I guess it is inductance because some resistors that sacrifice inductance (wirewound and thin film) to lower noise can cause 'damage' to sound. My preferred type thus happened to be metal film (of course not any metal film). Allen-Bradley is noisy and very audible at feedback position, but at 0.47 output resistor no noise or distortion perceived. I used to use low value AB for gate/base stopper, but I have better metal film replacement (secret!
)
Seriously? Have you compared the sound of inductors? So, why do they sound so dramatically different? Cause they have resistance?
There are fundamental issues in using electrical theory to explain what is audible. Lots of stuff simply remains unexplained. Practically everything relating to passive components.
It is perhaps more useful to stick to observations, even if subjective, than to come up with ridiculous oversimplifications. Like cable sound being fully explained by RLC
Another excellent option is wholesale denialism. I may not agree but it makes excellent logical sense and allows people to sleep at night and wake up still feeling like engineers.
Resistors by itself don't have sound. It is in an amplifier circuit that it can affect sound. Regarding amplifiers I have stated using simple one word: intermodulation. It is up to you to understand how small part with its small parameter changes can affect intermodulation or even stability of an amplifier. Lots of things are explainable, it is just not in public domain.
There are plenty of wikis and recent electronics textbooks covering the noise properties of various available types of resistors. However, components are evolving all the time, so it's a case of being up to date rather than reading outdated information in old publications but still parroted in some opinions expressed here.
For real figures, consult the manufacturer's current datasheets. These can be found at major parts distributor websites such as Digi-key, Mouser and Newark in the US. Otherwise, there are simplified guidelines and discussions published by many teaching sites. Some very generalised, others more informative and may cover all commonly available types. This may be all an audio DIY builder ever needs to know.
: resistors.Resistor Noise >> Resistor Guide
Ian et all, Skem was talking about known noisy *transistors.*. There is plenty out there on resistor noise.
It sounds like an plausible hypothesis, but does not explain why resistors retain their sound signature in all sorts of functions or circuits. A Shinkoh sounds always different to an AB or a Z-foil no matter if it is a grid stopper, a plate load, a riaa component, or part of an attenuator. That last example is sufficient on its own as attenuators clearly have a sound, irrespective of the amplification stages arround them. Don't you agree?
Those types seem to be pretty notorious for problems. There are two categories of “noise” - the noise that’s supposed to be there and the noise that’s not. I wouldn’t classify these as “noisy” because the offending noise is really a reliability problem. They start out fine from the factory and go bad (get noisy - REALLY noisy) in the field. Any time I see an A72x or C458 in a circuit it’s outta there - whether it’s bad yet or not.
Yes, and it's completely understandable why they would be motivated to share as little of their knowledge as possible. They are engaged in commercial enterprises, where their ability to compete largely rests on their accumulated personal knowledge, experience and judgement, beyond that found in standard engineering texts. That's why the DIY community is so fortunate that engineers such as Nelson Pass, John Curl and others do share some of their valuable knowledge with us.
That said, however, I'm uncertain as to what that observation might have to do with my earlier comment that the holy grail of specialty consumer audio engineering is obtaining complete knowledge of exactly which objective engineering parameters control exactly which subjective listening characteristics, thereby enabling perfect designer control over them.
Yes, I agree. But I don't think that it is a strange phenomenon. Just like when I watched illusion show by magicians, I might not know what is going on but i usually can see a few possibilities. Imo, there is only one thing to understand in audio, after that everything else falls nicely into its place. Problem is, it is my intention to refrain myself from getting involve with related discussions.
I think no one was disagreeing with you (not me or analog_sa). But i will easily disagree on your last statement.
Datasheets can only go so far. The -ONLY- way to test a particular device in an audio circuit is to (1) create a proper design, and (2) LISTEN to it. Develop a blind ABX test criteria to compare against a well-know reference. We've been doing this for 25+ years, and it's the only way to properly design high-end audio. By ear.
IC "audio opamps" can exhibit a wide range of sonic performance, with some indeed tending towards "grainy." The datasheet THD measurements of IC opamps are perceptually meaningless. There is even a modest correlation between ultra-low THD and poor sound quality (not always). Pretty sure this has to do with massive amounts of feedback used to achieve the ultra-low THD numbers.
Among the best perceptual active devices we've found (and use) are discrete FET-based operational amplifiers (bespoke). THD is high (in the .01 range, worst case), and noise isn't as good as, say, an LT1115, but sonic performance is better than virtually every IC we've tested.
IC "audio opamps" can exhibit a wide range of sonic performance, with some indeed tending towards "grainy." The datasheet THD measurements of IC opamps are perceptually meaningless. There is even a modest correlation between ultra-low THD and poor sound quality (not always). Pretty sure this has to do with massive amounts of feedback used to achieve the ultra-low THD numbers.
Among the best perceptual active devices we've found (and use) are discrete FET-based operational amplifiers (bespoke). THD is high (in the .01 range, worst case), and noise isn't as good as, say, an LT1115, but sonic performance is better than virtually every IC we've tested.
That makes a lot of sense and well explained.
I replace those terrible C458s with Fairchild KSC1845. Interesting to know what you replace them with and if KSC1845 are a good replacement?
The guy who designed the very popular Pass Labs HPA-1 lives about one mile from here. We know each other pretty well, and he has talked a lot about how he designs. Seems to me your assessment is about right: There is a lot of empirical knowledge, although measurement plays some role too. The empirical knowledge comes from many years of doing a lot of experiments and listening carefully to results. Also, most of the better designers know each other and share some of what they know with each other partly out of professional courtesy and partly out of reciprocity (if one shares some bit of knowledge then the other may wish to reciprocate), although they always keep some secrets to themselves.
Some have tried to share valuable information here in the forum, information that is not widely known. However, they tend to get told they don't know what they are talking about by people who only believe what is 'proven' by measurements and AES publications. So, they quit trying. Not worth all the abuse.
Last edited:
Perhaps, you can be specific regarding what exactly disagree with about that last statement and why?
With the release of consistantly subjectively musical sounding products from the likes of conrad-johnson design, Pass Labs, Lamm, etc., why would you guess that their designers have no better idea than some others? Rather, does it not seem likely with their product track records that they do, indeed, know something that some other audio engineers do not? Not only is empiracle knowledge still knowledge, it's knowledge that cannot be gained simply from from reading an engineering school text.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.