IIR vs FIR filters. Is the sacrifice in fidelity worth the better DSP of FIR?

My Mini DSP is being finicky with the infrared learning of my remote and I think I may need to replace the unit with a Mini DSP Flex Eight because the one I have is out of production. The newer models have the option of keeping 32 bit/96 kHz with IIR filters or 32bit/48 kHz with Dirac Live® which uses FIR filters. The additional processing power needed to run FIR filters reduces the throughput. Do the superior FIR filters, which can correct phase so it is linear (IIR filters cannot), improve the DSP more than to make up for the loss in fidelity? I ask because I have to chose which I want when purchasing the unit. I currently use IIR with Room EQ Wizard and run the measurements through Multi Sub Optimizer and plug those results into the 8 bi-quad filters of my Mini DSP. The FIR integration simplifies the process and uses Dirac Live to make the DSP happen. 

I would appreciate anyone’s experience! I don’t mind the extra work, I want what will sound best. Of note the system is all wired with balanced connections while the new Mini DSP model Flex Eight uses RCA. There is another Mini DSP model I could use that has balanced out but it only uses Dirac Live at 32bit/48 kHz. 




Personally I've never found phase correction worth that much.  Getting a glass smooth FR and integrating the bass well with the room is such a huge improvement over no EQ, but the extra steps needed for phase correction never did that much for me.

Yeah, I miss the 8 channel miniDSP myself.

Thanks Erik! The other model they have is the DDRC-88A. Reading through their literature they actually encourage the process I’m using with the 2x8 which is REW into MSO and then they do Dirac Live on top of those results and apparently the results are pretty impressive. I also had the idea of just getting another infrared board and replacing it on my current unit. I think they haven’t changed that. . . I wish they made this same 2x8 / 4x8. The results are awesome. Dirac seems to have a lot of support over many manufacturers, I want to weed through the chaff to find out if it really is great or just good and easy to use. 

I will probably catch some flak for saying this... And... If you are comfortable with REW what is the point of using Dirac too? Create some FIR filters and run them thru a convolution engine like HangLoose Convolver and be done with it...  If your worried about phase try rePhase along with REW...


I would say give HQPlayer a try, best pre processing digital option, this is software, you will need a computer to run the filters, conversions, modulators etc. it does have DSP in the digital domain at any resolution and it includes a convolver where you can feed your room corrections. You can find more info at audiophilestyle.com

I have been using HQPlayer for years and it is a game changed IMO, if you need more info feel free to PM me

I'm not associated nor I am a dealer or work for anything related to audio,

UPDATE 07/30/2023  I met Al Clark at AXPONA and I asked to be an early adopter of the dspNexus. It’s going to function at twice the fidelity of my MiniDSP. I can’t wait to hear the system operating at this level! I got the unit delivered last week and I’m working on building out the system in Audio Weaver software which will process the filters at 192K. You can build FIR or IIR filters with any other filters, delays, sinks, etc., you can imagine. One cool aspect is that I have to have my subs play -25dB to fully integrate with the system which costs fidelity but the dspNexus can do it through the hardware in each channel so there is no loss. I have to remeasure  my room at 192K and see if I can get the reference point correct for the filters to plug into Audio Weaver correctly. This is an adventure!

The dspNexus processes 8 channels and up-samples to 192K or 384K. It also performs jitter attenuation. The dspNexus executes time alignment in 5uS increments handles room correction through active crossovers. The design is modular so if better DACs come out they can be easily upgraded. The crossovers are custom designed in DSP Concepts’ Audio Weaver graphical design tool. Inputs include Asynchronous USB, S/PDIF, Bluetooth, Ethernet, WiFi and has a phantom input for microphone which can interface with a built in RPi4 to run room correction software etc.