If people are looking for some really good value film caps with higher values, Farnell are selling the LCR (amp Ohm) metalled Poplyprop Film caps at tremendous prices. They measure very well indeed almost exact for capacitance and low ESR as quoted. Nicely made with tinned copper leads insulated. £9 inc VAT for a 30Uf 100mmx45mm cap, this was a £30 cap. I havent A-B's yet but they look great !
You know what?
Many of us, fanatics especially, don't stop to consider the actual circuit and what is important. THe first thing to consider is the impedance level of the circuit. Most coupling applications eliminate any concern for ESR (and what do most of you measure and obsess over?). You can focus on what you might think is "the best" capacitor, but the truth of the matter is that there are many considerations at play here. Firstly, there are capacitors that are better for certain applications and they do not excel in every one. Secondly there are physical considerations (do the darned things fit in the space properly? Are the leads too large for the hole in the PCB?). Face it, if the part is too large you will damage the PCB traces, you may block airflow or expose that part to excessive heat from another part. Obvious, but also consider a capacitor is a conductor, therefore you are placing the signal close to other signals or radiating that signal into places you don't want to if the part is too big. Then of course some parts are heavier and not well supported. Yup, breaks connections and traces.
Now for another unpopular thought. Some parts are more linear than others. Also, some tend to short all by their lonesome, or they may have higher leakage currents (and tha may or may not be important). High impedance circuits reduce the importance of linearity, then you have got to understand that often the negative aspects of a part type only show when you develop a signal voltage across them. THe values of coupling capacitors are set so as not to allow a signal voltage across them. Okay, so what does that mean? It means that the part does not impart any degradation to the signal folks. What I just said was that a decent (any good quality capacitor) does not matter to the sound quality. It is romantic to think it might, but this is a world of physics and science and I've measured various parts in many configurations over decades. Yes, people and myself included have also listened to the effects (or total lack thereof).
We measure signals and distortions below 130 dB from -10 dBu without any trouble at all. The human body simply cannot detect levels that low, or sense or mystically devine problems for that matter. Then the s/n of your listening environment will not exceed 80 dB from quietest to maximum SPL, a much smaller range than I'm talking about for measurements! So all this "I can hear the difference" stuff is simply expectation bias. Yes, people are convinced they hear differences, but in that specific application you simply cannot unless the part is defective in some way. Could even be a bad solder connection as many are hobbyists not using good soldering stations and decades of soldering practice.
There are circuit applications where capacitor types do in fact make a big difference, but not in coupling applications. Copper vs steel? Straw man folks, the electrons do not care. Magnetic leads? Doesn't matter at signal current levels one bit. At higher currents the impedance is lower at lowish frequencies so again it doesn't matter.
So what parameters are important? For filters, the exact value (not for coupling applications). ESR for crossovers and power supply filters (but it can be too low!) as those are low impedance circuits (get the idea here?). Dielectric absorption is critical in some areas of a circuit. For most circuit positions and applications these things are not that critical. For example, a polyester capacitor might be okay for power supply bypass, or high K ceramics. The voltage doesn't change (by definition) so those negative characteristics don't impact the circuit.
People who comment on components who do not actually study and measure them, and also understand the engineering of the circuit aren't well positions to recommend or discuss the issue to be honest. Those who have merely performed listening tests are probably the worst off since they no longer consider things without bias. They get locked in and invested in some idea they read about and "confirmed" without actually understanding what is really going on.
Now, if telling people you use a certain type of part makes you feel good and is that important to you, go ahead and indulge. Why not? But stop at making recommendations to others for the simple reason that you're off in the weeds here. That thinking is so far from reality it isn't funny! My only concern is really letting you know the actual truth, and preventing others from wasting hard earned money on hopeful improvements. I don't sell parts and have no interest in the subject other than I actually do know how this stuff works from basic levels on how they are constructed to the materials used, and how that impacts an electronic signal (voltage or current). Also, no. The circuit doesn't know or care if that signal is music, a test signal or industrial control signal. The signal also doesn't anticipate a signal, but previous conditions may affect it, but this isn't what you think it may be or how it might affect music.
-Chris
Many of us, fanatics especially, don't stop to consider the actual circuit and what is important. THe first thing to consider is the impedance level of the circuit. Most coupling applications eliminate any concern for ESR (and what do most of you measure and obsess over?). You can focus on what you might think is "the best" capacitor, but the truth of the matter is that there are many considerations at play here. Firstly, there are capacitors that are better for certain applications and they do not excel in every one. Secondly there are physical considerations (do the darned things fit in the space properly? Are the leads too large for the hole in the PCB?). Face it, if the part is too large you will damage the PCB traces, you may block airflow or expose that part to excessive heat from another part. Obvious, but also consider a capacitor is a conductor, therefore you are placing the signal close to other signals or radiating that signal into places you don't want to if the part is too big. Then of course some parts are heavier and not well supported. Yup, breaks connections and traces.
Now for another unpopular thought. Some parts are more linear than others. Also, some tend to short all by their lonesome, or they may have higher leakage currents (and tha may or may not be important). High impedance circuits reduce the importance of linearity, then you have got to understand that often the negative aspects of a part type only show when you develop a signal voltage across them. THe values of coupling capacitors are set so as not to allow a signal voltage across them. Okay, so what does that mean? It means that the part does not impart any degradation to the signal folks. What I just said was that a decent (any good quality capacitor) does not matter to the sound quality. It is romantic to think it might, but this is a world of physics and science and I've measured various parts in many configurations over decades. Yes, people and myself included have also listened to the effects (or total lack thereof).
We measure signals and distortions below 130 dB from -10 dBu without any trouble at all. The human body simply cannot detect levels that low, or sense or mystically devine problems for that matter. Then the s/n of your listening environment will not exceed 80 dB from quietest to maximum SPL, a much smaller range than I'm talking about for measurements! So all this "I can hear the difference" stuff is simply expectation bias. Yes, people are convinced they hear differences, but in that specific application you simply cannot unless the part is defective in some way. Could even be a bad solder connection as many are hobbyists not using good soldering stations and decades of soldering practice.
There are circuit applications where capacitor types do in fact make a big difference, but not in coupling applications. Copper vs steel? Straw man folks, the electrons do not care. Magnetic leads? Doesn't matter at signal current levels one bit. At higher currents the impedance is lower at lowish frequencies so again it doesn't matter.
So what parameters are important? For filters, the exact value (not for coupling applications). ESR for crossovers and power supply filters (but it can be too low!) as those are low impedance circuits (get the idea here?). Dielectric absorption is critical in some areas of a circuit. For most circuit positions and applications these things are not that critical. For example, a polyester capacitor might be okay for power supply bypass, or high K ceramics. The voltage doesn't change (by definition) so those negative characteristics don't impact the circuit.
People who comment on components who do not actually study and measure them, and also understand the engineering of the circuit aren't well positions to recommend or discuss the issue to be honest. Those who have merely performed listening tests are probably the worst off since they no longer consider things without bias. They get locked in and invested in some idea they read about and "confirmed" without actually understanding what is really going on.
Now, if telling people you use a certain type of part makes you feel good and is that important to you, go ahead and indulge. Why not? But stop at making recommendations to others for the simple reason that you're off in the weeds here. That thinking is so far from reality it isn't funny! My only concern is really letting you know the actual truth, and preventing others from wasting hard earned money on hopeful improvements. I don't sell parts and have no interest in the subject other than I actually do know how this stuff works from basic levels on how they are constructed to the materials used, and how that impacts an electronic signal (voltage or current). Also, no. The circuit doesn't know or care if that signal is music, a test signal or industrial control signal. The signal also doesn't anticipate a signal, but previous conditions may affect it, but this isn't what you think it may be or how it might affect music.
-Chris
Chris, just for clarity on the copper point adding a copper GND plate to the 9000 was remarkable in the sonic difference, clearly GND was inadequate and in this case is fundamental to good sound, relying on a relatively thick steel chassis was not good enough. I only say this as delighted as I was with my DIY skills I was so taken by the difference I rebuilt my modified boards into a stock chassis and visa versa, which took 3-4 hours. It confirmed that this was substantive. Whilst comparing anyone on this forum to the knowledge of someone like Panasonic is laughable, they do have to make compromises and in some cases significant compromises and I suspect in the £1k player the hardware costs must be £100-150 max, which doesn't allow for anything other than cheapish components. Not one cap in the unit is a Panasonic FR, FC or FM and they are not even Panasonic.. No sign of any Oscon SMD's etc
Marantz and Ken Ishiwata liked copper too
Marantz and Ken Ishiwata liked copper too
That's all very nice, but you have fallen exactly into the trap I just mentioned.
Audio companies do what sells. Period. I've been in the audio industry for years and was trained at Marantz (first) and other companies. I know what goes on from inside, you haven't a clue obviously.
Look, I know you are certain about what you think you heard. I know you are saying what you believe to be true. But your brain is overriding what you actually sensed, and we are actually designed to do that for survival. But unfortunately you are incorrect. Simple.
The human body is incapable of perceiving these differences. We are terrible, unreliable test instruments. We are at our best comparing when the comparison has only msec of separation. Even then, ideas in our brain will override what is actually heard. That is called expectation bias, and no matter hw hard you argue you don't suffer from it - you do. I do, and I know better. I never assess my own work or experiments because I know better.
Audio companies do what sells. Period. I've been in the audio industry for years and was trained at Marantz (first) and other companies. I know what goes on from inside, you haven't a clue obviously.
Look, I know you are certain about what you think you heard. I know you are saying what you believe to be true. But your brain is overriding what you actually sensed, and we are actually designed to do that for survival. But unfortunately you are incorrect. Simple.
The human body is incapable of perceiving these differences. We are terrible, unreliable test instruments. We are at our best comparing when the comparison has only msec of separation. Even then, ideas in our brain will override what is actually heard. That is called expectation bias, and no matter hw hard you argue you don't suffer from it - you do. I do, and I know better. I never assess my own work or experiments because I know better.
Chris, I have never worked in an audio company so cant comment on that but I have been deeply inside many different companies assessing their business models and commercial operations (I spent 20 year in CF DD and advisory) so do have some idea about product costings.
If you are correct then its pointless buying anything new or old and just convincing ourselves that our system sounds good. I wish it was that easy, not sure what I am surviving though, I thought survival was more about assessing the environment for threats and the more sensitive we are the better ?
Do you buy products by asking your friends what they think or listen yourself as well ?
I am far from infallible and ofcourse mis judge, gain other views and re assess. But we are capable of assessment, there is far too much conspiracy theories on the internet for my liking.
If you are correct then its pointless buying anything new or old and just convincing ourselves that our system sounds good. I wish it was that easy, not sure what I am surviving though, I thought survival was more about assessing the environment for threats and the more sensitive we are the better ?
Do you buy products by asking your friends what they think or listen yourself as well ?
I am far from infallible and ofcourse mis judge, gain other views and re assess. But we are capable of assessment, there is far too much conspiracy theories on the internet for my liking.
Hi IWC Doppel,
Yes, costings can bring insight. Therefore you are well aware that product design is driven by the sales department these days. You supply what the market is demanding or whatever feature gives you a market advantage. That is what this is all about. Remember "wired with Monster" stickers? A current one for CD players is "stable platter" (and I love belt drive for CEC). Those actually impair CD performance by the way.
It is not pointless to buy a new product at all. You just have to know what to buy. Good products are far fewer, and robust products fewer still. This is true across the entire consumer industry no matter what the product is. The entire fight is to differentiate your product from others - as cheaply as possible. I think you must agree with that. A perceived improvement is all that is required. You tell people something is better, and they believe it. All you need is a good story to "back it up". Half truths, and improvements that do not actually bring an improvement are extremely common.
Electronics service and design is all about physics and reality. There is a black and white, but engineering is all about the grey area. You balance things off each other including cost, reliability and performance. These days the scales are heavily tilted to cost including in the distribution chain and the service chain (almost eliminated). This should get everyone's attention - big time. What that means is not many people understand the product, and it can't be easily repaired. It did not reduce your cost, that was pocketed. In my world I am forced to deal with reality as I also do industrial and test and measurement. All I can do is tell you the truth and I can't really be concerned beyond that point. I'm tired of fighting fables and romantic ideas.
I look at products that I'm either interested in that come to my attention via friends or the industry. I pay attention to certain manufacturers because I know their technical abilities and history of engineering. I also know which manufacturers just spin stories around cheap s**t. Those I don't bother looking at because they simply are not interested in bringing a valid product to market. THen I do listen and test on the bench, stuff that doesn't measure up go back - period. If it isn't made to last, it also goes back. I do rebuild and redesign equipment and most of mine falls into that category. These are typically 1980 - ish designs that were basically very good. Then I apply new engineering and components to the design, never changing the design a great deal. I measure and listen to performance before and after each change. I actually follow ISO procedures I learned in calibration labs (certified calibration tech as well). I am very, very careful, and do not advertise improvements. People who have heard my equipment have gone out and bought the same models, theen brought them in for the same work. Otherwise they enquire about doing something similar with their equipment. Sometimes this is not possible.
The information on the internet is about as reliable as a light paper water basket (untreated). People who do know what they are doing don't post or publish for one. They simply do the work quietly, and we do very similar things (physics always leads down the same path). We can be busier than we can handle without saying a word.
Anyway, audio reproduction is a cold, hard science if you understand and know what you are doing. I take great pleasure in accomplishing real improvements, and I love music (always playing on one of my many different systems). I have a very good system on the service bench as well. The problem for most is that the education is decades long, the equipment ghastly expensive (don't ask how much I have invested in 5 19" bays of equipment with more hanging out waiting for use 'cause it won't all fit). A lot is current equipment made by Keysight. So it is unreasonable to expect even an enthusiast to invest enough to really know what this is about. I have a rough idea how to design or service a car engine, but don't know near enough to comment on it. I have successfully rebuilt engines, and even my furnace, but professionals know far more. I would never attempt one of these things these days! Same for my finances, there are people who know every detail. I don't even come close, same for legal aspects of life.
We each have our profession or subject we excel in (well, some folks just drink beer and watch TV lol!). On the surface audio reproduction sounds pretty simple, but is often badly over-simplified, usually by people trying to be well-known or worse, to make a buck. Every industry has those people.
Not every new piece of equipment is substandard or made poorly (most are), but not all older pieces are well made and designed either. All old products need maintenance before using them. Life just isn't that simple. Wouldn't it be nice if it was?
-Chris
Yes, costings can bring insight. Therefore you are well aware that product design is driven by the sales department these days. You supply what the market is demanding or whatever feature gives you a market advantage. That is what this is all about. Remember "wired with Monster" stickers? A current one for CD players is "stable platter" (and I love belt drive for CEC). Those actually impair CD performance by the way.
It is not pointless to buy a new product at all. You just have to know what to buy. Good products are far fewer, and robust products fewer still. This is true across the entire consumer industry no matter what the product is. The entire fight is to differentiate your product from others - as cheaply as possible. I think you must agree with that. A perceived improvement is all that is required. You tell people something is better, and they believe it. All you need is a good story to "back it up". Half truths, and improvements that do not actually bring an improvement are extremely common.
Electronics service and design is all about physics and reality. There is a black and white, but engineering is all about the grey area. You balance things off each other including cost, reliability and performance. These days the scales are heavily tilted to cost including in the distribution chain and the service chain (almost eliminated). This should get everyone's attention - big time. What that means is not many people understand the product, and it can't be easily repaired. It did not reduce your cost, that was pocketed. In my world I am forced to deal with reality as I also do industrial and test and measurement. All I can do is tell you the truth and I can't really be concerned beyond that point. I'm tired of fighting fables and romantic ideas.
I look at products that I'm either interested in that come to my attention via friends or the industry. I pay attention to certain manufacturers because I know their technical abilities and history of engineering. I also know which manufacturers just spin stories around cheap s**t. Those I don't bother looking at because they simply are not interested in bringing a valid product to market. THen I do listen and test on the bench, stuff that doesn't measure up go back - period. If it isn't made to last, it also goes back. I do rebuild and redesign equipment and most of mine falls into that category. These are typically 1980 - ish designs that were basically very good. Then I apply new engineering and components to the design, never changing the design a great deal. I measure and listen to performance before and after each change. I actually follow ISO procedures I learned in calibration labs (certified calibration tech as well). I am very, very careful, and do not advertise improvements. People who have heard my equipment have gone out and bought the same models, theen brought them in for the same work. Otherwise they enquire about doing something similar with their equipment. Sometimes this is not possible.
The information on the internet is about as reliable as a light paper water basket (untreated). People who do know what they are doing don't post or publish for one. They simply do the work quietly, and we do very similar things (physics always leads down the same path). We can be busier than we can handle without saying a word.
Anyway, audio reproduction is a cold, hard science if you understand and know what you are doing. I take great pleasure in accomplishing real improvements, and I love music (always playing on one of my many different systems). I have a very good system on the service bench as well. The problem for most is that the education is decades long, the equipment ghastly expensive (don't ask how much I have invested in 5 19" bays of equipment with more hanging out waiting for use 'cause it won't all fit). A lot is current equipment made by Keysight. So it is unreasonable to expect even an enthusiast to invest enough to really know what this is about. I have a rough idea how to design or service a car engine, but don't know near enough to comment on it. I have successfully rebuilt engines, and even my furnace, but professionals know far more. I would never attempt one of these things these days! Same for my finances, there are people who know every detail. I don't even come close, same for legal aspects of life.
We each have our profession or subject we excel in (well, some folks just drink beer and watch TV lol!). On the surface audio reproduction sounds pretty simple, but is often badly over-simplified, usually by people trying to be well-known or worse, to make a buck. Every industry has those people.
Not every new piece of equipment is substandard or made poorly (most are), but not all older pieces are well made and designed either. All old products need maintenance before using them. Life just isn't that simple. Wouldn't it be nice if it was?
-Chris
I searched the topic, and havent found any discussion about it. Intermodulation influences how something sounds, more so than thd, and each component and circuit topology produces different. Specificaly we are talking about capacitors here, and capacitors do their fair share of imd, so i don't belive capacitors are all the same sounding. What is your knowledge/opinion on the matter?
To replace some 'lytics cathode decoupling caps, I've ordered these dirt cheap ones from Ali:
Cost is a few bucks a piece including shipping, capacitance is not too far off at 48.72, 48.45, 48.37, and 47.89 uF for a nom. value of 47uF, tempco is negative, so they're true MKP, and there's no detectable piezo elect. effect. They appear to be made by JB Capacitors of Taiwan, and enjoy OK-ish review from The Humble. It's an ongoing project, I haven't listened to them yet.
Cost is a few bucks a piece including shipping, capacitance is not too far off at 48.72, 48.45, 48.37, and 47.89 uF for a nom. value of 47uF, tempco is negative, so they're true MKP, and there's no detectable piezo elect. effect. They appear to be made by JB Capacitors of Taiwan, and enjoy OK-ish review from The Humble. It's an ongoing project, I haven't listened to them yet.
Chris, I do hear you so to speak
As an engineer (I started as an automotive engineer), I do want to see the truth and get comfortable with the science, but I'm also mindful that we are non linear beings and what to measure, how to interpret are obviously not easy questions. Experimentation is still for me worthwhile and a bit like the medical industry if we know the input and the output (hopefully, accepting some subjectivity), we don't need to know how or why, its a nice to know not a need to know
As an engineer (I started as an automotive engineer), I do want to see the truth and get comfortable with the science, but I'm also mindful that we are non linear beings and what to measure, how to interpret are obviously not easy questions. Experimentation is still for me worthwhile and a bit like the medical industry if we know the input and the output (hopefully, accepting some subjectivity), we don't need to know how or why, its a nice to know not a need to know
just change a caps an catode en old cerefine vs Oscon and you can ear a big difference ,more if you place a 100uf claritycaps ...is abusrd that AS half diyaudio can ear difference there are still some that negate ....
Why?
Edit:
The Bateman paper you quoted talks about "very low cost, miniature electrolytic". I'm talking about MKP, which is Metallized Polypropylene.
Last edited:
Hi Brijac,
In a nutshell, when you place a certain capacitor into an application where it isn't suited and it does cause distortion of a signal, that is an engineering problem. If you don't like the sound it creates, that isn't a surprise at all. We can measure increasing amounts of distortion long before any human can "hear" it. There is zero point in discussing how different parts sound if they were not suited to an application to begin with. One thing I will say with conviction and decades of empirical evidence to support it, the brand / model of a capacitor isn't nearly as important as the materials and construction of that capacitor (in this case). The dielectric is the most important thing, but the construction matters as well. If it isn't wound in the right ballpark for tension you will have problems. Therefore all hand made capacitors instantly fail the quality test. So do cheap ones made on older, poorly maintained machines. A good capacitor will have better consistency from unit to unit and over time. Read that as any decent brand of capacitor. Newer ones should on average be better than old ones.
In the higher impedance circuits we see, the lead material of the capacitor isn't important (steel vs copper vs copper plated steel). Get over it people.
I will not get into a discussion that requires me to teach anyone decades of knowledge, I do not have time for one. That and distortion and components cover volumes of books, the amount of information is immense! That would be an unreasonable request. I wish there was a simple answer, but if I advance valid conclusions there are many who would challenge the information and it would become a life's work to educate each person. No thanks! I can giv answers (and I have), figure it out from there on your own if you have questions.
-Chris
Okay, THD is different, not more or less so than IMD. I always test for both. They do have a different sonic from each other and it depends on severity as well.
In a nutshell, when you place a certain capacitor into an application where it isn't suited and it does cause distortion of a signal, that is an engineering problem. If you don't like the sound it creates, that isn't a surprise at all. We can measure increasing amounts of distortion long before any human can "hear" it. There is zero point in discussing how different parts sound if they were not suited to an application to begin with. One thing I will say with conviction and decades of empirical evidence to support it, the brand / model of a capacitor isn't nearly as important as the materials and construction of that capacitor (in this case). The dielectric is the most important thing, but the construction matters as well. If it isn't wound in the right ballpark for tension you will have problems. Therefore all hand made capacitors instantly fail the quality test. So do cheap ones made on older, poorly maintained machines. A good capacitor will have better consistency from unit to unit and over time. Read that as any decent brand of capacitor. Newer ones should on average be better than old ones.
In the higher impedance circuits we see, the lead material of the capacitor isn't important (steel vs copper vs copper plated steel). Get over it people.
I will not get into a discussion that requires me to teach anyone decades of knowledge, I do not have time for one. That and distortion and components cover volumes of books, the amount of information is immense! That would be an unreasonable request. I wish there was a simple answer, but if I advance valid conclusions there are many who would challenge the information and it would become a life's work to educate each person. No thanks! I can giv answers (and I have), figure it out from there on your own if you have questions.
-Chris
Hi nicoch58,
I'm afraids some of that is expectation bias. I do have experience with that exact situation.
Additionally, distortion from a tube amplifier is so much higher than a properly chosen capacitor that it swamps any other results you might want to hear. A cathode bypass capacitor can act partially as a filter element and have signal voltage across it, so you may hear some, especially poor electrolytic caps. However under a tube chassis you simply do not have room most often to mount a film capacitor of the same value. Bypassing an electrolytic is totally useless given the distortion mechanism. Just in case someone was about to mention that.
You have to look at an amplifier as a system whose performance is limited by various components - not just a few capacitors. So to pick on a few specific parts often will not improve performance nearly as much as claimed (again, more direct experience). Note I am not saying it is pointless to try, but you have to look at degree of difference and improvement. I yank out more oversized capacitors from chassis' and install decent electrolytics without suffering any meaningful performance issues. The parts are typically far too large and cause other issues, which is why they have to be pulled. I can't tell you how many customers were fleeced by internet reading non-technicians who "improve performance" of audio equipment while causing or missing real issues. This also happens with technicians who should know better and may well know they are taking a customer to the cleaners intentionally.
All,
In general, don't worry about using a film cap in place of an electrolytic cap. Besides, if you replace an old electrolytic with a good brand of electrolytic capacitor you will probably hear a difference. The old one likely had issues by now, and capacitors are simply better these days. Many old film caps have serious issues as well. Replacing them with good new types will also sound better and improve performance. No shock there either. It certainly isn't proof that the audiophile approved part is far superior to today's normal parts.
Reality check. AUdiophile capacitors are either produced by good manufacturers on their modern, well maintained equipment and will be equivalent to something they normally make (with a different label), or it may be made by the audiophile type company using old machines that are not as good. They do not have the volume to buy the good raw materials and maintain the equipment to ta level where it needs to be. Economics of scale. They will deny this tooth and nail, just so you know.
Please use common sense.
-Chris
I'm afraids some of that is expectation bias. I do have experience with that exact situation.
Additionally, distortion from a tube amplifier is so much higher than a properly chosen capacitor that it swamps any other results you might want to hear. A cathode bypass capacitor can act partially as a filter element and have signal voltage across it, so you may hear some, especially poor electrolytic caps. However under a tube chassis you simply do not have room most often to mount a film capacitor of the same value. Bypassing an electrolytic is totally useless given the distortion mechanism. Just in case someone was about to mention that.
You have to look at an amplifier as a system whose performance is limited by various components - not just a few capacitors. So to pick on a few specific parts often will not improve performance nearly as much as claimed (again, more direct experience). Note I am not saying it is pointless to try, but you have to look at degree of difference and improvement. I yank out more oversized capacitors from chassis' and install decent electrolytics without suffering any meaningful performance issues. The parts are typically far too large and cause other issues, which is why they have to be pulled. I can't tell you how many customers were fleeced by internet reading non-technicians who "improve performance" of audio equipment while causing or missing real issues. This also happens with technicians who should know better and may well know they are taking a customer to the cleaners intentionally.
All,
In general, don't worry about using a film cap in place of an electrolytic cap. Besides, if you replace an old electrolytic with a good brand of electrolytic capacitor you will probably hear a difference. The old one likely had issues by now, and capacitors are simply better these days. Many old film caps have serious issues as well. Replacing them with good new types will also sound better and improve performance. No shock there either. It certainly isn't proof that the audiophile approved part is far superior to today's normal parts.
Reality check. AUdiophile capacitors are either produced by good manufacturers on their modern, well maintained equipment and will be equivalent to something they normally make (with a different label), or it may be made by the audiophile type company using old machines that are not as good. They do not have the volume to buy the good raw materials and maintain the equipment to ta level where it needs to be. Economics of scale. They will deny this tooth and nail, just so you know.
Please use common sense.
-Chris