TZF Flanging and Rotary Speaker Sims

[Doug_H]

I'm forking this from the hoax thread, so it won't get hijacked into another subject.

(re: the Liqua Flange)

Said it before and I'll say it again: the standout flanging we are accustomed to is post-production flanging, generally pooling several channels/tracks.  That gives it the broad bandwidth in terms of content that lets every little one of those cotton-picking notches get heard.  While a guitarist obviously can't play multiple tracks at once, they CAN stick a fuzz in front of the flanger and increase the representation of the audible spectrum.  I'm certain there are plenty of folks who tried out a Hoax in the store too, and went "Well, interesting, but doesn't bowl me over".  Had they stuck a fuzz in front of it, they would have been baring their fangs at the music store sales staff growling "You will pry this guitar and pedal from my cold dead hands when I LET you, buddy!  Now back off!!"

Yeah, the fuzz was going to be my next step, Mark. I figured I probably didn't have enough harmonics to work with from a clean guitar to really distinguish the sound from my BF-2, which I rarely use. Next time I go to the store I'll try it again. I would like to try the Hoax at some point too although I suspect it may be too tweaky for me.

And yes, the Roto-Machine will likely change your expectations of what a chorus pedal should do every bit as much as the first time you heard a Dimension C.  The comment I hear most is that the Drive control gives a nice chewy delicious grind that people are unaccustomed to hearing from a pedal.  I got the sense that people would be willing to buy it for the distortion alone.  Although the separate slow and fast speeds, variable ramp-up/down and rotor balance controls are nice, the drive control is probably the nicest thing about the pedal.

I think the drive control sounds great in the way it compliments the rotary sound. The 122 filter on fast speed with some drive sounded like a B3 ready to explode... It was the most aggressive (and seemingly accurate) fast speed rotary sim I've ever heard. That being said, I didn't find the "distortion sound" that interesting in its own right- it was the way it complemented the leslie sound that I liked. Sounded very much like speaker breakup, which I'm guessing is what they were trying to imitate.

FWIW, I loved the vibey, phasey, and chorusy sounds it produced too. It covers a lot of ground, which makes sense since the vibe/chorus/phaser were inspired by the sound of a leslie anyway. And of course the ramping adds a lot of expressiveness too, allowing you to play the pedal similarly to a wah.

This one moved to the top of my priority list since I wanted something to use with my analog cx-3 as well as my guitar.

Doug

[Mark Hammer]

If you didn't do so already, you really owe it to yourself to try it in stereo.  If you have to bring in your own damn cables to the store, do it.  Where this thing really shines is in its stereo implementation. 

I'm fortunate enough to have a little Leslie (not a dual rotor job, though, I'm afraid), and was able to A/B the Roto against my Leslie, even going so far as to use them with the same amp (I drive my Leslie from my Princeton; the Leslie lacks its own amp).  Send the Roto out to two amps and it feels very similar to the spatial sense you get from a spinning rotor, the way it throws the sound around the room.  Very impressive.  I have no idea how it compares to other hardware or software-generate Leslie sims, but it does a very credible job and is a whole lot easier to lug around, even if it DOES weigh twice as much as a normal stompbox.  My buddy Tim, who makes the Retro-Sonic CE-1 clone came over to my house and had never tried a Leslie before.  Unfortunately I didn't have the Roto-Machine at the time so he couldn't compare, but this guy who THOUGHT he knew how a chorus was supposed to sound was quite taken aback.  I told him that he and every other guitarist owed it to himself to play through a Leslie at least once before they die.  He understood why after 5 minutes at it.  I'm certain the Roto would have impressed him just as much.  But ya gotta do it in stereo, my friend.

Same with those through-zero flange sweeps.  Hearing it in mono sounds great, but having this monster sound sweep across your sound field and feel like its turning inside out as it moves, like some sort of CGI movie monster walking through a wall and shape-shifting at the same time, is pretty damn indescribable.

[SeanCostello]

OK, but can it emulate THIS Leslie?

http://www.pat2pdf.org/patents/pat4017685.pdf

Trust me, you have to look at this patent: 3 rotating speaker drums, mounted on a platform that also rotates. Depending on the rates, you should be able to get complex chorusing, or an uber-cool version of a string ensemble chorus.

Anyway, does the Line-6 pedal emulate cabinet and/or room reflections? This is one of the aspects of Doppler shift that people tend to forget. A moving sound source will have a certain amount of Doppler shift relative to the listener, but there will also be different amounts of Doppler shift between the sound source and all first-order reflections of that sound source. Similarly, if the listener is moving at a given rate, the Doppler shift needs to be calculated between the listener and ALL sound sources - not just direct sound sources, but all reflected sound sources (to make it easy, first order reflections can be used).

In a Leslie, the rotary horn will produce a sinusoidal pitch change with direct relationship to the listener, but the changing angle of the horn to the cabinet surfaces, and walls of the room, will result in a number of sound sources with pitch changes that are out of phase with the direct sound path. Each of these sound sources will have different time varying amplitude and filtering, and the different sound paths will have different filtering depending on the distance from the horn, the material being reflected, etc. Of course, ALL of these first-order reflections will then reflect off of other surfaces, but these second and higher order reflections will not display any more Doppler shift, if we assume that the reflecting surfaces themselves aren't moving. Combine this with the rotating woofer, and you have a super complicated sound modifier.

Now, take all of these complicated time-varying reflections, and apply them to the device in the patent above. The result seems like the sort of thing that makes you want to hold a single chord on an organ for a few hours or so and just listen to it.

Sean Costello

[Mark Hammer]

I'm not so sure that would be a practical or desirable thing to do.  After all, the goal is to be a giggable pedal substitute for a big honking piece of furniture to be used on a stage of nonspecific dimensions, by people who are likely moving around, for people who are likely to be at least moving their head around.  My actual Leslie will sound different top my ears depending on where I'm using it and where I'm standing in that space.  So what is it that I'm supposed to mimic?

For me, the fundamental aspect is whether you find yourself getting lost in the swirl of the effect, and in this case you do...easily.  I think the only time I ever plum forgot the words to a song when gigging was in the summer of 1980 when I made the foolhardy gesture of playing through an amp set for slow tremolo on one side of me and a Leslie set for slow rotation on the other side of me (Neither synced to the other).  I walked up to the mike, opened my mouth, but my brain was being stretched all over the place like pizza or strudel dough, and words were just about the last thing I was going to be able to produce.    If a digital modelled effect can do that, what more does it need to do?

[markphaser]

 I don't think a Lesle produces TZF if so how?

 What i thought TZF was negative time delay , Zero through crossing point, Positive time delay does a leslie do this?

[Mark Hammer]

You're right, it doesn't.  The thread subject heading is misleading, because the discussion started on another thread when discussing TZF and eventually moved here.  Blame Doug, but go easy on him....he's a buddy.

[markphaser]

Its cool

[Doug_H]

Yeah- "TZF Flanging" and "Rotary Speaker Sims" - 2 subjects.

Doug

[puretube]

IMHO a L*sley cabinet
wudda produce TZF,
if the frequencies wuddan`t been splitted before the 2 radiating speakerz,
and
if the 2 speakerz wuddant be different...



(there is a reason, why Steve Winwood and Bernie Worrell use 2 cabinets on stage - even though again (forking to another "phase-" thread, here),
it are (hehe - that`s no good english...) 2 different animals,
once you compare/add/mix/interfere the sound of 2 channels electronically into one or 2 channels, or if you do so by ears in the soundfield...)


where does who go wrong?

[SeanCostello]

I'm not so sure that would be a practical or desirable thing to do.  After all, the goal is to be a giggable pedal substitute for a big honking piece of furniture to be used on a stage of nonspecific dimensions, by people who are likely moving around, for people who are likely to be at least moving their head around.  My actual Leslie will sound different top my ears depending on where I'm using it and where I'm standing in that space.  So what is it that I'm supposed to mimic?

Well, if you just emulate the direct sound path between a rotating horn and the listener, you would have vibrato, not chorusing. Assuming that the horn is fairly directional, it would simply produce a sound that goes up and down sinusoidally in pitch, and changes in amplitude and filtering. The horn and the woofer will not produce that much chorusing in their interactions, due to the crossover between the two (3rd order, -18 dB/oct).

In the Leslie, the horn is moving around in a cabinet that isn't that big. The sound will be reflecting off of the wood walls of the cabinet. If we just take the 4 sides of the box and ignore the top and bottom, you can see that as the horn moves closer to the listener, it is moving either towards or away from all 4 sides of the cabinet. These sound sources will produce reflections with different pitch shifts. Sum them together, and you have a very complicated chorus. Add the room reflections...

What I am trying to get at is that a Leslie horn does not produce its effect by itself, but relies upon first order reflections to create its chorus sound. A good emulator needs to at least address the reflections in the cabinet, even if it just lumps them into a signal that has a net sum of zero pitch shift, and simulates this by adding in the straight signal. Without the reflections, you just have a funky vibrato.

Obviously, there is no way of simulating the exact listening environment of a given space, and there is probably not much point. It is simply worth noting that the interaction of a Leslie with the reverberation of a real room will be different than just sending a Leslie emulator signal through a reverb. My guess is that a mono recording would show this to a lesser degree than a stereo or binaural recording.

Do the first order reflections need to be precisely emulated to make a good sounding Leslie emulator? Not necessarily. But if you've got the MIPS, why not do it? 100 MHZ is a LOT of cycles. My algorithms at work, even in C code, only use a small portion of the DSP cycles in a SHARC or Blackfin, and that is for things like pitch shifters and reverbs. If you have a dedicated DSP in a pedal, the only reason not to use as many cycles as you can is for power consumption purposes, which is probably dealt with by clocking the chip slower (it would be interesting to see what the ToneCore pedals clock the Motorola 56364 DSP at). Hell, why not emulate 2 Leslies in parallel? Or the Planetary Acoustic Phase Shift Mechanism in the patent I listed above? If you have multiple rotating sound sources, you can probably simplify your implementation of first order reflections, as the direct path signals will create a lot of complexity as is.

I need to go and hear this pedal. My guess is that Line 6 is addressing this to some degree, as they seem to do pretty darned good work.

Sean Costello

[Doug_H]

I agree that there are a lot of complex audio interactions with a leslie spkr/cabinet. Also, it's my understanding that the horn and drum spin/ramp at slightly different speeds too. How significant any of this is is subjective and part of the artistry of producing a good emulation.

I contend that for a $120 pedal that I can cart around on my pedal board, it does a damn fine job. Phasers, univibes, choruses, etc were developed in an attempt to mimick the leslie sound. They didn't really succeed at that, however they did succeed in creating new "paradigms" of time-based effects that we still enjoy today. This may end up being viewed as a finer grained "offshoot" of the family tree or it may fool enough people that it is essentially considered as the "real thing", sort of... Whatever the case, it's a great sounding effect I would love to have. It probably won't pass an A/B test with a leslie cabinet. At the same time it would easily nudge my chorus pedal off of my board.

FWIW, I took a class on a TI C6000 earlier this year. Audio is child's play for this chip, it would be like using a sledgehammer to drive a thumbtack for that. You won't use it with a battery either...

Doug

[SeanCostello]

FWIW, I took a class on a TI C6000 earlier this year. Audio is child's play for this chip, it would be like using a sledgehammer to drive a thumbtack for that. You won't use it with a battery either...

I just got a Verbzilla. I guess the answer to "how does Line 6 run the pedal on a battery" is "well, for about 15 minutes."

Was the TI C6000 floating point? I don't know much about the TI chips? The TI line seems confusing to me, as chips in the (seemingly) same product line have different native numerical formats.

As far as any chip being child's play for audio, depends on the task. There was a reverb being demoed at AES that used 7(!) 600 MHz Blackfin processors. Mind you, you need to run the Blackfin in double precision mode or better to get decent audio, which will reduce the available cycles by 1/2 to 2/3rds. Still, that is a hell of a lot of horsepower to throw at an audio task. For high end audio, the processors that run in the many hundreds of MHz and have SIMD, such as the new SHARCs, the TigerSHARC and the TI floating point processors, are still sought after.

Sean Costello

[Doug_H]

Was the TI C6000 floating point? I don't know much about the TI chips? The TI line seems confusing to me, as chips in the (seemingly) same product line have different native numerical formats.

Yes, their product line is horribly confusing. The "6000" is a line that includes a fixed point line and floating point line. So it depends on which specific model you pick. I was looking through the lecture notes and found a slide with a "fixed point roadmap" and "floating point roadmap"- all for the C6000 line. Each of the "roadmaps" is pretty confusing to follow in & of themselves.

As far as any chip being child's play for audio, depends on the task. There was a reverb being demoed at AES that used 7(!) 600 MHz Blackfin processors. Mind you, you need to run the Blackfin in double precision mode or better to get decent audio, which will reduce the available cycles by 1/2 to 2/3rds. Still, that is a hell of a lot of horsepower to throw at an audio task. For high end audio, the processors that run in the many hundreds of MHz and have SIMD, such as the new SHARCs, the TigerSHARC and the TI floating point processors, are still sought after.

Sean Costello

Well, there's audio and then there's audio...

Doug