the cap question

[davepedals]

so what's the real advantage of using mylar film, etc caps over ceramic caps?

[darron]

on caps and differences, there's a visual representation here on how the linearity of them can change....
http://greygum.net/sbench/sbench102/caps.html

[petemoore]

  so what's the real advantage of using mylar film, etc caps over ceramic caps?
  The 'ol Fender-amp-design-choice capacitor materials used CC's. Replaced with 'upgrade' materials, the result is reportedly hard, glassy edge replaced nice bright tone.
  The difference in a 'distortion circuit' to capacitor comparison is like 1 to googleplex.
  Main deal that'd be even detectable under distortion [say clipper] would be the component tolerance = frequency response targeting difficulty, if drifting or way off, a series>of>signal>staging>capacitors which alter voicing could throw freq response 'over there'...if they're all a leanin' that way of course.
  There have been reports that CC cap distortion in particular setting such as LP filter across FF Q2 B/C may impart particular wierdness..a faint high zizz-tone, confirmed only once that I know of, only apparent of course when FF was really cleaned up, [if..I like my FF rolloff capacitor, it's goes across the: 'it is a 1k to V+'  in my fuzzface in any case.
  http://fuzzcentral.ssguitar.com/axisface.php
  So for this for that, for-what-the other or what basically, HV, LV, lean/clean distortomean...
  Various types of real estate but it's all location location location, and what happens to be nearby.
 

[familyortiz]

In general, the improvement is in stability. Tighter tolerance and less drift. This matter in blocks like filters, both active and passive where you want repeatability from one design to another. That is not to say that ceramics are not suited for other uses like coupling/decoupling, timing and biasing where the wider tolerance range is not an issue.

[davepedals]

Thanks for the detailed info!

[darron]

on what petemoore has said on tolerance, here's a copy+paste off my website:




Tolerance

This seems to be the biggest selling point when people try to tell you that they use the best components. The idea is that if you want to match your original design, you want the values to be as close as possible. Why is this a problem? If, for example, you have high/low pass filter tone control, it will be made with a resistor and a capacitor. We will explore the values of a 1KΩ resistor and a 0.1μF capacitor. If they were both within ±10% tolerance, they have the potential to be from 0.9KΩ/0.09μF to 1.1KΩ/0.11μF. This changes the roll off frequency from approximately 1.97kHz to 1.32kHz - which could be somewhat noticeable.

I think it's a bit of a misconception sometimes that people think, for example, if every component was within ±10% tolerance, than it could all at up (10%+10%+10%+10% etc.) to a product that has potential to stray miles from the original. In reality, if every component in a circuit had a tolerance rating of ±10%, then the difference from one whole circuit to the next would be ±10%. Over all, it's not a shocking difference, but if you can keep the tolerances tight then you may as well.