TOPIC:

See image here. This is the current configuration of MiniDSP's implementation of Dirac. WHY is it not AFTER the routing/xover stages (red arrow)?
https://www.avsforum.com/cdn-cgi/image/format=auto,onerror=redirect,width=1920,height=1920,fit=scale-down/https://www.avsforum.com/attachments/screenshot-2022-10-12-at-17-48-09-png.3345802/

That is the case for the stereo Dirac platforms (SHD / Flex / DDRC-24) but it is not true for the multichannel Dirac platforms (DDRC-88A/D). For example see similar figure for the DDRC-88 platforms.

 

I don't know the exact reason why this is done other than it is obviously computationally less intensive to correct two channels and the platforms that only have two channel Dirac correction are designed as stereo input platforms.

Michael

This seems to defeat the purpose of utilizing this for sub integration though, as it applies correction PRIOR to setting the crossover and time delay. Yes?

Sub integration in all miniDSP products is a manual exercise, you set the crossover points, you set the delays. Assuming you do this properly Dirac supposedly has the ability to linearize the phase behavior of the x-over. I am somewhat skeptical of this claim as all miniDSP products have rather limited FIR taps and low frequency phase linearization requires a lot of taps.

If you move Dirac to the output channels this stretches DSP resources even further. Looking at the non-Dirac Flex8 for example they have FIR taps on the inputs as opposed to the outputs like 2x4HD and normal Flex. To me this is a good choice as it is your only hope of efficiently using FIR taps at a 96 kHz sample rate (would be great if they had the option to reduce the sample rate to 48 kHz for more efficient use of FIR taps!) to linearize phase.

Dirac has a relatively new product called Dirac Live Bass Control which will help with sub integration. It is currently only found on high end processors costing several times as much as any miniDSP product. I haven't researched it extensively but at first glance it looks like that to actually use DLBC on these processors ->  Online Store - Dirac  costs an additional $350 for the single subwoofer license and additional $500 for the multiple subwoofer license.

Michael

besposito wrote:

This seems to defeat the purpose of utilizing this for sub integration though, as it applies correction PRIOR to setting the crossover and time delay. Yes?

As @mdsimon2 describes, the purpose of Dirac Live on the stereo miniDSPs like the SHD is not subwoofer integration, but rather adjustment of the frequency response of a two channel system. More specifically, it is the time/phase integration of subwoofers that Dirac Live can't do.

besposito wrote:

This seems to defeat the purpose of utilizing this for sub integration though, as it applies correction PRIOR to setting the crossover and time delay. Yes?

The correction is applied prior to crossover etc., however, the measurements made in order to define the Dirac behaviour are made on the entire chain - crossover, EQ, time delays, etc. In other words, your workflow when setting up e.g., a DDRC-24 is to define the levels, crossover points and any eq/time delays FIRST, then you take the Dirac measurements using those values.