Master Thread: Are measurements Everything or Nothing?

[birdog1960]

Agreed! I own Bryston amps because of the reliability and the service. Not sure that @birdog1960 is going to find anything actionable on hifi brand and model reliability on a consumer info site. If so, please post! Reliability is way more controversial and difficult than electrical performance, so double good-luck. I bought Bryston because of the company's reputation and the 20 year warranty they offer, plus I used to sell them.

Another critical category, features. For instance, I own an RME ADI-2 FS DAC because of the features, even more than the performance. I also own a MiniDSP SHD that doesn't have quite the performance of the RME, but I don't care because I own it for the features. I certainly avoid chasing my tail in circles.

if it hasn't been done it easily could be. Consumer reports does frequency of repair polls on every car made and each system in the cars and yes, I realize their "lab" testing methodology leaves much to be desired. Even small market cars like high end porsches have ratings (spoiler: they're not good).

 

[amirm]

yes, but can you make a $400 yamaha receiver sound great with measurements at the thresholds of audibility under average listening conditions? Many people would like to know that (not currently in the market myself but will be when my receiver breaks). It's your forum but if the philosophy is to slay the dragon of snake oil claims from manufacturers then that question is important.

What the heck is the average?

 

[antcollinet]

yes, but can you make a $400 yamaha receiver sound great with measurements at the thresholds of audibility under average listening conditions? Many people would like to know that (not currently in the market myself but will be when my receiver breaks). It's your forum but if the philosophy is to slay the dragon of snake oil claims from manufacturers then that question is important.

If the $400 reciever had measurements below audibility - then it it is possible to get that under average listening conditions.

 

[solderdude]

Dr Floyd Toole has concluded that quote, listeners themselves are highly stable "measuring instruments". Apparently the data does not support your confident and unqualified assertion.

Measurement equipment can measure down to -130dB in distortion, your ears are lucky to reach -60dB.
Measurement equipment can measure from DC to GHz range. Humans are rather limited.
Measurement equipment can discern between 999.8Hz and 1kHz, bet humans can't.
Measurement equipment can resolve 0.01dB, humans are lucky to reach 0.2dB level differences.
In the evening the hearing is more sensitive than during the day, equipment is not.
When humans have listened to loud music they can't hear soft sounds anymore, for a while. Measurement equipment does not care.
Easy to fool the hearing, less easy to fool measurement equipment.
Ever had a recording sound crap one day and the next day it sounded fine ?

Humans can enjoy music, measurement equipment can analyze it at best.

My unqualified assertion seems to be supported by the data above and am quite confident the above is true.
I am sure what Toole wrote is in some context as to what speakers sound good and which one less. Not that listeners are highly stable measuring instruments.

 

[pma]

When humans have listened to loud music they can't hear soft sounds anymore, for a while. Measurement equipment does not care.

.... and human ear is able to hear speech or music under noise level, microphone recording equipment is not .

 

[birdog1960]

What the heck is the average?

avg living room space US: It's around 12 X 18 feet, creating a living area of 216 square feet, although the average living room size square feet can be up to 230 square feet in modern homes. I suspect it's less in Europe. Most folks have a good sized sofa, an upholstered chair or 2 and drapery. carpeting could be an independent variable. but perhaps everyman is not your audience. posters here could be polled to an average loudness used.

 

[solderdude]

.... and human ear is able to hear speech or music under noise level, microphone recording equipment is not .

But... a spectrum analyzer can detect continuous tones well below a noise floor that humans can not.
That said.. sonar guys can hear things in noise that are not easily spotted by algorithms, the same with morse code drowned in noise.

Human ears can follow one voice when mixed with other voices or music or listen to what one instruments does in a small ensemble.
Measurement gear cannot do that. That makes the hearing suited for listening to and enjoying music which is reliably can and for communication.
Modern speech recognition can listen to a single voice only.

So there are things hearing can do well, because that's its main function. But accurately 'measure' it can not nor can it beat measurement equipment.

 

[MAB]

if it hasn't been done it easily could be. Consumer reports does frequency of repair polls on every car made and each system in the cars and yes, I realize their "lab" testing methodology leaves much to be desired. Even small market cars like high end porsches have ratings (spoiler: they're not good).

While I would love to see a non-controversial assessment of hifi reliability and quality, I hope you realize how unreasonable this is. Not easy at all. Most consumer products don't have accurate field service records that can be collected. Consumers are typically in the dark on actual quality and reliability, both good and bad. Legal action can shed light on systematic issues with quality and reliability (and safety), but even then we are often shielded from data that would allow accurate assessment. Companies make cultural decisions to work hard on quality and reliability, and this allows them to offer good warranties and after-sales service and simple policies. (I use Bryston as an example - 20 year warranty, no questions asked.) But it isn't going to happen like you say.

 

[birdog1960]

While I would love to see a non-controversial assessment of hifi reliability and quality, I hope you realize how unreasonable this is. Not easy at all. Most consumer products don't have accurate field service records that can be collected. Consumers are typically in the dark on actual quality and reliability, both good and bad. Legal action can shed light on systematic issues with quality and reliability (and safety), but even then we are often shielded from data that would allow accurate assessment. Companies make cultural decisions to work hard on quality and reliability, and this allows them to offer good warranties and after-sales service and simple policies. (I use Bryston as an example - 20 year warranty, no questions asked.) But it isn't going to happen like you say.

so Toyota doesn't have better reliability than Jaguar? I don't believe it's unreasonable at all. And at least in cars, price is often inversely proportional.

 

[Jimbob54]

avg living room space US: It's around 12 X 18 feet, creating a living area of 216 square feet, although the average living room size square feet can be up to 230 square feet in modern homes. I suspect it's less in Europe. Most folks have a good sized sofa, an upholstered chair or 2 and drapery. carpeting could be an independent variable. but perhaps everyman is not your audience.

Size of room and desired SPL drive speaker choice, which drives amp choice. That should be the main determinant on price. It isn't of course in most set ups.

I suspect if you give the hive mind here that space and SPL requirement, they could give you a few good recommendations for what most would agree exceeds "transparency" if you do decent room treatment and room eq. I guarantee the dac will be a very small % of the bill or more likely built into the avr. The speakers would be by miles the most expensive part.

Even more likely you'd get recommended digital active monitors and subs so speakers would be 100% of the cost.

 

[MAB]

so Toyota doesn't have better reliability than Jaguar?

Nothing I said can lead to that conclusion.
You didn't read what I wrote. Or you purposely misunderstood and misrepresented what I wrote.

 

[birdog1960]

Nothing I said can lead to that conclusion.
You didn't read what I wrote. Or you purposely misunderstood and misrepresented what I wrote.

sorry. that was trite. but the polls I mentioned don't require any of the things you mentioned. And I don't see the manufacturers quibble much with the results.
Size of room and desired SPL drive speaker choice, which drives amp choice. That should be the main determinant on price. It isn't of course in most set ups.
I suspect if you give the hive mind here that space and SPL requirement, they could give you a few good recommendations for what most would agree exceeds "transparency" if you do decent room treatment and room eq. I guarantee the dac will be a very small % of the bill or more likely built into the avr. The speakers would be by miles the most expensive part.
Sounds very much like what I have and I noticed a very notable difference going from a $275 yamaha avr to a $550 (now $1700) denon in a larger than average room. and an even bigger difference going from mass market $400 fronts to boutique $2000 ones. Maybe there's something in between that would do the trick. Maybe not.

 

[Stinius]

.... and human ear is able to hear speech or music under noise level, microphone recording equipment is not .

exactly right

 

[Killingbeans]

Yeahh, but add 60dB of music on top of that noise level and... good luck.

 

[solderdude]

exactly right

Does that mean that human is able to hear speech when loud noise is present but a microphone cannot record that and a speaker not reproduce that ?
Or.... does that mean that speech, when recorded at the lowest bit level without dither becomes unintelligible ?
In a thread a music recording was severely clipped and turned into a 1-bit stream and was still recognizable.

 

[Stinius]

I can make a 120 db SINAD amp sound like crap by pushing it into clipping. This is why you want to pay attention to full set of measurements. And power doesn't come that cheap.

In my opinion you should drive the amplifiers you are testing into clipping as well as testing them with a square wave. It will tell you a lot more about the behaviour of the amp than a meaningless 1k 5W into 4 Ohm test.
Stein

 

[Stinius]

Does that mean that human is able to hear speech when loud noise is present but a microphone cannot record that

Yes

 

[pablolie]

Yes

Conversation attempts in nightclubs or loud pubs don't quite support that. We've all been there. :-D

 

[Stinius]

Conversation attempts in nightclubs or loud pubs don't quite support that. We've all been there. :-D

Nonsense
Place a mic in the nightclub or pub and listen to the mess you hear.
Stein

 

[killdozzer]

You fail to understand that often when people are describing a listening observation they are referring to “perceived” sound qualities or characteristics.
Some of the triggers which produce these effects are well know by music producers.
Some engineers seem to lack the ability to understand this aspect of psychoacoustics.

If someone is describing some gear as “faster-sounding” he or she means that the sound/music is perceived as such.
Instead of shaming and LOL it is more useful and polite to investigate why that is so (e.g. often a lack of room resonances or low bass can produce that perceived effect).

No, you fail to understand that if it's: "“perceived” sound qualities or characteristics", then it's not transients. I hope you do know and accept there's such a thing as bias? If you do, let me meet you half way and agree upon a working hypothesis that says there are such properties of the sound that can't be measured, but people can perceive them OR that perhaps they could even be measured if someone bothered but they are always neglected.

OK. So, we have bias on one hand and we have unmeasured sound properties on the other. How do you know which is which? How do you differ among the two? How do you decide on; oh, yes I liked that speaker, but I only thought I liked it because it was bias for sure, but this other speaker I genuinely liked because those properties surely couldn't be bias, no, those were unmeasured sound properties?

Also, if you could explain what do you mean by perceived in quotation marks? Are they perceived or not? Why do you need quotation marks?