Minidsp/Dirac/Bass management noob confusion

[phoenixdogfan]

Any idea when Dirac will release bass management for Studio?

 

[Bornsceptic]

To integrate a subwoofer, you need to:

1) Gain match the speakers to the subwoofer, SPL matching is also possible but not as effective. you will use the MiniDSP for volume control from now on.

2) apply crossover, usually low pass 4th order LR (24db/oct) at 80Hz to the subwoofer and high pass at 80Hz 2nd order BW (12db/oct) to the main speaker.

3) adjust the delay so your crossover region (80-120Hz) does not have any dips

4) lower the speakers' gain by 4 db to adhere to the harman target curve

5) apply your preferred method of room correction.

Bass management does all of these things automatically, while with a regular MiniDSP you need to do these processes yourself.

thanks for clarifying this. I think I’ll start with the purchase of a DDRC-24 and hopefully have some fun/headaches trying to learn all of this. I’m pretty sure in terms of improving sound this will get me much further than a new sub (although the joy of the actual purchase would be much bigger with a shiny new SVS 1000 pro compared to the small, boring-looking MiniDSP )

 

[MarkS]

Another noob question: can I use the DDRC-24 in the tape-loop of an old-fashioned receiver?

 

[fluid]

thanks for clarifying this. I think I’ll start with the purchase of a DDRC-24 and hopefully have some fun/headaches trying to learn all of this. I’m pretty sure in terms of improving sound this will get me much further than a new sub (although the joy of the actual purchase would be much bigger with a shiny new SVS 1000 pro compared to the small, boring-looking MiniDSP )

Have you got a calibrated measurement microphone already? You might want to start with a UMIK and get used to the process of measuring the system you have and trying to understand what you are seeing. If your audio comes from a computer then there are many different options to consider that might be more suitable and could allow you to test out what they can do before spending money on a hardware processor.

You might find that positioning your existing speakers to work better with the room has a big impact on the sound you are getting at your listening position. Correction systems are very useful but they cannot overcome poor placement and bad directivity. The desire to "fix" these sort of problems is what can give room correction products a bad reputation.

 

[Bornsceptic]

Have you got a calibrated measurement microphone already? You might want to start with a UMIK and get used to the process of measuring the system you have and trying to understand what you are seeing. If your audio comes from a computer then there are many different options to consider that might be more suitable and could allow you to test out what they can do before spending money on a hardware processor.

You might find that positioning your existing speakers to work better with the room has a big impact on the sound you are getting at your listening position. Correction systems are very useful but they cannot overcome poor placement and bad directivity. The desire to "fix" these sort of problems is what can give room correction products a bad reputation.

Yes I thought about that and that might actually be a good idea. Audio is however coming from RPi streaming with Moode (which works very smoothly) so no ”real” computer involved. That’s why MiniDSP and Dirac Live is such an appealing option. But as you say, there are seldom any quick fixes. But I read many accounts of people having success with taming at least the worst room modes in the bass area (where I perceive my problems are).

 

[fluid]

You cannot use any room based processing without a calibrated microphone in any sensible way. A Pi is a computer as far as this is concerned. As you are using Moode then look at CamillaDSP which is now integrated with Moode and can run quite a number of convolution and EQ options. There are a number of strategies that can be used to try and improve speakers based on in room measurements. Audiolense and Acourate are paid versions of software that I consider to be a step up from Dirac although they are not as simple to use. There is the free DRC_FIR which is very good but a steep learning curve. Alternatively REW can be used with single point or vector averaged multipoint measurements and Frequency dependent windowing in the frequency domain and rephase to tackle any time domain related issues, rephase can generate the filters in the format that Camilla accepts and REW can export it's filters in rephase compatible xml too.

If any of this is of interest to you I can point you in the right direction to start understanding it.

The miniDSP devices are tap limited and hardware quality limited but they do offer a simpler solution which is why they are popular.

 

[Bornsceptic]

Thank you! I will definitely reach out as soon as I pull the trigger on a UMIK and start my measuring.

So, only doing this with software, aren’t there limitations in terms of what can be separated for the L+R and the Sub? I don’t have a sub out on my amp either, so I’m not able to control the crossover. Wouldn’t a hardware crossover solution (such as the MiniDSP) be needed in this case?

 

[SKBubba]

Seems like you are half way to the cost of a Denon AVR X3700 that does everything except your laundry!

 

[pokerbarlo]

that's just the routing matrix.

Bass control is different from Bass managemnt it is Diracs more advanced solution and just available for mid to high end processors MiniDSP just has Bass management

 

[fluid]

Thank you! I will definitely reach out as soon as I pull the trigger on a UMIK and start my measuring.

So, only doing this with software, aren’t there limitations in terms of what can be separated for the L+R and the Sub? I don’t have a sub out on my amp either, so I’m not able to control the crossover. Wouldn’t a hardware crossover solution (such as the MiniDSP) be needed in this case?

The software is not the limiting factor but the number of channels available in a soundcard or hat. Camiila (if you chose to use it) can output to as many hardware channels as you have available. The Pi4 is really quite powerful as an audio processor in comparison to the standard hardware chips that are in DSP processors. 8 channels at high sample rates and long FIR filters is within it's capabilities.

Something like an ASUS Xonar USB interface is a cheap and easy way to get 8 channels of audio out of Pi at decent quality.

If you just have a standard stereo DAC then any processing will only act on the left and right channels. You can still use this to deal with reducing room mode peaks etc. but it is not as flexible as having multiple subwoofers and separate processing for each channel. If you have one sub you would cross it over in much the same way you do now, you could take the advice given above for integration and then over the top of that you can apply EQ. If you listen in a single position this could give you the possibility to effect a positive change on your existing equipment with virtually no cost. With only a single sub the bass throughout the room will be wildly different and the processing for one position may make it much worse elsewhere. The only real solution to that is directivity and multiple sources.

 

[Bornsceptic]

The software is not the limiting factor but the number of channels available in a soundcard or hat. Camiila (if you chose to use it) can output to as many hardware channels as you have available. The Pi4 is really quite powerful as an audio processor in comparison to the standard hardware chips that are in DSP processors. 8 channels at high sample rates and long FIR filters is within it's capabilities.

Something like an ASUS Xonar USB interface is a cheap and easy way to get 8 channels of audio out of Pi at decent quality.

If you just have a standard stereo DAC then any processing will only act on the left and right channels. You can still use this to deal with reducing room mode peaks etc. but it is not as flexible as having multiple subwoofers and separate processing for each channel. If you have one sub you would cross it over in much the same way you do now, you could take the advice given above for integration and then over the top of that you can apply EQ. If you listen in a single position this could give you the possibility to effect a positive change on your existing equipment with virtually no cost. With only a single sub the bass throughout the room will be wildly different and the processing for one position may make it much worse elsewhere. The only real solution to that is directivity and multiple sources.

This was really helpful, very clear and easily explained for someone like me Actually this was how I thought things worked, but often there is so much detail and technical jargon in discussions that the basic function never gets explained!

I like the idea of using the Pi for this, as well as having a modular approach to improvements. A UMIK and a Asus Xonar on the shopping list then!

 

[fluid]

This was really helpful, very clear and easily explained for someone like me

You are welcome

Here is a thread from diyaudio where scripts and a method are provided as well as sample settings to process impulses with DRC_FIR.

https://www.diyaudio.com/forums/ful...ectrical-loudspeaker-correction-networks.html

If you are able to use linux on a Pi then I doubt using this is beyond your capability to learn

With this and some knowledge there is no need to spend money on Dirac.

 

[Bornsceptic]

You are welcome

Here is a thread from diyaudio where scripts and a method are provided as well as sample settings to process impulses with DRC_FIR.

https://www.diyaudio.com/forums/ful...ectrical-loudspeaker-correction-networks.html

If you are able to use linux on a Pi then I doubt using this is beyond your capability to learn

With this and some knowledge there is no need to spend money on Dirac.

A follow-up question: I’m starting to realize that my amp (a Peachtree Nova) might be a limiting factor as well, since it has no sub-out (only pre-out) and no main in. How would this affect your mentioned solution (ie Rpi/Camiila —> Dac —>L + R (amp) and Sub(s)? Would I control the volume only with my Rpi/Moode audio? Because the Amp volume control would only affect the Mains, right? Is there a way around this, or do I just need to get an amp with proper bass high-pass management?

 

[fluid]

If you are using the high level input on the speakon connector then the sub receives the same signal and volume as your amplifier is putting out. The Volume on the sub will then set the relative level, the sub would change volume as you change volume on the amplifier but relative to the volume setting of the sub itself.

There doesn't seem to be a way to use stereo inputs on the RCA connectors. If you had an interface with more than two channels on your pi and you wanted to be able to EQ the signal to the sub separately you could go directly into the RCA input on the sub but you would have to do some bass management in the Pi to mix the L/R signals to mono. The crossover in the REL can't be switched off so you would have to factor that into the response.

Even with just a stereo connection you can EQ the full frequency range including the sub, but with only a single sub the EQ will only be valid in a small part of the room and not throughout the room as it could if you used multiple subwoofers and processed them correctly to achieve minimal spatial variation.

 

[abdo123]

Yes I thought about that and that might actually be a good idea. Audio is however coming from RPi streaming with Moode (which works very smoothly) so no ”real” computer involved. That’s why MiniDSP and Dirac Live is such an appealing option. But as you say, there are seldom any quick fixes. But I read many accounts of people having success with taming at least the worst room modes in the bass area (where I perceive my problems are).

if you have ABSOLUTELY NO OTHER INPUTS, then camila DSP can be very easily used on moode to do subwoofer integration (if you have a 4 channel plus DAC)

ofcourse you won’t have the ‘Dirac Magic’ this way.

 

[Bornsceptic]

Ok, understood! So my Velodyne Minivee might be better in this regard, as it has RCA stereo input and option to bypass internal crossover.

However, I’m more and more getting the feeling that I should just start selling off my ”budget audiophile” gear and get some proper stuff (like an amp with proper bass management and dual SVS subs) and be done with it!

 

[fluid]

Once you head down this sort of rabbit hole you will probably never be done with it

I would suggest that you learn to measure, place and equalize with the speakers you have before selling them and moving on. It will teach you a lot and the knowledge and experience you gain will give you a better idea of what you like and don't like and the causes for that. Take what you have as far as you can, then move on.

I can understand the desire to have a program or algorithm make life easy, there is no magic to it though. Arthur C Clarke's quote that "Any sufficiently advanced technology is indistinguishable from magic" could apply if you don't understand what is being done.

 

[onion]

https://www.minidsp.com/applications/subwoofer-tuning/ddrc24-multisub

Will this allow a minidspwupgrades to ddrc to do multisub bass management in dirac?

 

[fluid]

No the Dirac in the DDRC24 works on the combined result not separately. Quote from that application note

"However, much better results can be obtained by using multiple subwoofers. See our app note Tuning multiple subwoofers with miniDSP. (Do the multisub tuning first, then run Dirac Live afterwords.) "