Measurements & Sound Quality

[esldude]

4 hours ago, Ryan Berry said:

 

It's a good question and not one easily answered, I think.  For me, it comes down to a bit of both, really.  There's some elements that are "impure" that people associate with good-sounding music that can give an product that doesn't look as good on paper an edge: the pops of the needle as it moves through the groove on a record, that "musical" sound of your favorite amp that ends up being distortion, etc.  But then there's also a point where designing something to be as measurement-perfect as you can make it turns out being something that just isn't pleasant to listen to, no matter how long you try and live with it.  Having taken more than a few R&D units home to try and live with a potential product for a few weeks that looked great in measurements, I'm convinced there's just some things that there's no getting used to. 

Often, it just doesn't sound natural.  Sessions are often cut short with you feeling exhausted after spending any decent amount of time listening to it, whereas something tuned to sound as good as possible with the measurements being secondary (not ignored!) winds up with wanting to spend hours playing music.  Occasionally, they both line up and you have something that sounds AND measures good, which further confuses matters.  I can only speak from our experience, but I doubt that we're unique in experiencing this.  It makes high-end an interesting market.  People love to shop online any more, but it really takes going into a shop and comparing different products to really know you're making the right call.  The gear isn't cheap, for the most part, so I think we owe it to ourselves to do more than compare some numbers on a spec sheet and call it good enough.

I think you have the circle of confusion problem Toole has written about however. 

Most recordings are very compromised and tuned for the gear used in its recording and production.  The best cleanest highest fidelity reproduction of that often is not going to sound the best.  Not because measured linear results are somehow antithetical to musicality.  Just the whole polluted point of reference problem.  

 

I remember way back when comparing the same recordings on LP, CD, and reel to reel pre-recorded tape years ago.  Consensus was and in some areas is vinyl LP is the standard reference of musical great sound.  To my and my friend's surprise RTR and CD in the early days sounded very close to each other in general balance.  LP was the odd man out, way out.  Sometimes this was good, sometimes it wasn't, but it was clearly the most compromised medium.  

 

So do I make a product that is great with LP and fuss over the unmusicality of better measuring gear? I wouldn't.  It seems only a matter of degree vs making cassette an analog reference or 8 track.  LP could be better than those, but it wasn't a high fidelity medium like early RTR.  LP is more like the MP3 of its time.  

 

For most gear you really don't need to go into a shop.  That is why there are so few shops.  If the gear is clean and genuinely of high fidelity to the input.  You don't need to hear it. 

 

(exceptions are transducers.  On the replay end of it speakers.  You do need to hear speakers.)

[esldude]

2 hours ago, Ryan Berry said:

 

I'd obviously tend to disagree here.  Especially when we're involved in a hobby that gets down to using material A vs. material B for shelves and components, whether cables need to be elevated off the ground, whether speakers need spikes, etc. to achieve the best sound quality.  We tweak our systems constantly to get that (what we perceive) 1% better sound.  So the idea of not comparatively listening to the different pieces available seems a bit ludicrous to me, especially considering how the closer to the source you go, the bigger the effect tends to be on the overall sound of the system. 

The reason there's so few shops, in my opinion, is because it's a very specific market.  Similar to the reason we don't have nearly as many camera shops as we once had, convenience has supplanted quality for the general public when it comes to audio.  It's more convenient to shop online than it is to go into a store and deal with a salesperson, so most consumers will gladly give up quality for ease of completing a transaction.

So will the dealer have your shelving material, and cable elevators, and spikes and speakers you have?  And if you are comparing two components suppose the shelving material perfect for component A is the wrong one for component B.  I don't see how this makes any sense.  If it all is so touchy and easily effected a dealer is even less useful.  You'll have to buy something and see what you can make of it to know. 

 

Consumers have always valued ease of transactions and use.  That is why LPs sold in the millions and RTR tape in the thousands only.  RTR was the higher fidelity medium, but LP was cheap and convenient. 

[esldude]

15 minutes ago, STC said:

There is no way during playback of two speakers to recreate the soundfield or soundstage will ever be natural to us unless each instruments sound is confined to a single speaker.

Yep that will work.  I've done this using 5 musicians close miked in a damped environment.  Then play back over 5 speakers.  It can have a sound that is very real.  It sounds real in your room, not so much real like where it was recorded. 

[esldude]

30 minutes ago, STC said:

 

This is similar to the experiment by Briggs and AR. This also shows that you don't need to capture all the radiation pattern of the instruments to recreate realism.

Yes I was thinking of both of those as I posted.  

[esldude]

3 hours ago, Jud said:

 

Alex, there was (may still be) an incredibly thorough website by a Pennsylvania professor who built a couple of these. The fellow also was active in the DIY Audio thread about this amp.  Yes, these are the big ones.  The price may or may not have been exclusive of case, but as I recall there were people who managed to get sources for those that weren't too bad either.

 

I'm at a holiday party at the moment, but will see if I can find links, perhaps tomorrow.

A true audiophile.  While at a holiday party, takes time to post and respond to other audiophiles.  

 

Man, it is Christmas, well almost.  That lousy sound stuff can wait until later.  

 

Besides how can they have a proper Christmas in Australia?  It is like the start of summer there.  Just ain't right. 

[esldude]

11 minutes ago, Jud said:

 

Most of the guests at the party were pretty interesting, but this one fellow was kind of a bore, and he was just about to show us the sixth picture of a bobcat family he'd taken with his phone (we actually get a fair number of bobcats around here, but his bobcats were bigger and better, which he was proving to us via his photos), so I took the opportunity to use my own phone for a minute.   

Man, I got some big bobcat pictures.  Should I share them??????  Somewhere or another on the PC I've got the recording of a rare puma screaming in the distance.  Well it sounds like a woman screaming, but its a puma.  

 

Or I could tell you about feeding a Micky D's hamburger to a bobcat once.  Don't have pictures though.  

 

Okay, so posting about audio was the right thing to do.  Still ain't right to have Christmas in the summertime.  

[esldude]

3 hours ago, jabbr said:

 

That’s the crux of the question here. What electrically correlates with “sterile” or “clinical” or “lifelike” “live” etc?

Looking over the history of devices described that way, you usually are talking very low distortion, very flat in extended frequency response gear that gets called sterile.  You see this over and over.  

 

Life like and live are a bit more variable.  It can be foggy imaging, a bit of noise modulation, a softening of the recording such as happens naturally when recording to tape.  Now for those who like it you get more nice sounding names for it.  3D imaging, real depth, lots of space, soft and inviting sound on a black background.  

 

Extreme example.  Mike a trumpet with a good quality microphone.  Listen to it live yourself.  That flat sterile gear does a pretty good job of purveying the real sound of a trumpet which is in no way a polite instrument heard anywhere close.   Contrast that with the trumpet sound on Kind of Blue by Miles Davis.  That is actually a lousy rendition of a trumpet in terms of fidelity to the source.  But it is a beautiful wonderful musical recording that in some ways has no equal.  It is one of my favorite recordings.  It immediately reminds me of smoke filled small jazz clubs I've been in back when smoking wasn't illegal.  Even then it is a softer smoother than live sound compared to being in the back of those small clubs.  

[esldude]

3 hours ago, jabbr said:

 

Thats the standard answer. Is it true? Evidence? If that’s correct then why doesn’t most audio reproduction sound more “live”? I listen to a lot of live acoustic and amplified music and I don’t describe it as sterile, so if you are correct, what gives? I’m looking for authentic, uncolored, reproduction and I know it when I hear it  

The honest answer is I don't know. 

 

I think some of it comes down to the deficiencies of stereo in general.  And such recordings are close miked.  It sounds much like listening really close, but you normally don't listen that close. It can't sound live in totality. So some inaccuracies may seem more live.  But again I don't know.  

 

The best recordings that seem to get pretty close with clean gear and not sounding sterile are those miked at some distance with a couple of microphones.  But that isn't the recordings people are used to hearing and many find it boring.  The other way is to record only one instrument and play it back one instrument to a speaker.  That sounds clean, clear and like the instrument is in your room.  Again recordings in general are not done this way.  And you hear live in your room, not live in some other space. 

 

I do see every time you get flat and clean with your average recordings with lots of processing people call it sterile.  

[esldude]

3 hours ago, Ralf11 said:

didn't JBL do some studies of flat vs. slightly rolled freq. response ("warm")?  I have not seen the studies but have seen them referred to...

Actually that is not quite what they are saying at JBL.  

 

They shoot for a flat on axis response.  They shoot for a very similar response just off axis with a slight roll-off for the off axis sound.  The sloping response is an artifact of windowed measuring technique using sweeps indoors with reflections.  When Harman(JBL) measures in an anechoic chamber the on axis response is flat optimally.  If it is measured with windowed measurements that aren't anechoic then you get the gentle slope downward as you ascend in frequency.  

 

Their extensive testing of this indicates people prefer a flat anechoic response to speakers.  

 

The flip side to this is when people measure in a room and EQ to flatness.  It sounds bright, because the on axis response is rising to measure that way in a room with reflections.