AFMG Network Forum
AFMG Network Forum Index -> AFMG SysTune -> Tuning system with frequency response vs.Transfer funtion
Post new topic  Reply to topic View previous topic :: View next topic 
Tuning system with frequency response vs.Transfer funtion
PostPosted: Mon Aug 13, 2012 2:25 pm Reply with quote
r.b.
Member
 
Joined: 11 Jun 2012
Posts: 3




Hello All,

sorry if this has been answered somewhere else, i didnt find it ...

I have a feeling this is one of these stupid questions with an obvious answer, but i can't seem to get my head around it:

why are we tuning the system against the Transfer function, instead of the frequency spectrum (properly weighted) ?

we can easily display the reference spectrum A or C weighted...

then, match the the measured curve against it (e.g. with a 31 band EQ and with the proper settings in systune of 1/3 octave etc...)...

What is the problem with that approach, and why ist it preferable to flatten the TF instead (The freq approach seems to be much easier/faster)

Thanks,
Roland
View user's profile Send private message
PostPosted: Tue Aug 14, 2012 4:01 pm Reply with quote
Langston Holland
Member
 
Joined: 13 May 2007
Posts: 31




Hi Roland:

This may sound crazy, but I've been working on your questions pretty diligently for about 10 years now and I'm just beginning to grasp some of the fundamentals behind them. SysTune is kind of like a surgeon's scalpel - it looks pretty simple on its surface, but to know when, how and why to use it correctly takes a great deal of time, study and practice.

To give you a general answer to your question on "Transfer Function" vs. "Frequency Spectrum", the former "windows" or removes some of the acoustic data so you can more clearly see what the loudspeaker is doing in the room. The latter just shows all the sound going into the mic no matter when it happened, just like an old real time analyzer. As to "A weighting", etc., those are not target curves for perfect sound but an attempt by a couple of guys named Fletcher and Munson to describe how we perceive equal loudness across the audible frequency range.

There are two things that will help you the most in your pursuit:

1. Download the manual and read every page of it and do the measurements it covers.

2. Attend SysTune and Syn-Aud-Con seminars. The latter has the best on-line training available.

#1 will lead you to other documents and give you the tools necessary to understand them. Always do measurements to backup what you read.

#2 will result in the quickest way to get started in the right direction. Nothing beats experience.

SysTune v1.2 Manual: http://www.renkus-heinz.com/support/software-support/systune/SysTune-1.2_Manual.pdf


Last edited by Langston Holland on Mon Aug 20, 2012 1:05 am; edited 1 time in total

_________________
God bless you and your precious family - Langston
View user's profile Send private message
PostPosted: Wed Aug 15, 2012 12:29 pm Reply with quote
AFMG Pedro Lima
Forum Moderator
 
Joined: 05 Jun 2010
Posts: 266
Location: Germany




Dear Mr. Holland,

Your answer has been quite good, thanks. And yes, you're right when you say:

"Systune is kind of like a surgeon's scalpel - it's looks pretty simple on its surface, but to know when, how and why to use it correctly takes a great deal of time, sutdy and practice."

So, please allow me add just a few more comments in order to help other users clarifying things:

1. With regards the weighting cures, they have generally been invented so that broadband SPL values related more closely to the human perception of loudness. Hence the higher sensitivity around 4 kHz and low sensitivity for lower frequencies. That said, one should be very careful whenever trying to "match" a system frequency response using those curves. They were not originally intended to that purpose. That's why we are very hesitant about inserting the weighting curves into the TF Mag tab.

For instance, one in general should not try to give a subwoofer a boost of +40 dB at 31.5 Hz based on the sensitivity of the curve A for that region (see picture below). That's generally the region of a natural row off for most subs.



2. Secondly, a system Transfer Function is not only composed by it's Magnitude (or the system frequency response), but also by its Impulse Response (or the system temporal response). From a system Impulse Response several acoustical parameters can be extracted, such as the Reverberation Time, the STI, C50, C80, etc... This information cannot be extracted from a single channel measurement (using the spectrum). In addition to that, from an Impulse Response measurement, an experienced user can extract information with regards room reflections, apply windowing and also check the system phase (from the phase graph). Finally, in order to use the newly developed Delay Analysis plug-in of Systune, one should measure the system Transfer Function.

I Hope That Helps.

Best Regards,
Pedro
View user's profile Send private message
PostPosted: Wed Aug 15, 2012 2:17 pm Reply with quote
Langston Holland
Member
 
Joined: 13 May 2007
Posts: 31




"To help other users"? Heck - I learned from your post as well Pedro. :)

_________________
God bless you and your precious family - Langston
View user's profile Send private message
PostPosted: Wed Aug 15, 2012 8:01 pm Reply with quote
hadisumoro
Member
 
Joined: 01 Oct 2007
Posts: 39
Location: USA




http://karlnordstrom.ca/research/wp-content/uploads/2010/09/Time-and-Frequency-in-Loudspeaker-Measurements.pdf

Hadi
View user's profile Send private message
PostPosted: Mon Oct 22, 2012 11:21 am Reply with quote
r.b.
Member
 
Joined: 11 Jun 2012
Posts: 3




Pedro and Hadi,

Thank you very much for your precious input!

it helped me get closer to an understanding of what's going on.

The guy who wrote the article really has an amazing understanding of Physics.

Two side notes:

I wasnt trying to boost a sub +40 db at 40 hz, that's why i originally said to "match the curve" : in the case of A i would have tried to set the sub at -40 db (pink noise)

The idea is to use a room plus eq to transform the signal into what the human ear then decodes as "flat" (pink noise).

Since the human ear does it differrently for each loudness, use the appropriate curve, and yes they are averages and old an whatnot. but, well, so are the ears of our listeners ;o)

Of course this is just a start, but i usually don't have the luxury of doing half an hour of pink noise measurements, i get 15 minutes tops and mostly 5. (indoor DJ setups in pubs sometimes with guests present, not open air FOH)

So i need to zero in on that eq settings pretty quickly and i wanted use the frequency domain as a starting point to get into the ballpark.

The eq simulator doesnt help me, since it doesnt support 31 band eq which everybody uses. (btw, pls build that in!)

anyway i will continue to experimant aorund it.

Thanks again for your input, and sorry for the late reply, somehow the mailing notfication didnt reset when i read the first response, so in just saw the last ones...
View user's profile Send private message
AFMG Network Forum Index -> AFMG SysTune
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
All times are GMT  
Page 1 of 1  

  
  
 Post new topic  Reply to topic  


Powered by phpBB © 2001-2003 phpBB Group
Theme created by Vjacheslav Trushkin
Variation by CodeWeavers and AFMG