Master Thread: Are measurements Everything or Nothing?

[Katji]

Look up the meaning of "analytical".

Amplifiers/whatever do not do analysis, they cannot "be analytical." No logic. Logic needs computer/software, to me. Chip logic is far from analysing anything, let alone sound.

 

[surrie]

I've had the opportunity to audition the Benchmark AHB2 for a week and it was the best amp I ever heard on all objective criteria: the sound was neutral and balanced, the soundstage wide and bass went extremely low. However, when I had to return the Benchmark, and swapped it for my trusty old Quad 303, suddenly the music came back! I am sure that the Quad measures much worse than the Benchmark, but for music I prefer the Quad over the Benchmark.

Has anyone had a similar experience?

That got me thinking about the metrics that we are measuring. We are aiming for total transparency, but maybe we are missing something? Should we aim for more? L:ike headphones where there is a reference frequency response, could there be something similar for power amps?

My suggested approach would be as follows: select a couple of amps that are famous for their subjective sound quality. Measure everything that can be measured. Then try to find the characteristic that they all share, but is missing from amps that measure well on transparency and power ratings, but don't sound great subjectively.

What could be the missing metric?

 

[JSmith]

Hello and welcome to ASR.

What could be the missing metric?

To be quite frank... there isn't one, everything possible is measurable and quantifiable apart from ones own perception (which varies wildly due to being human).

The likely reason you may prefer the less performing item is it may have a distortion profile that you find pleasing, which is fine. However you can't alter that... IMO it's better to have a transparent system, then alter from there as required to taste, rather than having a certain "sound" baked into a product.

I hope you find the resources here interesting and informative.


JSmith

 

[tifune]

I've had the opportunity to audition the Benchmark AHB2 for a week and it was the best amp I ever heard on all objective criteria: the sound was neutral and balanced, the soundstage wide and bass went extremely low. However, when I had to return the Benchmark, and swapped it for my trusty old Quad 303, suddenly the music came back! I am sure that the Quad measures much worse than the Benchmark, but for music I prefer the Quad over the Benchmark.

Has anyone had a similar experience?

That got me thinking about the metrics that we are measuring. We are aiming for total transparency, but maybe we are missing something? Should we aim for more? L:ike headphones where there is a reference frequency response, could there be something similar for power amps?

My suggested approach would be as follows: select a couple of amps that are famous for their subjective sound quality. Measure everything that can be measured. Then try to find the characteristic that they all share, but is missing from amps that measure well on transparency and power ratings, but don't sound great subjectively.

What could be the missing metric?

please start here

Audibility thresholds of amp and DAC measurements

Introduction A recurring theme on ASR is whether or not the various measured qualities of the devices are audible. In this post, I'll present some clear and visual thresholds for when those imperfections can be considered a potentially audible concern. I will not explain the basics of amp/DAC...

www.audiosciencereview.com

 

[Vacceo]

It is quite possible you enjoy distortion. There is nothing wrong about that. What matters is knowing what kind of distortion you're getting on your amp and work from there.

 

[mhardy6647]

It is quite possible you enjoy distortion. There is nothing wrong about that.

 

[Vacceo]

It is perfectly fine to enjoy Duchamp

Again, what matters is knowing what is going on.

 

[Mart68]

QUAD 303 in full spec working order won't have audible distortion or noise so assuming that and assuming it's not being pushed beyond its limits it shouldn't be doing anything different to the Benchmark.

This is another one of those situations where o/p needs to do some blind and level matched testing to see if there really is a difference that he can reliably identify. Only then is it worth looking into what might be the cause of the difference.

Lots of subjective reviews out there saying the benchmark is clinical and boring, just having read one of those is enough to create an unconscious bias that will alter one's perception.

 

[surrie]

The likely reason you may prefer the less performing item is it may have a distortion profile that you find pleasing, which is fine. However you can't alter that... IMO it's better to have a transparent system, then alter from there as required to taste, rather than having a certain "sound" baked into a product.

That thought has occurred to me as well. To test that hypothesis, I added a tube pre-amp to my setup when I still had the Benchmark. It did nothing to improve music, it only added distortion...

Maybe the comparison with reference frequency response is sending the wrong message. What I really wanted to ask, is: do we really measure everything that can be measured? Is there nothing to be measured that can explain the perceived differences between a transparent-only amplifier and one that sounds genuinely musical? For example, the amount of current that is pushed to the speakers.

 

[surrie]

QUAD 303 in full spec working order won't have audible distortion or noise so assuming that and assuming it's not being pushed beyond its limits it shouldn't be doing anything different to the Benchmark.

This is another one of those situations where o/p needs to do some blind and level matched testing to see if there really is a difference that he can reliably identify. Only then is it worth looking into what might be the cause of the difference.

Lots of subjective reviews out there saying the benchmark is clinical and boring, just having read one of those is enough to create an unconscious bias that will alter one's perception.

I didn't do a blind test, so you are right about that. On the other hand, I did not read any subjective review before auditioning the unit. I always try to avoid that and read the reviews afterwards.

 

[RHO]

Maybe you prefer the Quad for sentimental reasons?
Your brain could translate that in a feeling of better music enjoyment.

 

[anphex]

Maybe there was just a volume mismatch as in you played your old amp louder after pluggin it back in.
And as others already said, there is the thing where some people like distortion. In my case, I like NOS filters. Meaning using a DAC without any roll off filter. And boy did I see some brandished pitchfork when I started a thread about this.

So yes, it's totally possible that your old amp just suits your system better. Might also be due to your preamp (output sensivity or impedance) going along better with your amp. Or, you just like a small amount of distortion in your sound.

 

[Mart68]

I didn't do a blind test, so you are right about that. On the other hand, I did not read any subjective review before auditioning the unit. I always try to avoid that and read the reviews afterwards.

yep I try to do that too. I was interested in the Benchmark and one came up at a good price second-hand and I nearly bought it, but I'd already inadvertently read some of the reviews and they put me off. Bought another Krell instead because I know for sure I like them (and it was cheaper)

 

[DSJR]

I've had the opportunity to audition the Benchmark AHB2 for a week and it was the best amp I ever heard on all objective criteria: the sound was neutral and balanced, the soundstage wide and bass went extremely low. However, when I had to return the Benchmark, and swapped it for my trusty old Quad 303, suddenly the music came back! I am sure that the Quad measures much worse than the Benchmark, but for music I prefer the Quad over the Benchmark.

Has anyone had a similar experience?

That got me thinking about the metrics that we are measuring. We are aiming for total transparency, but maybe we are missing something? Should we aim for more? L:ike headphones where there is a reference frequency response, could there be something similar for power amps?

My suggested approach would be as follows: select a couple of amps that are famous for their subjective sound quality. Measure everything that can be measured. Then try to find the characteristic that they all share, but is missing from amps that measure well on transparency and power ratings, but don't sound great subjectively.

What could be the missing metric?

Being an owner of a good restored Quad 303 (and a similar sounding set of II's) along with vintage 1970's Crowns which still measure quite well even today and which I believe are spiritual ancestors of an amp like the Benchmark, I'm going to stick me oar in and comment thus -

Sure you like the 303! So do I as it's charming in sound BUT, it's the measured flaws (very low output current even into 8 ohm loads is one major aspect) and outdated single rail cap coupling to the speakers in this design which I believe is adding distortion or driving errors which gives the sonic 'charm.' I'd also suggest that perhaps recorded venues all tend to sound more the same with the 303? The Crown I love (barely warm running after hours and no temperament into standard loads at all) sounds a bit 'precise and dry' just as it always did at the time, but different recording techniques, venues and so on from album to album as well as track to track are clearly audible. Unless I'm severely mistaken, I suspect the Benchmark is exactly the same, albeit with very much enhanced distortion and noise figures. The 303 is probably acting as a subtle compressor and, charming as the sound is, this 'flattering' isn't really what's there in the recording (I've 'heard' so many amps now).

I could offer a bit more, but I'm going to venture into personal subjective vibes here. Just maybe (treading on red hot coals to suggest this), an Artera Stereo power amp might make for a great balance between the two, the 606 onwards era Quads being I strongly suspect more like the Benchmark... Can you try one of these - the Artera has been tweaked further and there's healthy output into 4 ohm loads, which earlier models didn't like much at all (professional 500 series a possible exception).

 

[Midwest Blade]

I think over the years we have all been bombarded with the notion that amplifiers have a sound or in some way contribute to what we are hearing. Endless subjective reviews have audio hobbyists chasing their tails over the latest and greatest, system matching and all sorts of other flabby ideas. Neutral, warm, round, smooth, fast, controlled, clear, bright, dull….the list of adjectives go on and on but tell us little.

Your speakers and whatever you listen to have the largest bearing on what you are hearing. Objective performance reviews of electronics provide us with the knowledge that what we put in is generally what will be coming out.

I have a Quad 405-2 from the mid 80’s, refreshed the guts a bit over 3 years ago. Still performs like new. Your 303 was designed to work with the Quad ESL speakers, an unbelievable difficult speaker to drive properly, your amp is a good piece of kit and may just need a check under the hood.

 

[Vacceo]

I think over the years we have all been bombarded with the notion that amplifiers have a sound or in some way contribute to what we are hearing. Endless subjective reviews have audio hobbyists chasing their tails over the latest and greatest, system matching and all sorts of other flabby ideas. Neutral, warm, round, smooth, fast, controlled, clear, bright, dull….the list of adjectives go on and on but tell us little.

Your speakers and whatever you listen to have the largest bearing on what you are hearing. Objective performance reviews of electronics provide us with the knowledge that what we put in is generally what will be coming out.

I have a Quad 405-2 from the mid 80’s, refreshed the guts a bit over 3 years ago. Still performs like new. Your 303 was designed to work with the Quad ESL speakers, an unbelievable difficult speaker to drive properly, your amp is a good piece of kit and may just need a check under the hood.

Well, some do. I´m thinking Pass Labs, but those cases are quite public showing that they do, indeed, generate particular forms of distortion.

 

[Kal Rubinson]

I didn't do a blind test, so you are right about that. On the other hand, I did not read any subjective review before auditioning the unit. I always try to avoid that and read the reviews afterwards.

Good. That limits the bias to one's own and that is unavoidable (for all of us).

 

[surrie]

Being an owner of a good restored Quad 303 (and a similar sounding set of II's) along with vintage 1970's Crowns which still measure quite well even today and which I believe are spiritual ancestors of an amp like the Benchmark, I'm going to stick me oar in and comment thus -

Sure you like the 303! So do I as it's charming in sound BUT, it's the measured flaws (very low output current even into 8 ohm loads is one major aspect) and outdated single rail cap coupling to the speakers in this design which I believe is adding distortion or driving errors which gives the sonic 'charm.' I'd also suggest that perhaps recorded venues all tend to sound more the same with the 303? The Crown I love (barely warm running after hours and no temperament into standard loads at all) sounds a bit 'precise and dry' just as it always did at the time, but different recording techniques, venues and so on from album to album as well as track to track are clearly audible. Unless I'm severely mistaken, I suspect the Benchmark is exactly the same, albeit with very much enhanced distortion and noise figures. The 303 is probably acting as a subtle compressor and, charming as the sound is, this 'flattering' isn't really what's there in the recording (I've 'heard' so many amps now).

I could offer a bit more, but I'm going to venture into personal subjective vibes here. Just maybe (treading on red hot coals to suggest this), an Artera Stereo power amp might make for a great balance between the two, the 606 onwards era Quads being I strongly suspect more like the Benchmark... Can you try one of these - the Artera has been tweaked further and there's healthy output into 4 ohm loads, which earlier models didn't like much at all (professional 500 series a possible exception).

The reason that I own a 303 is exactly what you say: I used to listen to ESL's. But then I moved to another house where the ESL's no longer fit. So I sold them and ended up with the smallest speakers that still produce good sound (Boenicke W5 SE). I have been on a search for a matching amp ever since.

Thanks for the heads-up on the Artera, could well be the amp that I'm looking for.

 

[Head_Unit]

2) To me it´s also curious that the vast majority oh the recommended speakers are cone type speakers. Very few horns, panels, omnis.... Why??
...bad measurements of Volti speakers or ATC. Volti has many good reviews...(see Stereophile)...It´s possible to find pleasure on a "imperfect" sound??
4) Is there any high efficiency speaker (99 db or more) with good measurements?? Any horn out there with good test bench measurements??

Music is mixed on forward radiating cone speakers, maybe with horn mid/highs. So omnis and panel are thus in some sense not accurate, that's one factor. Now they can sound very good (OMG giant MBLs playing a reel-to-reel of "Rapper's Delight" maybe the best thing I've heard ever), however they are especially sensitive to the room and placement so don't work for everybody. Also some panels really can't do bass, maybe not play super loud (unless giant panels with huge amps maybe). Omnis and panels also tend to be more expensive and more niche, "normal people" won't find them so easily.

The Volti, good reviews, the frequency response was not at all flat, very peculiar. Horns can be like that, there is *something* about their sound which can be very compelling. However why should the basic response be [vomit]? Can't we have both? Then to your #4: I'd expect high end JBL should be smoother measuring than horns of old days; newer regular line Klipsch as well. I still dream about horn speakers, I still hope someday to build maybe 12" two-way with giant horns. I just wish there was a good multichannel DSP crossover solution for that.

 

[57gold]

Seems like some folks believe that the top 25 or so DACs on ASR's SINAD list, the ones starting at the top and down through the "excellent" category are all going to sound the same playing music driving an amp driving speakers in a high resolution system. They all test great using this measurement and others, yet feature different DAC chips, USB receivers, some balanced some single ended circuits, some with robust linear power supplies for digital and analog stages versus outboard switcher power dongles, unique circuit implementations of the both digital and analog stages...hard for me to believe.

Like if I took the OKTO Stereo and the SMSL DO 100 and listened to them in my system playing a range of music - jazz, blues, rock, classical...I wouldn't be able to hear a difference between them? The $250 DAC and all the economies made to produce it versus a $1500 product...its all just the casework and the source country labor cost differential?