Spectrum analysis on three different Guild electrics

parker_knoll

Senior Member
Joined
Apr 27, 2006
Messages
1,453
Reaction score
410
Location
London
I set up a spectrum analyser plug-in in Logic to test the effect of different vibratos on the sound (see here) and I thought I might as well use it to look at some of my other Guilds, so I plugged in the following:

2001 Starfire III-90
2000 M65 Custom Shop
1989 X160

Guitars are recorded completely clean and direct into Logic - no attempt to flatter the sound :)

I'm not comparing like for like, I'm just interested in what we see visually in the spectrum analysis and how it matches what our ears tell us.

Here are the results:

1. Guild Starfire III-90
2001
Ernie Ball Rock'n'roll Pure Nickel 10-46, about a month old
Seymour Duncan Antiquities
Aluminium Bigsby bridge on aluminium base

Sound sample

SFIII-90.jpg


2. Guild M70 Custom Shop
2000
GHS Brite Flats 10-46, new
DeArmond 2K, 1990s
Schaller ATM Bridge

Sound sample

M65.jpg


3. Guild X160
1989
DR Tite Fit 10-46
Guild HB1
Aluminium Bigsby bridge on rosewood base

Sound sample

X160.jpg



Here's the three guitars overlaid:

Guilds-overlaid.jpg


Discussion

The main takeaway for me here is that it confirms what my ears told me, namely that the M70 has a lot of what is sometimes called 'sparkle' or 'air' or similar, namely activity in the 5k to 8k range, and by contrast the X160 has literally none, and the SF III-90 is in the middle. Although it looks relatively small, these high frequency sounds don't need much level to be perceived. Some people don't like high frequencies. Personally I do, and I also know they are hard to reproduce, easily lost, and impossible to replace when lost. Incidentally, a change to a roller bridge on the M70 sucked all those high frequencies right away.

The X160 is an all maple guitar with an aluminium bridge. It's possible the HB1 pickups are just not capable of picking up frequencies in that range; the Guild M55 I tested previously was actually quite strong in that area. If someone else wants to test an HB1 equipped guitar it would be interesting to see the results. In this case, when the guitar is not picking up anything in this frequency range there is no way to put them back in there - there's no signal to boost.

Interesting to me is that the X160 and the M70 both have a mid-range scoop between 1k and 2k, although much more pronounced in the M70 (makes it sound Fender-y). This is a very significant EQ region for guitars and I'm sure is part of the HB1s distinctive sound, different from most mid-heavy humbuckers.

Next string change I'll test the M70 with normal roundwound strings to see if we get even more HF content, since the semi-flat GHS Brite Flats are supposed to be 'mellow', by which we usually mean reduced HF content.
 
Last edited:

parker_knoll

Senior Member
Joined
Apr 27, 2006
Messages
1,453
Reaction score
410
Location
London
incidentally, I'd love to put the roller bridge back and see what exactly it was doing to the sound but I sold it and I'm not committed enough to buy another just for that reason. I'll test the X160 with its stock wooden bridge at some point.
 

cupric

Senior Member
Joined
Apr 20, 2018
Messages
1,930
Reaction score
1,362
Guild Total
3
Interesting! To me they definitely look similar. Unfortunately I use a small tablet so a wide screen interpretation is impossible.
 

parker_knoll

Senior Member
Joined
Apr 27, 2006
Messages
1,453
Reaction score
410
Location
London
Interesting! To me they definitely look similar. Unfortunately I use a small tablet so a wide screen interpretation is impossible.
look at the three overlaid. if you can zoom in enough to see the scale, the key section for me is between 4k and 8k
 

shihan

Senior Member
Gold Supporting
Joined
Jan 18, 2013
Messages
1,551
Reaction score
813
Location
Ventura CA
You and mad scientist GAD should get together; you guys could do some serious Guild analyzing!
Very interesting to see the results graphed together. I too would have said they all look very similar. Fun fact about the frequency range difference. Thanks for sharing.
 

HeyMikey

Enlightened Member
Joined
Jul 12, 2018
Messages
5,500
Reaction score
4,843
Location
MA
Guild Total
9
Neat stuff. Thanks parker
 

GGJaguar

Reverential Member
Joined
Jan 17, 2011
Messages
21,741
Reaction score
32,029
Location
Skylands
Guild Total
50
Fascinating, thanks for the analyses. I also have a Bigsby saddle on a rosewood base on my '89 X-160 Savoy as well as my '97 SF-III so the results you observed are of great interest to me. I await results using different strings. :)
 

Nuuska

Enlightened Member
Joined
Jan 18, 2016
Messages
7,697
Reaction score
6,068
Location
Finland
Guild Total
9
Hello

Interesting - how do you strum the guitars to keep consistency?

Could - if you have them - show us those curves in a form, where there's only a line - not the filled area underneath?

Or doesn't your unit have that option?
 

parker_knoll

Senior Member
Joined
Apr 27, 2006
Messages
1,453
Reaction score
410
Location
London
Hello

Interesting - how do you strum the guitars to keep consistency?

Could - if you have them - show us those curves in a form, where there's only a line - not the filled area underneath?

Or doesn't your unit have that option?

Hi Nuuska

I'm using a free plug in called SPAN which has the benefit of an "average" function, so what i did was loop the section with the strumming and let it play until the frequency analyser had got a stable image using the average function and then snapshotted that. It doesn't do a line, though. At least, I don't think so, but I can have a fiddle around with it next time.

Other analysers are available and the inbuilt one in Logic does do a line, but does not take an average, so it looks quite different one second to the next.
 

parker_knoll

Senior Member
Joined
Apr 27, 2006
Messages
1,453
Reaction score
410
Location
London
Anyone got a hog Starfire and a maple Starfire they want to record totally dry for me?
 

Nuuska

Enlightened Member
Joined
Jan 18, 2016
Messages
7,697
Reaction score
6,068
Location
Finland
Guild Total
9
Averaging function sounds very good - the longer the better.
 
Top