Arguably, this is Edge's 'greatest delay song' because the delay is so clear and audible and yet it can't easily be
separated from the guitar signal, they are so well woven together. In almost every other song where
a long delay is used, a casual listener can probably tell that some kind of delay is being used ('Streets', for instance).
In 'Bad', for the most part, you would probably never know that half of the guitar notes you hear are from the delay.
Anyone who claims that Edge is not a 'great' guitar player, should try playing the main riff to this song
with the right delay. Your timing has to be perfect or it sounds awful.
The main riff is something of 'tongue twister' to play with the delay if you're not used to it.
It's important to keep the low notes palm-muted and to hit them on every beat to help keep time. The high notes should ring
out. Edge picks all of the notes as downstrokes except the high notes. Play all of the notes in even time! (The delay adds the rhythmic element of the song and its 2 or 3 riffs).
(A) (let the high notes ring) (D)
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 ..
(palm muted ---) (palm muted ---) (palm muted)
(these notes should not ring, they should sound very staccato)
Solo: (all notes are 1/8th notes, in even time)
1 2 3 4 5 6 7 8 - 9 10 11 12 13 14 15 16
Since the delay fills in the other half of the notes you hear, the above (slow) riff sounds like these 16th notes,
but he doesn't actually play this! (The red notes are the notes added by the delay unit, 3/16th behind):
1 & 2 & 3 & 4 & 5 & 6 & 7 & 8 & - 1 2 ..
And the delay fills in the rest.
- Starting at time 0.00 at the beginning of the CD track, the sample runs from [2.2 - 3.5 secs].
- In the sample you hear the first high 'E' note (12th fret, E string) Edge plays in the song.
There's clearly a mic bleed (~5ms, 1-2 repeats) surrounding the attack (see the main page for more info). Then there's a slight delay
at 220ms (approx 3/32 @ 95 bpm) with no echo around it. Then there's the
main delay that you hear in the song, at about the same volume as the initial attack, with the same
room echo as the initial attack.
- At Live Aid, they played the song at 102 bpm. Edge's main delay was set to 440 ms (exactly 3/16 @ 102).
There was also a secondary delay, as in the album version, at 330 ms.
Here's an mp3 of me playing the 2 main riffs from the song,|
with no delay and, later, with delay.
(I turned the delay on a few seconds into the recording).
I used a Fender Telecaster (middle pickup)
into a $100 Line 6 Toneport on my PC with the setup on the right.
I had to use a Korg SDD-3000 for the delay because
the Toneport delay options don't (yet) include a digital
delay with modulation.
Here's the sample clip I studied for this song
(it's slowed down 8x maintaining the same pitch and matches the waveform in Figure 1)
Figure 1: Waveform of the sample clip
- There are 3 delays. This is the most likely signal path:
Guitar -> Signal split to A/B:
A -> 467ms (1 repeat) == 3/16 @ 95 bpm
with modulation, 90-100% of the inital signal -> amp #1 (loud) + Mic bleed/room echo
B -> 221ms (1 repeat) == about 3/32 @ 95 bpm -> amp #2 (softly)
(Note: I think the delay settings he used for Live Aid and other live shows
sounds better, where this 'shorter' delay is set to 3/4 the time of the main
delay above, instead of 1/2 the time. If he had used that setting on the
Unforgettable Fire, this delay would have been set to 350ms). + NO mic bleed/room echo
Note: It's important to have exactly 1 delay repeat on the input signal and no faint further repeats.
This is especially true in the 'solo' section (part 3 of the clip of me playing), there is
a distinct rhythm that the delay makes with the actual attack. If the delay even faintly repeats
again, that rhythm is lost and the overall sound becomes flat.
It's also important for the delay to be the same level as the input signal. On some delay
units, this may mean turning the feedback to 0 and setting the dry out less than the delay out
Additional note: The 467ms (3/16) delay is the important one! You can play the song with just this. The other
delay is not neccessary, but is mentioned here because Edge uses it, even though you can't really
hear it in the album or live recordings.
Interestingly there's no echo on the 221ms delay. This may be because he set up the amp
that was playing this delay (after his signal split) in a place that had no room echo and no bleed over from other amps. In 'The Unforgettable
Fire' video (available on the 'Live at Slane Castle' DVD), you can see he has one amp set up
outside of the castle -- so maybe that's what we see here...
Recorded with: 1976 Gibson Explorer (maple) - the Z-shaped guitar he used for the 'Boy' album
(click for email)
HI there, well, i completely checked out you heavy work here. Really Amazing
what you`ve done. But i must admit, sorry, that i can`t hear anything in the
soundclips to lead to your conclusions and some U2 riffs you played are not
correctly performed. Maybe sometimes it`s better to first learn how to play...I
m sorry, but thats the way it is. However your work is really interesting and
-- ZORO, Apr 14, 2005
The Edge es un gran profesional y una persona admirable por su incansable
auto exigencia, me gustar a sugerirle talvez qure alternara en nuevas canciones
instantes sin DELAY, que el sabraaventarlos sabiamente.
-- carlos parra,
Apr 25, 2005
Here are some live recordings of Edges work performed by a U2 tribute band that
has put a lot of work into sounding identical to U2... www.notu2.com
-- Joe Cumia, May 4, 2005
i think you get a great work in doing this job.but to my ears this doesnt sound
right.....sorry. i dont think this short delay is actual there...the samplers
didnt sound right,you can play any u2 song with just one delay and sound
great(but you have to play using exactly the same tecnique the edge is using or
you will sound awfull) the delay times you have write in this site is
correct,theres no mistery,since you just have to count the bpm in any other
song or version and use some math to get the time i sorry you have a tremendous
job in there,but...is wrong
-- estevao, Aug 25, 2005
I can only use two delays,so i deleted the short "5ms" delay and it sounds pretty great to me.Its
obvious that the edge uses some hardware to get his "exact sound".If you look behind him he has two
fridges full of processors.Great job though i was just listening to bad and it made me think how
his sound was created.So i jumped on to the net and found this site first off.I'm pretty impressed
as you go into great detail and once i programmed the times into my rocktron chameleon PRESTO! im
the edge but with only $200 million less in the bank .Whether this is exactly the right way ,who
cares?Its given me something to play with and may be develop my own new sound and thats whats
important! keep up nthe good work J.F.
-- famous, Aug 25, 2005
whack to ya! that's bang on, you've done your eccer with the gizmo
-- dave e., Sep 17, 2005
I appreciate your work here but the samples say it all - no, no, no. BAD is the
best example. A simple dotted 8th modulated delay nails it. Your sample proves
the point that you're making a mountain out of a molehill. The Edge is
brilliant because he does so much with so little - no need to complicate it,
-- b dotson, October 18, 2005
Greate site. Thank you :)
-- James, Oct 29, 2005
You have done some solid work here. I think that in some instances, a short delay coupled with a long delay may not be the answer to the master's riddle.
Bad has alot more actual picking going on than some might realize. If you actually try to pick the part the way it is sounding to you, then add the longer
delay I think the results are surprising. The longer delay needs a little more regeneration than 1 repeat, but not much more. I think the key here is
the constant rhythmic picking of the "drone" strings (open A & open D) along with the notes at frets 11/12 (C# & D) and frets 7/9 (D & E). I believe
Edge does use up and downstrokes (which makes it even trickier to hit both strings evenly and consistently) and the delay fills in the spaces. All you
need to do is listen to the begining and you see he IS picking this way. He starts on the open A picking in time 4- 16th notes to the beat
(down/up/down/up), then adds to it. The delay fills out the !
part, but he keeps the rhythmic motion not by adding delays, but by returning to pick those open string/fretted note combinations. This is what allows you
to have that continuous, almost sustained quality to the part without relying on the delays to do the work for you. Besides, if you used multiple delays
or kept the regeneration level greater than 1-2 strong beats, you do not get the same sound. I think The Edge is a greater guitarist than most
realize for this fact, his effects color his sound. They contribute to the end product. They do not create it...he does. Thank you for the site, it's
good to see people thinking and sharing ideas about music that matters.
-- Falcone, Nov 12, 2005
hey, im not sure if im thinking correctly, but i think some of the "room echo"
and "short delay" you speak of may in fact be caused by his amp itself. The vox
Ac-30 that edge uses so often has a delay knob i believe, or if it isnt a knob
then it is just a natural part of the amps sound. but the amp naturally has
delay to it, so i think that may account for some of the multidelaying that
-- Tyler, Dec 10, 2005
i think its is great....i hv my own band/act and what i do is try to start out
to emulate the edge, but i end up somewhere else that totally becomes
mine....even though i was trying to be the Edge.....i ended up being me....
its a natural accident but its great to read the notes here and see how things
are done...using delays and all that.....thx theauguststory.com
-- FZE, Mar 5, 2006
Hello, first of all why are we trying to solve this after 20 odd years lol..
2006 and we cant figure out Bad from 1985? LOL. Looks like we are gona have to
go back in time and see what they were using .. problem is U2 has always been
very secretive about their work.. why do you think they have had the same
people working for them for 20 years?? .. TRUST.In fact there was a theory a
while back that Daniel Lanois was at the stadium in the back doing overdubs for
edge. in some parts you can hear it clearly that there is some sampling going
on.Now it's a sampler or another guitar going i dont know .. here is what i use
Digitech 2112 SGS. Cubase SX 3 PSP 42 VST. and PSP 84 VST my Bad version is an
analog delay set to 440 and 50% smear and 0 feedback in the Digitech unit with
some chorus ( it completes the BAD sound) and some reverb. Then the signal goes
into my edirol Multitrack unit into cubase sx 3 wich i use sometimes for
outboard effects..the PSP 42 is set to 441 ms and no feedback and 2$ modulation. I think the PSP 42 is the best anolog
sounding delay VSt effect out there.it is smoth and has lots of features
including the famous filters wich the KORG sd 2000, 2200. 3000. are famous
for.. Best tip is to GET a KORG!!!!!! and an AC30 VOX amp. that is the begining
of the u2 sound.. remember edges using metal picks too i have one made of
aluminum and one from a floppy disk metal slider.. trust me it works in getting
that chingy sound. i think the problem is we need to get analog sound gear not
digital.. in streets that all changed.. but you can still hear the korgs
going.. GREAT WORK...
-- Carlos Q., Apr 14, 2006
Hey, great work here! I love the attention to detail and all of the time you must have put in analyzing the waveforms, etc. Just a few
suggestions/cautions for readers: First, the edge's picking is often more complicated than what's going on in the dry (or non-delayed) parts of the
posted samples. Simply put: You can't use more complicated delay settings to compensate for some really nuanced and subtle playing. I don't think the
Edge started out with such complicated technology (which is a good thing because if he did, he wouldn't have developed his unique style). Try playing
Bad or Streets or whatever WITHOUT delay. When you have something that sounds pretty good and it sounds like you're actually picking some of the
delayed notes, then add some delay (for Bad, a simple delay at about 440-460ms with 1 strong feedback at 100 percent volume and another at about 50%
volume for instance). Second, play WITH your delay. That is, don't just try to copy the ex!
act BPM or delay settings of the song and then play to what you're hearing on CD. Set your delay close to what sounds right, turn off the CD and play
what feels right at the tempo that feels right given your delay settings. It doesn't really matter if you're at 440 or 442 or 460ms; you can make BAD
sound really good by working with your delay and your own tempo (and you'll probably end up putting a bit of your own style in as well). Third, use
compression and know how to use it. This is essential. Fourth, remember that the Edge would be dissapointed in you if all you're trying to do is copy
him. I dialed in the Streets settings suggested here, played the intro once, and then spent two hours coming up with my own riffs using that exact
setting. Again, thanks for the info!
-- Liam, May 9, 2006
Hmmm, very good comments all round, I found Falcone's one to be quite interesting with regard to the up-down strokes and the longer repeats ... I'm adding simply because I've just finished a little writeup on our band site about the gear I happen to use, since at our last gig we had a punter who insisted on taking quite a lot of photos with his phone(!), not to mention a few others taking a closer look. Anyways, the use of this gear was inspired 80% by the Edge ... (I am a fan of over 20 years) and 20% David Gilmour. Very similar in terms of their relentless pursuit of 'hi fidelity' sound as it were. So, the gear I use: RECORDING Digitech RP20, functioning as an A/B Splitter, and Overdrive across a 2 amp setup ... L/R outputs go to a Custom Patchbay/Router ...which then goes off to ... [A/Left Channel] Korg SDD-1000 Digital Delay (A Channel) Boss GL100 Guitar Driver (A Channel) >> Mix/Direct/Wet to inputs 1-3 on Presonus 8 Channel Firepod >> Router also sends the Direct and Wet ouputs to a 100W H/H Amp Ch 1 and Ch 2 respectively. Each Channel on this amp has seperate volume and tone controls, which is handy - multiple ways of mixing the signal all the way to the amp to get exactly the right mix. [B/Right Channel] Korg SDD-3000 Digital Delay Ibanez DUE400 Processor (used pretty much solely for Compressors) >> The SDD-3000's outputs go to the Presonus's channels 4-6 >> THe Router sends the mix output to a 50W Marshall Amp THE LIVE SETUP Bringing that lot into a more compact version, I made a custom pedalboard which has 2 levels, containing ... (CLEAN System) Korg AX30G Effects Processor (Clean settings & Main Backup) Morley Wah A/B Splitter for a or b Guitar (one std tuning, the other "D" tuning for our originals. One is a Fender Strat, Rosewood, with Seymour Duncan SSL-5 in the bridge position, the other a Strat with a EMG-HZ in the Bridge) (OVERDRIVE/Lead System) Boss SD-2 Dual Overdrive (Rhythm / Lead Overdrives) Marshall Guv'nor
Plus Overdrive (Overdrive booster) Boss DD6 Digital Delay (set to a fast delay normally) Boss BF-3 Flanger Line 6 ToneCore EchoPark Delay (set to a slow delay/80% Wet) [This has a great Tape delay mode] For amplification live, I use the H/H with the 2 inputs, as well as a Fender 300W Bass amp + Cab, which also has 2 inputs. It allows stereo for both the clean and the Overdrive sections of the board, as both sections have l/r outputs (the overdrive section of my board has pedals that feature stereo outputs) I had an engineer friend design a custom switching device as well, that handles multiple input/output configurations. Put simply, it allows the korg ax30g's stereo outputs on the bottom level of the effects board, my clean sounds, in addition to the top level of the pedalboard, the overdrive section of the effects board and containing mostly stompboxes but STEREO stompboxes, to go into this box and from there go to up to 4 amps, in this case being both inputs 1 and 2, on both amps. This switchbox also allows both sections of the board to be used at the same time. So it's basically a or b, or a+b (a: clean section b: overdrive section. This is particularly useful in a song such as bad, hence the post ... (finally!): with the left/right going to corresponding left right amps, on inputs 1 on both amps, I can take another lead out of there and put them into inputs 2 on both amps. So, a total of 4 outputs from the pedalboard to the corresponding 2 inputs on both amps. So, thats the l/r of the korg to 2 amps, and the l/r of my overdrive section to also go across 2 amps. The possibilities are great: utilising total clean for most of Bad, I get the bass tone of the fender, with the bass-mid-top end of the H/H coming through inputs 1 on both amps, with a true stereo sound. The Korg has modulation delay - of course. Then for the 'solo' section, I engage the overdrive section of the board, which using a very slight mild overdrive allows the sound to
beef up again - not using it for the distorted characteristics, more like a simulation of driving the sound harder as I would normally do using the rackmounted Korg's with the db boost they feature, and that goes across both amps, via the 2nd inputs on both of them. I also have the delay settings slightly longer than 3-4 repeats. Why? Both units, the line 6 pedal and the Korg AX30 feature ducking, which is a very handy feature. It allows a drop off of delay 'level' in relation to the initial attacking signal, which means that the delays can come shining through in the general rhythmic areas, so it fills out the sound, but the high 'e's and the 'b's will sound like they only have sounded out twice. I do think a combination of up and down is used - and try this if you want ...
In this short example I play the octave a with upstrokes, and the a flat - the 11 - and the open A's on the down. Obviously the high e is upstroke. I've tried to give the impression of the repeats that happen (in my setup anyway) by surrounding those with '()'s. Basically, the way I pick it above, with the particular settings I use on these boxes and my home-protected(!) Korg SDD delays, sound pretty bang on I have to say. It also allows an interesting effect to be obtained - those of you who have listened to this song in many formats (my personal favourite is the Wide Awake in America version btw) will notice that sometimes he seems to 'buck' the delay. What I mean is this: on listening to different moods that Bono is prone to do this song in, it's either going to be an intense and emotionally powerful but a controlled ascent into the climax of the solo, rhythmically speaking, OR it is going to ride a wave. Sometimes that wave might be more urgent or 'bouncy' I suppose I could call it. And what I mean is that you can emphasise certain delayed patterns within the main delay that is going on and give the effect of really going up or down with the songs urgency level. You know how when he comes off singing the "I'm wide awake ... I'm not sleeping" and the song calms itself, and so does the guitar part ... with a tad more variation in the up down configuration, you can ghost certain parts of the phrase and bring others to the fore. I've also found that after the addition of my SSL-5 pickup to the Strat, because of the nature of wiring scheme and the polarities and the effec this pickup has with others, it has effectively also boosted the outputs of the neck and mid pickups (std issue pickups at this stage). So, clean on this Strat now sounds BIG anyway. Coupled with the particular type of delay the korg unit's put out , it is a very, very rounded sound. I've wondered if he's also had in his amp setup the combination of clean on one amp / set of amps, and a mild overdrive on another amp / set of amps. This produces great effects where you can have the clean, shimmering quality of a good clean sound with the bite and cut of a mildly overdriven sound combined at once. Throw in the effect of using the rough end of a pick (or dimpled Herdims) and you have a very big and wide, fat ... well, an Edge tone :)
-- Alan, Sep 26, 2006
I think your 467ms delay is slighty slow. In my opinion 420ms sounds much closer. I find myself "waiting" for the delay at 467ms.
-- Chris, Sep 26, 2006
Finally someone has gotten it right. You have to play so little to make Bad sound like the real thing. It's obvious if you listen to the unique syncopation of the higher echoing notes. Faking it never rings the same. Great job. I figured it out by experimenting hour after hour back in the 80's. It's fun to see your systematic approach.
-- Brad, Sep 29, 2006
um this is a cool website and everything, but for this song and where the streets have no name i use a behringer digital delay that has alot of echo and you can actually get it really close to sounding like the edge. i've listened close to the cd's and i'm actually really satisfied with the sound plus it only cost me $30 so that's pretty cool
-- u2edge40, Oct 18, 2006
In this link you can hear perfectly how The Edge plays Bad. there s just
him without the band showing his rig .You can clearly hear Bad s Delay .It
s around 2 48 : http://www.youtube.com/watch?v=gSPG73zwnsA
-- Gabriel, Jan 5, 2006
MY SETUP: FENDER STRATOCASTER, DYNACOMP, SD1, EH MEMORY MAN, DIGIDELAY IN PARALLEL, 2 X VOX AC15TBX UK. TIME: EH MEMORY MAN 450 ms STRONG MODULATED (CHORUS SW)2,3 REPEATS DIGIDELAY 225 ms MODULATED ("5" POSITION) MANY REPEATS HIGH VALUE MASTER GAIN IN ORDER TO OVERDRIVE THE INPUT OF AC15TBX BUT ONE OF THE MORE IMPORTANT THINGS IS THE HERDIM- PLETTRO... USED IN REVERSE MODE... PERFECT SOUND OF THE EDGE... BYE
-- LUCA, June 22, 2007
Does anyone have a tab for the short harmonics edge sometimes plays at the very start of bad . Its in a similar fashion to the ones he plays at the start of streets sometimes .
-- Ger Hickey, Aug 26, 2007
make it simple 1st delay 430 ms at 100% 1 to 2 repeats the second delay is about 276 ms 50% with 1 to 2 repeats or more all modulated. i added a reverb hall so it sound big and very shimmering and of course used dunlop or any gritted back picks to sound like edge.
-- ago24, July 17, 2009
Really, the song is a round -- like "Row Row Row Your Boat." The delay lets one musician effectively start the round at multiple times, and the song harmonizes with itself like rounds do. Of course, the notes are pleasant and bright as well. But that's probably where the inspiration came from.
-- Carl, Nov 1, 2009
This is not too accurate, even though most of your other work was and I
really respect what you've done here. I find this song is good to play
with the settings of "Where The Streets Have No Name". Well, there is
definitely none of the choppy delay that you included here, and you only
play the diads once as quarter notes. The 2 repeats are done by the
delays. It really sounds much more accurate.
-- Stefan, Jan
Your website is fantastic. Can't thank you enough for the education.
I'm 35, and 2 years into playing guitar. Much of what you have is still
over my head, however I have a question for you. Have you seen "It
Might Get Loud?" Specifically, one of the extra scenes where Edge is
sound-checking? He goes through a few classic Edge tunes, one of them
being "Bad." While I find your settings for "Bad" perfect for playing
the original recording of the song, he plays it slightly differently in
the live "Wide Awake In America" version and even more so in the scene
I mentioned from "...Get Loud." (that latter version is my favourite...
more delay and more palm muting it seems like). I use Garage Band on an
iMac and more recently I've also started using Amplitube on my iPad. I
was wondering if you could help me get the tone and delay i hear him
playing in It Might Get Loud? ps: if you haven't seen it yet.... GET
-- Andy, Aug 11,
Like what you are trying to do but, as someone who has played this live over the years, the mp3 isn't sounding correct.
I think you are maybe over-analyzing and have the "secondary" delay incorrect.
Its pretty straightforward. The "main" delay is the dotted eighth (or "TRIP-O-LET") to whatever meter you use
(I just foot-tap it in on my memory man...if not a live show) with feedback aroundd 20-30% (3- 4 repeats total,
with normal fading i.e. the first 2 repeats are what matters). A bit of
filtering (for the crunchiness Edge gets). Minimal decay. The scondary delay is just 1 repeat in quarter note time.
It is essential that the quarter note meter is matched to the dotted eighth mmeter.
So I would normally set secondary first....simple "slapback" of one (or faintly a second) repeat in time with
the song meter (quarter note). Then add the main delay in "TRIP--O-LET" (doted eighth) to that same meter. Not sure what the
exact settings are if pre-set - I have them pre-set
but theey not handy as I write this - but you can just use dotted eighth setting to your 4/4 meter,
then add the secondary as quarter note slapback. Keep the secondary lower in the mix. There's a pretty good tutorial on youtube
Add a comment
-- Stymie, Nov 29, 2014
All text and pictures copyright © 2004 Tim Darling.