Watts Up...?

[514077]

I pity drive-unit designers - be it Balanced Armatures, micro Dynamic Drivers, or fullsize loudspeaker drive units, of various kinds.

These devices are expected not only to mimic real instruments, but often using MUCH smaller diaphragmatic area than real instruments utilise to achieve their sound.

I mean, just as one example, consider a loudspeaker tweeter - a 1 inch soft-dome, for example. That thing is expected to reproduce the sound of a cymbal that may be 15inches in diameter, and at the same SPL!!

Poor Rob goes to so much effort to reconstruct an analogue signal with breathtaking, eye-watering, accuracy. and yet has to suffer the indiginity of that signal being transduced by extremely compromised transducers!

Exactly. People expect stage sound in their homes. Here's to the poor designer's stiff upper lip.

 

[jayz]

I mean, just as one example, consider a loudspeaker tweeter - a 1 inch soft-dome, for example. That thing is expected to reproduce the sound of a cymbal that may be 15inches in diameter, and at the same SPL!!

From the point of view of Science or more specifically, the fundamental Physics behind acoustics, those two are almost opposite to one another. If you take a percussion instrument, it is the vibration of a specific material attached to an enclosure or stand of a specific size and shape, that produces sound by various bits resonating. In an ideal speaker, the cone is infinitely strong and has zero weight allowing it to move air according to electrical signal but it is not supposed to create a sound of its own by resonance. Imagine if they made tweeter cones from the same material used to make cymbals, that will probably be the worst tweeter ever. Also don't underestimate the power of accoustic transducers/motors, there are relatively small devices out there which can even cause temporary deafness .

 

[Dobrescu George]

From the point of view of Science or more specifically, the fundamental Physics behind acoustics, those two are almost opposite to one another. If you take a percussion instrument, it is the vibration of a specific material attached to an enclosure or stand of a specific size and shape, that produces sound by various bits resonating. In an ideal speaker, the cone is infinitely strong and has zero weight allowing it to move air according to electrical signal but it is not supposed to create a sound of its own by resonance. Imagine if they made tweeter cones from the same material used to make cymbals, that will probably be the worst tweeter ever. Also don't underestimate the power of accoustic transducers/motors, there are relatively small devices out there which can even cause temporary deafness .

*I think getting that to sound realistic, like the real instrument, penny for penny was the issue, not a small thing causing a loud noise. I'm fairly sure you can get a very simple system to go loud at a certain frequency, it is most enticing that our daily drivers are able to reproduce music faithfully, not just to make noise

 

[JaZZ]

Don't underestimate the wide frequency spectrum of cymbals! It doesn't just consist of treble, lower frequencies form a considerable part of their sound – hence it's not just the tweeter that's challenged. And what makes them sound so exquisite is the multitude of wafting vibrations within the brass plate causing an erratic radiation pattern that's almost impossible to be reproduced through loudspeakers with their poor radiation characteristic (especially in the upper treble). Headphones are incapable of capturing the threedimensional radiation pattern as well. Of course much of the problem is also the lack of resolution and transient accuracy of most sound transducers.

 

[Mython]

Don't underestimate the wide frequency spectrum of cymbals! It doesn't just consist of treble

Of course; I was merely offering a deliberately brief example, but point taken.

In any case, no matter how astonishingly well an ADC & subsequent DAC perform, they are still at the mercy of one or more transducers (microphone at the beginning of the recording=>playback chain, and drive unit(s) at the end of the chain, and all of these very frequently employ diaphragms that are disproportionately smaller than the instruments being played in the recorded performance.

It's actually amazing that they work so (relatively) convincingly, but it must be peculiarly frustrating for people like Rob, to create such high-performance ADC/DACs that remain at the mercy of the performance of these other factors.

But I'm going to step back now and be quiet, as this is really Rob's thread.

Cheers for your remarkably diligent efforts, Rob

 

[MRC001]

Don't underestimate the wide frequency spectrum of cymbals! It doesn't just consist of treble, lower frequencies form a considerable part of their sound – hence it's not just the tweeter that's challenged. ...

Agree - cymbals are great musical signal when listening to a system to assess the consistency of response from midrange into treble.
And castanets are great for assessing the treble into extreme HF. They have significant energy up to and beyond 20 kHz. I can no longer hear pure 20 kHz tones, but listening to castanets still makes it obvious when a system doesn't have linear frequency response extending that high.
Both are also very good for assessing transient response.
Of course, this assumes a top quality recording with good acoustics and mics, with sparing use of eq and compression. That's unfortunately, and sadly, rare.

 

[castleofargh]

Good question

Sometimes, I wonder, I am used to live instruments and music, but the more I analyse some of the recordings I'm listening to, the more I realise that some of the mastering was actually done considerably smoother than the thing sounded live

I'm talking only about rock / metal music, with large dynamic compression and lots of guitars and things, lots of it sounds different when recorded. I fear that with metal music, often the masters erase some of the brightness that should be there, leading to music that sounds falsely smooth. Makes me wonder if revealing that recording alone isn't enough and I actually end up needing more and more EQ simply as a result of their fault

Which kinda makes me wonder what is the most recommended way to add some brightness without introducing noise. As in, what is the cleanest EQ supported by a software, and such?

that's pretty easy, something like Pro-Q has been "good enough" to mix tracks and make albums for years, so I assume it's good enough for an audiophile to EQ a headphone to his preference ^_^. thought I'm not sure how many people can tell that apart from most parametric EQ using the same settings and filters.

in general I'm of the idea that everything changing the FR is an EQ so there is no point being scared of what we specifically call an EQ. when an amp outputs some rolled off trebles, how is that different from an analog EQ? the amp probably has more noise, and disto, but the specific frequency response changes, are they not those of an analog EQ? a crossover or a low pass filter are obviously the very definition of EQ. so we'll use many EQs anyway, knowingly or not, it's just a matter of adding one more to correct a FR that is almost guaranteed not to be flat thank to headphones or room+speakers.

 

[Dobrescu George]

that's pretty easy, something like Pro-Q has been "good enough" to mix tracks and make albums for years, so I assume it's good enough for an audiophile to EQ a headphone to his preference ^_^. thought I'm not sure how many people can tell that apart from most parametric EQ using the same settings and filters.

in general I'm of the idea that everything changing the FR is an EQ so there is no point being scared of what we specifically call an EQ. when an amp outputs some rolled off trebles, how is that different from an analog EQ? the amp probably has more noise, and disto, but the specific frequency response changes, are they not those of an analog EQ? a crossover or a low pass filter are obviously the very definition of EQ. so we'll use many EQs anyway, knowingly or not, it's just a matter of adding one more to correct a FR that is almost guaranteed not to be flat thank to headphones or room+speakers.

Oh, I should look into it.

Also, I agree that everything does one form of EQ or another, what I was wondering is what sounds the best.

Interesting case:

I dislike Foobar's EQ implementation, all of them induce more noise and distortion than the built-in EQ app of my workstation, which relies on some Sound Blaster X-Fi MB5, basically an ESS DAC + a custom AMP circuit, but the difference between Foobar's and its implementation is so high that I actually keep using the EQ on this one even though it has less bands and less precision

 

[castleofargh]

I don't want to stay off topic for too long, but in foobar you can use a VST wrapper which brings you tens of other EQ. the limit from foobar is that you have to stay at 32bit and some heavy EQ settings(for the more advanced ones) can be pretty CPU intense in 32bit. the 64bit version or VST3 version(which I believe are always 64bit compatible? would need to check that) can often make the same job easier for your CPU.

I have DMG equilibrium (Bob Katz sell speech convinced me despite how I learned most of how to use an EQ watching proQ videos on youtube^_^), and IMO it's absolutely overkill for stereo EQ and consumer usage(same as proQ2) but for sure you have all the tools and options you can dream of.
but free or not, parametric EQ is the way to go IMO. I don't remember ever using foobar's default EQ to be honest.

 

[x RELIC x]

@Dobrescu George, I thought I’d chime in here. One thing I find with the smoothness that Rob talks about (Mojo excluded) doesn’t come across as a smoothing of the entire frequency spectrum but rather a removal of exaggeration, if that makes sense. If listening to well recorded music I expect the soft sounds to sound soft and the hard sounds to sound hard. I expect warm sound to be warm and bright to be bright. As Rob says, yang and yang. This variety and nuance is what I find Rob’s designs do very well. For a lot of ‘bright’ gear I find that the subtleties to be lost and everything comes across as bright and hard, which in the end significantly reduces my personal enjoyment of the music.

It’s analogous to dynamically compressed recordings IMO. This is a great video that demonstrates dynamically compressed music vs maintaining high dynamic range in the track. The author of this article calls it wimpy loud sound, which is somewhat comparable to how I find other DACs to reproduce music.



This is the link to the article:

https://www.computeraudiophile.com/ca/ca-academy/dynamic-range-no-quiet-no-loud-r643/

I’m afraid that a lot of what we choose for our preferences has a lot to do with the type of music we listen to and how well it’s been recorded/mastered. Often I see people choose brighter gear and then turning up the volume to compensate for nuance/dynamics because there isn’t the contrast required to really feel the impact as it should be felt. A lose / lose situation IMO. Of course this is just my musings on the matter and just one aspect of audio reproduction. YMMV, and of course I am not addressing personal preferences at all in this post.

 

[Dobrescu George]

@Dobrescu George, I thought I’d chime in here. One thing I find with the smoothness that Rob talks about (Mojo excluded) doesn’t come across as a smoothing of the entire frequency spectrum but rather a removal of exaggeration, if that makes sense. If listening to well recorded music I expect the soft sounds to sound soft and the hard sounds to sound hard. I expect warm sound to be warm and bright to be bright. As Rob says, yang and yang. This variety and nuance is what I find Rob’s designs do very well. For a lot of ‘bright’ gear I find that the subtleties to be lost and everything comes across as bright and hard, which in the end significantly reduces my personal enjoyment of the music.

It’s analogous to dynamically compressed recordings IMO. This is a great video that demonstrates dynamically compressed music vs maintaining high dynamic range in the track. The author of this article calls it wimpy loud sound, which is somewhat comparable to how I find other DACs to reproduce music.



This is the link to the article:

https://www.computeraudiophile.com/ca/ca-academy/dynamic-range-no-quiet-no-loud-r643/

I’m afraid that a lot of what we choose for our preferences has a lot to do with the type of music we listen to and how well it’s been recorded/mastered. Often I see people choose brighter gear and then turning up the volume to compensate for nuance/dynamics because there isn’t the contrast required to really feel the impact as it should be felt. A lose / lose situation IMO. Of course this is just my musings on the matter and just one aspect of audio reproduction. YMMV, and of course I am not addressing personal preferences at all in this post.

Thanks, that is entirely true.

I also used this article / similar articles to teach some of my mastering friends to not apply dynamic compression unless it is absolutely necessary...

Of course, we cannot blame our tastes, but that is a large issue with mine, listening to metal music that's mastered to sound compressed and such, I want to further exaggerate some parts of it.

I agree that it'd be better with less dynamic range compression actually... To further exaggerate the issue, because it is so dynamically compressed, sometimes I go ahead and push the volume very loud or use gear that has better impact just to bring more impact in my experience.

This leads to many other discussions, but basically, I want to hear the impact of a live cymbals in death metal, along with bass notes, but the master usually erased both in many albums, leading to a very poor experience for somebody who is used to live death metal and live metal. If you're a metallica fan, their latest albums, Hardwired to Self Destruct, are far different from Death Magnetic. There is far less dynamic compression, and it sounds somewhat normal again, while Death Magnetic was so compressed in its original release that you couldn't hear anything besides clipping and distortions all over, along with almost no impact, the whole album was a bomb. Of course, the HDTracks / subsequent masters did solve the issue, but you probably understand the issue with metal music.

If my understanding is correct, Mr. Rob's latest works aren't tuned similarly to Mojo and Mojo shouldn't make a benchmark to learn where Chrod products are going signautre wise?

 

[x RELIC x]

If my understanding is correct, Mr. Rob's latest works aren't tuned similarly to Mojo and Mojo shouldn't make a benchmark to learn where Chrod products are going signautre wise?

Based on what I hear between the Mojo, Hugo2, and DAVE they have similar tuning (they are in the same tonal family) but as you go higher up the chain you get more resolution and timing benefits. Relatively speaking, I find the DAVE to have the most detail while also having the most impact and bottom end. The Hugo2 isn’t too far behind but isn’t quite as impactful and detailed (relative to DAVE). The Mojo has a similar tonality but sounds a little more ‘compressed’ (still great in it’s category IMO).

 

[Rob Watts]

.....
If my understanding is correct, Mr. Rob's latest works aren't tuned similarly to Mojo and Mojo shouldn't make a benchmark to learn where Chrod products are going signautre wise?

Yes Mojo is not my reference - that would be BluDave. And BluDave has not been tuned - or tweaked in terms of sound quality - at all. I am not trying to create a particular sound - the intent is to try to make it as transparent as possible - that is so that it neither adds or subtracts anything to the performance. The intent is to close down the huge (but actually getting considerably smaller) gap from reproduced audio to the sound of unamplified acoustic music. With BluDave the development process has been entirely and solely driven from the aberration point of view - in short, it's a question of defining aberrations (differences from the ideal), listening to those aberrations to see what the sonic consequences are, then doing as best as possible job in minimizing all of them. I am on a constant hunt to find out what the aberrations are - and this comes from detailed measurements, simulations of aberrations, and most crucially thinking through every single aspect of the conversion process - and then exploring the assumptions that are built into these aspects.

The M scaler illustrates this process; we are trying to recover the original bandwidth limited analogue signal that is inside the ADC when it is being sampled. We know absolutely and with complete certainty that the only way of recovering the un-sampled signal perfectly is to use a sinc function interpolation filter - and my work over the last 38 years has been trying to establish how much of a loss a finite tap length filter (that is one that is clearly different to a sinc function) makes to the sound quality. So here I have identified a difference, then worked relentlessly to improve the accuracy of the filter, to get it closer to the ideal sinc function. And I will keep on going until I get to a stage where doubling the tap length no longer alters the sound quality. Now with the M scaler, I know that going from 0.5M taps to 1M taps is very audible, so this quest is not finished. With 1M taps and the WTA algorithm I have a filter where all of the coefficients are identical to an ideal sinc function to at least 16 bits. But what we do not know - and nobody on the planet knows - is at what level of accuracy is needed so that doubling the tap length has no change sonically.

Now within the BluDave there are a myriad of similar issues - analogue and digital - but I guess this one is the most important sound quality wise. And I have identified these aberrations - and still looking for more I might add - and reduced them to be as small as possible given the design constraints (cost, technology levels etc). And we end up with the sound of BluDave having worked on minimizing all the aberrations - there has been no point where I have tweaked something to get a particular sound quality. So BluDave represents my reference point, and other lower priced products are sound quality optimised from the BluDave reference.

So am I certain that BluDave adds or subtracts nothing to the sound quality? No - absolutely not. Nobody has heard the perfect DAC. And this is one of the motivations behind the Davina project - I will be able to find out how accurate BluDave is against the ideal perfect DAC; and this will be extremely powerful information.

 

[Dobrescu George]

Yes Mojo is not my reference - that would be BluDave. And BluDave has not been tuned - or tweaked in terms of sound quality - at all. I am not trying to create a particular sound - the intent is to try to make it as transparent as possible - that is so that it neither adds or subtracts anything to the performance. The intent is to close down the huge (but actually getting considerably smaller) gap from reproduced audio to the sound of unamplified acoustic music. With BluDave the development process has been entirely and solely driven from the aberration point of view - in short, it's a question of defining aberrations (differences from the ideal), listening to those aberrations to see what the sonic consequences are, then doing as best as possible job in minimizing all of them. I am on a constant hunt to find out what the aberrations are - and this comes from detailed measurements, simulations of aberrations, and most crucially thinking through every single aspect of the conversion process - and then exploring the assumptions that are built into these aspects.

The M scaler illustrates this process; we are trying to recover the original bandwidth limited analogue signal that is inside the ADC when it is being sampled. We know absolutely and with complete certainty that the only way of recovering the un-sampled signal perfectly is to use a sinc function interpolation filter - and my work over the last 38 years has been trying to establish how much of a loss a finite tap length filter (that is one that is clearly different to a sinc function) makes to the sound quality. So here I have identified a difference, then worked relentlessly to improve the accuracy of the filter, to get it closer to the ideal sinc function. And I will keep on going until I get to a stage where doubling the tap length no longer alters the sound quality. Now with the M scaler, I know that going from 0.5M taps to 1M taps is very audible, so this quest is not finished. With 1M taps and the WTA algorithm I have a filter where all of the coefficients are identical to an ideal sinc function to at least 16 bits. But what we do not know - and nobody on the planet knows - is at what level of accuracy is needed so that doubling the tap length has no change sonically.

Now within the BluDave there are a myriad of similar issues - analogue and digital - but I guess this one is the most important sound quality wise. And I have identified these aberrations - and still looking for more I might add - and reduced them to be as small as possible given the design constraints (cost, technology levels etc). And we end up with the sound of BluDave having worked on minimizing all the aberrations - there has been no point where I have tweaked something to get a particular sound quality. So BluDave represents my reference point, and other lower priced products are sound quality optimised from the BluDave reference.

So am I certain that BluDave adds or subtracts nothing to the sound quality? No - absolutely not. Nobody has heard the perfect DAC. And this is one of the motivations behind the Davina project - I will be able to find out how accurate BluDave is against the ideal perfect DAC; and this will be extremely powerful information.

Woah

Makes me wonder if you shouldn't be also making an ADC - so that the whole process is as good as it can get, since listening to a lot of music I'm sure that the microphones and the ADC steps are probably not as good as the DAC step phases (?)

For reference, this type of song has its cymbals smoothed out audibly from the ideal presentation:

 

[x RELIC x]

Woah

Makes me wonder if you shouldn't be also making an ADC - so that the whole process is as good as it can get, since listening to a lot of music I'm sure that the microphones and the ADC steps are probably not as good as the DAC step phases (?)

For reference, this type of song has its cymbals smoothed out audibly from the ideal presentation:

Lol, he is! It’s called Davina and Rob has posted about it quite a bit already.