Why do standard size pedals sound better than mini?

[R.G.]

It may possibly be that some SMD circuits have some indefinable something that's worse in some way than a through hole version. Maybe.

Not all SMD parts are alike. As Jack pointed out, some SMD resistors are noisier than through-hole carbon film - generally, the thick film ones - but some SMD parts are not noisier. The metal film types come to mind here. This is not a cataclysmic difference, but is measurable with proper instrumentation. Through hole caps have wildly different characteristics, not only to through hole types, but to one another. I an SMD design sounds different to a through hole design, I'd bet on poor choice of dielectric types in the design. I'm pretty sure I can measure the difference there, even with my home instrumentation and some thought. SMD semiconductors will show the smallest difference, with the SMD parts being in general superior because the semiconductor die inside is the same as the TH part, just with shorter and lower-parasitic leads.

So yeah, there are possible differences. But are the differences audible (a) in any case or (b) in ALL cases. I'd say the answer to (a) is yes, and to (b) no.

In general, a human can't trust simple swapping tests that they perform because they cannot eliminate their own internal self biases. Psychologists developed the whole blind and double blind testing methods because they got fooled by their own testing and over time had to figure out ways to eliminate the biases of testers in general. This has been found to be effective.

It has been effective particularly in audio quality tests. Hifi  tweakos, as I mentioned, failed blind testing spectacularly. People who claimed to be able to hear the differences in, say, ordinary copper wire and silver wire or ordinary copper wire and oxygen-free copper wire often performed worse than random guessing or coin tossing in simply telling if one was different from the other. This was the case in testing [insert claim here] as well. It was so bad that dedicated hifi tweakos developed the idea that the process of testing itself caused the differences to vanish, or that humans detected differences that were impossible to measure as reasons for why even participating in a test was laughable. They decided to belittle any testing that exposed the idea that they were just making up the differences, or worse that they could not really discriminate the tiny, tiny differences that they claimed were so very, very irritating.

I can accept that you may be the exception to the repeatable results found by blind testing. It is much like the old saw - the race is not always to the swift, nor the contest to the strong,  but that's the way to bet. I suspect that some SMD circuits do indeed use poor (in a real, repeatably measurable sense) part selection for the task at hand. That does not necessarily prove that all SMD circuits will have the same flaws.

It is also possible that careful testing of SMD versus non-SMD circuits carefully executed to eliminate differences other than the parts selection may turn up heretofore unsuspected physical principles that do have a vast effect on perceived audio, maybe. If so, the careful testing was valuable. If the testing isn't done, there's no way to improve all audio in the future.

[Mark Hammer]

SMD semiconductors will show the smallest difference, with the SMD parts being in general superior because the semiconductor die inside is the same as the TH part, just with shorter and lower-parasitic leads.

Good point from someone who knows.  We tend to forget that the difference in size between TH and SMD semiconductors is not necessarily a reflection of the "guts", but a reflection of an industry-standard form factor to accommodate any and (nearly) all possible "guts" in a single epoxy-with-pins package.  No different than a music CD can have a 4 minute single, or 60+ minutes of content, in the same form-factor.

So yeah, there are possible differences. But are the differences audible (a) in any case or (b) in ALL cases. I'd say the answer to (a) is yes, and to (b) no.

Sums up my feelings on the matter nicely.
These days, one need be careful about drawing overly strong and broad inferences based on limited data.  Misconceptions spread over the net like ebola in this day and age, and common sense can have a hard time catching up.

[stallik]

I'd like to add into the melting pot, the variability of (my) ears. My preferences may be fixed or change with my mood but the old lug holes tend to filter various high frequencies apparently at random and i have no control over that. When I A/B different builds of the same pedal, i may pick one which sounds much better but, as sure as eggs are eggs, I'll pick a different one the next day. It's the same as me preferring one guitar to the next on successive days

I've also noticed that if the volume increases even slightly in one build, it inevitably rises to the top of the list

[ElectricDruid]

I've also noticed that if the volume increases even slightly in one build, it inevitably rises to the top of the list

I've seriously considered adding a subtle boost to all my pedals just to take advantage of this effect. If the overall gain is just a little bit above unity (maybe +2, 3, 4dB or so?) there's a sudden "presence" when you turn the effect on. It grabs you by the ears more. If the volume increase is subtle enough that you don't really hear it as obviously louder, you'll interpret it in other ways, but mostly positively ("more clarity", "more presence", "cuts through", etc). Obviously this doesn't apply to stuff like fuzzes with a Gain and a Volume control, but it might work on a Phaser, Chorus, Flange, Filter, etc.

[GGBB]

I think we may be missing the forest for the trees. Small enclosures simply contain far less mojo than big enclosures. It's still physics, but the simple kind.

[R.G.]

I've also noticed that if the volume increases even slightly in one build, it inevitably rises to the top of the list

I've seriously considered adding a subtle boost to all my pedals just to take advantage of this effect. If the overall gain is just a little bit above unity (maybe +2, 3, 4dB or so?) there's a sudden "presence" when you turn the effect on. It grabs you by the ears more. If the volume increase is subtle enough that you don't really hear it as obviously louder, you'll interpret it in other ways, but mostly positively ("more clarity", "more presence", "cuts through", etc). Obviously this doesn't apply to stuff like fuzzes with a Gain and a Volume control, but it might work on a Phaser, Chorus, Flange, Filter, etc.

This is a result that I remember from reading about music reproduction testing. Any ampliifer/speaker that resulted in the sound being slightly but non-obviously louder than an otherwise same reproduction was judged better in the usual terms of hifi tweako discourse - better clarity, notes less smeared, a "space" between the noted, more three-dimensional sound stage, quieter quiet passages, less noise, etc.  There were some ugly situations when the two sound chains being tested were occasionally revealed to be the same sound chain, but with a sub-db boost on one setting.

It is nearly impossible for the human ear to accurately diagnose subtle volume boosts. You get descriptions like a clear start with an oaky finish and overtones of parsnips and rutabaga.   

[ElectricDruid]

It is nearly impossible for the human ear to accurately diagnose subtle volume boosts. You get descriptions like a clear start with an oaky finish and overtones of parsnips and rutabaga.   

My lawyers will be contacting your people, R.G., seeking compensation for the injuries I sustained falling off my chair laughing at your post.

T.