When optimizing, go for Lowest 3rd harmonics .. not THD!!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Lumba Ogir said:
Peter,
I`ve seen the gear used in the recording trade. I did not vomit but was very close.

Like on the playback side or in just about any industry there are good equipment and there are bad equipment. I don't think that come as a surprise to most people.

That said it would be interesting if you could add some substance to that highly emotional and subjective post. I mean if you feel sick looking at some gear it really tells more about you than about the performance of the gear.

Gear are best judged by blind listening tests and measurements. Doing it that way you eliminate all funny things that happens in the brain when you know what brand resistors are in the circuit and what topologies are used.


/Peter
 
Some mention was made above about testing by listening to different waveforms of the same frequency. I thought I'd mention a trap some people (me included) have fallen into when comparing, say, a 10kHz sinewave and a 10kHz squarewave -- matching the levels of the contained sinewave fundamental.

If you play both at the same peak voltage (such as a 1V peak sinewave and a 1V peak square wave), your fundamental component levels will be 2dB mismatched -- and different levels certainly make things sound different. If you took the harmonic components out of a 10kHz @ 1V square wave (leaving only the sinewave component), you'd find that the remaining sine wave level was 1.27V peak (i.e., LARGER than the square wave peak). That's because all the odd harmonics reduce the peak levels of the square waveform.

So if you did this listening test with matched peak levels, try it again with the square wave's voltage at 78.5% of the sinewave's voltage. And don't do the test with high power and a tweeter, for a lot of us playing a 10kHz sinewave that can be clearly heard will be enough to heat or damage a tweeter (you don't usually get high level continuous 10kHz waves in music). Headphones are probably a better way to do it.
 
I wish Santa Klaus bring you a full bag of harmonics emptyness dear Lineup

A gift for this Christmas...no harmonics...no distortions.... a fourier wonderland.

I hope you enjoy sound without them...but i am afraid you gonna junk them

ahahahah.

Merry Christmas.

and Happy new year.

Send to our common friend, Halojoy, a deep hug!

Carlos
 

Attachments

  • santa klaus animated.gif
    santa klaus animated.gif
    71.1 KB · Views: 208
Administrator
Joined 2007
Paid Member
Just get a pair of headphones connected up to a function generator ( pad them out with a resistor ) and with square wave selected just slowly increase the frequency. You WILL reach a point where the sharp rasping harmonic sound becomes less and less and then finally dissapears. Don't deafen yourself, just start with a fairly quiet level and increase the frequency. If you then switch back and forth between sine and square although the "level" will sound different the timbre should be the same.
 
Pan said:
Is that beliefe grounded on actual listening tests or from the way you have been thinking on sampling theory? Wouldn't it be cool if you knew that higher resolution was equal to higher fideliy instead of believe? If you find some evidence please tell me about it. I thought the same as you.. until I started looking for proof and I found none.

I have compared the sound of CD and SACD on two different players (one slightly expensive, the other EXPENSIVE :D) and would say that the sound of the SACD's was more detailed and refined with more ambience.

Pan said:
But we don't hear what a waveform looks like on the scope. A scope can have GHz bandwith but our ears has much more limited bandwith. Also how common is it with wide bandwith microphones? Not very common but sure, they do exist.[/B]

I would think that the waveform on a scope is a much closer representation of what we hear, compared to a spectrum analyzer. I believe our hearing is sensitive to the shape of a wave.


Pan said:
Try to put up some tests and see what you learn from it. Use a wide bandwith signal and slap on different LP filters until you hear the difference. Again such tests have been done and there is nothing that indicates that we need much more than 20kS/s for normal music.

Listen to sines, squares and triangular waves at frequencies from a couple of hundred Hz up to 15kHz and see how they sounds more and more similar the higher up you go. Be aware though that nonlinearities in teh playback gear may skew the reults slightly. Same for a el-guitar with overdrive/high gain. As you work yourself up on the neck the buzz dissapear and the notes sound clean. The dist is there on a scope but our ears filter the HF out.
/Peter [/B]

My experiments with amplifiers suggest that the FR must be much higher than 20Khz to sound its best, I would think the same will apply to CDP's etc.

I have read about an experiment that was done where brainwaves was monitored in response to different frequencies, the results showed that the brain detected a difference when an audible freq combined with it's 'unuadible' harmonics was presented compared to the same signal without harmonics. The high freq signals alone wasn't detected.
 
I have read about an experiment that was done where brainwaves was monitored in response to different frequencies, the results showed that the brain detected a difference when an audible freq combined with it's 'unuadible' harmonics was presented compared to the same signal without harmonics. The high freq signals alone wasn't detected.

There is also this theory:

The brain might not be able to pick out the finer details after just one cycle of a waveform. But one cycle after another, perhaps the brain finally "realizes" the details of the waveform.

It has to do with aliasing such as on a computer monitor. Look at a simulation waveform such as below. You can see a reproduction of the waveform as a "heterodyne" of the original because of the aliasing.

Perhaps our brains are just like this. Sometimes our brain can't compute it all at once but if it looks at the distorted waveform and the "heterodyne" it can see much finer details than it could after only one cycle.

I believe this is what was shown in the experiment mentioned above.

- keantoken
 

Attachments

  • heterodyne.png
    heterodyne.png
    20.3 KB · Views: 185
Mooly said:
Just get a pair of headphones connected up to a function generator ( pad them out with a resistor ) and with square wave selected just slowly increase the frequency. You WILL reach a point where the sharp rasping harmonic sound becomes less and less and then finally dissapears. Don't deafen yourself, just start with a fairly quiet level and increase the frequency. If you then switch back and forth between sine and square although the "level" will sound different the timbre should be the same.

I haven't tried but do you think even headphones can reproduce a 10 or 15Khz square wave without starting to change it into sinewave?
 
Lumba Ogir said:
Peter,
simplicity. Short, straight, clean signal route and processing. 25 thousand knobs don`t give transparency. Not a chance.


I promise you that 1) not all recordings is done that way and 2) no one will ever hold a gun to your head forcing you to use that "25 thousand knobs" approach.

I love processed music as well as puristic acoustic music and when I record music myself I use good mic's into good preamps into good AD's. The only knob the signal ever "sees" is the gain staging in the mic preamp. Since I love the way real instruments sounds (I play several myself) and choose my instruments from the balanced voice the instruments has intrinsically I don't want anything altering the sound when recording or play back. The acoustic event has all the beauty I could ever wish for and I only want the gear to handle the signal from the mic to the speakers without getting in the way.

Why is it you choose the worst possible scenario/alternative to make some kind of point?


/Peter
 
Pan said:
I love processed music as well as puristic acoustic music and when I record music myself I use good mic's into good preamps into good AD's. The only knob the signal ever "sees" is the gain staging in the mic preamp. Since I love the way real instruments sounds (I play several myself) and choose my instruments from the balanced voice the instruments has intrinsically I don't want anything altering the sound when recording or play back. The acoustic event has all the beauty I could ever wish for and I only want the gear to handle the signal from the mic to the speakers without getting in the way.
/Peter

Peter, do you sell these CD's? It is good to see there are still people that care about recording quality.
 
Andre Visser said:
I have compared the sound of CD and SACD on two different players (one slightly expensive, the other EXPENSIVE :D) and would say that the sound of the SACD's was more detailed and refined with more ambience.

Do you know for sure that the redbook layer and the SACD layer used the exact same master?

Did you level match? SACD and redbook often have different levels.

It's easy to draw wrong conclusions. :)


I have also a couple of hybrid SACD/CD's and I do find the SACD layer to sound better but still I cold not hear any difference when I inserted a 24/44.1 PCM AD>DA chain between the SACD output and my amp. What conclusions can be drawn from that? :)


I would think that the waveform on a scope is a much closer representation of what we hear, compared to a spectrum analyzer.

Several engineers I know that have looked into this with blind listening tests during the years have come to the opposite conclusion.. and so have I. You can manipulate the signal a lot so it looks totally different on a scope without hearing any difference, however a roll off of about 1/10 of a dB at 10kHz and 2-3/10 of a dB at 20kHz is easily heard. These small differences often are described as something else than a "roll off" such as cleaner, higher resolution and so on. This is probably the major reason for audible diffrencec in cables, only people are not aware of it and prefer to believe in vodoo and snake oil instead of accepting science. :)


I believe our hearing is sensitive to the shape of a wave.

Well, I can't argue with that. A scope trace of a piece of music that is lowpassed at 2kHz will look very different and also sound very different.

My experiments with amplifiers suggest that the FR must be much higher than 20Khz to sound its best, I would think the same will apply to CDP's etc.

But why? Off course we don't want to roll off audibly below or at 20k but other than that do we really need so much more? Remember also that an amp is a min-phase device and a typicall CDP is a linear phase device.


I have read about an experiment that was done where brainwaves was monitored in response to different frequencies, the results showed that the brain detected a difference when an audible freq combined with it's 'unuadible' harmonics was presented compared to the same signal without harmonics. The high freq signals alone wasn't detected.

I think I know that test and it has received a lot of critique if I remember correct.

Also remember that most music has very little HF content and both mic's and speakers have limited HF output so one really has to question the need for gear able to playback 100kHz.

I'm open but so far I have seen no evidence and the bandwith of the other gears and the output from the source are such a limitation anyway.

/Peter
 
Andre Visser said:


Peter, do you sell these CD's? It is good to see there are still people that care about recording quality.

First I want to thank you for a civilized debate and second, nope, no CD's for sale. I plan on starting a business after new year though so maybe in the future I can offer some. Possibly I can post some clips in the beginning of next year.

I'm mostly interested in folk, easy-jazz, solo and smaller ensembles but also choir and organ.


/Peter
 
Lumba Ogir said:
Peter,

Unfortunately, that`s the common scenario. I`m sincerely glad you represent a very rare exception and I wish you success.

Well I guess you're right (especially for the pop/rock/dance sceen)but still there are a lot of good stuff to be found out there. But it sure would be nice if even more of the modern music was handled with more care. The loudness race has destroyed plenty of music. :-(

Thanks for your wishes! :)


/Peter
 
Hi everyone,
look what high-order harmonic distortion products do to the precious signal waveform. Your highly dependable software would show exactly the same THD level for (b) and (c).
Just another illustration of how worthless the THD standard is (SMPTE and CCIF not a bit better), so don`t be so delighted every time, maybe Nirvana is not reached after all.
 

Attachments

  • thd.png
    thd.png
    55.1 KB · Views: 167
illustration of how worthless the THD standard is

yes. this is one of my points with this topic
your attachment shows this issue clearly.
even if such high levels of distortion is unusual these days.

And yet :D
90% of the peoples in this forum uses THD
for basic compare of amplifiers and other gears

THD, TOTAL level of distortion, can give only a small hint.

But tells not much about the sound quality
or even circuit quality. As two identical THD figures
can be so very different when looking at harmonic frequency details.
 
Bravo mfc, you are right.

Very, very true. I believe that in an instrument such as a piano, the correct harmonic balance gives rise to what we would call a "musical" sound having to do with mathematics. In such an instrument I don't believe it is as important how pleasant it sounds to the ear but how the notes are tied together by the harmonics. Perhaps Mozart would be able to inform us on this concept.

Anyways, although I do believe that sound and its perception can perhaps be explained completely in non-subjective terms, I know that there are things that we don't know yet about the subject and things that we won't know for a very, very long time. For now all we can do is give it the sound that "sounds best" and infuse it with our quality. If it's not done with scientific precision, then maybe it'll at least be done with quality.

Merry Christmas to all,
- keantoken
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.