A few more experiments:
Silmic + Nichicon KZ. This is no good, edgy and boring, nothing can save the KZ and FG in competition with better bypass caps.
Silmic + Nichicon ES. The shouty quality of the ES on vocals doesn't go away, so no.
Silmic + Kemet. Not better than Kemet alone
Silmic + Kaisei. Strangely this seems to work - quite a nice sound and vocals are OK. Competes with KG Gold Tune.
Kemet C4AQ, 30+30uF. Combinations to make a larger value seem to work fine. Still the best cap here but this is DC Link polypropylene.
Not tried yet - solid polymer. I'll order some this week. I'm also tempted by some Vishay DC Links.
Silmic + Nichicon KZ. This is no good, edgy and boring, nothing can save the KZ and FG in competition with better bypass caps.
Silmic + Nichicon ES. The shouty quality of the ES on vocals doesn't go away, so no.
Silmic + Kemet. Not better than Kemet alone
Silmic + Kaisei. Strangely this seems to work - quite a nice sound and vocals are OK. Competes with KG Gold Tune.
Kemet C4AQ, 30+30uF. Combinations to make a larger value seem to work fine. Still the best cap here but this is DC Link polypropylene.
Not tried yet - solid polymer. I'll order some this week. I'm also tempted by some Vishay DC Links.
Please remember one fact. Each capacitor surface is an antenna. It will either transmit or receive whatever electrical signals are in or near it. Just exactly like a wire, however it has much greater surface area.
Another inconvenient fact is that many times you are blocking air flow. Equipment designers are already allowing higher than desirable operating temperatures on some components. Blocking airflow in one place can restrict overall ventilation in another. This can be unintentional on your part, but it is still a factor sometimes.
Adhesives and tapes can cause other issues, and quite frankly, you end up running wire in places it shouldn't be. Wire dress is actually engineered in good products, and a constant battle for me to correct in modified equipment. Once cleaned up, measured and subjective performance is always improved.
Another inconvenient fact is that many times you are blocking air flow. Equipment designers are already allowing higher than desirable operating temperatures on some components. Blocking airflow in one place can restrict overall ventilation in another. This can be unintentional on your part, but it is still a factor sometimes.
Adhesives and tapes can cause other issues, and quite frankly, you end up running wire in places it shouldn't be. Wire dress is actually engineered in good products, and a constant battle for me to correct in modified equipment. Once cleaned up, measured and subjective performance is always improved.
Anatech, I hear you and have upgraded the fan and heat sink on the processor chip significantly and have measured the board and chip temperatures and they are below stock, I also endeavour to sheild as much as I can. I have a stock player as a reference and go forward and backwards if I am unsure. I'd love more space and the ability to move fixed things around. I'm sure you'd hate my unit for layout or the significant location compromise but I have subjectively A-B'd before moving to the next step and indeed I tried the player some 6m ago against a high end Audio Note DAC 5 and a belt drive Audio Note transport / Raspberry Pi streamer.
The conclusion of all three of us was the best transport (into the DAC 5) was the AN belt drive, followed by the 9000, followed by the Raspberry Pi (TiDAL or Qobus) and the 9000 model player on its own was better than the streamer and DAC5, the internal modded player was better in the bass than the DAC5 more dynamic but not as harmonically 'sweet' and as tonally accurate but it was close. One preferred my DAC, I was in between, depended on the track, but I didnt want to give up the dynamics and the other preferred the DAC5
The difference on audio and picture against my stock player is significant, for me it's the difference between great enjoyment and a flat, undynamic boring player. The picture with the power supply and board upgrades and clock is notably better too.
Not much more I can say really.
The conclusion of all three of us was the best transport (into the DAC 5) was the AN belt drive, followed by the 9000, followed by the Raspberry Pi (TiDAL or Qobus) and the 9000 model player on its own was better than the streamer and DAC5, the internal modded player was better in the bass than the DAC5 more dynamic but not as harmonically 'sweet' and as tonally accurate but it was close. One preferred my DAC, I was in between, depended on the track, but I didnt want to give up the dynamics and the other preferred the DAC5
The difference on audio and picture against my stock player is significant, for me it's the difference between great enjoyment and a flat, undynamic boring player. The picture with the power supply and board upgrades and clock is notably better too.
Not much more I can say really.
Hi IWC,
I understand where you are coming from.
These days we have rather excellent instrumentation. We can actually predict what something will sound like reliably if you can interpret the spectrum graphs properly. Measuring 30 dB below what the human body can even perceive or sense is easy these days, I do it several times a day on the bench. There are some that will argue vehemently against this idea, but sorry these are the facts. Science and technology has come a very long way since the 1980's when several audio ideas were formed and persist.
I extremely often see that audition guided type improvements may (often do) fall off the rails at some point. Our brains are fantastic at overriding what we actually hear. So therefore we need a guide to keep us on track. We are not test instruments and do not have any built in reference. We are however, pretty good at comparing sound quality. We do have to be very careful though, we are easily mislead by what may be in our heads.
My girlfriend / partner has amazing hearing and doesn't read audio anything. She just knows what she likes. The woman is Irish, so she tells it as it is. So, I never judge my own work as I can't hope ever to be impartial. I am usually too hard on myself (overcompensating). I use my instrumentation to guide my efforts and also listen. I use a pair of Klipsch THX 6000 speakers on the bench to assess what I am doing. Then for the acid test, if the equipment is performing well enough, it is auditioned by others. Ann is an uncanny ability to pinpoint problems (I've gotten to a point where there aren't any big ones before auditions). The instrumentation always guides improvements in the right direction. I'll be honest, auditions have become less necessary these days. We still need them in order to describe to others, and as witness statements.
Like you, I keep a stock unit and the "victim" unit so I have a reference. I also have "as found" measurements to go by. This all keeps me honest. In assessing equipment for others, it isn't uncommon to find there are actually problems induced by the improvements. Once eliminated, reviewers do agree it sounds better. With some good equipment, that actually means returning the equipment to bone stock condition (some manufacturers know exactly what they are doing). I've also seen more than my share of equipment failure or misbehavior due to inappropriate component selection. Normally this means that "the best" component was used even though it didn't fit or didn't suit the job at hand.
For capacitors and resistors (and semiconductors), I have specialised test equipment to characterise them. This is done at several frequencies and levels depending on the application. I hate to say it, with knowledge of what characteristics are important in each circuit position, you can find the best component by measuring it. Reading the manufacturer's data sheet (not white paper!) has valuable information you need to have as well. There is skill involved, and experience. But it is not magic.
Designing by ear does not work. Some audio designers will offer old ideas if they are trying for market position or if they actually do not understand what they are doing. That is extremely common by the way. I can think of a few well known designers that spout garbage. I will not name them as that will lose the message in a free-for-all argument.
I understand where you are coming from.
These days we have rather excellent instrumentation. We can actually predict what something will sound like reliably if you can interpret the spectrum graphs properly. Measuring 30 dB below what the human body can even perceive or sense is easy these days, I do it several times a day on the bench. There are some that will argue vehemently against this idea, but sorry these are the facts. Science and technology has come a very long way since the 1980's when several audio ideas were formed and persist.
I extremely often see that audition guided type improvements may (often do) fall off the rails at some point. Our brains are fantastic at overriding what we actually hear. So therefore we need a guide to keep us on track. We are not test instruments and do not have any built in reference. We are however, pretty good at comparing sound quality. We do have to be very careful though, we are easily mislead by what may be in our heads.
My girlfriend / partner has amazing hearing and doesn't read audio anything. She just knows what she likes. The woman is Irish, so she tells it as it is. So, I never judge my own work as I can't hope ever to be impartial. I am usually too hard on myself (overcompensating). I use my instrumentation to guide my efforts and also listen. I use a pair of Klipsch THX 6000 speakers on the bench to assess what I am doing. Then for the acid test, if the equipment is performing well enough, it is auditioned by others. Ann is an uncanny ability to pinpoint problems (I've gotten to a point where there aren't any big ones before auditions). The instrumentation always guides improvements in the right direction. I'll be honest, auditions have become less necessary these days. We still need them in order to describe to others, and as witness statements.
Like you, I keep a stock unit and the "victim" unit so I have a reference. I also have "as found" measurements to go by. This all keeps me honest. In assessing equipment for others, it isn't uncommon to find there are actually problems induced by the improvements. Once eliminated, reviewers do agree it sounds better. With some good equipment, that actually means returning the equipment to bone stock condition (some manufacturers know exactly what they are doing). I've also seen more than my share of equipment failure or misbehavior due to inappropriate component selection. Normally this means that "the best" component was used even though it didn't fit or didn't suit the job at hand.
For capacitors and resistors (and semiconductors), I have specialised test equipment to characterise them. This is done at several frequencies and levels depending on the application. I hate to say it, with knowledge of what characteristics are important in each circuit position, you can find the best component by measuring it. Reading the manufacturer's data sheet (not white paper!) has valuable information you need to have as well. There is skill involved, and experience. But it is not magic.
Designing by ear does not work. Some audio designers will offer old ideas if they are trying for market position or if they actually do not understand what they are doing. That is extremely common by the way. I can think of a few well known designers that spout garbage. I will not name them as that will lose the message in a free-for-all argument.
Instead of Silmic you should try Cerafine + Others Andy but must get the 50v version. Lightweight in bass but lots of transparency with smooth treble
Last edited:
I'm sure your technical skills have been much admired, but please don't carry on labouring the point. You've repeated this again and again. We all know where you are coming from by now. And if you think conservatoire trained musicians are unintelligent you're way off the mark (I have lots of data). While you have been gathering expertise in measurements, others including myself have a lifetime of playing live music to draw on. As a professional musician it would be unimaginable for me not to audition equipment extremely carefully using very familiar tracks selected for precise musical reasons. I also happen to be an academic psychologist specialising in music and musicians so I'm well aware of how the brain works, and particularly in relation to music. We measure and assess in different ways with different skill sets. So can we agree to differ on this point?
I made some jottings on trusting your ears and although this is arguably not scientific, it's something I think is actually scientific to the extent there could be something that could be explained but we might not have the budget, need or desire to do so. Science as we know evolved and challenges and even quantum recognises that things might it be what they seem perhaps my system is either utter rubbish or absolute bliss and I dont know until I listen 
Here's a good reason why you should listen and whilst being fully aware of your brains ability to mislead. Trust your ultimate measurement tool, it's very, very good. its the whole reason we bother in any case.
I have been experimenting with parallel mains filters, using a box to plug separately into a spare mains plug (I'm running out, even though I went OTT when wiring to room up !). The filters in this case are just caps across the line in differential mode. This is a unit plugged in near electronics to reduce mains noise and NOT in line, I have NEVER got on with inline means filtration and tried many solutions along the way, that all went back.
So I have built 2 very similar, one using more Panasonic X rated caps. My brother built a similar unit but in a different box and slightly different capacitor choice but again similar. All have the same 30uf and 12uf caps. the cascade down to 10nf, whilst not identical, the cascade adds up to 52-54uf and is just differential mode. I AM NOT RECOMMENDING ANY DIY by the way.
Do they work, yes. Are they transformational. definitely not, BUT they have qualities I didn't expect. Sonically the differences are there but small, musically they seem bigger. I keep engaged and it seems to continue to draw you in musically as opposed to somehow doing that less or the reverse without them. Here we go my brain again. How can that be a thing ?!
Okay why do I say that ? Because it's actully harder than usual to evaluate what they do and it IS in some respects subtle. So lots of A-B by myself and I am comfortable I can hear my version 1, vs my version 2 and none and both V1 and V2 together. So my brother was here at the weekend with his version, so now we have three boxes to listen and compare. All boxes have had over a month of burn in and this DID make a perceivable difference.
So on Saturday evening, 1/2 glass of wine in. We set them up plugged in behind the listing equipment and agree a blind test shoot out. Richard my brother will 'name' them A,B and C, then play A, then B then C. I then ask for any after that point. I cant see, but as its possible I get a feeling about the location of the switches and possibly guess which is which with some known or subliminal information, we agree he will move around turning on and off more than he needs to. All plugs are within 18" of each other anyway and they are nearly directly behind the sofa. I also covered my ears as he made the changes to avoid distraction anyway.
I listen and request a number of changes, then conclude which I prefer and why. I also say I am 99% sure I can actually hear which is which. My reasons why were aligned with the my listening before. So getting this wrong now would be embarrassing as not only would I be wrong but I would have found differences that were clearly inconsistent. The differences I am hearing with these 'swaps' are way less than any power supply capacitor. Which is more than enough for me not to need to A-B IMO. Also I don't want to solder and resolver or build a mule every time I make a change, but thats a separate point.
I was nervous and to some extent surprised I was 100% accurate. I got the why's right and which was which, given how small the changes are between them it goes to show how good our hearing is. So how on earth would I measure this ? I have no doubt they would be indistinguishable the capacaitance is almost the same, the values are either the same or extremely close (I ran out of 10nf and used a 22nf in one version) I can hear differences between the same design with some of the caps being only a different make in a mains filter. This is a parallel mains filter !!!! not even in line... or in a any of the 9 power amps we are listing to, the 4K player, the miniDSP, the preamp and processor etc.
This has reminded me to have confidence in your own judgement. How many either have very low resolution systems, so badly set up they aren't even close to hearing subtle benefits or are so convinced something wont make a difference ignore what they hear ? Or more usually quote ASR and never ever A-B or experiment.
Do I think I have golden ears, NO ! Im nearer 60 than 50, I spent many hours in my youth listing to Thelonious Monk at 120db peaks through 105db/W horns. BUT I do take care to evaluate carefully and are very self critical, and I do go back and undo, however painful if I am not sure.
Could I walk in 1-2 hours later and tell you which one is plugged in, I seriously doubt it. But I have no doubt I'd get it right again with A,B,C blind test.
I await for the flack, the statistical probability, the fact that it's not double blind, my brother lied and so on....
So why am I saying this ? It's because I would have been seriously fed up if I had spent years believing 'simple' proofs, conspiracy theories and never experimented for myself. Many people out there are discouraged from listening to evaluate and we all think a 'flat' SPL curve is the pinnacle of perfect sound. We are very insightful non linear creatures with X billion years of getting this hearing thing right, don't dismiss it !
Darwin or God would be disappointed if you did
Here's a good reason why you should listen and whilst being fully aware of your brains ability to mislead. Trust your ultimate measurement tool, it's very, very good. its the whole reason we bother in any case.
I have been experimenting with parallel mains filters, using a box to plug separately into a spare mains plug (I'm running out, even though I went OTT when wiring to room up !). The filters in this case are just caps across the line in differential mode. This is a unit plugged in near electronics to reduce mains noise and NOT in line, I have NEVER got on with inline means filtration and tried many solutions along the way, that all went back.
So I have built 2 very similar, one using more Panasonic X rated caps. My brother built a similar unit but in a different box and slightly different capacitor choice but again similar. All have the same 30uf and 12uf caps. the cascade down to 10nf, whilst not identical, the cascade adds up to 52-54uf and is just differential mode. I AM NOT RECOMMENDING ANY DIY by the way.
Do they work, yes. Are they transformational. definitely not, BUT they have qualities I didn't expect. Sonically the differences are there but small, musically they seem bigger. I keep engaged and it seems to continue to draw you in musically as opposed to somehow doing that less or the reverse without them. Here we go my brain again. How can that be a thing ?!
Okay why do I say that ? Because it's actully harder than usual to evaluate what they do and it IS in some respects subtle. So lots of A-B by myself and I am comfortable I can hear my version 1, vs my version 2 and none and both V1 and V2 together. So my brother was here at the weekend with his version, so now we have three boxes to listen and compare. All boxes have had over a month of burn in and this DID make a perceivable difference.
So on Saturday evening, 1/2 glass of wine in. We set them up plugged in behind the listing equipment and agree a blind test shoot out. Richard my brother will 'name' them A,B and C, then play A, then B then C. I then ask for any after that point. I cant see, but as its possible I get a feeling about the location of the switches and possibly guess which is which with some known or subliminal information, we agree he will move around turning on and off more than he needs to. All plugs are within 18" of each other anyway and they are nearly directly behind the sofa. I also covered my ears as he made the changes to avoid distraction anyway.
I listen and request a number of changes, then conclude which I prefer and why. I also say I am 99% sure I can actually hear which is which. My reasons why were aligned with the my listening before. So getting this wrong now would be embarrassing as not only would I be wrong but I would have found differences that were clearly inconsistent. The differences I am hearing with these 'swaps' are way less than any power supply capacitor. Which is more than enough for me not to need to A-B IMO. Also I don't want to solder and resolver or build a mule every time I make a change, but thats a separate point.
I was nervous and to some extent surprised I was 100% accurate. I got the why's right and which was which, given how small the changes are between them it goes to show how good our hearing is. So how on earth would I measure this ? I have no doubt they would be indistinguishable the capacaitance is almost the same, the values are either the same or extremely close (I ran out of 10nf and used a 22nf in one version) I can hear differences between the same design with some of the caps being only a different make in a mains filter. This is a parallel mains filter !!!! not even in line... or in a any of the 9 power amps we are listing to, the 4K player, the miniDSP, the preamp and processor etc.
This has reminded me to have confidence in your own judgement. How many either have very low resolution systems, so badly set up they aren't even close to hearing subtle benefits or are so convinced something wont make a difference ignore what they hear ? Or more usually quote ASR and never ever A-B or experiment.
Do I think I have golden ears, NO ! Im nearer 60 than 50, I spent many hours in my youth listing to Thelonious Monk at 120db peaks through 105db/W horns. BUT I do take care to evaluate carefully and are very self critical, and I do go back and undo, however painful if I am not sure.
Could I walk in 1-2 hours later and tell you which one is plugged in, I seriously doubt it. But I have no doubt I'd get it right again with A,B,C blind test.
I await for the flack, the statistical probability, the fact that it's not double blind, my brother lied and so on....
So why am I saying this ? It's because I would have been seriously fed up if I had spent years believing 'simple' proofs, conspiracy theories and never experimented for myself. Many people out there are discouraged from listening to evaluate and we all think a 'flat' SPL curve is the pinnacle of perfect sound. We are very insightful non linear creatures with X billion years of getting this hearing thing right, don't dismiss it !
Darwin or God would be disappointed if you did
Hi IWC,
Never once have I said that you shouldn't listen to the product, not ever. However, to be sure you are on the right track you absolutely do need properly measured information. You will find they agree by the way as long as you are not biased during listening tests.
I have seen time and time again where groups of educated and skilled people follow each other (there is a clue in there) and come to a conclusion that later dies (a very hard, long painful death). I can't tell you how many times I have looked at equipment certified to be in perfect working order to find not only defects, but over 1% distortion. Technicians that maintain they don't need to spend the money on a simple THD meter - only to show they cannot in fact tell if equipment is working properly by listening to it.
Look at the 1970's for this classic example (a capacitor one at that!). Tantalum capacitors were supposed to be wonderful, so much better than electrolytic capacitors. Were they? Nope, in fact they are much worse, never mind the fact they like to short. High end manufacturers all jumped on that bandwagon, and early modification gurus ripped out perfectly good (and superior) electrolytic caps to install these terrible things. So forgive me as I have watched this time and time again over the decades. The power of the mind.
Finally, I did not say what you did impaired the performance. What I did was alert you and others who read these threads to the possible negative outcomes if you don't take everything into account (like they did when they designed it). Sure, some things are built also considering cost. But also availability in production quantities and failure rates too.
Some of the worst equipment I have ever seen was designed by ear alone. They also didn't really sound that good.
Never once have I said that you shouldn't listen to the product, not ever. However, to be sure you are on the right track you absolutely do need properly measured information. You will find they agree by the way as long as you are not biased during listening tests.
I have seen time and time again where groups of educated and skilled people follow each other (there is a clue in there) and come to a conclusion that later dies (a very hard, long painful death). I can't tell you how many times I have looked at equipment certified to be in perfect working order to find not only defects, but over 1% distortion. Technicians that maintain they don't need to spend the money on a simple THD meter - only to show they cannot in fact tell if equipment is working properly by listening to it.
Look at the 1970's for this classic example (a capacitor one at that!). Tantalum capacitors were supposed to be wonderful, so much better than electrolytic capacitors. Were they? Nope, in fact they are much worse, never mind the fact they like to short. High end manufacturers all jumped on that bandwagon, and early modification gurus ripped out perfectly good (and superior) electrolytic caps to install these terrible things. So forgive me as I have watched this time and time again over the decades. The power of the mind.
Finally, I did not say what you did impaired the performance. What I did was alert you and others who read these threads to the possible negative outcomes if you don't take everything into account (like they did when they designed it). Sure, some things are built also considering cost. But also availability in production quantities and failure rates too.
Some of the worst equipment I have ever seen was designed by ear alone. They also didn't really sound that good.
we can spend a lot of money or spend a little, in the end when we listen, we listen to the music not to the sound of capacitors that you put in your amp...
and no matter what you do, the live original performance will always sway emotions in you that makes you have goosebumps, so as long as our feet are firmly planted in the ground, we should all be fine...
anecdotal essays do not sway me, nor bother me, unless they insist that i should hear what they hear, therein lies the trouble, be happy with what you hear, that is what it is all about...
and no matter what you do, the live original performance will always sway emotions in you that makes you have goosebumps, so as long as our feet are firmly planted in the ground, we should all be fine...
anecdotal essays do not sway me, nor bother me, unless they insist that i should hear what they hear, therein lies the trouble, be happy with what you hear, that is what it is all about...
that is why i am very careful to lead people into believing something i believe in, i can be wrong....
i have people come to me saying that he was invited to listen to a system costing millions but to him the sound did not justify the millions spent...
Im with you here 100% it's the classic board room discussions where a journey is often decided by one starting point that gets supported inflated and never truly challenged or properly assessed. The 'process' looks right the conversation palatable but little rigour to really dis deep enough to genuinely support
Well, it's for me to decide if we disagree. The mind is one thing but what is important in auditioning acoustic instruments is a very well developed musical ability to differentiate subtle details in tonality, as professional musicians routinely do. It's not enough to simply listen to tracks and decide what you like the sound of.
You don’t need to be constrained by your lack of live musical experience - there is always something you can do about it. Admittedly musical talent is up to 50% genetic but you can always work with what’s left. Ideally you would start a musical instrument around age 7, but late learners can make progress with daily practice. Your aim would be to play in an orchestra alongside the extended family of live acoustic instruments for long enough to make them extremely familiar to your musical memory - two or three years should do it. And then you would want to play in a jazz group for another couple of years to get familiar with the drum kit and percussion. Naturally, you would want piano skills to be able to distinguish between a Steinway, a Bechstein and various other concert grands - having a grand piano at home would help (a Bechstein in my case). This would all require advanced sight reading skills and an ability to play over complicated chord changes - I mean, someone could call Giant Steps as an encore. You would also need advanced skills in tonality and tone to be able to choose a good musical instrument at the right price. About thirty years should do it. And then you’d be in an excellent position to audition acoustic instruments with comparable skills to trained professional musicians.
It's not just capacitors this can be true of most things, at least until you get a level of satisfaction that is ahead of the frustration. You will always be forced to leave something on the table. I just want to make sure I eat most of the main course