• 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!

We don't talk about Dirac enough in this forum.

I agree the Dirac Live software is clunky and unclear. When I first started using it (the webpage was a little different) I couldn't for the life of me figure out WHAT to download. I'm running it on a Mac and host it with SoundSource (which is excellent btw), but there is a bug in DL3 where it won't play sound unless you already have sound playing before you run DL3. So I have a 30min noise file I play in Quicktime, then load DL3 and then it works for running a measurement/calibration. From there its pretty smooth sailing and the plugin works perfectly hosted in SoundSource from there on out.

Software hassle aside, it is an absolute GAME. CHANGER. I want to yell from the rooftops to everyone wasting their time and money on speakers with no correction, use this! I've played around with REW a good bit, demoed Sonarworks and IK ARC3, but none of them work as well as DL. DL just sounds "right". I use the Olive-Tool target curve, load it up, and BAM, perfection. I correct full range as I find it tightens up the image as well.

BullBuchanan, you can correct with a subwoofer. I do this by using an REW RTA with periodic pink noise to roughy dial in the sub level and xover point. Then I run the DL calibration as normal. The subwoofer plays along with each left and right channel and it fixes everything up nicely. You don't have to do the REW RTA step, but that'll take some of the heavy lifting off of DL.

Yea, I just did the EQ manually through REW using the method you described with a moving mic. Granted, maybe you can get slightly better EQ results with Dirac, and they may be doing other things under the hood with timing and phase that you aren't getting with REW, but it's hard to say since they hide what's actually going on. My results were pretty spectacular using that method, and honestly seemed better than my one pass with Dirac - but part of that (maybe a large part) could be that I did my setup with "Subwoofer Plus" instead of just the mains via DL.

I appreciate that someday DL may offer a low-hassle solution for folks that want plug and play, but right now it's not really the case and they're charging a massive premium for it. Considering you get Audyssey XT32 included with your $400 receiver, I'm just not sure that as of today their value prop is that much stronger to warrant all the out of pocket + the hassle.

I hope that they can convert me someday though, because I'd like a global solution not bound to a box.
 
the ridiculous issue about Dirac is that having a license won't guarantee access to new versions. You have to buy a new license to have the new nonsense they release!
 
Stopped playing with rew and rephase after purchasing dirac
 
the ridiculous issue about Dirac is that having a license won't guarantee access to new versions. You have to buy a new license to have the new nonsense they release!

That's the responsibility of the manufacturer, the same thing with new versions of Android on phones.
 
Fully agree about the Dirac UX being less than optimal - my MiniDSP SHD Studio did not want to cooperate with the Dirac application or my computer and kept on throwing random errors (and yes, the reversed orientation got me too). But once it's dialed in, the results are fantastic. I just hope I don't have to calibrate it again for a good while...
 
Hopefully DLBC is released soon for PC. In the meantime, it's possible to setup your crossovers with other software or use the one in your AVR, assuming you're using an AVR.
 
Yeah, nothing is perfect I suppose. I've spent several weeks with REW. Amazing and flipping free, can't exactly complain about anything there, and quite powerful if you take a lot of time learning it.

But, I downloaded the trial PC standalone of DL, spent a half hour setting up and figuring out how to output it, then another 15 doing measurements...and Viola! Results are phenomenal for the effort involved, and while subjective, I think it does sound "right". Very, right. I've been messing with it for a week and it may be a little kludgey but its darn simple. I've had no weird glitches, and its my understanding that version updates are part of the liscense.

I'm seriously considering going with a Minidsp SHD to help further with sub integration and to have DL reside on the device rather than having a laptop open, and selling my analogue preamp to pay for it. It is indeed a game changer.
 
the ridiculous issue about Dirac is that having a license won't guarantee access to new versions. You have to buy a new license to have the new nonsense they release!

I should point out that I started in 2015 with version 1.1 on my PC. Last year, I was able to upgrade to DL 3 for nothing for both my PC's. I'll gladly pay to upgrade to Bass Management if they ever make it available for PC's. My critique with Dirac involves the opacity of their website, and the way they give short shrift to the original customers with regard to providing updates and also support.

The critique is not with the product itself which I consider the absolute gold standard, especially if you don't want to spend half a month bringing yourself up to speed on how to get the most out of it.
 
Do you tend to listen at near-field or far-field?

My couch is 10 feet/3m from the main speakers.

My experience with them (MartinLogan electrostats) is that the listening experience is like nearfield for an extended distance, as the side/floor/ceiling reflection level is quite subdued.

My initial audition in 1994 was in a showroom that was about 20 feet wide and 50 feet long. Even at the far end they had an "in your face" quality that I liked (from my prior experience running an in-your-face PA)
 
My understanding of Dirac is that I either have to run it off a dedicated PC (do not want) or from some licensed device like MiniDSP (do not want, unless I trip over enough cash for SHD or SHD Studio). The PC software won't produce a FIR filter file that I can use with another a convolver of my choice... or will it?

As for REW, I've found it works very well for squashing big bass humps, which itself can be a game changer, but I've never figured out how to use it for anything more sophisticated. It would be great to see a tutorial from someone who uses it for serious room correction.
 
My understanding of Dirac is that I either have to run it off a dedicated PC (do not want) or from some licensed device like MiniDSP (do not want, unless I trip over enough cash for SHD or SHD Studio). The PC software won't produce a FIR filter file that I can use with another a convolver of my choice... or will it?

No it won’t. It also not just a FIR filter. It’s a clever hybrid of FIR and IIR filters. Obviously that’s something that could theoretically be exported out of the software, but for IP reasons they obviously won’t let you do that.
 
My understanding of Dirac is that I either have to run it off a dedicated PC (do not want) or from some licensed device like MiniDSP (do not want, unless I trip over enough cash for SHD or SHD Studio). The PC software won't produce a FIR filter file that I can use with another a convolver of my choice... or will it?

No offense, but if you use neither a PC or a dedicated device. then how are you listening to music?

a tablet connected to a DAC?
 
How many useable taps do you reckon the newer miniDSP SHD has? Couldn't really find anything in their website and forum. My guess is that it's probably hovering around 6144 taps for stereo (48kHz ~64ms delay) like the OpenDRC-DI -- or more likely twice that... any more and the total delay might be very noticeable.

...

Even with an ordinary miniDSP 2x4 or MCH dac without Dirac or other automated room correction software, one should be able to achieve similar or better results -- in both frequency and time domains -- provided one is using a PC and doing manual correction and/or convolution (maybe something like rePhase). The cost, of course, is time figuring out things... as this whole 'optimization' thing can be a rather long and tedious learning/refining process. Then again, to achieve the best possible results from any 'room correction' software or DSP, one really has to optimize everything else first (e.g. room acoustics, listener and speaker placement).
 
Last edited:
No offense, but if you use neither a PC or a dedicated device. then how are you listening to music?

a tablet connected to a DAC?

Headless Linux box in the loft serves music files. RPi4 streams to the DAC in the listening room.
 
How many useable taps do you reckon the newer miniDSP SHD has? Couldn't really find anything in their website and forum. My guess is that it's probably hovering around 6144 taps for stereo (48kHz ~64ms delay) like the OpenDRC-DI -- or more likely twice that... any more and the total delay might be very noticeable.

...

Even with an ordinary miniDSP 2x4 or MCH dac without Dirac or other automated room correction software, one should be able to achieve similar or better results -- in both frequency and time domains -- provided one is using a PC and doing manual correction and/or convolution (maybe something like rePhase). The cost, of course, is time figuring out things... as this whole 'optimization' thing can be a rather long and tedious learning/refining process. Then again, to achieve the best possible results from any 'room correction' software or DSP, one really has to optimize everything else first (e.g. room acoustics, listener and speaker placement).

The SHD has a fixed delay of ~25 ms, so at a 96 kHz sample rate you are looking at an upper bound of ~4800 taps/ch.

The OpenDRC-DI (miniSHARC) uses a 333 MHz SHARC ADSP21369 and the SHD uses a 450 MHz SHARC ADSP21489 so it makes sense that the SHD has more taps available.

All of this does bring up an interesting question, do different Dirac implementations have different capabilities depending on DSP power?

Michael
 
I'd like to know how much boost Dirac is likely to apply - neither my Amp or sub has a lot of spare power.
 
how much boost Dirac is likely to apply
sometimes it can try to apply a lot, like 8-10 dB, depending on circumstances
 
Back
Top Bottom