• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

My DRC Setup - Am I Doing This Right? + Alternatives

dfgoiuj

Member
Joined
Mar 22, 2022
Messages
43
Likes
5
Hi all,

For some years now I've been doing the following to correct my room and speakers:
  1. Multiple position measurements in REW using independent left and right sweeps with ECM8000 mic (connection via Citronic CULINK 48V XLR USB interface; I'm looking to replace this)
  2. Export individual impulse responses to 32-bit float WAVs
  3. Combine all left channel WAVs in Audacity and save to single WAV, do the same for right channel WAVs
  4. Convert to PCM with SoX and use DRC Designer create the filter, load into Equalizer APO
  5. Measure result in REW and EQ to taste
My audio chain looks like this:

Audio source -> TOSLINK In (cheapo USB soundcard @ 48kHz) -> Windows PC (actually a VM on Proxmox)-> TOSLINK Out -> Topping E30 DAC -> Sub (93L ported Adire Audio 12") -> Active Speakers (Beoplay S3)

It sounds pretty good to my ears. Very balanced, and the sub provides very impressive bass. I am looking to upgrade the speakers, but that's another topic.

I suppose I'm wondering if I could be doing this better somehow, and if I could get better results from a paid software suite. I've also been wondering about dedicated hardware such as MiniDSP. Particularly to act as a crossover to separate sub from speakers properly.

Thanks
 

Rednaxela

Major Contributor
Joined
Mar 30, 2022
Messages
2,052
Likes
2,676
Location
NL
I've also been wondering about dedicated hardware such as MiniDSP. Particularly to act as a crossover to separate sub from speakers properly.
If you have access to a multi channel sound card, you may be able to emulate this with what you have.
 

Rednaxela

Major Contributor
Joined
Mar 30, 2022
Messages
2,052
Likes
2,676
Location
NL
Hope others more qualified than me can help you with the other questions.
 
OP
D

dfgoiuj

Member
Joined
Mar 22, 2022
Messages
43
Likes
5
If you have access to a multi channel sound card, you may be able to emulate this with what you have.
Interesting, I hadn't thought about that, thanks. I don't have such a card at the moment, but it's good to know there's an alternative.

I've just been watching Mitch's video on room correction, and I'm wondering if multiple measurements are the best approach after all.
 

Eetu

Addicted to Fun and Learning
Forum Donor
Joined
Mar 11, 2020
Messages
763
Likes
1,180
Location
Helsinki
I'm wondering if there's a reason you don't average the measurements within REW? Vector averaging retains time-domain info.

I'm not familiar with DRC Designer but go from REW to rePhase.

If you are doing convolution with a lot of taps then MiniDSP is not the best solution since the processing power (taps) is limited. As suggested above I would look at multi-channel DACs instead. Of course a hybrid solution is possible too, where you do the crossovers/routing with a MiniDSP but still use a PC to apply FIR filters.
 
OP
D

dfgoiuj

Member
Joined
Mar 22, 2022
Messages
43
Likes
5
I'm wondering if there's a reason you don't average the measurements within REW? Vector averaging retains time-domain info.

I'm not familiar with DRC Designer but go from REW to rePhase.

If you are doing convolution with a lot of taps then MiniDSP is not the best solution since the processing power (taps) is limited. As suggested above I would look at multi-channel DACs instead. Of course a hybrid solution is possible too, where you do the crossovers/routing with a MiniDSP but still use a PC to apply FIR filters.
I tried to use the inbuilt averaging but it didn't let me export the result as an impulse response, unless I missed something. I'd prefer that workflow as it's one step less.

According to Equalizor APO, the filter is 65k taps. What I haven't figured out yet is why this doesn't translate to huge latency.
 

Eetu

Addicted to Fun and Learning
Forum Donor
Joined
Mar 11, 2020
Messages
763
Likes
1,180
Location
Helsinki
I tried to use the inbuilt averaging but it didn't let me export the result as an impulse response, unless I missed something. I'd prefer that workflow as it's one step less.

According to Equalizor APO, the filter is 65k taps. What I haven't figured out yet is why this doesn't translate to huge latency.
Yes, you should find an option to export impulse response as WAV in the File->Export menu.

Hmm, not sure how EQAPO handles latency. Now that you mention it I recall having more noticeable lag when using convolution filters of the same length on a Raspberry Pi.

But for comparison, MiniDSP handles only 2048 taps/channel IIRC.
 
OP
D

dfgoiuj

Member
Joined
Mar 22, 2022
Messages
43
Likes
5
Yes, you should find an option to export impulse response as WAV in the File->Export menu.

Hmm, not sure how EQAPO handles latency. Now that you mention it I recall having more noticeable lag when using convolution filters of the same length on a Raspberry Pi.

But for comparison, MiniDSP handles only 2048 taps/channel IIRC.
That's the option I use to export the individual measurements as impulse responses, but it throws a 'no data to export' error when trying to do the same with an averaged measurement.
 
OP
D

dfgoiuj

Member
Joined
Mar 22, 2022
Messages
43
Likes
5
I wonder that this is working for you, DRC-FIR says there should be at most 8 samples between left and right measurements. If you move your mic around during measuring it will obvisouly not work. https://audiosciencereview.com/forum/index.php?threads/room-correction-linux.25894/post-1036796
Interesting. The filter makes a huge difference to my ears when produced with the method I posted, but I'm keen to try a single measurement again and see how it compares. When I first started playing around with DRC I would take a single measurement, but I wasn't blown away by the results. I decided to try taking multiple measurements and preferred how it turned out. However, it's entirely possible my poor results were due a mistake somewhere, especially as the process (and taking measurements in general) were new to me back then.
 

Eetu

Addicted to Fun and Learning
Forum Donor
Joined
Mar 11, 2020
Messages
763
Likes
1,180
Location
Helsinki
That's the option I use to export the individual measurements as impulse responses, but it throws a 'no data to export' error when trying to do the same with an averaged measurement.
That's probably if you have used RMS average? Vector averaging should work.
 
D

Deleted member 46664

Guest
For some years now I've been doing the following to correct my room and speakers:

Audio source -> TOSLINK In (cheapo USB soundcard @ 48kHz) -> Windows PC (actually a VM on Proxmox)-> TOSLINK Out -> Topping E30 DAC -> Sub (93L ported Adire Audio 12") -> Active Speakers (Beoplay S3)

Okay you have a self-powered sub and powered speakers.

Have you tried actually plugging this into the mini-trs jacks on the back of your Windows PC?

download.jpg

In the sound applet for your default Speaker Out device you should be able to select 2.1 audio in which case you connect your powered mains to the green jack and the sub to the brown jack. From there you should be able to configure your setup first using the gains in your speakers and second with the gains in your sound control panel.

Yeah yeah, I know motherboard audio sucks ... but have you ever tried it?
 
OP
D

dfgoiuj

Member
Joined
Mar 22, 2022
Messages
43
Likes
5
I've not tried it, but that's pretty cool. The computer I'm using is a small form factor Dell Optiplex which doesn't have multichannel audio. It's around ten years old, but built to last, and compact enough to be hidden away. I am thinking of swapping it for something smaller and cooler running, though. Something like a NUC, or this RPi 3 I've got if I can figure out how to turn it into a DSP box.
 
Top Bottom