Jemsite Pickup test

[Devo]

Hey folks I was reading the post on 'thoughts on the tone zone' and it got me thinking.

I think its about time we should do a test for ourselves on what our pickups are actually putting out.

Between all of us we have a huge resource of pickups in different mounting configurations and in different body woods and with a little thought and effort we could actually start looking at what these differences are.

To get started we need some waveform analysis software, something that can do a fourier transform for a start. We can already get the DC output of all our pickups straight from the manufacturer. I think well probably develop this idea further as we go-along but if anyone could help out with identifying some (preferably free) software to start then we can move on to how to submit sound samples, analysis and compiling some sort of database.

Hopefully we can get a few of us together to try and do this... should be very interesting!

Steve

[red5]

Sounds interesting, but I'm too lazy

[j.arledge]

i think we should be able to just send in clips of the pickups we have and give a breif description of the guitar and gear along with how the clip was recorded. You'd eventually build up a large enough data base of clips you could hear any one pickup used in different settings and under different conditons.

[Devo]

That would be good .... but too large and variable to be attainable...

the simplest place to start is just with a standard test... like say an open D string plucked once and recorded. After the initial attack (~0.1 of a second) the wave will decay to a relativly stable longlived form. If we analyse say 0.5-1 second of that, that should give us more than enough information. For example from that we should be able to see how many overtones and fundamental frequencies it picks up, which ones of the possible fundamentals are most prominent. Might even be able to spot a pickup by its signature overtones... who knows...

[giannifive]

Originally Posted by Devo

That would be good .... but too large and variable to be attainable...

the simplest place to start is just with a standard test... like say an open D string plucked once and recorded. After the initial attack (~0.1 of a second) the wave will decay to a relativly stable longlived form. If we analyse say 0.5-1 second of that, that should give us more than enough information. For example from that we should be able to see how many overtones and fundamental frequencies it picks up, which ones of the possible fundamentals are most prominent. Might even be able to spot a pickup by its signature overtones... who knows...

I agree that this is the more scientific way to do it, but I think we'd have to take way too many measurements to make the results useful for someone trying to choose a new pickup. I think a more useful pickup test would be for each person here to record a few samples of each pickup (I'm assuming humbuckers here):

1) clean, serial
2) clean, parallel
3) dirty, serial
4) dirty, parallel

And each sample should contain chords and single note stuff. Of course we also have to keep track of the position of the pickup (neck/bridge), orientation (adjustable pole pieces toward neck or bridge, if only one coil is adjustable), guitar body wood, neck wood, fingerboard wood, amplifier and gain settings, etc.

I'd be willing to contribute once my new 7-string pickups arrive

[giannifive]

Originally Posted by Devo

That would be good .... but too large and variable to be attainable...

the simplest place to start is just with a standard test... like say an open D string plucked once and recorded. After the initial attack (~0.1 of a second) the wave will decay to a relativly stable longlived form. If we analyse say 0.5-1 second of that, that should give us more than enough information. For example from that we should be able to see how many overtones and fundamental frequencies it picks up, which ones of the possible fundamentals are most prominent. Might even be able to spot a pickup by its signature overtones... who knows...

The relative strengths of the overtones are strongly affected by where the string was plucked, the pick used, the style in which the string was plucked, the bridge (floating or fixed), the body wood, etc. My guess is that with many different players doing the recording it will be too non-uniform to see a trend. That's just a guess, of course. I think it would be interesting to do, but I'd bet people would be more willing to submit clips of playing.

My 2 cents...

[Algiman]

There are far too many variables at work here.

[Devo]

All these are great ideas... and im aware of the huge number of variables... and introducing things like amp settings etc makes it too difficult to do at the moment...

Im really interested in what info we can get from a single pickup... so to see if its feasible we really have to start small... with something really reproducible... like plugging the guitar straight into a soundcard/audio device with the bridge pickup set up exactly 4mm away from the D string and plucking at the 12th fret with a pick... allow the string to 'normalise' for 1 second and use the following 0.5 of a second to analyse the tone.

I bet we can pick up PUP nuances from a single string. Im going to do it anyway to see how it turns out as I have about 14-18 different Dimarzio pickups loaded into guitars right now. So Ill start with them. If I can find the software... It would be great to put some actual data behind what we think we can hear...

Im also thinking of using the same technique to test cables... see how different ones affect the guitar signal.... I have my favourite cables and I think they sound better... but I havent a clue why and Im interested in finding out..

anyway... got to try find some audio/waveform analysis software...

[Algiman]

That could be quite interesting from the scientific point of view, but players want to know what these things sound like used in anger. How about a couple of standard patches for common amp modellers? That is probably the most consistant "rig" you could get, unless there is some software that would be freely available to let you plug straight into a sound card, as you suggested.

[Devo]

Originally Posted by Algiman

There are far too many variables at work here.

the whole gain settings,series/parallel, etc is not at all achievable, I agree... I also dont think its useful because people will choose pickups on other peoples playing style and ability in each sound clip... My idea is more about trying to come up with a reproducible test for a single pickup that can tell us something about how it behaves... definatly achievable...

[Algiman]

Absolutely, hence my suggestion about a standard patch.

[j.arledge]

Originally Posted by Algiman

Absolutely, hence my suggestion about a standard patch.

yeah but what about the recording methode, even if you had a patch how could you insure that everyone used the exact same way to record, also computers and computer sound cards are different .

there could be no uniform way so that every clip was recorded the same way unless it was all done on the eact same quipment using everything exactly the same.

i think the patch idea is great, but it wont solve the fact that people have thier own ways and style. I think it actually would be better to just have a few different clip of people using the same pickup with maybe the same gain/amp and all that but just letting them apply thier "style" to the clip. It would give a better idea to what the pickup can and can't do.

[Algiman]

+ different set ups, different picking techniques, different string gauges ..........

it would always be a compromise.

[Drew]

I say accept that, then, and have everyone post clips through their own rigs and own setups - we all optimize our gear to compliment our fundamental guitar tone anyway, so I think this might give us a more accurate picture of the sorts of tones a pickup is capable of.

[Jemwielder]

If it sounds good who really cares about the numbers?