Looking for Dirac ART and Trinnov users

[umamiaudio]

Would anyone here be kind enough to help compare the performance of my MIMO speaker calibration software to Dirac ART or Trinnov? Mine is fully automatic and also takes your distortion measurements into consideration as well.

I know it probably seems a little bit outrageous and silly to think that a solo developer could have any chance of edging out Dirac or Trinnov, but we all know what happened when David met Goliath.

It currently runs as VST / AU. https://umamiaudio.com

 

[EWL5]

Does your software support infrabass (frequencies below 20 Hz)?

 

[umamiaudio]

Yes, it looks at everything down to DC. We do not simply roll off every below 20 if that is what you're asking. In fact, I think rolling off everything under 20 is a pretty terrible idea and a quick way to ruin time domain response.

 

[kifeep]

Does this apply a separate filter to each node in the support matrix? The below from your website sort of sounds like there's a single filter for each group vs one for every speaker in each group:

Also, FWIW @umamiaudio, since Dirac ART only works on specific hardware, the number of members who also have a VST/AU host to compare the two may be limited. I'm very curious to hear any member's experience with this whether directly compared to ART or not. I don't currently have a way to test it myself, but have been thinking about moving to a MIMO system and considering my (limited) options.

 

[umamiaudio]

I should correct that and make it more clear sorry, it applies per speaker filter for each group, so for example, if you're doing stereo with four subs, that's two groups of 5 speakers per group and 10 filters and delays in total.

You can also join other users on discord! https://discord.gg/fUGUVQDc

 

[kifeep]

Got it, thanks. I have a couple related questions.

From the website:

Perfect Soup optimizes each group as a unit: it manages the crossover between primary and support speakers, aligns their timing, balances their levels, and corrects the combined frequency response, all while accounting for how the room affects each speaker differently.

Does Perfect Soup actually crossover speakers to subs (i.e. with complimentary HPF and LPF) or is more like ART where any available support speaker can help fill in at any frequency it's capable of?

Similarly, do support speakers only provide additive support or are they also used destructively to help tame ringing?

 

[umamiaudio]

The answer is it can do both and depends entirely on the measurements. For example if your speakers have very little low end output then it will act more like crossover. If your speakers are all full range, then it will act more like ART. It also gives weight to the distortion of your speakers as well so if say for example if you have a high performance sub and sub par main speakers, it's going to allow more load to transfer the subwoofer rather than the main speaker. It does all this while considering time domain response as well as magnitude response. The supports are used both additively and destructively to reduce ringing / decay / room modes.

The design philosophy is "no-tradeoffs" , "improvements only" , and "do no harm" -- which is all to say the measurements should either stay the same or improve as you add more speakers to your groups, and there's no "Oh this and that got better but this and that got a bit worse too". If you add a cheap sub a group or place it in a poor spot, it just won't be contributing much, but it will not ruin the sound.

 

[Fidji]

Trinnov user with WF here - how would this work?

 

[G|force]

I'm intrigued.
What is the typical memory burden running AU in Logic Pro on an ARM Mac? Also, where are you located?

 

[umamiaudio]

What is the typical memory burden running AU in Logic Pro on an ARM Mac? Also, where are you located? <=== for a 20 channel system, about ~300mb of ram for the AU instance.

I'm from Thailand, but I'm in Portugal right now.

 

[umamiaudio]

Trinnov user with WF here - how would this work?

Yes it would work. The measurement wizard uses regular ASIO / Core Audio drivers and the calibration loader runs as VST/AU. The AAX is coming soon (or you can download the dev version of PT, then you can run it now). It would be amazing if you could try it and let me know your thoughts compared to WF.

Basically you run the wizard which walks you through the process of taking measurements which gets uploaded to a server where the heavy work is done. The calibration file is then sent back to the user. When you open up your DAW and insert the calibration loader, the calibration file should already be waiting for you in a dropdown.