Long-term scientific "blind" acoustic string test

[MacGuild]

I simply think - after having nearly 50 years of sound engineering + over 50 years guitar playing under my belt - that there is no scientific way to define sound quality. Fine approximations on various details - no denying there.

And as far as I know - there is no "standard" good sound - on any instrument.

Instead of trying to prove my statement w technical data I quote a finnish man, who invented the Ortoperspecta 3-speaker sound system in 1960:s and more importantly also found out important things about TIM - transient intermodulation distortion - he once wrote: my free memory . . .

"If the measurements of sound show good but our ears hear bad - there are two alternatives : either we have not measured enough many parameters - or we have been measuring wrong parameters . " - Tapio Köykkä

Totta. All very true, @Nuuska.
The reason I like what @dwasifar is doing here is because it is not unlike something I wish I had done myself, starting a long time ago, in the sense of making detailed observations about various string sets. Like you and many people who frequent Let's Talk Guild, I have been playing for decades and have had dozens of various guitars, which means, like everyone else, I have gone through countless sets of different strings. What I failed to do was document a lot of that variation, all those string/guitar combinations. And I wish I had. You know, perhaps a spreadsheet with columns for which guitar, which strings, general opinions and observations. But because I had a loose and lazy way of tracking string history, most of that information got lost over the years. There are times when I wish I had retained it.
I like trying a lot of different strings, it is all part of staying excited.
Besides, this thread has stimulated a lot of interesting conversation, which is why we all come hear in the first place! I will very happily read dwasifar's observations over the next year. As a string geek, I could read this stuff all day. And I like the subjective human element. Being told by machinery what is closest to ideal bothers me, these days more than ever. I spent decades working with machinery, machines are tools and nothing more. I would much rather have a conversation about hex cores or note decay with you, Nuuska, than with an audiometer.
Stay warm, friend!

 

[dwasifar]

String set A, two day update.

I played them for a while this evening and I have some further observations.

For the first day or so, the overtones were so hard and metallic that they sort of blended into a kind of faint jangly buzz, audible when playing anything that hit the higher strings more than gently. I believe those overtones were responsible for the tuning irregularities. They're fading now, the jangly buzz is no longer present, and the tuning is rock stable. The balance from string to string is pretty even until you get to the plain strings, which are a little less peaky than the lower strings. It could be that this will even out as they age a bit more. I said earlier that they bend well, and they do, but they're pretty slick-feeling and sometimes it feels like I have to dig into a bend to keep the string from slipping out from under my finger. Also I notice they are markedly squeaky, more so than I'm accustomed to. You would think slick and squeaky would be mutually exclusive, but apparently not so much.

The overall tone of the set currently leans toward brilliant rather than mellow, articulate as opposed to smooth. It's hard to say whether this is a good thing or not; I suppose it depends on what you want. It's kind of like listening to a record through studio monitors: they're very precise and detailed, but not always euphonic. For example, there's one song I play ("All Blue" by Cheri Knight) where I go up to an open D at 5th fret (xxo775) for emphasis between the lines of the chorus. With the previous strings (Martin) that emphasis was not very noticeable; with these strings it just leaps out. So they're definitely good if you need that kind of projection in the upper register, but along with that comes a forwardness, a hardness, that may not be appropriate for quieter or more melodic play. Bass notes are definite, but hard-edged, and sound secondary to the mids and highs, which is not at all what I ordinarily expect from this guitar.

When fingerpicking, which I do with flesh and nail rather than fingerpicks, the forwardness can be more of a benefit; it helps to balance out the natural mutedness of the fingers.

Second day grades:

• Construction B+ (this will remain unchanged unless a string fails in some way)
• Volume A
• Playability A- (deducting half a grade for the squeaks)
• Tone B-

I'm leaving Tone at B- because although they have their virtues, I'm not convinced it's the tone I want. Maybe this will change at the next evaluation.

Here is the part where I embarrass my future self by trying to guess what strings these are. The last time I did an extended test like this one, my batting average on these guesses was humiliatingly bad. I'm going to guess these are the La Bella.

 

[gscratch]

You have more patience than I have. One thing:
"After testing - give the strings back to you wife with a test number written on paper to go back into the package.

This way you can not tell which strings are left when you eventually get there. First after all sets have been evalueted - she can show you the packages.
"
is exactly what I was going to say.

 

[dwasifar]

You have more patience than I have. One thing:
"After testing - give the strings back to you wife with a test number written on paper to go back into the package.

This way you can not tell which strings are left when you eventually get there. First after all sets have been evalueted - she can show you the packages.
"
is exactly what I was going to say.

Yeah, I've addressed that. It's a valid suggestion, but it would also result in there being no useful results for readers until the entire test was complete. People will lose interest if they have to wait a year to find out what String Set A is.

 

[Nuuska]

Or . . . we all might start guessing which set is which - you have given the list - we could try to identify them - maybe organize some kind of competition about it. After a year - when last set was evaluated - we could send our written opinions to someone, who'd be wiling to be non-partial about this. Then you would publish the results. And the judge would check who got closest.

Prize for the winnah - all the used sets of strings

 

[dwasifar]

Or . . . we all might start guessing which set is which - you have given the list - we could try to identify them - maybe organize some kind of competition about it. After a year - when last set was evaluated - we could send our written opinions to someone, who'd be wiling to be non-partial about this. Then you would publish the results. And the judge would check who got closest.

Prize for the winnah - all the used sets of strings

Set it up

 

[Nuuska]

OK - since I have plenty of strings = no personal desire to participate - and plenty of time - I can volunteer to be the one to whom all participians can send their verdict.

Quickly thought rules:

- You do your tests and report them using numbers or alphabets - not both . . .
- Participants keep their tracks and send their verdict to me AFTER last set evaluation.
- When time is up - I declare - no more verdicts. Please pm me . . . - and remind a week before we are there.
- 24h after that you publish your results.
- I will compare verdicts to your result and declare winner - and if participans agree - I can declare everybodys results.
- You contact the winner and send the used strings.

- if anybody feels the urge to reward the winner w something else - be my guest



Is this a deal?

 

[Nuuska]

I would like to say, that it would be best, if the participants would allow me to publish their verdict - in order to avoid later speculations of unjustice. Naturally not mandatory - but in that case any aftertalks should be ignored.

Is this ok ?


Just trying to set this up so nobody should have anything to bicker about after it's done.

 

[dwasifar]

OK - since I have plenty of strings = no personal desire to participate - and plenty of time - I can volunteer to be the one to whom all participians can send their verdict.

Quickly thought rules:

- You do your tests and report them using numbers or alphabets - not both . . .
- Participants keep their tracks and send their verdict to me AFTER last set evaluation.
- When time is up - I declare - no more verdicts. Please pm me . . . - and remind a week before we are there.
- 24h after that you publish your results.
- I will compare verdicts to your result and declare winner - and if participans agree - I can declare everybodys results.
- You contact the winner and send the used strings.

- if anybody feels the urge to reward the winner w something else - be my guest



Is this a deal?

Not sure you recall that each set will be named at the two week mark. If I understand your plan correctly, it might need to be modified to accommodate that.

 

[Nuuska]

Naturally I expected you do not reveal the names. - Just give them a number or whatever - but not the real name.

 

[dwasifar]

Naturally I expected you do not reveal the names. - Just give them a number or whatever - but not the real name.

No, this was just discussed the other day. At the two week evaluation, I share the actual name of the set under test.

If I wasn't going to share what brand each set is as I go, there would be no point in sharing my evaluations at all until everything was finished. Otherwise what are you going to do with the review information in the meantime?

Let's say you've just read my String Set A evaluation and want to buy some (or avoid them). You can't go to Guitar Center and say "gimme a pack of 'String Set A' today." You'd have to wait a year to know, and nobody is going to sit waiting patiently for that long to get the one piece of information that makes the review useful.

 

[Nuuska]

You are right. Then - if we'd like to have a guessing game - it should be done for each set separately - and at the end of the year thaparticipant who has most correct guesses is the winner.

 

[dwasifar]

Hi folks,

An update.

First: String Set A is settling in nicely and sounding more musical all the time. I'm happier with them now and I think they will do well at the two-week mark.

Second: I have added two more sets to the test. Why? Because I needed to order a humidifier recharge kit and I didn't want to pay Strings by Mail's $4 shipping charge just for that one thing, so I added the following:

Maker	Name	Model
Black Diamond	Phosphor Bronze Light 12-53	N600L
Magma	Phosphor Bronze Med/Lt 12-54	GA140PB

I will edit the first post to add these to the main list. I also bought a set of GHS Americana in 13-56 to try on my Taylor.

And speaking of 13s, this is not part of the test, but I might as well mention it here: I restrung the Taylor today with a set out of my I've Been Meaning To Try These box, namely Nicola Strings phosphor bronze 13s. These are another Sfarzo/GSM product, and are notable for two reasons: 1) They are extremely unmusical right now, so I hope the metallic janglyness wears off, and 2) they are 13-58s. Yes, 58. All the other strings in the set are pretty normal gauges for a set of 13s: 13-17-26-36-46-58. I miked it, and yep, .058 it is. That's so weird; I've never seen any other set like that. Right now they have so much jangly overtone that they make the guitar sound like a 12-string, but maybe they'll settle in.

 

[dwasifar]

Hi folks.

Technically today is when the two-week review of String Set A is due, but I recently refreshed the humidifier in the case and I know that affects tone for the first few days, so I'm putting it off until Monday. Stay tuned.

 

[GGJaguar]

I'm putting it off until Monday

This could affect your annual performance rating. Looking forward to the results!

 

[HeyMikey]

Watching the thread and looking forward to your test results.

 

[dwasifar]

Hi folks.

Well, we are at two weeks, and I think Set A is sounding as much like they're going to sound as they're going to sound.

They've settled down to be much warmer than they initially came across, but there is still some residual peakiness in the midrange, which gives them a slight nasality. I don't know if they would ever get to the point of being called mellow, but the brightness has settled down quite a bit and jumping up to that open D at 5th fret I mentioned earlier does not cut through like it did when they were first put on the instrument. In the course of breaking in, they've lost some volume, meaning the low bass is not as definite as it was, but they also don't squeak as prominently. Overall they do still seem a little hard-edged. And I don't think they've got a long life ahead of them. I'll keep them on for another week or two but I think they're going to be ready to be changed not long after that.

Two week grades:

• Construction B+ (this will remain unchanged unless a string fails in some way)
• Volume A-
• Playability A-
• Tone B

So now is the moment where I go find out what strings these actually are. Getting up to check it right now.

[Jeopardy music plays]



I'm actually sort of surprised. I expected when the Pearse came up they would be rated better.

I'll write one more post about them before they eventually come off.

I had guessed La Bella and I was wrong. I'll be wrong a lot as this process grinds along. I was dreading that they would be another Martin set and I'd be embarrassing myself by talking about how two identical sets were different, exposing my subjectivity and negating the entire test. That could still happen, I suppose.

 

[dwasifar]

Final report on the Pearse strings. They've mellowed a lot in the two weeks since I posted last, and they actually sound better now than they did when I put them on, but at the expense of volume. That's nice to know, but really it doesn't do me much good; I can't wait three weeks for new strings to break in.

They might be okay to leave on another couple of weeks, but the test must continue, so they'll be coming off tomorrow to make way for the next set.

Final grades:

• Construction B+ (this will remain unchanged unless a string fails in some way)
• Volume B+
• Playability A-
• Tone B+

 

[dwasifar]

The test continues today with the next blind set, which we will call string set B.

These strings have what is possibly an identifying characteristic (the color coding of the ball ends). I'm sure I've seen this particular color coding before, but I do not remember where. Please don't remind me if you know; just keep it to yourself until I'm done with the two-week evaluation.

Measurements and observations:

• Actual gauge measurements: 1st .012, 2nd .016, 3rd .025, 4th .0325, 5th .042, 6th .054.
• All strings are 46" long. Core wire protrudes from wound strings a stupidly long 7.5", leaving a usable wound length of 38.5". Hex core.
• Deflection of the sample 6th string on the testing jig is 43mm, with the weight placed 30mm from the string end.
• Ball ends are color coded black, white, and brass (two of each). Ball end windings are well wrapped, no gaps. There is one very slight pointy end near the ball of the 6th string, where the end of the wrap wire is fastened, but nothing that would catch in the bridge slot or puncture a finger while stringing.
• The wrapped strings look excellent. The wrap is 100% perfect from end to end, no irregularities, no marks or spots or flat areas. Color is typical phosphor bronze. It may be my imagination, but the 6th string looks just a bit lighter in color than the others.
• Plain strings are also excellent, completely smooth end to end, no detectable faults at all.

Tone and playability:
I am really, really impressed with these right out of the gate. They sound much more musical than fresh strings generally do. There is a hint of new string jangle, but only a hint; mostly they just sound clear, pure, and well-balanced. They came up to tune easily and they were stable right away; the pitch does not drift from overtones the way I've seen some (most) strings do when new.

In light of the slightly heavier gauges on a couple of the strings compared to the previous set, it is not surprising that the deflection measurement is lower, but oddly this does not translate to harder fretting or bending; they play easy and bend smoothly, and are not especially squeaky.

My concern is that if they sound good (instead of bright) now, they're going to sound lifeless (instead of mellow) when they break in. But we'll see. Initially they're making quite a good impression.

Initial grades:

• Construction A
• Volume A-
• Playability A
• Tone A

 

[Nuuska]

• Deflection of the sample 6th string on the testing jig is 43mm, with the weight placed 30mm from the string end.

Do you cut the plain core off before measuring deflection?