Question About LFE Simulaiton for Atmos w/2 Subwoofers

whoareyou

Member
Thread Starter
Joined
Dec 23, 2021
Posts
260
More  
Main Amp
Benchmark AHB2
Additional Amp
ATI AT525NC
DAC
Exasound E38
Computer Audio
JRiver
Front Speakers
Psb synchrony one
Center Channel Speaker
PSB synchrony one C
Surround Speakers
Revel m105
I'm trying something new with subwoofers. Previously I treated the 2 subs as 1 speaker and simulation would show the subs w/Atmos setting with the 10db amplificiation clearly visible.
With 2 subs configured as separate speakers (see setup image), the simulation shows each sub but with lower than 10db amplification. Is this as intended since each speaker is contributing to the 10db boost?

Also, when configured in this manner is the simulation only able to show the subs individually or is there a way to show the combined effect just for LFE channel?
Thanks
 

Attachments

  • Setup.png
    Setup.png
    81.2 KB · Views: 32
  • FrequencyResp.png
    FrequencyResp.png
    312.9 KB · Views: 29
Hi, yes the reduction is intended. The som of the two subs give the 10 dB boost.

You can not see the combined effect for the LFE channel and they are not corrected together here. You could e.g. export the sim to REW for a closer examination there.

The two subs are however corrected towards the same target, and unless there’s a non min phase cancellation on one that is not similar on the other the two will add up pretty well. You can spot a cancellation between the two as a close to 190* difference on the wrapped phase.
 
Why is left full range and right small?
 
For whatever reason, my room likes the left speaker's placement better and that speaker's bass response is smooth enough down to around 25 - 30 hz. Right speaker is a little rough and needs the SW support to fill a larger dip at around 40hz.

But entire reason for the new setup with 2 individual subs vs combined sub, is that I'm trying to reduce number of convolution paths. Since the left didn't need the bass off-loading, I removed it and reduced paths by 1 more. Probably negligible from performance standpoint, but makes no difference sonically, so I did it anyway since the goal is to reduce paths.

I really liked the combined SW setup, but it resulted in 35 convolution paths and that got a bit limiting at times i.e. with HLConvolver it limited number of filters I could load at same time, or it limited sample rates I could use.
 
Back
Top