• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

JBL LSR 308 in the house

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
6,783
Likes
2,873
Location
Riverview, FL

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
6,783
Likes
2,873
Location
Riverview, FL
Just for the record, my tendency is to calibrate the room levels to the levels on the CD, for general "loud enough" listening pleasure.

If a channel of CD maxes at 96dB - then two speakers, playing the two tracks, should max around 99dB for uncorrelated signals, and 102dB for correlated signals.

Since I DRC flat - it makes a nice reference to observe the raw levels of the tracks vs the combined and DRC'd levels in the room, and usually comes quite close.

An RTA peak trace of the DRC'd room sound should match the peaks on the raw source pretty well, and does... Works better with the ML than the JBL though, their traces look a little more ragged for some reason.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
6,783
Likes
2,873
Location
Riverview, FL
Oh, Z-weighting... well that's an unusual choice. Throwing my whole reference set out of whack!
http://www.cirrusresearch.co.uk/blog/2011/08/what-are-a-c-z-frequency-weightings/



Well...

For my purposes, which is not measuring "safety", and not measuring "equal loudness", and is not particularly concerned with sub-sonic frequencies, and is measuring recorded music, which has already been subjected by the tonmeister to shaping by his ears, I use Z mostly.

I'm not measuring my response, but response of the system to excitation, and assuming the excitation (when it is music) has already been "curved" by someone's prior listening and knob twiddling.

The goal of all parts of the system - player - DAC - pre - amps - speakers - microphone - is to generate "flat".

Why would I measure with a curve? Should I boost 30hz by 40dB because A-weighting is closest " to reflect the response of the human ear to noise"?

Hint: boosting 40hz by 30dB is not what I want to listen to. I've experimented with that - curve to ear sensitivity doesn't work for me.

C and Z are close enough there's no reason that I see to go with C.
 

watchnerd

Major Contributor
Beer Hero
Joined
Dec 8, 2016
Messages
4,407
Likes
1,475
Location
Seattle Area
http://www.cirrusresearch.co.uk/blog/2011/08/what-are-a-c-z-frequency-weightings/



Well...

For my purposes, which is not measuring "safety", and not measuring "equal loudness", and is not particularly concerned with sub-sonic frequencies, and is measuring recorded music, which has already been subjected by the tonmeister to shaping by his ears, I use Z mostly.

I'm not measuring my response, but response of the system to excitation, and assuming the excitation (when it is music) has already been "curved" by someone's prior listening and knob twiddling.

The goal of all parts of the system - player - DAC - pre - amps - speakers - microphone - is to generate "flat".

Why would I measure with a curve? Should I boost 30hz by 40dB because A-weighting is closest " to reflect the response of the human ear to noise"?

Hint: boosting 40hz by 30dB is not what I want to listen to. I've experimented with that - curve to ear sensitivity doesn't work for me.

C and Z are close enough there's no reason that I see to go with C.
Oh, I'm familiar with it, I just never use it because so few things are measured with Z-weighting that it creates extra work for me to normalize the data.

This usually comes up when I'm trying to deal with different microphones.
 

fas42

Major Contributor
Joined
Mar 21, 2016
Messages
2,818
Likes
135
Location
Australia
Not being a mic jockey, I hadn't appreciated these weighting aspects, especially in the higher frequencies - I had always assumed that that people were talking of measuring the flat response in a situation. If the measuring device is "distorting" the reading, and the reader is not aware of that no wonder confusion reigns at times ...
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
6,783
Likes
2,873
Location
Riverview, FL
I hadn't appreciated these weighting aspects
As I understand it:

Measurement microphones are intended to to measure "flat" or unweighted... They may also include a calibration file to correct for their unavoidable un-flatness. Their output may be displayed with or without other weighting schemes applied depending upon what it feeds and the capabilities of those devices.

SPL meters have curve choices, usually A and C (at minimum). I would think that is like a "filter", applied to the otherwise flat-ish output of the attached "measurement" mic.

REW's SPL meter gives the choices of A, C, or Z (unweighted).

REW's microphone setup options "corrects" for whether or not the initial data from the measuring device is C or unweighted, to permit the use of C-weighted SPL meters as an input, and permits the use of a Calibration file for use with measurement mics so equipped.

Noise regulations are based on the weighted measurements, since it more accurately reflects the dangers of SPL at frequency for regulatory sound exposure (health and safety) purposes.

Performance microphones (for recording vocal or instruments) have curves and polar patterns based on their inherent capabilities or tuning for specific purposes.

You can roam around here - http://www.acoustic-glossary.co.uk/definitions-l.htm - for more than you need to know...
 
Last edited:

DonH56

Major Contributor
Technical Expert
Patreon Donor
Joined
Mar 15, 2016
Messages
2,688
Likes
2,783
Location
Monument, CO
I always use unweighted response for audio system measurements, A weighted only when required for specific tests (e.g. OSHA or to verify manufacturer's specs), and C only because my SPL meter does not "do" flat. In audiophile fora I have always seen system measurements based upon unweighted ("Z") response. Audio components often list several specs and sometimes mix things up, e.g. unweighted frequency response and weighted SNR (to provide the best-looking numbers for marketing, natch).

C deviates substantially from Z at low and high frequency extremes. I actually use a calibrated mic, either a UMIK-1 with cal file (from CSL), or my Earthworks M30 (which has more extended response). Z has meant "flat" over the years but the EIA (I think? maybe IEEE?) defined it explicitly a while back because "unweighted" was interpreted differently by different manufacturers so mic and system roll-off was not consistent. Most handled the HF portion OK, i.e. flat to 20 kHz, but the LF corner varied (some used 20 Hz, some up to 30 or 40 Hz, and some lower -- my mic rolls off to -3 dB around 5 Hz). I believe the standard now is flat from 10 Hz to 20 kHz with +/-1 dB (not sure the flatness spec; IIRC it is tighter than +/-3 dB but not sure it is +/-1 dB).

FWIWFM - Don
 

watchnerd

Major Contributor
Beer Hero
Joined
Dec 8, 2016
Messages
4,407
Likes
1,475
Location
Seattle Area
Performance microphones (for recording vocal or instruments) have curves and polar patterns based on their inherent capabilities or tuning for specific purposes.
This is correct. Performance microphones are definitely not flat.

You don't even need a calibrated measurement mic to detect this. It's blatantly obvious to the unaided ear. There are microphone review sites with regular shootouts that compare the same sounds recorded through different mics.
 

fas42

Major Contributor
Joined
Mar 21, 2016
Messages
2,818
Likes
135
Location
Australia
Thanks, gentlemen ...
 

Blumlein 88

Major Contributor
Joined
Feb 23, 2016
Messages
5,681
Likes
4,659
This is correct. Performance microphones are definitely not flat.

You don't even need a calibrated measurement mic to detect this. It's blatantly obvious to the unaided ear. There are microphone review sites with regular shootouts that compare the same sounds recorded through different mics.
You can look here at the recordinghacks mic database for specs and response graphes of nearly any recording microphone you can come across. They aren't flat in general. No matter the cost.

http://recordinghacks.com/microphones


Here is a $3800 Neumann microphone with omni capsule.

http://recordinghacks.com/microphones/Neumann/TLM67



Now to my surprise, I took an inexpensive omni mic and created a calibration file for REW using a published graph of response like the one above. When I compared it to my recently acquired Umik1 it was a near perfect match. I would have thought the published curve too good to be true.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
6,783
Likes
2,873
Location
Riverview, FL
In post #1, I showed a step response. I inverted it to make it "look" a little better.

Here's the "raw" (red) and "DRC'd" (blue) step response measured at the listening position without any "invert the signal" post-processing in REW (or in the source), just presented as a curiosity:

upload_2017-1-14_9-22-46.png
 

Similar threads

Top Bottom