[bigshot]

Is that an insult?

You're done. Please take your seat.

 

[gregorio]

My point is that testing with scientific tools that can provide accurate measurements is great. But once those measurements start measuring things beyond the range of human hearing, the measurements become nothing more than numbers on a paper. They won't make your stereo sound any better.

That's not necessarily true. For example, you may have some serious distortion at say 40kHz. Admittedly you will not hear this but a whole load of energy at 40kHz may cause non-linearities in a speaker or amp, which is not designed to handle much energy at 40kHz, and cause unpredictable problems within the hearing range. So in this case, measurements beyond the range of human hearing would be useful to identify where problems are happening and would certainly be more than numbers on a page. There are many other examples where measurements outside the range of human hearing can identify non-linearities which could impact the performance of equipment; noise distribution in dither algorithms, phase artefacts of anti-alias, decimation and anti-imaging filters, etc. All of these measurements could be used in the initial design or repair of your stereo and therefore make it sound better.

G

 

[bigshot]

How would I ever get non-linearities at 40kHz when I'm playing a CD in my living room?

 

[jcx]

NOS DAC? - some people are proud of thier "filterless" direct I/V - you will have ultrasonic image frequency output with sin(x)/x amplitude from a zero order hold - NOS TDA1541 with R or xmfr "I/V" fed RedBook 44/16 will have 40 KHz image frequency output with ~ 4KHz music signal

 

[Uncle Erik]

How would I ever get non-linearities at 40kHz when I'm playing a CD in my living room?

You never know, Bigshot, a non-linearity at 40kHz might upset your dog.

Oh, if there's a big issue at 40kHz, simply put together a low pass filter. A cap and resistor on each channel and a plastic box with jacks. Maybe $10, unless you want to cryo it and put gigantic vibration-control feet on it, etc. You could simply cut off anything over 20kHz and it would never be a problem. Just like the very common rumble filter, but for the high end of frequencies.

As for using ears as the end-all of testing, it reminds me of the old Folger's ads. For those who haven't seen them, I'm sure a few have made it to YouTube.

Folger's had a series of ads where people were fooled into thinkin they were drinking a very expensive coffee because Folger's was served in a high-end place.

The same thing goes on with audio. Assume you are listening to something fancy and expensive and that's what you'll hear. What you perceive is connected to expectations. Expectations are rolled into everything you hear.

Wilson Audio's speaker demo with an iPod as source fooled everyone, too.

 

[gregorio]

How would I ever get non-linearities at 40kHz when I'm playing a CD in my living room?

If your DAC or amp had a fault or there may have been a problem in your model of DAC during the design process which needed to be resolved before production or if your DAC is capable of 96kHz sample frequency or if it's an upsampling DAC, it's performance would have to have been tested. Try to read my last post again, bearing in mind that I'm just giving you a hypothetical situation to demonstrate why measurements outside the hearing range are required and can be useful.

G

 

[ffdpmaggot]

Alright, even though I started the thread, I'm unsubscribing from it, so you guys can keep arguing whatever your arguing at this point, but its just gotten too abstract and apples-to-oranges for me to keep up with it.

 

[Uncle Erik]

Gregorio, do you know of any examples where your hypothetical actually happened?

Why wouldn't a low pass filter solve it?

 

[gregorio]

Gregorio, do you know of any examples where your hypothetical actually happened?

Why wouldn't a low pass filter solve it?

Hi Erik. I think you mis-understand. Bigshot stated that measurements outside the hearing range were pointless and he didn't see how they could improve his stereo system. I gave some hypothetical situations about where such measurements could be useful. A LPF would solve any of these HF problems but first you would have to run a frequency analysis to identify that it is a HF problem. For example, a few years ago I had a student (I used to be a lecturer in music tech) come to me with a problem test mix (24/96), with some strange low level tones. I stuck on a frequency analyser and couldn't see anything particularly wrong except a large peak around 40kHz. We went through the mix and discovered he had used a type 1 noise shaping dither algorithm on every single channel, reduced them to 16 bit and then mixed them. Bit of a workflow misunderstanding! After some testing I discovered that the rather poor quality amp did not like the high level of 40kHz and was causing distortion (IM tones) in the hearing range. The techs switched out the amp and the IM distortion could not be replicated. Obviously I had the student remove all the dither. There are various other examples, including a famous one with Rupert Neve and a malfunctioning capacitor causing a problem at 53kHz on a mixing desk. Best one though was when a colleague was high shelf boosting 16kHz on a track not realising that there was already a high shelf boost in another part of the chain at +18dB @ 25kHz. The first sign of a problem was when the tweeters in the Blue Sky's melted!

G

 

[SpaceTimeMorph]

Quote:

Gregorio, do you know of any examples where your hypothetical actually happened?

Why wouldn't a low pass filter solve it?

 
A low pass filter would work.  But unless you scoped the product prior to hooking it up for the first time this could probably happen to you.

From NwAvGuy:
 
ANOTHER “DESIGN BY EAR” EXAMPLE: I bought a power amplifier from a high-end manufacture similar to NuForce in some ways. After I started listening to it, the tweeters failed in the speakers it was driving. I measured the amplifier and discovered an overly protective current limiting circuit (triggered by my 4 ohm speakers) that created nasty high frequency bursts of powerful and inaudible noise. It was these bursts of ultrasonic noise that destroyed the tweeters. And in contacting the manufacture, much like with NuForce, they seemed completely unaware their amp had a problem. All they could talk about was how it sounds great and pointed to subjective reviews (without measurements) that also said it sounded great. Well, unless perhaps you’re a dog, you can’t hear a 25 Khz burst that destroys tweeters—that requires proper testing and measurements. Their amp, like NuForce’s DAC, also didn’t meet its published specifications and apparently had never been properly measured.

 

[bigshot]

If I buy a CD player, I don't expect it to produce any frequencies above redbook. If it does, it's probably broken. There's nothing up at 40kHz on a CD to reproduce anyway.

I think people just like making things complicated. All this theoretical talk does is confuse people who just want their stereo to sound really good. It isn't all that different from audiophile hoodoo in effect, even if that isn't the intent.

Plain old practicaliity and common sense seems to be in short supply on both sides of the aisle.

 

[Uncle Erik]

Thank you, Gregorio. Very interesting.

My next project will most likely be a preamp. I was going to build it according to the schematic, but I'll add a couple filters to limit it to 20Hz-20kHz. Won't cost much and won't make it much harder to build.

 

[wink]

 
Quote:

If I buy a CD player, I don't expect it to produce any frequencies above redbook. If it does, it's probably broken. There's nothing up at 40kHz on a CD to reproduce anyway.

I think people just like making things complicated. All this theoretical talk does is confuse people who just want their stereo to sound really good. It isn't all that different from audiophile hoodoo in effect, even if that isn't the intent.

Plain old practicaliity and common sense seems to be in short supply on both sides of the aisle.

Love the attitude.
Just proves the old adage that:-      IGNORANCE IS BLISS........................

 

[SpaceTimeMorph]

Quote:

If I buy a CD player, I don't expect it to produce any frequencies above redbook. If it does, it's probably broken. There's nothing up at 40kHz on a CD to reproduce anyway.

I think people just like making things complicated. All this theoretical talk does is confuse people who just want their stereo to sound really good. It isn't all that different from audiophile hoodoo in effect, even if that isn't the intent.

Plain old practicaliity and common sense seems to be in short supply on both sides of the aisle.

This is something I was thinking about... the audio voodooism runs so deep that it has to be countered with equally deep science, which leaves some people behind.  I agree all you should have to worry about is hooking a CD player or whatever up, but unfortunately that isn't always the case.  40 Khz tones/bursts/whatever shouldn't be present, but they do come into play somewhat.  The point is though, if audio manufacturers published reliable specs/measurements or there were enough checks placed on them by reviews, then this is something that you wouldn't have to worry about (much like to you don't need to worry about a torque curve in a car unless you want to).
 
Besides aren't you on the sound science forum?

 

 

[gregorio]

If I buy a CD player, I don't expect it to produce any frequencies above redbook. If it does, it's probably broken. There's nothing up at 40kHz on a CD to reproduce anyway.

I think people just like making things complicated. All this theoretical talk does is confuse people who just want their stereo to sound really good. It isn't all that different from audiophile hoodoo in effect, even if that isn't the intent.

Plain old practicaliity and common sense seems to be in short supply on both sides of the aisle.

Obviously a Redbook CD cannot contain frequencies above 22.05kHz but all it takes is a single malfunctioning component (like a capacitor for example) in your CD, amp or speakers and it could produce distortion frequencies anywhere, if it's in the hearing range you may spot it easily, if it's outside the hearing range (above or below) there maybe no or very limited audio clues, until a tweeter or a bass cone goes bye bye.

I don't agree with your statement that people like to make things complicated, I think the opposite is true. I think most people want to turn the keys in the ignition and just drive and to a large extent, modern technology allows them to do that but the reality is that under the hood modern digital audio systems are very complex. You think consumer playback is over complicated, you should try out the other side of the fence, the recording world, you can multiply the complication many times over. To an extent we face some of the same problems the consumer faces, no recording engineers I know of have PhDs in Electronic Engineering, Mathematics and Digital Signal Processing, so there's still scope for snake oil and misinformation. The marketing of snake oil to the pro-audio world is generally more scientific and sophisticated but it can still be difficult to get accurate expert opinion which is not biased by employment considerations, advertising revenue or product endorsement. Like with driving a car, there are times when there is no choice but to get "under the hood" and/or develop some knowledge if you need make an informed decision on product choice, want to change or improve something or just want to be a well informed driver.

Thank you, Gregorio. Very interesting.

My next project will most likely be a preamp. I was going to build it according to the schematic, but I'll add a couple filters to limit it to 20Hz-20kHz. Won't cost much and won't make it much harder to build.

Good luck, that's a pretty serious project. Not that I'm an expert on electronic engineering but from my experience there seems to be a lot of scope for subtly (or not so subtly) reducing SQ with filters, there always seems to be a trade off in electronic engineering.

G