What criteria do you use to optimise your system?

Kindly bear with me for the introductory story…

About five years ago, I abandoned ‘live’ as a goal for my system’s music reproduction. (I use ‘live’ analogously for what happens in the recording studio.) ‘Live’ as a goal leads to an endless and ultimately unsatisfying journey. Instead, I have since pursued the most authentically possible reproduction of recorded music. This may sound academic at first glance. But it makes a big difference in practice. If I really wanted to reconstruct the ‘live’ experience, the system would have to create an illusion that ceases to be one. I no longer consider that achievable. My system and listening room were suitably advanced to draw this conclusion. Limitations of the recording and playback chain make the “vanishing illusion” impossible. Starting with the microphones, through the extensive production technology and ending with the stereo speakers. At the very latest, the possibilities end with the reconstruction of the vibe in the hall, the stadium, the club, the studio. Credible clapping or murmuring over loudspeakers is an insufficient surrogate for actually being there.

What did the change from ‘live’ to ‘authentic reproduction of recorded music’ mean for me in practice? First of all, it took quite a while to let go of the ‘live’ idea. The desire to experience ‘live’ while listening to music was strong and took time to pass. After it was gone, it felt like a bit of a release. Mentally and in listening practice, I replaced the live event with the recording. A latent layer of dissatisfaction while listening to music disappeared. It turned into “It’s good the way it is.” This freed up bandwidth to listen more deeply into the music itself.

This much about one perspective on how to optimise a system. Later I want to write more, but would like to learn from your experiences first…

What criteria do you use to optimise your system? What principles do you follow? What concepts do you use? When listening to upgrades, what questions do you have in mind?


This is a very complex and ultimately quite personal question. I have been thinking about it off and on since you posted. If I can find a reasonably brief way of responding I will post. If not…


+1 to Pete’s note. I bashed the heart icon soon after you posted, Marco, and have been mulling it over since. (Well, that, and packing a house. Gah.)

There are so many different things to take into account. I think you’ve nailed most of them already, and love the thought process you took to get there.

Similarly to you, I figure* that most recordings have had a great deal of work put into making the end result sound “just so.”

That work could take many forms:

  • record the whole shebang in one take, and apply as little processing as possible (I understand labels like Chesky take this approach)
  • various different recordings for each instrument (single mic, multiple mics etc.), with the mix created afterwards
  • no actual “recordings” (much of my beloved electronica falls into this group)
  • …and so on.

Each of the end results could be evaluated differently. There’s little point looking for a “live” rendition of something that was never live, for example.

In the years that I’ve been comparing ways of optimising my system, it’s always been in the service of finding which combinations and setups are likely to have “Future Ben” grinning the most as he listens vs. compares. It’s often not the most impressive-sounding setup that does that. It’s the one that has my foot tapping earliest in a track, or the one that keeps me finding yet another album to listen to.

Looping back to how to evaluate a system’s sound, though, I think it’s a safe assumption that artists (and recording engineers) just want their music to sound fantastic when people like us play it back. (I’m deliberately ignoring the pursuit of loudness and what it’s meant for dynamic range.) They’ll do what they can to put that on the disc/file they sell or stream to us.

When my system manages to play music of all genres back to me in a way that makes me happy — I’m…err…happy. I once read a note along the lines of “if your home system makes you grin as much as your car’s system does on a nice drive, you’re on the right track.” I liked that. If the artists and engineers all decided that something was good enough to sign off on, then there’s probably some enjoyment to be had in listening to that piece of work. If I don’t enjoy it, then it could be me, it could be my playback system, or it could be that they all had terrible taste :wink:

I guess the best optimisation metric I can think of is “Number/duration of grins for a given configuration” :stuck_out_tongue:

*All this is from the point of view of someone who has almost zero experience of recording audio — I may well be making myself look very silly. That’s never bothered me before, though, and I love the topic, Marco :+1:t2:


That is it for me too!

Coming from the professional audio world, both live and in the studio, I know what “they” did/ could do, and what is left in a recording.

The budget granted for a recording by the record company/ publisher determines in a large way what you can expect from the end result. It determines the quality of the studio, the producer, the engineer(s), and the time they get to finish it. Or equally for a live recording.

If you have ever been into a control room, or as a engineer in a live venue, then you can forget to hear the same result at home. Not even close.

But we all have to do with the recordings published. Some of them are really good, and as Ben says, if they make you grin, then it is OK.

Live music is loud. Live music recordings should be played LOUD. If your gear and room and neighbours can have that, enjoy!

To add: loud can be done in several different setups. Nearfield listening is one of them, a way they also use in studios. The monitors and you are in equal small triangular distance of each other. This is what I (have to) do.


Agreed on volume. If something was supposed to be loud, I’ll try to play it loud. If it was live and unamplified I’ll anchor the volume in a way that makes sense for the material. “No one can sing that loud” or “the guitar strings would probably break if it were played that heavily” — that sort of thing.

My dream is to have a system replaying violin music that actually sounds like a violin.


There are many different but perfectly valid ways of optimising a system which will ultimately depend upon the owner’s objectives. I am a classical music listener where my approach may be fundamentally different to, say, someone trying to emulate a club sound system to play (and move to) dance tracks.

Classical music does have a recognisable and more or less consistent way in which the music is presented in recordings which, in turn, normally relates to the tradition of how performers may be grouped in live concerts or recitals. Other genres often have no relation to live performance at all, being synthesised in the studio and so the resulting presentation of the music by your system may be difficult to judge or validate .

It may not be possible to reproduce live sound in an absolute or specific sense but, for me, the sound of live acoustic music recalled from decades of concert going and how the music itself communicates are nevertheless the main arbiters when judging how well I may have reached my objectives.

So I am not really equipped to recommend any particular form/technique of optimisation as being universally applicable. However I suggest these points for consideration in the hope that they may be useful:

  1. Decide what your objectives are. Don’t approach the task randomly.
  2. Sound and music are related but are not the same. Do not mistake good sound for good music.
  3. Knowledge can be useful or dangerous. Be wary of trying to optimise one theoretically important aspect to the exclusion of all other considerations.
  4. Nearly all of us need to make compromises even those with the luxury of a dedicated music room. Don’t become obsessed with those compromises and contract audio nervosa.
  5. Most of us do not have a dedicated music room so bear in mind that you (and maybe your partner) have to live with the system and what you have done to accommodate it. You will probably be doing things in the room other than listening to music. Maybe you will actually spend more time looking at it than listening to it!
  6. Don’t judge the result just by listening to your best sounding recordings. Include the worst sounding too (has it/they improved?).
  7. At the end of the day remember that this is home entertainment and not a scientific experiment.

Thank you Pete, Ben, Erno, Katzky for your valuable replies. I find the different angles fascinating and the reasoning is illuminating.

Based on your posts I began writing a more in depth reply. Coming soon…

Thanks again for the contributions. It is good to see how differently we approach the question.

First, let me briefly explain why a more detailed thought experiment on system optimisation might be useful. Visual art and music speak to a deeper level of happiness within us than, say, enjoying a lime sorbet or an episode of Game of Thrones ever can. That’s why I think listening to music in a concentrated manner at home goes beyond the concept of home entertainment. Just as art on the wall serves the purpose of decorating the room, but also deeply appeals to our intellect and emotions.

I see the audio hobby as a journey. Translated into a week- or month-long hike, this would mean that I decide where the stops might be and which place is the final destination. Compass and map help me to arrive at the desired places. If I then make detours, they are intentional and for pleasure or curiosity. Plus, I get to the final destination in the end. If, on the other hand, I often ask not only for the direction but even for the right destination along the way and everyone tells me something different, then that can also be interesting - probably also marked with dead ends, detours and circles though.

As Ben @all2ofme and Erno @Ermos said above - “Number/duration of grins for a given configuration” ingeniously covers it. That’s probably already the better compass than the one suggested below. My best attempt to cast everything into just one sentence would be “spending as many meaningful moments with music as possible”.

The question of optimisation criteria can be broken down into parts and thus made manageable. With which contents someone fills the framework described here is completely individual. However, I consider the framework transferable to all high-end journeys. The idea is adapted from the field of software design.

The whole thing starts with use cases. This is where it is decided how the system will be used. Everything from secluded indulgence to enjoying music together with friends can be captured. Possible use cases could be, for example:

Enjoying music . immersing oneself in the music . dancing . mirroring or triggering emotions . experiencing stories . venting . finding refuge . achieving healing . filling silence . facilitating routines . show off . nurture relationships . view with pride of ownership . evaluate . tinker

Part of the requirements result from the use cases. Another part of the requirements arise from the inclinations and aspirations of the users.

Aspirations could be:

(self-directedly) put together the ‘best’ system one can on a given budget . playback of recorded music as authentically as possible . spend as many meaningful moments with music as possible . immerse oneself as deeply as possible in music.

Inclinations would possibly include:

following a concert intellectually . intuitively perceiving music . leaving everyday life behind with music . tracking the sound of a recording . maintaining a music collection . experimenting with hi-fi.

If one adds the preferences for music reproduction, most of the requirements already become apparent. Requirements can be classified according to importance, e.g. essential (non-negotiable), important, nice to have or dispensable. This makes the decision easier in edge cases. An essential requirement will then not give way to one that is only ‘nice to have’. Here I have only listed examples of essential and important. Following Pete @PAR I distinguish between music presentation and sound. The difference can perhaps be illustrated with an example. Sound would describe what it sounds like when the drummer hits the bass drum - the reproduction of the moment when the mallet hits the skin. Music presentation in this context would mean how the drummer plays together with the band and how the bass line creates the groove.

Examples of music presentation requirements might be:

essential . it makes music as opposed to it shines with sound . the artist’s intention is revealed . the music presentation does exactly what the particular performance was likely to be about - e.g. groove, stimulate, soothe, arouse emotions, stimulate the intellect, appeal to a deeper level within us, make the heart leap with joy . be able to listen deeply into the music . the music moves one internally . the music entices one to groove . easy intuitive access to the music as opposed to intellectual access.

important . easily follow the individual musicians during the piece . the composition remains intact . the musicians play with each other and it becomes obvious how they play with each other . remain composed and accessible in complex passages . be taken out of everyday life . discover what the music is about - emotions? poetry? structures? time? . be able to rest in the music . be able to follow the individual plot lines . the inner context of the music is revealed . open up or unlock the music . the colouring of the music becomes immediately clear and is sustained e.g. elegiac, lyrical, joyful, sinister, angry . if given, be able to learn from the music . as short a time as possible until one can direct ones attention exclusively to the music and the body . as captivating and satisfying as possible . Reactions such as curiosity, being touched or forgetting time.

Examples of requirements for the sound:

essential . accurate in timing . the colours of the music become clear when the recording allows it . gripping . engaging . long listening sessions are enjoyable . one looks forward to music listening and does it often . gets under the skin

important . different recording scenarios remain intact . reproduces drive + punch where it is in the music . analogue . organic as opposed to sliced and diced . lyrical where it is lyrical as opposed to “here are the facts” . PA when it was PA . visceral playback . free of “fabric softener” . playback electronics largely disappear.

Requirements for the equipment

essential . low-maintenance . failsafe . technical issues such as number of inputs or outputs are appropriate . fits in the room . WAF

important . proper industrial design . tolerable resale values

For the actual optimisation, it is also interesting which yardsticks we use, how we find and select upgrades, and whether we want to make overriding considerations.

Yardsticks can be:

Concert . other live music . visit to recording studio . visit to fairs . listening tests at acquaintances or dealers . the heaviest possible power amplifier :wink:

Selection of upgrades via:

trusted testimonials . experience of others who have gone before . press . forums . trial listening . own experience

Design concepts

Elegance . the whole before the detail . Focus on how mind and body react (emotions, reactions, thoughts) . Intuition says “yes” . Source first . Empiricism

Of course I don’t walk around with a checklist. Most of it is automatic and intuitive. After all, experience also plays a big role. However, this is an attempt to make what is going on explicit, repeatable and discussable. It’s a complicated compass, but one that works reliably, at least for me.

What are your thoughts on this?


Idk… I was fortunate enough to dine at Alinea during its prime. That was a transcendent experience. Also… I spent a month living in a ghost town in central Alaska. It took two days to hike out to the nearest town. When I got there, I went to this restaurant and ordered some real food. The server brought me some bread and butter. The butter had a bit of sugar in it. I don’t remember anything about my order but that bread and butter remains the best food I’ve ever eaten. That was nearly 30 years ago and I can still remember the sensation.

Sadly I lack ambition required to optimize anything. Can’t even bother to ensure the shelves on my rack are level. The most I’ve done is a half assed attempt to prevent power cords from touching signal cables. I did spend $2.5 million worth of bitcoins to buy stillpoints back in 2013. Currently can’t be bothered to actually try to use them.

My primary strategy has been to only buy Naim brand products. They’re the perfect solution for apathetic audiophiles. Last thing I want to do is try to decide questions like “which bnc cable sounds best?” Naim makes everything you need and you’ll know that you’re hearing what the engineers intended. The upgrade path is clearly defineD so you just need to determine how much money you want to spend. Finally it’s a cult and once you’ve purchased an expensive Din cable you’re never going to leave.

Prior to this summer, the only upgrades I’ve ever done have been to my source. Started with nd5 quickly moved to ndx. Then the Chord Hugo was all the rage so I used that via ndx. Later it was replaced with a Berkeley Alpha and briefly a Berkeley reference. This year got a Bartok and it was the first source I felt deserved an upgrade downstream. Followed the Naim formula and moved up about five levels in preparation for my incoming Rossini


I traversed the Naim path for a long time and to the very end. After having built a basement listening room from scratch with a company that designs and builds recording studios, there was not much more to upgrade except speakers and maybe a unicorn record player. The speaker candidates would have been so rare and heavy, that I would have had to travel the globe to listen to them. That was the moment where I decided this is taking too much emphasis in my life. Sold the Naim system, gave away my record collection and now only use headphones with Vivaldis. I compromise on certain aspects by doing that, but the enjoyment of music is of the same or higher quality. And I get to see the garden and library while listening. :smile:


I bought Naim 90% because I thought it would be fun to join a cult. After looking into the many cults in existence Naim felt like the safest. I went 100% Naim from floor protectors under my speakers to power cords. Even though I’m a bit famous for having sold $30 million of bitcoins to buy the system, I still enjoyed it. My daughter and I spent 100’s of hours listening to music together. She now has an excellent understanding of the history of 20th century Western music and I’ve learned what “the kids” are listening to these days. Weird thing is she’s 15 and listens to a lot of jazz while I’m now heavily into Billie Eilish.

Prior to this I was 100% headphones my entire life. My thought process was that I might one day afford the best possible headphone system but there is no upper limit on speaker systems. At this point my speaker wire retails higher than my speakers did in 2013. They are the last thing I need to upgrade in my slightly morbid quest for moderately higher fidelity. Unfortunately Naim no longer makes speakers so I’m stuck. The options here are practically limited to Wilson. Unfortunately no home demo is possible. If I do get Wilson speakers then I’d sell off the Naim system and replace with a Diablo 300 as recommended by my local dealer and use the new Linn Klimax streamer as my source. The Rossini is going to go to headphones only.

This is all pretty ridiculous and wasteful considering I can only listen to my hifi a couple hours a day before my octogenarian neighbors start treating me as a reasonable person would a war criminal. Also my family are prone to recommend I listen at 60 dB which is just pointless. I generally listen at 70 dB. Ultimately I’ll probably revert back to 90% headphones and the speaker system will mostly serve as an annoyingly expensive sound system for television & movies.

The Linn dealer is going to setup room optimization with my current speakers. As a believer in technology, I’m slightly optimistic that the improvement afforded by this will put decreased pressure on the need to buy new speakers.

Hmmm… what’s my point? Not sure anymore… The rest of the world would be happy if I listened to headphones. With my budget and space, headphones will sound better and I’ll definitely enjoy it more. Possibly I’ll end up like you in the not too distant future.


If you want to join a cult that is less expensive than Naim, Linn and dCS, a cult that leaves your family and neighbour in peace…go to chruch, sometimes they sing and even play organ :laughing:


:notes: :innocent: :musical_score:

Live music is better anyway. Let’s sell and go. :wink:

1 Like

The music in my church almost convinced me that there is no God.

Matthew 22:14 - For many are called, but few are chosen. :grinning:

1 Like

Unfortunately I’ve been to my “local” temple. Turns out Buddhists aren’t big into singing. We are, however, really into long stretches of being uncomfortable. It was all so depressing tbh after realizing I’ll be reincarnated as ant since I burned one alive with a magnifying glass when I was five. Really I was just amazed that I could create fire with the thing and the ant was a casualty of curiosity. Felt guilty for years then I read Camus so it’s all good.

1 Like