Jump to content

Every Component has an audible effect, when actually heard.


Recommended Posts

We live in a time when there is extensive choice available to us to choose how our audio system can reproduce audio. We don't have to be dictated to by measurements alone, and steered toward equipment that measures well,  but then getting it home,  sounds well short of what we expected.

 

Worse would be contacting the manufacturer and finding a non caring response, like "it measures well, and we don't ever listen to what we manufacture". .... Rather our expectation should be to take home a piece of equipment, that has received praise from others, that indeed measures well, but has gone one very important step further,  to have the manufacturer deeply involved, just as we are listening,  and doing everything humanly  possible, by choosing every component internally in the equipment , to deliver the best possible audio experience for us. 

 

We should only choose equipment,  having this caring for us, ( no doubt difficult to do ) ...  correct outlook.

 

 

  • Like 2
Link to comment
Share on other sites

Hi S-C, your opening post seems to be about whether a manufacturer "cares", or "listens". 

 

However the thread title is "Every Component has an audible effect, when actually heard."

 

Could you elaborate, please?  (I'm not sure I quite understand what subject matter you intend that this thread should cover.)

Link to comment
Share on other sites
3 hours ago, MLXXX said:

Hi S-C, your opening post seems to be about whether a manufacturer "cares", or "listens". 

 

However the thread title is "Every Component has an audible effect, when actually heard."

 

Could you elaborate, please?  (I'm not sure I quite understand what subject matter you intend that this thread should cover.)

Every manufacturer should have a facility to measure and listen to not only the finished product ,  but uniquely have ability and level of care  to hear the contribution every single component contained in that device, is  then making to audibility. 

 

In this manner components such as discrete parts bjt,jfets mosfets , shunt devices,    Vref/R regulators, series regulators  cascode stages, differences of P channel vs N channel fet devices, circuit board layout  etc, that have better ability with audible effect  than other components, in the design,  are chosen, .... then actually heard.  

 

This better, quite difficult approach,  then presents a product that typically gets feedback from customers, that it has improved the reproduction of audio, in their audio systems, ... which should be, every manufacturers goal.  Consumers should search out manufacturers that listen to their products, like this one, In this video Paul ironically learns from a visit to a audiophile - no doubt then taking that listening experience and what was conveyed by its owner as to the detail of what was making that possible,   to further his own product.   

 

Any product is the combination of a huge number of components. each component uniquely changes the audio presentation of the next component, due to relationships that are never in individual datasheets, rather arise because of the combination of one part to the next. Each component has an audible effect, when actually heard. 

  • Like 2
  • Thanks 2
Link to comment
Share on other sites
3 hours ago, stereo coffee said:

Every manufacturer should have a facility to measure and listen to not only the finished product ,  but uniquely have ability and level of care  to hear the contribution every single component contained in that device, is  then making to audibility. 


Agreed,

but in the real world that is not the case…. Lots of dodgy stuff out there, and there’s lots badly designed and construction based due to costs.  Some components out there that are designed specifically to reduce or eliminate unwanted effects that can harm the audio path do exactly the opposite…..  Ive seen this over my journey on SNA..   Jitter is one subject that seems to dominated SNA where several major players have got on this site selling there craft at ridiculous prices, and you can see this in the classified….. or ASR. 

 

 

3 hours ago, stereo coffee said:

 

In this manner components such as discrete parts bjt,jfets mosfets , shunt devices,    Vref/R regulators, series regulators  cascode stages, differences of P channel vs N channel fet devices, circuit board layout  etc, that have better ability with audible effect  than other components, in the design,  are chosen, .... then actually heard.  

 

 


The design and layout all contribute to SQ.   Passive components such as capacitors and different types of resistors and pcb traces can really impact circuit SQ…..    Take resistors etc,   We know that metal film resistors have a fraction of thermal noise compared to carbon with a whole host of other benefits, however the Audiophile community seem to prefer a specific brand and series and it’s all carbon film!   All SS; BJT, jfets etc are all different in SQ.  and there inherent advantages and disadvantages are layed out bare to hear.

 

3 hours ago, stereo coffee said:

.  

 

This better, quite difficult approach,  then presents a product that typically gets feedback from customers, that it has improved the reproduction of audio, in their audio systems, ... which should be, every manufacturers goal.  Consumers should search out manufacturers that listen to their products, like this one, In this video Paul ironically learns from a visit to a audiophile - no doubt then taking that listening experience and what was conveyed by its owner as to the detail of what was making that possible,   to further his own product.   

 

Any product is the combination of a huge number of components. each component uniquely changes the audio presentation of the next component, due to relationships that are never in individual datasheets, rather arise because of the combination of one part to the next. Each component has an audible effect, when actually heard. 


Unfortunately Paul doesn’t practice what he preaches.   Check out his latest DAC utilising the latest ESS Sabre chip.   You can’t  tell me that destroying the performance of this chip with heavily increased distortion, with low S/N is going to make a good sounding dac? 

Edited by Addicted to music
  • Like 1
Link to comment
Share on other sites

35 minutes ago, Addicted to music said:


Agreed,

but in the real world that is not the case…. Lots of dodgy stuff out there, and there’s lots badly designed and construction based due to costs.  Some components out there that are designed specifically to reduce or eliminate unwanted effects that can harm the audio path do exactly the opposite…..  Ive seen this over my journey on SNA..   Jitter is one subject that seems to dominated SNA where several major players have got on this site selling there craft at ridiculous prices, and you can see this in the classified….. or ASR. 

 

 


The design and layout all contribute to SQ.   Passive components such as capacitors and different types of resistors and pcb traces can really impact circuit SQ…..    Take resistors etc,   We know that metal film resistors have a fraction of thermal noise compared to carbon with a whole host of other benefits, however the Audiophile community seem to prefer a specific brand and series and it’s all carbon film!   All SS; BJT, jfets etc are all different in SQ.  and there inherent advantages and disadvantages are layed out bare to hear.

 


Unfortunately Paul doesn’t practice what he preaches.   Check out his latest DAC utilising the latest ESS Sabre chip.   You can’t  tell me that destroying the performance of this chip with heavily increased distortion, with low S/N is going to make a good sounding dac? 

But where cost of all the extensive listening assessment required is borne by the manufacturer,  then to the consumer it is free. It's an extremely difficult process to develop a product in this way, and many will straight away take the cost argument, to not do it.  But to others who really care, it is vital to provide the best possible sounding equipment

 

There are alternatives in every application, that remove the non listened to, short cuts on components  others may take. Convenience of connection as example such as found with single cable digital equipment, is replaced with individual connection of Bitclock, Data, and Left Right Clock , that would straight away explain itself, when auditioned. PS audio  ESS sabre dac,  proves the value of consumer feedback to any product release, to do it better next time, and spend more time listening. Straight away I would see the reconstruction filter as problematic, and the one size fit all approach the DAC design has taken on. Hopefully Paul is also providing valuable feedback to ESS sabre, if each want to remain in confidence with consumers.

 

It is not easy to assess one single component, let alone two or 57 or more, when they all have relationship changing each other, but when listening to every component, it goes without saying,  that ease is what is removed, and that is, to such a manufacturer.. what is good.  

 

 

 

Link to comment
Share on other sites
1 hour ago, muon* said:

Look at valve, vinyl, old dacs.

 

Ultra low S/N is not always the answer for great sounding systems :)

If each of those good historical examples that have inherent noise issues challenging signal level, could with far more effort at the same time also have ultra low S/N, as well without sacrificing audio quality.... there is the challenge.  How would we do it ?  We might have to resign to only have low S/N ...but there is no harm in trying.

 

Firstly we would examine every past effort in this regard, and with very extensive research, and listening hopefully arrive... with yet a better solution. Here is as example ,what worked very well in the past.  Forward to 7.25 to hear the S/N difference

 

  • Like 1
Link to comment
Share on other sites

An often referred to aspect of audio equipment, is one piece of equipment vs another has good ability with portraying depth in a recording. Referring to a live recording , lets say Jakob Bro "Bay of Rainbows"  the depth on offer is not just the overall recording but quite a combination of many aspects. The venue of recording contributes as do  if we listen it is individual instruments each displaying their own ability to resonate. A good audio system will be able to separate for you, all of these aspects.

 

What is required though to hear this separation of where depth occurs. We would begin with a source component with good ability with dynamic range capability. If we were referring to a CD player, our experience might not be as  rewarding with a 16 bit player, as it would be with a 24 bit capable device. If it was a MP3 file vs a HiRes Flac file we might spot the same difference that one form of file was allowing us to hear more into the recording than another form, just like having available more dynamic range . But what else contributes ?  we could not leave our attributing with all that we hear just to the type of file, no there is a lot more contributing. in CD players or DAC's, and  this takes the form of a current to voltage conversion stage - also called a I/V stage.  But behind the I/V stage is a power supply supplying current and voltage to each stage in the CD player. We would hope our financial outlay might permit separation of a digital supply separately to a analog stage.... but why... the answer begins with how grounding of each area is done. This article gives us a plan of what is not so good, vs very good.    

https://www.analog.com/en/analog-dialogue/articles/staying-well-grounded.html

 

Link to comment
Share on other sites

We also need to examine ( basically everything ) but lets then look at what some sections of industry accept as sufficient , vs what others might view as well worth while implementing if the possibility of better sound quality is the design goal....    Yes price and parts costs begins to dictate for some manufacturers sadly what is actually achievable...

 

For others what is actually achievable has higher relevance than stooping to cost cutting, with parts that measure well, but then find the finished product lacking ability to achieve required audio result ....its yet again up to you to seek out such manufacturers who go one step further and apply component choice wisely, related to achieving the best possible audio quality.

 

As example, a very common part related to power supplies is a bridge rectifier, its low in cost, and provides a result, but is it the best that industry can really do. ?   No they can do much better, so why don't they ?  The answer lies again in how many manufacturers are evaluating their equipment both by measurement and listening.

 

The graph shows a LT4320 rectifier, ironically higher in cost than a bridge rectifier but then presents as  saving cost concern of heatsinking   A good designer will see quite a bit more though than just the sales pitch of the LT4320 manufacturer. Such a review might ask what are the thermal effects on diodes, what causes thermal effects. He or she might also ask comparison of different mosfet types when using the 4320, listen to each and make decision of which part to use. it might also lead to examining reverse bias leakage, and its effect with audio circuits, and moving to better well known alternatives.

 

 

LT4320-18084.png

Link to comment
Share on other sites
54 minutes ago, stereo coffee said:

As example, a very common part related to power supplies is a bridge rectifier, its low in cost, and provides a result, but is it the best that industry can really do. ?   No they can do much better, so why don't they ?

Because, I would think, "industry" wouldn't expect there to be any discernible difference in the audio output. As for heat sink size, if the bridge rectifier is passing a very substantial current then chances are the enclosure for the device will be pretty large anyway. As long as there is a reasonable distance between the power supply circuity and sensitive audio circuitry, conventional design wouldn't expect there to be a problem.

 

I can see your point that an LT4320 could be a better design choice for reducing heat dissipation. However I don't see sticking with a traditional bridge rectifier as being particularly likely to affect audio quality.

 

If you're aware of an actual instance where audio quality was significantly compromised through use of a bridge rectifier, I'd be interested to know about it!

 

Edited by MLXXX
  • Like 1
Link to comment
Share on other sites
6 hours ago, MLXXX said:

Because, I would think, "industry" wouldn't expect there to be any discernible difference in the audio output. As for heat sink size, if the bridge rectifier is passing a very substantial current then chances are the enclosure for the device will be pretty large anyway. As long as there is a reasonable distance between the power supply circuity and sensitive audio circuitry, conventional design wouldn't expect there to be a problem.

 

I can see your point that an LT4320 could be a better design choice for reducing heat dissipation. However I don't see sticking with a traditional bridge rectifier as being particularly likely to affect audio quality.

 

If you're aware of an actual instance where audio quality was significantly compromised through use of a bridge rectifier, I'd be interested to know about it!

 

It's the "wouldn't expect" and "particularly likely"  part in your reply that explains all, because how will they really know unless  one component  relative to the other,  is listened to. ?  No what invariably happens is a decision that is based on mundane cost cutting, alone, occurs,.

 

Whereas what should happen is the better approach that this thread voices, namely that research is done to both measure and listen,  to all of the alternatives for the part involved,  regardless of cost, difficulty and complexity.  ... Partial answers, based on unrelated criteria like cost,   are achieved more easily, than full answers,  that take much more effort

 

Within that alternative overall design , individual components are similarly compared. If that is properly done the now revealed cost cutting of the "wouldn't expect" and "particularly likely"  compromised approach is actually heard, and can be discarded.

 

 

 

 

 

 

 

 

 

 

  • Like 2
Link to comment
Share on other sites
1 hour ago, stereo coffee said:

It's the "wouldn't expect" and "particularly likely"  part in your reply that explains all, because how will they really know unless  one component  relative to the other,  is listened to. ?  No what invariably happens is a decision that is based on mundane cost cutting, alone, occurs,.

 

Whereas what should happen is the better approach that this thread voices, namely that research is done to both measure and listen,  to all of the alternatives for the part involved,  regardless of cost, difficulty and complexity.  ... Partial answers, based on unrelated criteria like cost,   are achieved more easily, than full answers,  that take much more effort

 

Within that alternative overall design , individual components are similarly compared. If that is properly done the now revealed cost cutting of the "wouldn't expect" and "particularly likely"  compromised approach is actually heard, and can be discarded.

 

 

 

 

 

 

 

 

 

 

Here is an example of a successful speaker manufacturer who uses a mix of objective and subjective tests to design and select components that are most appropriate to the musicality they are wanting to achieve.

 

* Extract from HiFi Plus - Meet Your Maker feature on Neat Acoustics 

 

Music runs like a red thread through every product Neat makes too, from the smallest Iota to the largest Ultimatum; the products are designed more as a result of careful listening tests with a varied musical programme than by the usual cycles of objective testing and measurement. The latter is not dismissed, but the most important consideration in making a good Neat Acoustics loudspeaker is music first, last, and always.

 

That’s not to say the company is entirely free from technological development. The company uses a specially made corner anechoic chamber for driver testing and development, and works closely with a range of loudspeaker driver manufacturers to deliver custom-made versions of drive units to meet Neat’s exacting specifications. But those specifications are more to do with how the end result sounds musically than how it performs on the test bench. Because, ultimately it reflects what a loudspeaker has to do for a living – make music sound good. “We tend not to work in a theoretical way”, says Bob Surgeoner, “we tend to work intuitively.” “Because we’ve been using the same drive units in combination for some time, design is more of an iterative process.” It’s about finding the right part – the Peerless bass unit we use in Motive wouldn’t work with the Ultimatum, and the custom-made unit in the Ultimatum isn’t as good as the off-the-shelf model for the Peerless when used in the Motive. It’s all about context.

 

Reference: https://hifiplus.com/articles/meet-your-maker-neat-acoustics/

  • Like 3
Link to comment
Share on other sites

I first got into 'hi-fi' seriously around 94 having amassed around 1000 LPs by then, mostly from dedicated car booting. I used to buy HiFi World mag in the UK and saw the plug for World Audio Designs. Lots of newbies bought into their products only to find that the PCBs were crap and track lifting was usual. They used carbon resistors and lots of other components that were far from optimal. What difference in price would it have meant to use quality produced PCBs, better components - 10-15% no they were built down to a price something that has plagued a lot of British products since WW11. The Germans never abandoned quality construction, compare where engineering is in both countries today.

 

On another thread I posted the interview available on You Tube with the speaker designer Peter Comeau who briefly owned WAD before being headhunted to go and work in China. This is the man responsible for the early Heybrook designs and Mission and lately for some really good speakers for Wharfedale. He listens to individual components and experiments with designs but as he says until you listen you cannot know how the whole thing is going to sound. Some very expensive valve amps when opened up have very cheap components and can be a rats nest. 

 

Howe many pieces of gear have more time spent on how the item looks than on how it functions. i/connects that look really sexy, FFS they won't be seen they function behind the equipment. Look at the cost breakdown and you will see where the majority of production is - on how it looks. 

 

T/tables - I cannot believe how much some of these eye candy pieces cost but in actual engineering terms the Japanese super decks of the 70s and 80s' are still superior.

 

I look at amps, irrelevant of type s/s, valve, class D and what do I see flat rectangular boxes that have little if any attempt to dissipate heat. Over 14 years ago I and my Dutch mate Harry (it was his idea) to use computer fans to move the heat generated by mosfets and power resistors - why didn't the manufacturers do that. It would be easy to actually design enclosures for a lot of gear vertically (aiding heat dissipation) and needing only one shelf instead of 3 or 4. The resistance to using the superb Z foil resistors, to spending the money on the PSU which is so important to 'good' sound - no, lets make the box look sexy the dummies will fall for that.

 

There is so much resistance to change in the audio world - the inertia effect. Why do box speakers have the x/overs inside the box this is completely irrational, they should be outside, very easy to create something easy on the eye but no one does because - we've always done it this way and no one has complained. As Peter Comeau says since the 70s' or later 'eye candy' appeal is as important as how a speaker sounds. My friend Harry was in the specialist audio shop in Groningen (they hand make the de Graaf caps) and a punter walked in and said "I've got €20,000 for a sound system, I could'nt give a toss what it sounds like, I just want a system that makes my neighbours jealous". A few weeks later Harry was back in that shop and in walked the same punter, he was grinning. He said I'm really happy with what you guys selected for me, all my neighbours are really jealous, thank you very much. The guys who ran the shop said "great but how does it sound" answer "oh sod the sound I could'nt care less about that". As Harry said you had to be there to believe it.

 

When you see all those very expensive speakers with the beautifully curved cabinets with excellent real wood finishes and the same audio effects could have been achieved by using curved corner pieces in construction to deal with standing waves. Dead easy on a production line but then the punters would'nt want to pay eye watering sums because the cabinets/boxes would'nt be so sexy would they?

  • Like 4
Link to comment
Share on other sites

The problem is that you never listen to a component, only to a system, even changing one single cap can only be evaluated by listening to a complete system. Measurements are a very good guide, but again, are only measurements of one component in the chain, the exception being speakers. The problem I have though is that you can change an amplifier or a turntable in a system and what you'd measure at the speaker is substantially the same, but the sound of the system can be considerably different

  • Like 3
Link to comment
Share on other sites
10 minutes ago, proftournesol said:

The problem is that you never listen to a component, only to a system, even changing one single cap can only be evaluated by listening to a complete system.

 

Except for cables. The way they sound can be tested in isolation.

 

Here’s a test in progress:

 

image.gif.0b26681146538d7c82e119dbea0a4bf3.gif

  • Like 2
Link to comment
Share on other sites
2 hours ago, proftournesol said:

The problem is that you never listen to a component, only to a system, even changing one single cap can only be evaluated by listening to a complete system. Measurements are a very good guide, but again, are only measurements of one component in the chain, the exception being speakers. The problem I have though is that you can change an amplifier or a turntable in a system and what you'd measure at the speaker is substantially the same, but the sound of the system can be considerably different

However a good manufacturer will have a system  set up where individually  each component can be assessed. A typical day for me if doing this, might compare a 2N5457 jfet, to 20 or 30 other jfets  to arrive at which was best in that parts equipment allocated position. Extremely slow and requiring patience and enormous commitment, it eventually has  reward to actually find what sounds best  

 

An example audiophiles actively do in a similar way is op amp rolling, where different types of opamp are listened to  to have worse or better characteristics, when actually heard.

 https://www.headphonesty.com/2018/03/1667/

Link to comment
Share on other sites

4 minutes ago, stereo coffee said:

A typical day for me if doing this, might compare a 2N5457 jfet, to 20 or 30 other jfets  to arrive at which was best in that parts equipment allocated position. Extremely slow and requiring patience and enormous commitment, it eventually has  reward to actually find what sounds best  

Of course any batch of  manufactured semiconductor devices will not consist of perfect;y identical units.  And different batches of the same device might vary again.

 

I recall that years ago one could purchase matched pairs of PNP and NPN output transistors for output stages.

 

And today some manufacturers offer matched pairs of microphones.

 

It sounds to me that you would expect a manufacturer not merely to experiment with semiconductors at the circuit design stage, but to measure a batch of units of the same semiconductors and select only those units that fell within a narrow range of performance.

Link to comment
Share on other sites
23 minutes ago, stereo coffee said:

However a good manufacturer will have a system  set up where individually  each component can be assessed. A typical day for me if doing this, might compare a 2N5457 jfet, to 20 or 30 other jfets  to arrive at which was best in that parts equipment allocated position. Extremely slow and requiring patience and enormous commitment, it eventually has  reward to actually find what sounds best  

 

An example audiophiles actively do in a similar way is op amp rolling, where different types of opamp are listened to  to have worse or better characteristics, when actually heard.

 https://www.headphonesty.com/2018/03/1667/

Agree, but the choice of components in the reference system will also influence the assessment outcome., just as the assessor's 'desired sound' will also influence the choice of components in the reference system. It's an imperfect system, I don't have a problem with that.

Link to comment
Share on other sites
8 minutes ago, MLXXX said:

Of course any batch of  manufactured semiconductor devices will not consist of perfect;y identical units.  And different batches of the same device might vary again.

 

I recall that years ago one could purchase matched pairs of PNP and NPN output transistors for output stages.

 

And today some manufacturers offer matched pairs of microphones.

 

It sounds to me that you would expect a manufacturer not merely to experiment with semiconductors at the circuit design stage, but to measure a batch of units of the same semiconductors and select only those units that fell within a narrow range of performance.

Both measurement and listening would be involved.   Such a manufacturer, would of course choose a component and manufacturer of that component , with good consistency to begin with.    Indeed measuring and verifying published parameters  prior to listening would occur , but such a manufacturer would use listening as the higher order arbiter of choice, rather than measurement alone.   ...Difficult... certainly .... but it is, what  brings the result needed.

Link to comment
Share on other sites
10 minutes ago, proftournesol said:

Agree, but the choice of components in the reference system will also influence the assessment outcome., just as the assessor's 'desired sound' will also influence the choice of components in the reference system. It's an imperfect system, I don't have a problem with that.

If continually worked on, it gets to a point where all components,  have been chosen by listening.   Eventually it reaches a point  that realism of audio reproduction is before you.   If manufacturing such a product, its then necessary to do it all again, time after time.

Link to comment
Share on other sites

I find when familiar with a system changing the one cap (two caps L/R channels) we can come to recognise the differences and the character of each type of cap.

 

Same with resistors, and dare I say wire also, I have tried different wire on the inputs of my amp (about 13" length) and the fly leads (several inches long) from the outputs of my op amps to output caps in my cd player, and can hear the differences between say Neotech solid core copper, a soft annealed solid silver and Duelund solid silver wire.

  • Like 1
Link to comment
Share on other sites

A listening panel of 16 or 17 year olds might be good, or even younger, because of their still intact high frequency hearing. 

 

Or at least such young people could help in the listening evaluations.

 

They's also be likely to have excellent short-term and medium-term memory if it took a little while to change between A and B versions; and they might be able to keep half a dozen different versions of sound in their heads; speeding up the selection process.

Link to comment
Share on other sites

I think it would be beneficial to many manufacturers if they had the ability to listen and compare their products with other brands.

 

It is all well and good to measure and listen to an item for years, tweaking and improving but add another brands equipment and it might show that a lot of time has been wasted.

 

Also what are manufacturers listening on? I have heard a few stories of items being made and yet the equipment used to develop them (DACs, amps etc) are quite poor relative to what is out there.

 

I quite like in D'agostino interviews where he talks about his time after Krell and going to listen to a whole heap of gear from many brands and realising it made him happier than what he had been working on at Krell that just measured well. I do take that with a grain of salt considering what happened though. 

 

I believe he also evaluates individual parts and spends a lot on R&D and good on him for being in a position to do so.

Link to comment
Share on other sites
1 hour ago, MLXXX said:

They's also be likely to have excellent short-term and medium-term memory if it took a little while to change between A and B versions; and they might be able to keep half a dozen different versions of sound in their heads

There's a lot results showing audio memory is hopelessly small, but I've never seen anything saying that doesn't apply to youngsters.

Link to comment
Share on other sites
7 minutes ago, davewantsmoore said:

There's a lot results showing audio memory is hopelessly small, but I've never seen anything saying that doesn't apply to youngsters.

 

Nor would a high frequency acuity limitation not apply for youngsters. 

Link to comment
Share on other sites
 Share





×
×
  • Create New...
To Top