximations out there Goldwave (when I compared it) was way off.
Yes, the original seems to have been archived off. I used a thermal annealing algorithm (unfortunately in in-house software) that did the same thing after many iterations but I could just leave it in the backround. You make an initial guess using the ordinary bi-quad formulas, this works but is in reality far off what can be achieved with tweeking. I exchanged emails with Bob, his "error" was very trivial in that he used 50.5 Hz were it should have been 50.05Hz (or is that visa-versa?). In fiddling with that I derived the ones for the next sampling rate up.
EDIT - I might publish a MATLAB/OCTAVE script to do any RIAA type equalization in Linear Audio but it will have to wait a couple of issues.
EDIT - I might publish a MATLAB/OCTAVE script to do any RIAA type equalization in Linear Audio but it will have to wait a couple of issues.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.