Accuracy in HiFi

I should not confuse accuracy as in measured and accuracy as perceived which is where my rational was going.

You are absolutely correct Jerry in that transducers are always in themselves inaccurate, this especially applies to cartridges and even more so loudspeakers. So if you are trying to build a totally neutral accurate system your on a hiding to nothing because the inaccuracy of the transducers will always add some distortion.

The same goes for amplifiers. The voltage amplifier will produce less distortion although power amps can contribute a fair bit of distortion.

Adding these factors together makes it difficult ever to have a accuracy in a system especially if the listener is convinced that because their measured performance looks good it should therefore be an accurate system for music reproduction.

Haha at least your being honest to yourself.  

Like I said earlier in the thread if the differences between recordings are obvious, you're probably close enough for it not to matter much after that. I mean differences in the character of recordings, not the quality of them. If you don't even notice that when moving from one album to the next then the system's a long way from accuracy. Which isn't to say it won't be a good listen of course, but that's beside the point.

As for a trumpet really sounding like a trumpet - well it's never going to because even on the most accurate system in the world it's a recording of a trumpet you're hearing. So not only will the sound have lost something in the recording process it will also have had some compression applied and possibly other effects too. Compression also being the reason you don't have to worry about 'real' dynamics.

So all those issues are taken care of at the recording stage. Recordings are designed to sound good, not to be accurate to the live sound. Some Classical music, and acoustic jazz probably the exceptions but a lot of those recordings are still manipulated in some way. They'll certainly have some compression applied if nothing else.

My point is being missed.

What I'm saying is that if you were to give some people real, genuine, accuracy - which is open to however you want to quantify it - they won't want it - because that's how we are. Some people like macaroni cheese - I can't abide it. As I said give some people accurate brass, which can be quite an asault on ears, and they won't want it. And the rest.

You can look at in in terms of software and hardware. And then there's the personal subjective processing to finish it off.

You can record something's to be as accurate as possible, within the limitations of the recording media. If that's what you want. Simple vocals can be reproduced with uncanny realism/accuracy. Or you can bugger-about with them with multi-tracking or whatever processing you fancy, making them less accurate, less realistic. The word simple comes to mind here. The simpler the capturing process (mic into a simple one valve mixer into tape as opposed to mic into a mixing console with wadds of eq options) the greater integrity the end result will have, I would suggest.

Making an amplifier that accurately amplifies it's input, is, on the face of it, a not too difficult process, if you have the knowledge to do so. But that then doesn't mean to say you're automatically going to like what it actually sounds like. In the 70s when the "spec wars" raged, the Japanese were making amps with impressive specs, compared to homegrown efforts. But according to the mags (whether true or not) they were accused of sounding flat and sterile (not a superlative you hear much of now). And this was said to be put down to the liberal dollop of feedback applied to lower distortion. Sansui were the boys... Not content with feedback in their amps - they also deployed feedforward! Beat that! No idea how they sounded but the top of the range, the AU 919, I think, seems to be regarded by those that have had one.

Processing/subjective assessment. What can you say? We all have the say auditory process, but that doesn't mean we all hear the same thing, or all like what we hear, as obviously we don't.

Don't know about records being made to sound good, as opposed to accurate. Good on what? Decades ago it used to be said that the 45 single was mixed to sound good on s transistor radio with it's speaker limitations. No idea how true that is. There are specialist labels that go out of their way and exist solely to "accurately" capture and manufacture an artist's material as best they can.

I done know... A system must be pretty bad if it cannot differenciate between recordings. My mobile has no problem there. A bad recording is apparent on whatever it's played on. As they say, you can't polish a turd. Here's something. Why is it that when you hear a piece of music for the first time and you connect with it, it doesn't matter what the source is, it doesn't matter the limitations it might have, you filter that out and just connect with what you're hearing?

At the end of the day, it's all just an illusion.

Think I've waffled-on enough.

And what about 3D/Binaural? An illusion of "accurate placement" in a soundfield. Another illusion. As Herb, the man says, accurate - to what?

Again this is conflating two completely different things.

Recording - let's say we decide to record a bloke singing and playing his acoustic guitar. There's lots of ways that can be approached. We can try and record it as true to life as possible -so simple stereo mic set up, or maybe a single ambient mic. We can add no effects or compression to the recording. Or we can mic him up to buggery and the add all sorts of processing and effects to the sound. Lets say we do it both ways and have two different sounding recordings, A and B.

Now A is clearly going to be closer to the sound of the actual event of him singing an displaying his guitar in an acoustic space.

But this is completely irrelevant to playback. For playback both recording methods are valid. All recording methods are valid.

For a 'high fidelity' playback system we are only concerned that it reproduces the A or B recording as accurately as possible. So the A recording - let's say - should sound raw and dynamic and the B recording should sound slick and processed. If the playback system makes the B recording sound raw and unprocessed then it is not accurate.

It's not the job of the playback system to change the sound of the recording - whichever recording we listen to. It is for the owner of the system to have character, not the system itself (to paraphrase LKJ Setright).

if the recording has been processed to sound good on an average car stereo, or a transistor radio, then that is how it should sound through a hi-fi system. In reality most recordings are balanced so that they don't sound terrible on either. A compromise, yes. But that is in the concern of the recording process, not the playback process. The two are not connected.

As for some of the Jap amps of the late 1970s sounding flat and sterile despite 'measuring perfectly' you have to consider what it is for an amplifier to 'measure perfectly'.

Just because it has THD 1Khz at 0.00001% does not mean it 'measures perfectly.' It takes a vast amount of measurements to fully characterize the performance of an amplifier. If our (possibly mythical) flat and sterile Jap amp were to be fully characterized with measurement there is no question that we would discover that in fact it does not 'measure perfectly.' And that is why it sounds flat and sterile.

I prefer detail and intimacy to massive thundering scale and agree sometimes everything can get swallowed up by the bass excited by room nodes. In a nutshell I like near field listening but not so near that its like listening to a pair of cans.   

I'll have one last go.

Forget about the media. Forget about the recordings. That can go out the window. What's left? Your system, your room, yourself. Right?

You have a cd player. The perfect accurate source for the man in the street. Perfect Sound Forever, Philips described it as. And they all sound the same being so perfectly accurate, well, as near as damn it, don't they? Well, we won't go there for now will we. It's enough to say you have your accurate source.

Amplifier. Obviously one measurement is only going to tell you what's happening in relation to one given set of parameters, that's sort of the point some critics were making back in the day. Obviously low distortion isn't the whole story. The brain/ear interface is rather adept at filtering out or filling in what isn't there. Anyway, you have an accurate amp. You got it off your mate Billy (Whizz). Billy has Golden Ears. He knows accuracy when he hears it. It was the most accurate amp he had, until he got his new one, which sounded... more accurate.

Speakers. Well everyone with two ears (well one's enough) knows the easiest way to change the sound presentation of your system is simply to change your speakers (or room). So, surely, that has to be where accuracy takes a nose-dive. You got your speakers off Billy. They did sound more "accurate" in his room, even with your amp and cd. When you got them, a pair of Yamamoto's SL250s, you tried them in three different positions. You settled on them in position 3 - either side of the chimney breast and away from the corners: best image placement and bass. Not as accurate as Billy's sound, but life's a compromise. Billy had a listen and he preferred the sound from position 1, along one wall. Billy' a flash bastard so we tried his Italian Solid Silver speaker cables - £6000s worth. Had another go with positioning. This time Billy prefers position 3, and perversely enough, you now prefer position 1. So, 6 different sounds. Which is the "accurate" one? And we haven't gone hairshirt yet with red capacitors sounding more "accurate" than blue ones? Shakey stones improving "accuracy" or does my system sound more accurate on an empty stomach?

And I don't think it's a question of Steve G not getting it about accuracy. I think it's more like after all the gear he's had in his time, he doesn't feel the need to worship at that particular alter or don that particular straightjacket. Over the decades there's been tons of great-sounding "coloured" gear. And so you get "system synergy". Taking an overall holistic view. It's a valid approach. It might not be "accurate", but it's equally as valid, if ears are anything to go by.

Forget about all of that though. My point was simple. Which is, you can give someone 100% accuracy, in whatever terms you care to - and they won't want it. That's the way we are. Anyone remember panel reviews? When did you ever get a group of 6 listeners (or whatever) all agreeing as to the sound quality of an item? Never.