Accuracy in HiFi

'The Dust Particles of Information', that's a good level to aim for.

'Bad toilet training..' 

Yep, i'd agree with that. There has been a lot of gear here that made things sound stunningly good, but they made everything sound the same at the same time. When you get transparency and accuracy, the system should deliver whatever its being fed. Simple, but i don't think people have genuinely heard this kind of system often enough to recognise this.

I know it was a revelation when i heard transparency for the first time. Took a £100k system to open my eyes but it also had that sterility and uninvolved attitude thats no good to man nor beast. Its about balance

If it was sterile and uninvolving then it wasn't even close to accurate. Is that the same Chord/Kef system that you also said was hard and strident? 

The biggest upgrade you make to a hi-fi is a better sounding recording, whatever format. If it all sounds similar there is a problem. (Garbage in, Garbage out). I've heard some mighty expensive sterile systems demonstrated almost exclusively digital source.

If it is strident then there is almost certainly intermodulation distortion - this is nothing to do with digital/analogue and most probably the amplification. Speakers do produce IMD but mainly in the bass where we least notice it.

Uninvolving - as mentioned above - is due to noise and distortion masking the details of the recording. This could be the fault of any of the components individually or a result of poor matching. It isn't likely to be the speaker's fault although some speakers can be a bit too woolly-sounding none that are remotely competent and working correctly are.

I agree that vinyl fronted systems, whatever other faults they might have, almost never sound sterile and uninvolving (although it is possible to manage it - I have heard such systems on occasion). This is mainly to do with the innate properties of vinyl 'sexing-up' the sound with added reverb and so forth and this can mask other issues in a way digital does not.

Really the digital/analogue thing has nothing to do with accuracy, it is possible to get a vinyl sourced system to a more than acceptable level of accuracy with enough money and experience/knowledge thrown at it. It is easier to get a digital system to sound sterile and uninvolving though.

I think it important not to confuse detail retrieval with accuracy. No-one makes a recording that sounds hard and strident. Just because you can hear everything in the mix clearly is not an indication of accuracy, just clarity. If it doesn't sound good to listen to then it is not accurate to the recording. (excepting the very few recordings deliberately intended to sound 'nasty' or something your mate recorded in his garage on a ghetto blaster).

Okay so 'hi-fi' in common usage has come to mean a stereo system of any description or quality when it originally meant something more defined than that. That's fair enough. But the thread is about 'accuracy' and I don't think that word is open to interpretation.

I reckon that the more accurate a system is the more enjoyable it will be to listen to. I can't prove that of course. But it has been my experience. You always get people who say 'I listened to a perfectly measuring system and it was boring/ unsatisfactory/ fatiguing etc. But when you drill down it turns out that it wasn't a perfectly measuring system, just that it was digitally sourced and the amp had THD of 0.0001% at 1Khz. These are always the people who say that they have no idea about measurements and don't care in any case so we can probably forgive them for thinking the THD at 1Khz is a complete measurement of an amplifier. And of course no speaker or room is perfect.

Anytime people start talking about bad recordings I ask them to name a few. They never do, probably because they know someone will chirp up and say 'I've got that and it sounds fine on my system.'

This is the problem with 'fake' hi-fi. Shedloads of THD and IMD, but play a simple acoustic mix on it and these flaws subjectively enhance the sound. Play something with more tracks in the mix and the IMD makes it sound terrible 'Bad recording!' Or play some rock music and all the extra THD makes it sound as rough as a badger's arse. 'Bad recording!'

Fake hi-fi is a rabbit hole but I guess most of us have gone down it at some point. The system dictates what music you can listen to and you end up only buying 'audiophile' recordings of a bloke hitting a block of wood with a stick while a bell chimes because it is the only stuff that doesn't make you want to switch it off and put the telly on.

And then it's made worse by the long-standing but seemingly widely acepted bolllocks that the better the system is the more of your music that will become unlistenable becuase 'it's showing up the flaws in the recordings'. Totally arse about face.

Could you let me have the name of the artist hitting the block of wood whilst a bell chimes, I quite fancy that!

I think you have a good point there Jerry. How can reproduced recorded music be accurate or at least how can you tell it is accurate unless you were there at the original recording session. So much can be added or taken away in post production and as Herb Reichert pointed out some music is purposefully recorded without reverb in separate isolation booths so that when it is reassembled it does not bleed together to create a noisy indistinct mess.

I think recorded music is never accurate to the original performance because it is usually messed with in some form or other to sound how the studio engineer wants it to sound so that it will be ok played back on a massive variety of equipment for earbuds to full range high end speakers.

Then you have the problem of all the inaccuracies of your own system and the acoustic environment that you play it in. I am sure it never sounds like the acoustic venue that the original recording was played in.

No recorded music is a replica of the real event! And that's got nothing to do with hi-fidelity replay in any case.

The idea is high fidelity to the recording not to the performance. The finished recording is the starting point. What went before is irrelevant. I don't understand where this misconception came from.

Even if it is bad recording high fidelity's aim is to reproduce it as accurately as possible. Not try to improve on it. Not to try to make it sound like live musicians or real instruments. Just to reproduce it as closely as possible i.e with minimum noise and distortion and with a flat frequency response.

And the aim is high fidelity not absolute fidelity as reproducing the recording with 100% accuracy is not really possible. Your not looking to get bang on just as close as you can.

How do you know if it's reasonably close? As Jez said, when you are playing different recordings and what you hear is the character of each individual recording rather than the character of the system. Or at least when it is the character of the recording that dominates, anyway.

Regarding bloke with wood block, stick and bell music. I don't know what it is called but they play it at hi-fi shows. There's a bird warbling on it too, off and on.

If I want to lsiten to soemthing similar - which is rare - I stick on Pat Metheney 'As Falls Whichita'. You;d like that James, if you don't know it already and it's a lot better than bloke with wooden block. Don't know if that's any help.

Erm, it is, for me.   That is my objective.

I suspect that this is where hifi disappears up its own arse in the deliberate separation of accuracy to the recording and accuracy to the original event [assuming there actually was one - there often is in the music I mainly listen to].

E.g. I want to be fooled into thinking I am at the concert hall listening to the Chicago Symphony Orchestra play a Bruckner symphony.

The 'system' is my audio gear and room, the music delivery system [Qobuz streaming], and the recording equipment and engineering.

It just happens that I only have control over the first of those 3 parts, and the best way on average to get the best out of all the contributors is to have as neutral a home playback system as possible.

It's quite plausible that each individual recording can be made more accurate (i.e. to make allowances for inaccuracies in music delivery, and even more for the original recording engineering) by judicious use of EQ on the playback system.
Take this to a reasonable extreme - a jazz or classical recording from, say, 1925 or 1930.  Played back 'accurately' as per the hifi definition of reproducing the recording as well as possible, it's gonna sound pretty damn awful in real terms - and the one thing you can be absolutely certain of is that it sounded nothing like that at the actual performance!  But careful use of EQ can bring it back closer to plausibility.
The same principle extends to any recorded performance, only (hopefully) less EQ will be needed for a modern recording than for a 1930 recording!

Well you're on a hiding to nothing in that case, because it is simply impossible to accurately reproduce a live instrument in its entirety, on a Hifi system.

But I applaud your endeavour to do so.

Oh yes...I totally agree..

As soon as you put speakers on the end of a system, its game over.

Are we talking accuracy or transparency?

Well accuracy is measurable and not really up for opinion due to it's very nature.

Neutrality on the other hand......