Better than True By-Pass?

[Guitar Toad]

What if every pedal you built had a Simple IC Buffer or other decent buffer circuit, then the foot switch selects bewteen buffer-only and buffer+effect, thus every pedal would be a buffer. Wouldn't that be better than true by-pass? There must be some pedals that already do this...I just don't know which ones they are....just a thought.

[Mark Hammer]

That is the rationale used by a great many existing manufacturers.  If I'm not mistaken Roger Mayer pedals give you the choice of buffered or not buffered bypass outputs.  Visual Sound pedals also have a buffer and TB bypass I believe.

[JHS]

The old Korg pedals use a circuit with an I-buffer that buffer the guitar in bypass mode and the FX when the pedal is engaged.

JHS

[R.G.]

Buffered bypass is not always better or always worse. It depends on the pedal - old fuzz face style germanium pedals certainly depend on the pickup impedance and not having anything between the guitar and the distortion for some of their sounds. That being said, there are very few pedals for which the guitar pickup driving the effect directly is of any consequence whatever, and those pedals are primary archeological ones and their clones.

For a more in-depth discussion, you can read my recent article in Musician's Hotline "Should you bypass True Bypass?"

The merits of true bypass are that it prevents any form of loading down the guitar signal and causing treble loss the "tone sucking" you'll hear of. Also, it is guaranteed to let you work in bypassed mode if your battery is dead.

Buffered bypass sidesteps the loading by ensuring that no matter what, the pedal does not load down the guitar. If a pedal's input impedance is high enough, it simply does not matter whether it is connected to the guitar signal in bypassed mode or not. It doesn't load, period.

It sidesteps the "works when your battery is dead" issue as well, because if your battery is dead, you can replug the cord. That's what you'd probably do with a dead-battery true bypass pedal too, unless it's knit into a mondo pedalboard. However, mondo-pedalboards are the boards most likely to have wall-powered supplies and least likely to be all battery, so this becomes a very-much secondary issue.

As to what pedals are buffered bypass -

Ibanez. All of them with JFET switching.
Boss. All of them with JFET switching.
Visual Sound. All of them. The buffer in the VS line is not a simple transistor emitter follower like the Boss and Ibanez, but rather a buffer designed both by theory and ear to give good results.

There are probably many others. And yes, Roger Mayer won't do a true bypass. He's adamant on the issue that only buffers are worthwhile, from an interview with him a ways back.

In the issue of bypassing, Mother Nature intrudes a lot. There are no perfect bypas buffers - or stompswitches, or relays, or JFETs, or whatever. There are only things that work better or worse, depending on the application. There is no substitute for knowing both Mother's Rules and how to apply them.

[Mark Hammer]

There are "samples" of different bypass on the CD accompanying Dave Hunter's guitar pedal book, and the Visual Sound sample of straight guitar, through the buffer to the output, is impressive.  What I've never heard, though, and what I'd like to hear, is a high-quality sample of, say, 8 Boss/DOD pedals in series (or anything else using FET switching), all in bypass mode, compared to the same length of patch cord/cable in a straight-wire (TB) configuration.  I think that would be informative for a lot of people.  The third reference point would be the same guitar going into one buffer and then through 7 TB pedals.  In other words, something that lets the listener identify the consequences of having many cascaded buffers, vs no buffering over the same number of stages (and feet of cable), vs a single buffer stage between guitar and the world.

What gets lost in a lot of the discussion regarding buffers is the difference between the single pedal and the many-pedal context.  I think you'd agree, Brother K, that they ARE different.

[wampcat1]

Good stuff RG, also, you might put the article url up here.

Mark, I've done that before and there is a "slight" difference, I'll put a soundclip up here in a minute since I already have all the equipment ready.

Brian

[Guitar Toad]

RG-

Thanks for that excellent post. I'll need to look at your article. Is it a current article or is it archived? Hopefully, it discusses the guitar signal loading vs unloading and compares the various true by-pass schemes.

My apology for that terrible word choice in the thread title...I could have selected a better word than better. 

Mark Hammer...YES...I need to hear what 8 BOSS pedals sound like in bypass...that would be a greatly informative sampling...

[twabelljr]

     Here is the link to R.G.'s article:  http://www.musicianshotline.com/education/200603_educationcenter_techviews.asp

[Paul Marossy]

Also, it is guaranteed to let you work in bypassed mode if your battery is dead.

I don't use batteries in many of my effects, but I do like this fact!

If a pedal's input impedance is high enough, it simply does not matter whether it is connected to the guitar signal in bypassed mode or not. It doesn't load, period.

That's my viewpoint. I hear of people true bypassing Boss pedals, and I wonder "why?!". I haven't measured the input impedances of any my various Boss pedals, but I would venture to say that they are typically high enough to not load the signal. 

[wampcat1]

Ok, a soundclip:
http://www.indyguitarist.com/soundclips2/TB-demo-306.mp3

I didn't have 8 boss pedals right now, but I have a ton of buffered pedals. For the soundclip I'm using a boss od2, dd3, ibanez ts5, dano eq, arion chorus and octave pedals, I'll explain in the mp3 more.

bw

[Mark Hammer]

I wish they had a boy-scout badge for that one.  Excellent work, Brian!  Shows both pros and cons of buffers, and the "dark side" of cables.  Nice, and required listening.

If I was able to correctly link the switch clicks (i.e., the acoustic sound of the switch picked up by the mic) and the portions of the sample.  A bunch of buffers in series (and I certainly can't speak to their respective quality) tends to eat some of the girth and low end, while saving much of the top end.  That's sort of what one would expect with all those series caps.  No single one shaves off THAT much bottom, but add 'em up and they start to matter.  The hiss level did not seem that audible, but I'll reserve judgment since it is a mic'd amp.  It may, or may not, be the case that hiss becomes audible if going into a board or sound-card with wider bandwidth.

At the same time, with realistic cable lengths, it is clear that buffering the signal helps immensely with preserving clarity.  If I was Grant Green or someone like that, I suppose it wouldn't matter, but for anyone playing single coils it obviously does matter.

One of the things the sample suggests is that it may well be helpful to replace input caps with larger values.  For instance, a .022 instead of a .01, a .1 or .15 instead of a .047, just to keep a little more bottom.  If you're a wah-fuzz-chorus kind of person, then it likely makes no difference, but if you have a music-stoe display case on your pedalboard, it might help.

[bwanasonic]

The main reason I often place buffered FX in a bypass loop is control. If I want a buffer, I can place a TBP one where I want in my signal chain, without having to worry about interactions with older "vintage" style FX (FF, RM, etc.). I always have a buffer on, but it's where I want it in my signal chain.

Kerry M

[bioroids]

Let's not forget that the Fet bypass system provides some distortion and coloration of its own, I don't know how much, but this often is forgotten in True Bypass/Buffer discusions.

Another problem with true bypass is that the bypassed signal can sound duller than the effect signal. Say for example in a delay pedal. When you switch in the delay, ideally the clean sound should stay exactly the same. But no, because the effect path is buffered, and it passes along all the meters of cable without loosing high end. So in the end the bypassed signal, in comparison with the effect, sound a little like "tone sucked".

I think the best may be to have a mechanical (carling or whatever) switch, and an input buffer that is internally set. If you want complete true-bypass you have it, or if you want the buffer you select it with a jumper inside the box, and voila, you have it too. That can't be too difficult to implement.

Luck

Miguel

[Guitar Toad]

On the subject of loading and unloading the guitar signal, is it correct to say that an ideal buffer for a dual-coil pickup would be different than the ideal for a single coil guitar? The SC buffer should have a lower impedance than a DC guitar. Is that right?