You are here

Interpreting audio measurement data - suggestion for an article

For feedback and suggestions about the SOS magazine, app, web site or forums.

Re: Interpreting audio measurement data - suggestion for an article

Postby Eddy Deegan » Fri Jul 26, 2019 11:53 pm

It may be beyond the scope of the article but one of the things in my case is that was seeking to find an acceptable balance between the extremes of things that are as perfect as possible and things that compromise that, such as the space available in a domestic room, the materials it is constructed from, budgets and the equipment one needs to accomodate in the environment.

A useful outcome (one of of many) for me was an understanding of the compromises I would be making to accomodate equipment in certain parts of the room, how much attention I should pay to them (or not) and consideration of the money I was willing to spend treating the room to achieve the balance.

The needs and desires, some negiotable some not, of the person who is seeking to improve the acoustics of an environment are I think very significant factors in any consultation process and speaking for myself I was very happy with the plan that emerged!

I may be diverging from the point of the article of course. I'm looking forward to reading it a lot :thumbup:
User avatar
Eddy Deegan
Frequent Poster (Level2)
Posts: 3506
Joined: Wed Sep 01, 2004 12:00 am
Location: Brighton & Hove, UK
Some of my works.
Please consider supporting the SOS Forum Album project.
 

Re: Interpreting audio measurement data - suggestion for an article

Postby blinddrew » Sat Jul 27, 2019 10:37 am

I'll make sure to download the latest version of REW then. I'm reasonably comfortable with the maths side of things (it's 25 years since i finished my engineering degree but I like to think that the ability to think in the right way is still there :) ), and I've done a lot of reading on studio building and the fundamentals of sound/audio following some book recommendations on here. But, like Eddy, I know that I will be starting from a position of compromise and then making more from then on.
But it will be better than what I have now! :)
User avatar
blinddrew
Jedi Poster
Posts: 9517
Joined: Sun Jul 05, 2015 12:00 am
Location: York
Ignore the post count, I have no idea what I'm doing...

Re: Interpreting audio measurement data - suggestion for an article

Postby Hugh Robjohns » Sat Jul 27, 2019 3:57 pm

I have a great deal of respect and admiration for acousticians because their business -- more than any other in the audio world, I think -- involves a very great deal of interpretation and experience. There really is as much art as there is science involved!

We all readily accept that a hugely experienced recording engineer will be able to make a far better recording, apparently with much less effort, than we mere mortals can achieve. (Especially so in the world of classical recording.)

But the same is true -- but with tinsel and gold knobs on -- for acoustic design.

There are indeed some fantastic tools about -- like REW and others -- and they can draw very pretty pictures. But actually knowing what those pictures mean and how to interpret the results correctly, and then to understand what to do with that information is a whole different ball game. And one made a whole lot harder by the fact that moving the measurement mic 6-inches one way or the other can change the pictures dramatically!

This requirement for interpretation is not a unique condition to acoustics measurements, of course. It applies everywhere that scientific measurements are taken.

I know very well indeed that if I inadvertently have an inappropriate setting on the Audio Precision test set when I'm running bench tests on review gear I can get totally misleading results very easily. So it's absolutely vital that the tester has a good understanding and lots of experience before embarking on the testing. I think you almost need to know what the result should be, before it appears! :D

To highlight this very point, I recently had to explain (to a microphone design engineer) how to properly use a dual-channel oscilloscope for analysing a floating balanced source because he took the (very misleading) results he was seeing at face value without realising that it was a complete nonsense -- caused entirely by the inappropriate way he was making the measurement!

To be fair, it was an understandable mistake, but he now knows better because I questioned a statement he had made.. and thankfully he had the humility to listen to what I was saying. And I only recognised the problem and knew the right way because I had the luxurious benefit of being taught how to use the equipment correctly in that kind of situation by the BBC 35 years ago! Sadly, many others I've tried to help along the way have just blustered and BS'd... and still know no better! But that's another story!

Anyway, as has been said above, there are no absolutes in the real world of acoustics -- it's always all about compromises due to the physical building structure or fabric, the practical requirements of the space (especially in multi-purpose domestic rooms), and the available budget. So knowing what measurements to make, how to make them properly, how to interpret them, and how to use them to arrive at the best compromise really is a real art.

I don't want to put people off measuring their rooms -- it's a very interesting thing to do and it might lead to a whole new career! But I would urge some caution just because optimising the acoustics of a space it's not quite as simple and straightforward as it might appear.

Max is extremely good at it, but he's been doing it for a very long time and I'm sure has many tales to tell of the challenges and surprises he's had along the way!

H
User avatar
Hugh Robjohns
Moderator
Posts: 26330
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Interpreting audio measurement data - suggestion for an article

Postby Dennis J Wilkins » Sat Jul 27, 2019 5:10 pm

Good Morning (well, here in Colorado),

I wanted to note a couple things , and just like that a post from Doctor Hugh (congratulations, Hugh!) popped up which states a number of facts I was about to state (time machine? The Doctor?)!

Although I've been an engineer for longer that most (electronics and systems engineering) and worked with audio on the side for 50 (!) years, digging into room acoustics has been an eye (ear?) opener. As Hugh says, acoustics is as much art as science, and is extremely complex. There is no magic solution to turn a bedroom into Ocean Way or Abbey Road studios, but there are ways to greatly improve a room and tools to quantitatively assess acoustics at the beginning, middle and end of a studio upgrade project (assuming there ever is an end!).

As Hugh said, there are today some excellent, powerful tools to assess room/speaker acoustics, but "knowing what those pictures mean and how to interpret the results correctly, and then to understand what to do with that information is a whole different ball game". I'm hoping my experiences with improving a very bad (square) room and the results of measuring both frequency and time domain metrics can help others to make significant improvements. However, if one needs a really excellent studio and has the money, there are a number of professional acousticians who will provide the best possible solution, from making measurements and providing suggestions to complete construction projects.

For most 'home' recordists working within the bounds of a given room, speaker placement, listening position and acoustic treatment are probably the only physical improvements that can be made, but electronic correction (ARC or Sonarworks) can really clean up the low end which is quite impossible to solve in a small room by treatment alone.

And as Hugh said, moving a measurement mic a few inches can change the response significantly, especially in an untreated room. And even more surprising to me, is even with a mic at a fixed location, on a stand, moving one's head a few inches can significantly change the response at high frequencies!

Anyway, I am trying to cover all the above, and issues like the results of inadvertently using an inappropriate setting or test file, in my article. REW is a superb program, but as Hugh noted, even a small 'misuse' of such a tool can result in misleading conclusions. But as I've found in my slow, several-year project, even a small square room can yield extremely flat response with excellent imaging and acceptable decay characteristics.

Everyone have a fine weekend!
User avatar
Dennis J Wilkins
Poster
Posts: 19
Joined: Fri Aug 05, 2011 12:00 am
Location: Colorado USA
Dennis - Colorado USA

Re: Interpreting audio measurement data - suggestion for an article

Postby Hugh Robjohns » Sat Jul 27, 2019 5:13 pm

:thumbup: 8-)
User avatar
Hugh Robjohns
Moderator
Posts: 26330
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Interpreting audio measurement data - suggestion for an article

Postby Dennis J Wilkins » Sat Jul 27, 2019 6:39 pm

Back from some errands - forgot to mention REW 5.2 will be released about mid-August, otherwise it's still beta (and not available on the REW site which still has 5.19). And as I did mention, using the previous swept-sine measurement files with 5.2 messes up the results (in like manner to Hugh's observations about inappropriate settings and measurements). So if you'd used 5.19 or earlier and have measurements saved in files, you might want to keep a 5.19 version around. Actually, John recently added a check on the file type, and 5.2 will refuse to use older test files, so again, keep an older version around if you want to analyze older test data.

Acoustics are terribly complex and measurements are "touchy" but with some experimentation and patience (believe me, you will need patience!) you can make excellent assessments of a room. And hopefully make significant improvements.

I gotta run again. No rest for the weary!
User avatar
Dennis J Wilkins
Poster
Posts: 19
Joined: Fri Aug 05, 2011 12:00 am
Location: Colorado USA
Dennis - Colorado USA

Re: Interpreting audio measurement data - suggestion for an article

Postby Dennis J Wilkins » Sat Jul 27, 2019 7:18 pm

Not intending to reply to my own reply, but right after I posted the last message, I received notice of a new REW beta update, 5.2 beta 16, which appears to have more changes than any of the last few betas. I'll give it a run-through soon.

BTW, version 5.19 is perfectly fine to use and wouldn't limit measurement capability in any significant way, but I felt an article should reflect the latest version (which has confounded my progress!). Seems this has happened to me a few times before with other software . . .
User avatar
Dennis J Wilkins
Poster
Posts: 19
Joined: Fri Aug 05, 2011 12:00 am
Location: Colorado USA
Dennis - Colorado USA

Re: Interpreting audio measurement data - suggestion for an article

Postby blinddrew » Sat Jul 27, 2019 7:33 pm

I did think about taking some measurements of my current room for comparison purposes, but as it would add no value I went outside and dug up more of the garden instead. My back is telling me I may regret this in the morning. :)
Thanks for the updates and the sanguine advice.
User avatar
blinddrew
Jedi Poster
Posts: 9517
Joined: Sun Jul 05, 2015 12:00 am
Location: York
Ignore the post count, I have no idea what I'm doing...

Re: Interpreting audio measurement data - suggestion for an article

Postby Dennis J Wilkins » Sun Jan 19, 2020 8:42 pm

My apologies for the delay . . . hard to believe I last posted here in July, but several factors delayed the article on the audio measurements topic. My original draft was longer than my SOS editors wanted and no doubt difficult to trim, everyone has been busy with editing and writing other reviews, and Room EQ Wizard 5.2 has yet to be released!

As I tested each new beta version, there were mostly small changes/fixes - until about a month ago when the time-domain measurement tools were drastically changed, and I realized the new capabilities pretty much obsoleted the method I had described. Since time domain measurements (sound-decay vs time) is a key topic of the article, and the method I had devised with John Mulcahy's help was a bit tedious, the new RT60 Decay plot and the new T60M measure are a vast improvement to these measurements. So the delay is actually a good thing!

Hopefully REW 5.2 development will settle down soon and the new measurements will provide a faster and more consistent indication of sound decay characteristics. Fixing poor room decay will still require significant acoustic treatment, but being able to measure decay times consistently down to the bottom few octaves will help to determine when you have reached a reasonable compromise (as Hugh had said, room acoustics are always a compromise!).

Hope everyone is having a fine 2020!
User avatar
Dennis J Wilkins
Poster
Posts: 19
Joined: Fri Aug 05, 2011 12:00 am
Location: Colorado USA
Dennis - Colorado USA

Re: Interpreting audio measurement data - suggestion for an article

Postby blinddrew » Sun Jan 19, 2020 9:12 pm

Thanks for the update Dennis, as it happens I'm still working on my room anyway!
Nearly there though... ;)
User avatar
blinddrew
Jedi Poster
Posts: 9517
Joined: Sun Jul 05, 2015 12:00 am
Location: York
Ignore the post count, I have no idea what I'm doing...

Previous