Rocket caps blind test

[arjepsen]

Hey.
I just finished a rocket, and prepared to do some testing on how various caps would sound.
I recently got my hands on a Two Notes torpedo reload, and used it to record a dry guitar signal, and then just sent that to the amp, and recorded it back through the reload.
So I recorded a variety of caps - all with the same setting on the amp, playing back the same clip, so the only difference from clip to clip is the caps I changed.
(amp settings: volume at 10:30 -ish, treble at 9 o clock, bass at 12 o clock cut at 3 o clock)
The caps I changed are C8 C10 and C11.
Here's a list of the caps I recorded:

A: The standard panasonics as in the official BOM
B: Xicon MPPs
C: Jupiter Condenser yellows
D: Mallory 150 (old yellow type)
E: CDE 716 Orange Drops
F: Sozo Standard Mustard
G: CDE DME's

I'd like you guys to do a blind test, and let me know which ones you like best, so here's a link to downloading the files.
Their numbers doesn't mach this list...
Here's a link to a RAR file containing the wav files:
https://www.dropbox.com/s/4t4jsmmwk5oqu ... t.rar?dl=0
You will need something like winrar to unzip it.

tomorrow I might do another round with a higher gain setting.
Let me know what you guys think.
/Anders

[R.G.]

This is certainly a step in the right direction - blind testing instead of anecdotes.

However, there are some caveats here.
-> the value of the actual caps being tested must be matched to a high degree - well under 1% IMHO - for the test to not be attributable to cap value.
-> You really ought to do only one cap at a time. Caps in various parts of the amp affect the sound to different degrees.
-> A statistically valid test would do a number of samples of each cap to help eliminate variations within the brand and value. The more samples tested the more representative the results can be. One test of anything is an anecdote.
-> You really ought to to ask whether a difference is heard or not between some reference and the cap being tested, and the unknown cap being tested for "difference" should include the reference.

Designing tests for audio quality is notoriously difficult- but highly worthwhile. The hifi nutzos have been proven flatly wrong so many times that the "golden ears" pretty much won't participate in fair tests any more, claiming that the fact that it's a test can hide the subtle differences they claim to hear. I chuckle at that a lot.

Oh, and you need to do some serious thinking about whether the process of recording and playing back the clips through literally any computer A->D and then back through D->A and random unknown speakers may cause error in the test.

[xtian]

Ruh roh, no like your RAR:

UNRAR 3.60 beta 6 freeware Copyright (c) 1993-2006 Alexander Roshal

/Users/aaron/Desktop/Cap Test.rar is not RAR archive
No files to extract

Can ZIP or .gz or something?

[10thTx]

Sort of a similar quest for tone regarding various caps but done in a different fashion. Certainly not scientific and with a large amount of subjectivity but maybe still useful?

Kind of lengthy reading with 157 replies on this thread but perhaps some thoughtful information here: "Do capacitors sound different?"

http://el34world.com/Forum/index.php?topic=19534.0

With respect, 10thtx

[chief mushroom cloud]

how come no ceramic disc caps?

[arjepsen]

I agree that this test would be nice to do really thoroughly, one changed cap per turn, closely spec'ed tollerance etc.
However, I did this test with the caps i had available - I don't have the money to buy large quantities
Mainly I did this test to see if I could hear any notable difference between what I had on hand - I just felt like sharing it

[Stevem]

I would highly recommend you listen first thru decent Headphones in terms of clean volume check outs to pick up on small changes that your speaker may cover up and in doing such for one test try this.

Place a .22uf and a .022 uf cap , both with a 600 volt rating in parallel, then wire these two in across the last power supply filter in the preamp section and give a listen.

Then remove them and place them across if first power supply filter after the recto and give a listen.
I have found that bigger cased filters even though they have a tad higher ESR do work better for the plate and screen filters where they seem to have more reserve and make for better punch, that is if your looking for such.

Expensive audiofile type power supply filters provide no advantage if you place the above posted two caps across, again that's if your looking for that kind of definition in your tone.