Parts quality does matter, yes, but..

[hiktaka]

on some designs, it doesn't matter as much as on another designs.

I've built some DIY; but I found that sound difference between exotic parts vs
cheap parts aren't same for all designs.

Of few that i've built:
JLH 10W is the least sensitive, even on the breadboard using scrap parts sound pretty good.
Mini-Aleph and F5 are moderately parts-affected. Difference are noticeable, but with cheap parts, it's not that bad.
The most picky for parts is Hiraga Lemonstre. Change a resistor or two (with cheap replacement) and degrade badly.

Any opinions?
Thanks,

[DF96]

I will get shouted at for saying this, but my opinion is that given reasonably decent component quality then the better the circuit design the less it will be fussy about components.

[counter culture]

Some designs require matched parts, some do not. Some parts of circuits are critical with regard to what type of part is used, some are not.

In some places it's not clear exactly what influence part selection may have and it's difficult anyway to know whether the reputations some parts enjoy is deserved.

I pick parts according to known factors affecting functionality and by price (how cheap can I get it). What else can one do apart from abandon oneself to the vagaries of internet mythology? Not my cup of tea, vagaries.

[KatieandDad]

Manufacturers will aim towards a fair design with fair parts. ASFAIAC DIY is concerned it will always be a compromise. To a certain degree parts will make a difference, that should be counteracted by good design.

I've got an Aleph 4 where the specified capacitors were very poor in relation to reasonably priced alternatives. Resistor changes in this deisgn make no difference IMHO, as long as you start with half decent components, this one uses 1% vishays from RS.

Any capacitor in the signal path will make a difference but you don't have to GO STUPID.

I fronted mine with a B1. There was ne real difference between PIO's and good MKPs in the signal path.

[CBS240]

It all depends on the function. A general purpose transistor is just that, general. Price depends on other factors. There is a SOT-563 duel transistor package that I use quite a few of. They are generally ~20 cents ea. I found my supplier has a different brand with similar properties for 5 cents. 2.5 cents per transistor is not bad. One could argue that resistor matierial has different noise specs. It depends on the function of the resistor in the circuit as to wether or not it makes a difference. Some resistive material can catch fire, others do not.

[lanchile]

I go for Design and parts quality! combine both and you get a killer amp

[Tony]

Originally Posted by DF96

I will get shouted at for saying this, but my opinion is that given reasonably decent component quality then the better the circuit design the less it will be fussy about components.

i am not shouting, in fact i agree with you whole heartedly....

"it is not much the hatchet, but the Indian throwing it...."

[Ian Finch]

This might just be the corollary of DF96's post above but, as a general observation, simple design amplifiers do seem critical on component choices, grades and quality. More elaborate designs, containing refinements that ensure optimal circuit operation, are much less so.

Importantly, simple amplifiers can often have very unique and desirable sound qualities, so it is then a matter of choosing what sort of amplifier you really want and this leads to a trade-off of parts cost against complexity. Through ignorance, and I guess the conclusion that expensive parts must then result in better sound quality, some guys just buy the biggest and most expensive parts and brands available in the hope that this will confer better sound quality.

I think that's where the idea falls apart.

[Coconuts 500]

Be kind to your wallet; do more blind testing. Observe how the myth that price and performance are related comes crashing down. Enjoy your new found enlightenment.

[hiktaka]

IMO, here are some of parts function that have significance of by parts quality

- Input C: very important, very noticeable difference.
- R in signal path: noticeable, but a cheap one won't be bad.
- R & C in bias circuit: doesn't matter.
- Output cap: the big one doesn't matter, oddly bypass C (with smaller uF) does.
- Semiconductor: in a simple design, yes. In a complex design, no.
- Passive load choke, output L: should be, I have less experience.
- PSU parts: Trafo should have biggest impact

I won't even go to talk about cables & interconnects

Thanks.