Single string pickups - 6 of 'em

[Mark Hammer]

Mine works fine, to the extent that these things ever "worked". It has been promised at a set price to "The EH Man" to add to his collection. I like to honour my promises. I'm just waiting for the Canadian dollar to drop again so I can make a couple of bucks in the transaction. While I'm waiting for that blessed day, I should probably cobble together a guitar without a bridge pickup so I can ram that little sucker snugly up against the bridge and hear it under optimal conditions.

Just glad I didn't leave it out where our rabbit could get at it. We have a pet rabbit who has the free run of the house. Normally he is extremely well-behaved, but a month ago I left my GK-to-MIDI-brain cable hanging off the side of my gear and he chewed the thing beyond repair. Happily, I have the schematic, but finding a comparable hunk of multi-conductor cable is gonna be as much of a bear as resoldering all those pins.

[David Schwab]

I used to have a rabbit, so I know what you mean....

[jrfrond]

Having worked extensively on all forms of digital modeling ad nauseum here in the shop, just take this as food for thought:

Much of the issue with distortion patches on modeling amps and effects (no names please!) is that polyphonic information is fed to a single processor, which then converts it to serial information, does it's processing thing, and reassembles it, all with improper harmonic phasing. This is why modeled distortion sounds like ass, even on perfect 4ths and 5ths.

If you could rig a hexaphonic pickup to feed six SEPARATE, identical modeling boxes, with all of the processors driven off the same clock in sync, modeling would probably sound pretty good, since you would have cleaner time arrival of harmonics. In theory anyway.

This is one practical use of a hex guitar pickup that I could think of.

Didn't Bartolini do a quad-out bass pickup a long time ago?

Call me an old warhorse, but I like my passive pickups with either single or double coils AND all of their inherent faults. Then again, I don't dig all of the Phtoshop work in Playboy either!

[David Schwab]

This is why modeled distortion sounds like ass, even on perfect 4ths and 5ths.

Some of it sounds very good. Like this: Digital amp model

The parts that don't work well is the slightly distorted crunch tones. Digital does good clean and very distorted, but in between is a problem.

The problem with amp modeling is with the modeling. It's getting better al the time tough. They just need to tweak it more.

Also all the presets that come with Pods and such suck, just as the factory sounds on synths also suck. You have to go in and teak them.

I will say though that my clip is not a Pod... I really don't care for them.

If you could rig a hexaphonic pickup to feed six SEPARATE, identical modeling boxes, with all of the processors driven off the same clock in sync, modeling would probably sound pretty good, since you would have cleaner time arrival of harmonics. In theory anyway.

I talked abut hexaphonic fuzz earlier in the thread. It wont sound like an amp, because you don't use an amp for each string. It has a very clean tone lacking intermodulation distortions, which are quite prevalent in real amps.

It's a cool sound though, and even very fuzzy/fizzy distortions sound good.

Didn't Bartolini do a quad-out bass pickup a long time ago?

All his original designs had one coil per string. This was for tonal reasons, but also allowed multiple outputs. I don't think he's made any hex or quad pickups in a long time.

Then again, I don't dig all of the Phtoshop work in Playboy either!

You have No idea! It's not just Playboy.. it's EVERY photo of a person you see in print. I did some freelance photo retouching work for Bloomindale's last holiday season. We aren't talking about removing blemishes and things, we are talking about making fingers longer, ears smaller, noses straighter, moving eyes, and general thinning of legs and arms, and removing every wrinkle in clothing.

I have some examples on my personal website.

[jrfrond]

Nice website Dave.

My wife is a professional photographer, and I use Photoshop a lot as well, so I am painfully aware of it's overuse. You can't believe anything you see in a pic anymore! Still and all, one of the coolest programs ever.

Like I said (I think): digital modeling is like Photoshop to me. More like a good lithograph, actually. It all looks good until you get up close, and then you can see all of the pixels. The same analogy applies to modeling.

My "theory" posted above is not too far from the mark. Several years ago, when Tech 21 was still in NYC, I went to go visit my friend Andrew Barta, the company CEO and engineer. We sat in his lab with a Pod and a SansAmp, armed with test equipment (scope, distortion analyzer, FFT) and figured out why the analog SansAmp sounds pretty good, and the Pod did not. We were able to see the out-of-phase harmonic arrival. Harmonics need to have a phase relationship with the original signal, in addition to their mathematical relationship. Screw up the phase, add some distortion, and you've got some ratty sh*t on your hands!

[David Schwab]

Thanks. I need to update the site, but I never get around to it. I've been in graphics and lithography since 1980, but I'm ready to move on.

The Sansamp is just some overdriven op amps and a simple filter for speaker emulation. Sounds fine, but it doesn't sound like an amp. A Rat sounds good too, but also doesn't sound like an amp. I like the Rat better. The Sansamp was always too mushy sounding for my tastes. I agree about phase angles and such though. Cheap A/D converters can make a mess of things.

I'm not crazy about the Pod. It sounds plastic. I have heard some real nice amp sims though.. especially in software. Digidesign has a new one called Eleven. They say they spent a few years looking at why amp and speakers sound like they do, and pointed out how the other amp sims just imitate a generic tube preamp stage. They went though every stage and modeled it. The speakers too. The clips I've heard sound really really good.

I just think they haven't figured out all the little details yet.

The clip I played was of a very overdriven Fender Twin sim that's in my Roland VM-3100Pro digital mixer. You can really tweak the various attributes of the amps they have in there. They don't tell you a lot of what the various amp types are for a given patch, but there's a lot of cool tones in there. You can even change the mic type and angle.

When used in a recording, I sure can't hear the difference between that and a real amp mic'd up.

Here's a clip I just recorded testing out a prototype of new neck pickup design I made today. Sounds like a single coil, huh? But where's that awful hum... Also a Twin patch.

single coil neck test

[Corvus]

Gibson has a hexaphonic design: http://www.gibson.com/DigitalGuitarN...onDigital.html

And Dave Bunker has a brilliant design for his touch-style guitar/bass instruments: http://www.gibson.com/DigitalGuitarN...onDigital.html

[Joe Gwinn]

Much of the issue with distortion patches on modeling amps and effects (no names please!) is that polyphonic information is fed to a single processor, which then converts it to serial information, does it's processing thing, and reassembles it, all with improper harmonic phasing. This is why modeled distortion sounds like ass, even on perfect 4ths and 5ths.

If you could rig a hexaphonic pickup to feed six SEPARATE, identical modeling boxes, with all of the processors driven off the same clock in sync, modeling would probably sound pretty good, since you would have cleaner time arrival of harmonics. In theory anyway.

Interesting. This would work only if the delay through the six channels was the same, at all frequencies of interest.

More generally, if a single channel system exhibits "linear phase" behavior, then the harmonics will all stay in sync.

Linear phase? What this means is that the phase (in degrees or radians) varies linearly with frequency, which means that the end-to-end delay in seconds is the same at all frequencies of interest. This is easily arranged, but linear-phase filters do not have sharp cutoff frequencies.

[jrfrond]

I know that the digital modeling topic is a tad OT here, but I that one of the problems with digital modeling has nothing to do with the sound or design at all. It's the nomenclature applied to the patches. You call up the ubiquitous Blackface, Plexi and British patches (Fender, Marshall and AC-30 respectively), and if you know these amps at all, you find that they just bear a vague sonic resemblance to them. I say let the consumer name their own patches, because most of them don't sound like what they are supposed to be. In most cases, they are overexaggerated.

[David Schwab]

I know that the digital modeling topic is a tad OT here, but I that one of the problems with digital modeling has nothing to do with the sound or design at all. It's the nomenclature applied to the patches. You call up the ubiquitous Blackface, Plexi and British patches (Fender, Marshall and AC-30 respectively), and if you know these amps at all, you find that they just bear a vague sonic resemblance to them. I say let the consumer name their own patches, because most of them don't sound like what they are supposed to be. In most cases, they are overexaggerated.

Yeah, that's what I was saying about the modeling used... the concept is good, but often the models aren't. It's quite possible that units like the Pod just don't have the resolution and/or necessary models to accurately pull off these patches.

Another thing to consider is do you really need an exact copy of an amp? Same is true of pickups. I think as long as you can wrangle quality usable tones out of a rig, that's what counts. Maybe you can't cop the tone of your favorite artist's recordings, but you shouldn't really do that anyway. Just find unique tones that work for you. What's making a lot of music so damn boring is the same handful of guitar sounds are being used ad nasium.

That said, check out the Digidesign Eleven. They have a video on the creating of the plugin, and in the background they have a sound track of various "sound-alike" songs featuring various famous tones... they are pretty damn exact. I guess the more resources at your disposal, the better the models that can be done.

They talked about how they got a collection of real vintage amps and studied the speakers and output transformers (down to the model numbers) to get everything right. They also claim to have the best speaker cabinet emulator.

After I get a new Mac I plan on getting this plugin for my ProTool setup.

That brings up another point... with the digital modelers, you have a small amount of latency. The faster the CPU and I/O, the lower the latency, but this latency would certainly translate into some kind of time domain/phase shifts I'm sure.

[David King]

That brings up another point... with the digital modelers, you have a small amount of latency. The faster the CPU and I/O, the lower the latency, but this latency would certainly translate into some kind of time domain/phase shifts I'm sure.

I thought latency was a matter of needing to wait for 1/2 wavelength to determine frequency. I guess it depends on what you want the dsp to do next.

[jrfrond]

Half of the digital modeling thing is marketing. "Use this and you'll sound like _________". It's all such a load of crap! Even if you possessed there actual Tele and Supro amp that Page used on the first two Zep albums, you wouldn't sound like him. Different fingers, different brain, etc. However, year after year at NAMM, this is the crux of marketing, making players (and non-players) believe this will make them sound like that.

A couple of years ago, my brother was given a Digitech Jimi Hendrix pedal at NAMM. Having both worked in Electric Lady Studios here in NYC in the 70's, we are well-versed on the Hendrix sound. When we got back to his place that night, we plugged it in, using a Strat and 4 x 10 Bassman RI. We ran through all of the patches. Needless to say, it sounded nothing like REAL Hendrix tone. More like a caricature thereof. I can get a more realistic Hendrix sound using my Boss pedals! Yet, the "sound like Hendrix" vibe was all over the packaging and promo material. THIS is how "stuff" is sold year after year, ad nauseum, in the music business. The whole "Signature" and quasi-signature form of promotion (and patch programming) really makes me want to vomit!

[Mark Hammer]

I think the problem with many models of amps, cabs, pickups, guitars, etc., is that the target of the model is not the component over the entire range of its use but rather how it sounds when it distorts.

Maybe I'm making too much of it, but I have one of these Zoom Amp Modellers ( http://www.amptone.com/zoomgm200.htm). The amps are largely indistinguishable from each other until you crank the gain, at which point they distort in qualitatively different ways. Of course, that's only one aspect of the limitations. The other is akin to advertising one colour TV on a different one: How would you know what the advertised one looks like if you're seeing it through the picture tube of the watched one? I play my GM-200 (on those few occasions when I do) through my little tweed Princeton. No matter how good the model is, do I have any right to expect it to sound like a Vox or Marshall through that amp and 8" speaker? In a sense you want any amp and speaker used for such a purpose to be ruler flat and clean so that you ONLY hear the model. But what self-respecting guitar player owns one?

[David Schwab]

I thought latency was a matter of needing to wait for 1/2 wavelength to determine frequency. I guess it depends on what you want the dsp to do next.

I think you need one whole cycle, then you have to wait for the whole process to finish, and then wait for the signal to go to the D/A converter. The software is modifying the signal in real time, it's not like playing though analog filters and stuff.

When I'm recording in something like Cubase or ProTools, I get at the minimum about 12 ms latency, and sometimes as much as 24 ms if I'm monitoring though software plugins (12 in and 12 out). Since my Mac is about 7 years old, I sometimes have to increase the latency in order for things to play back smoothly, but by that point I'm not dealing with live signals mixed with the digital playback. A faster system could get maybe 8 ms minimum. But there's always a delay.

I was playing in a band with a guy that ran his guitar to a PowerBook running GuitarRig software. You could feel the delay when you played, but it was hard to hear it.

[jrfrond]

It seems like any standalone, processor power-hungry, affordable musical device has latency issues. When Muse Research came out with the Receptor, an application-specific computer to play synth plug-ins, I tested one out to see if it would work as the core of a super electronic drum rig (I am a pro drummer by trade). The latency issues were just too bad to deal with. I could easily feel the delay between the pad hit and note trigger. Very disconcerting. From a player's standpoint, latency is a huge issue.