Home Theater Forum and Systems banner

1 - 4 of 4 Posts

·
Registered
Joined
·
17 Posts
Discussion Starter #1 (Edited)
Hi Folks,

I measured Genelec 1031 Speakers with different sound optimizers/equalisers in a broadcast truck for a student projekt. Guess that might be interesting for many of you too ;-)

Now I made Waterfall graphs and I think/hope I analyzed it right.
There are quite big differences in the very early "decay", roughly between 90 and 200 Hz.

I'm not sure if its really like that - and if it is - why?

Or is it due to analysis artefacts/mistakes?


My first guess was that room modes are not excited so much if the EQ dampens the Bass - but the Trinnov is not reducing about 120Hz and above that its even amplifying a bit.



All measurements done in sweetspot/hearing position, pretty loud weighted sweep, small capsule Pressure Mics

Waterfall Analysis Preference were the same for all.
Preferences for SPL/MAgnitude were different due to different IR. So the SPL/Magnitude is not 100% right in the Waterfalls.
E.g. Trinnov IR start 5ms before Peak and seems only correct if I choose a "Rise Time" of 5ms - but then time resolution is worse.... anyway thats not the problem.

Please have a look at the graphs.
(Noise that look like resonances at 30, 50, 100 Hz is caused by the Aircon (not 100% sure but 95%)
Genelecs 1031 no EQ/filters at all
Genelec links.png

1031 witk KRK Ergo
KRK links.png

1031 with Trinnov filter with preference and view x=0
Trinnov with preferences.png

1031 with Trinnov filter
Trinnov links.png

1031 and Trinnov Magnitudes + Trinnov Transferfunction
Trinnov Magnitudes.png


Any ideas or further questiones to make sth clear ?
 

·
Super Moderator
Joined
·
9,186 Posts
Not sure I get the purpose of the experiment? Waterfalls are primarily useful for gauging improvements in low frequency absorption treatments (e.g. before and after). They tell you nothing about a speaker.

In addition waterfalls are not useful below about 3-400 Hz. Typically you want to see one with a 300 ms window and horizontal axis down to 35-40 dB, which is the ambient noise level in most residential rooms.

There are quite big differences in the very early "decay", roughly between 90 and 200 Hz.

I'm not sure if its really like that - and if it is - why?

Or is it due to analysis artefacts/mistakes?
It will be a function of the speaker’s magnitude response in that frequency range. For instance, all things being equal (e.g. all speakers level-matched before measurement) a speaker with response that sags somewhat in that area will show quicker decay in that range. Conversely, one with emphasis in that range will show longer decay time. See here for a discussion on signal levels and decay times.

Regards,
Wayne
 

·
Registered
Joined
·
17 Posts
Discussion Starter #3
Well the prupose was to compare before/after "optimization" with different Gear that is sold to improve your sound. KRK Ergo and Trinnov Optimizer are "Room Acoustic Optimization Systems" Gear/Hardware like Lyngdorf, Audissey etc.
And those Systems are promoted to improve sound, phase, time correctness, tightness, whatever comes to the mind of the sales department ;-)

So it's always the same Genelec Speakers with a different Equalization/Filter between Speaker and Soundcard.

I'm not sure if I interpret you answer/question right.

My question/"problem" is not the magnitude in Waterfall Slice no. 1 - but the difference between Slice 1 and Slice 2 that varies between the 3 measurements/Waterfalls.
Trinnov Slice 1 Level around 92dB.
Trinnov Slice 2 (1ms later) Level around 80 dB.
If I extend the Rise Time to 5ms or 10ms the difference is less but still quite big.

So I ask myself if I did a mistake in the analysis or if there is some kind of DSP Magic going on, though I'm quite sure there is no such thing as magic in Audiogear, but Physics and Math are deep and I cant dive to the ground of it...
 

·
Super Moderator
Joined
·
9,186 Posts
My question/"problem" is not the magnitude in Waterfall Slice no. 1 - but the difference between Slice 1 and Slice 2 that varies between the 3 measurements/Waterfalls.
Trinnov Slice 1 Level around 92dB.
Trinnov Slice 2 (1ms later) Level around 80 dB.
If I extend the Rise Time to 5ms or 10ms the difference is less but still quite big.
You’re right, something is definitely amiss there. There’s no way the sound would decay 12 dB in a few milliseconds and then go back to “normal” decay.

Regards,
Wayne
 
1 - 4 of 4 Posts
Top