The Acoustic Guitar Forum

The Acoustic Guitar Forum (https://www.acousticguitarforum.com/forums/index.php)
-   General Acoustic Guitar Discussion (https://www.acousticguitarforum.com/forums/forumdisplay.php?f=24)
-   -   Tuners: accuracy vs. precision... (https://www.acousticguitarforum.com/forums/showthread.php?t=325578)

philjs 01-17-2014 04:59 AM

Tuners: accuracy vs. precision...
 
In another thread discussing tuners that people use there seems to be quite a bit of confusion about what accuracy is, and the importance of the distinction between accuracy and precision. I thought this was important so am posting this as a new thread...this image should help:

http://www.carolina.com/images/teach...-precision.jpg

A tuner should be both accurate AND precise. To my mind, there are very few tuners that are both...and those that are are strobe tuners.

Phil

AndrewG 01-17-2014 05:07 AM

All the accuracy and precision I need is in an A440 tuning fork and my ears.

mattyfez 01-17-2014 05:19 AM

Quote:

Originally Posted by AndrewG (Post 3778972)
All the accuracy and precision I need is in an A440 tuning fork and my ears.

Yarp, I use a reference tone and ears, no need for an expensive tuner.

Luke W 01-17-2014 05:24 AM

Quote:

Originally Posted by AndrewG (Post 3778972)
All the accuracy and precision I need is in an A440 tuning fork and my ears.

Not the most practical on a noisy stage though.

AndrewG 01-17-2014 05:29 AM

Quote:

Originally Posted by Luke W (Post 3778979)
Not the most practical on a noisy stage though.

I agree, although my gigging days are over so it isn't an issue any longer.

mc1 01-17-2014 07:11 AM

another point about tuners, besides there accuracy and precision, is how that translates into usability. for example, i have a tuner that claims to be extremely accurate and precise, but the way the display works makes it hard to use that accuracy and precision.

neat diagram, phil, i'll have to think about it. the "accurate not precise" kind of stands out to me. (as well as the "not accurate not precise"). they seem to be the interesting ones.

HHP 01-17-2014 07:15 AM

Quote:

Originally Posted by mc1 (Post 3779028)
another point about tuners, besides there accuracy and precision, is how that translates into usability. for example, i have a tuner that claims to be extremely accurate and precise, but the way the readout works makes it hard to use that accuracy and precision.

neat diagram, phil, i'll have to think about it. the "accurate not precise" kind of stands out to me. (as well as the "not accurate not precise"). they seem to be the interesting ones.

I think a couple of the graphics are mis-labeled.

philjs 01-17-2014 07:16 AM

Quote:

Originally Posted by mc1 (Post 3779028)
neat diagram, phil, i'll have to think about it. the "accurate not precise" kind of stands out to me. (as well as the "not accurate not precise"). they seem to be the interesting ones.

Think of it this way: The idea is to hit the center of the target (accuracy) AND to place your shots as close together as possible (precision). If all of the shots are close together then you have precision but ONLY if the shots are all close together AND at the center of the target do you also have accuracy.

Phil

philjs 01-17-2014 07:17 AM

Quote:

Originally Posted by HHP (Post 3779034)
I think a couple of the graphics are mis-labeled.

No, they're not...read my reply to mc1 above.

Phil

martind42 01-17-2014 07:23 AM

Quote:

Originally Posted by Luke W (Post 3778979)
Not the most practical on a noisy stage though.

Hey, if Segovia could do it in his early punk phase, so can you. Suck it up.

HHP 01-17-2014 07:23 AM

Quote:

Originally Posted by philjs (Post 3779036)
No, they're not...read my reply to mc1 above.

Phil

Maybe in a symbolic sense, the one labeled "accurate not precise" is neither in a literal shooting sense. More illustrative of accuracy and calibration.

Luke W 01-17-2014 07:32 AM

Quote:

Originally Posted by martind42 (Post 3779042)
Hey, if Segovia could do it in his early punk phase, so can you. Suck it up.

Tuning or punk, you can only pick one :)

fazool 01-17-2014 07:34 AM

I like that diagram. I think I'll use that in other conversations, too.

In engineering terms, we think of accuracy and precision when measuring physical things (like frequencies).

If you are measuring the dimensions of a mechanical part using two different calipers or micrometers, and tool-A can read out to 0.0000" but tool-B can only read out to 0.000" inches we say that tool-A has greater precision. That's analogous to your bullseye diagram, where you have a very tight cluster.

If tool-C can read within +/- 0.0005 and tool-D can read within +/-0.001 we say tool-C has a tighter tolerance and, therefore, is more accurate.


Your diagram is a nice illustration of the concept.

I had tremendously bad luck with a crappy strobe tuner. I'm not sure (yet) that strobe tuners are generally better. Why are they? I understand the level of precision and response time of the light is better, but there is a power circuit driving the light at a specific frequency. If that wavers, then the accuracy is lost. A piezo sensor has a circuit reading a frequency. If that wavers, accuracy is lost. I think they are probably pretty comparable, no?

Anyway, I greatly disliked my crappy strobe tuner.

I had an in-line digital tuner and hated that as well.

I used a Android microphone app with pretty good luck, then got Snark piezo tuners and have been very happy.

devellis 01-17-2014 07:43 AM

The "accurate" and "precise" concept refer to bias and consistency (or validity and reliability). An unbiased (accurate) device gives a correct value on average. A precise device gives highly consistent values.

mc1 01-17-2014 07:50 AM

Quote:

Originally Posted by philjs (Post 3779035)
Think of it this way: The idea is to hit the center of the target (accuracy) AND to place your shots as close together as possible (precision). If all of the shots are close together then you have precision but ONLY if the shots are all close together AND at the center of the target do you also have accuracy.

Phil

however, what makes "not accurate not precise" interesting is that the dots seems all closer together than the "accurate not precise" dots. does this perhaps imply more precision? i can see that the "accurate not precise" dots are all around the actual target. i get the point, so to speak, but i'll still want to think about it a bit.

and to get really far out, these diagrams, i'd assume, are four shots aimed once and shot 4 times without any correction. another interesting thing is the sample size of 4, which i suppose is just enough to make the point.

Quote:

Originally Posted by fazool (Post 3779060)
I like that diagram. I think I'll use that in other conversations, too.

In engineering terms, we think of accuracy and precision when measuring physical things (like frequencies).

If you are measuring the dimensions of a mechanical part using two different calipers or micrometers, and tool-A can read out to 0.0000" but tool-B can only read out to 0.000" inches we say that tool-A has greater precision. That's analogous to your bullseye diagram, where you have a very tight cluster.

If tool-C can read within +/- 0.0005 and tool-D can read within +/-0.001 we say tool-C has a tighter tolerance and, therefore, is more accurate.

...

this is more the way i understand accuracy and precision, although i might say accuracy is how capable a device is of meeting its claimed precision.


All times are GMT -6. The time now is 01:16 AM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.
Copyright ©2000 - 2022, The Acoustic Guitar Forum

vB Ad Management by =RedTyger=