Can Dirac make things worst ?

[Worth Davis]

OP - I found that all room correction needs good placement and stuff to begin with. Move them around until they are level matched with a ratshack meter is a good starter. Check your wiring/config and make sure nothing else is doing any config like a receiver with levels/distance or some other device.

Try a few different filters - you can save and load at will. I tend to save "full dirac" "dirac to my highest cut that needs to be made" "diarc to 300 or 500" I like the Dirac to 1000 on my system, I definitely do not like full dirac which boosts the heck out of my super high treble, I do have a curve which naturally falls off with the tweeters that sounds great. You can limit the correction by removing orange set points, dragging the right side of the entire graph down to like 500. If it wont go far enough you need to delete a set point in the way.

I dont think dirac will fix a room that is 50db spread across with phase issues on its own...neither will REW. I do like REW to Roon with just cuts, but Dirac is a lot easier to deal with and tweak curves, load them up...and unless I charge myself less than $1/hour - Dirac is a lot cheaper than REW.

 

[Flak]

As RayDunzl correctly pointed out PappyBlue is not measuring what Dirac Live is measuring... he is measuring the sum-response which is not controlled in Dirac Live, and the “issue” here is the cancellation.
Using REW and rePhase as suggested by daftcombo is a possibility but not exactly stupid simple for most users...
the good news are that we are most likely going to address this very issue in the upcoming bass management upgrade for Dirac Live 2.x

 

[Music1969]

the good news are that we are most likely going to address this very issue in the upcoming bass management upgrade for Dirac Live 2.x

Nice. When is the macOS version planned to be released?

 

[arboleda]

I’m new to Dirac. I’ve got a minidsp SHD and have done some measurements. My room is symmetrical but the speakers are on a desk in one corner. I’m pretty sure that phase issues are causing me some issues under 100 hz. The two speakers show different valleys at different areas under 100hz. Therefore I have a phase issue due to placement right? Note that I’m not mixing a sub into this (yet).

Two questions:

1 - phase issues are frequency specific, right? My speakers could be in phase at 50hz, out of phase at a higher frequency, and back in phase at further higher frequencies in terms of the non-direct audio waves right (those bouncing off other parts of the room back to me).

2 - Is Dirac correcting for frequency-specific phase issues?

 

[RayDunzl]

1 - phase issues are frequency specific, right? My speakers could be in phase at 50hz, out of phase at a higher frequency, and back in phase at further higher frequencies right?

Yes.

My left and right go 180 degrees (inverted/cancelled) out of phase at 48Hz at the listening position.

2 - Is Dirac correcting for frequency-specific phase issues?

If you can run a sweep with REW and capture the DAC output you can see what Dirac does to the frequency response and phase.

Without "correction" those would be two flat lines.

 

[arboleda]

Ok, got it on the question of Dirac phase correcting on frequency (yes).

i take it that first graph is with Dirac or other correcting software on, and REW showing the phase degrees at frequency. Is that measured from REW at the output from the DAC? no speakers/mic involved Right? Or is that with speakers/mic?

 

[RayDunzl]

i take it that first graph is with Dirac or other correcting software on, and REW showing the phase degrees at frequency. Is that measured from REW at the output from the DAC? no speakers/mic involved Right? Or is that with speakers/mic?

The first graph is mic at the listening position, left and right speakers,

The second graph above is the electrical output of the DAC, one channel.

REW Sweep -> miniDSP OpenDRC-DI with AcourateDRC IIR and FIR filter data -> DAC -> preamp -> analog cable back to the PC soundboard input... two years ago

 

[arboleda]

Sweet. I’m learning.

 

[TonioRoffo]

Dirac Live 2/3 have an option to disable phase correction at playback. Leaving it turned on produced a very big audio stage but problems pinpointing instruments. Turning it off was way better. This was after doing a careful 9 measurements in focused mode. Might be I did smth wrong.

 

[Flak]

Dirac Live 2/3 have an option to disable phase correction at playback. Leaving it turned on produced a very big audio stage but problems pinpointing instruments. Turning it off was way better. This was after doing a careful 9 measurements in focused mode. Might be I did smth wrong.

I don't know of a possibility of turning off phase correction only... anyhow, please avoid making measurements in too small a space. Even for the “Tightly focused” listening environment, it is important to spread out the microphone positions in a sphere of at least 1 meter of diameter. Too small a space will result in over-compensation that will sound very dry and dull.
The measurement points should have a distance of at least 30 cm (12 inches) between each other.
(by the way, I personally prefer the wider sofa option)

 

[TonioRoffo]

I don't know of a possibility of turning off phase correction only... anyhow, please avoid making measurements in too small a space. Even for the “Tightly focused” listening environment, it is important to spread out the microphone positions in a sphere of at least 1 meter of diameter. Too small a space will result in over-compensation that will sound very dry and dull.
The measurement points should have a distance of at least 30 cm (12 inches) between each other.
(by the way, I personally prefer the wider sofa option)

Sorry, my post was in error. It's the delay compensation which (for me) causes a lot of problems. Can be turned off here:

 

[AndrewDavis]

I know this is an old thread but, is there a consensus on whether to move room furniture like the coffee table before taking the room measurements? Or just leave as is?

 

[Doodski]

I know this is an old thread but, is there a consensus on whether to move room furniture like the coffee table before taking the room measurements? Or just leave as is?

Do what you want. I moved it out of my way completely full time when I had a stereo that was worthy of such tuning the room.

 

[tw99]

I know this is an old thread but, is there a consensus on whether to move room furniture like the coffee table before taking the room measurements? Or just leave as is?

You should have the furniture set up as you plan to use the room.

 

[CTRLM]

I know this is an old thread but, is there a consensus on whether to move room furniture like the coffee table before taking the room measurements? Or just leave as is?

I've been using Dirac for years with a MiniDSP DDRC-22D. I always measured with everything in place including my couch, coffee table to the side etc. I always liked the result but only to a point......very happy with the actual tone, timbre or whatever you want to call it but there was always a "closed-in" type of presentation compared to no correction. I lived with this as the compromise was still an improvement.

Recently and after reading something from Mitchco on emptying the room to avoid comb filtering, I tried moving everything out of the room forward of the speakers, that included the couch, other chairs, tables etc. even if they were not directly in between the speakers and the listening position. I used the middle measurement setting - the one up from chair in Dirac 3, which I think is called focused sofa? Anyway it's the one with 13 measurement points. I was very careful to get the the first microphone point exactly where my head would be then followed the other points without the same level of attention ie. not measuring between them, as I have read (and we all know the internet doesn't lie) that the remaining points are not so critical. As is the usual case in this hobby, the result sounded pretty good at first but it soon became apparent that it wasn't - dry and very little depth to the sound, any sort of bass/mid bass extension was missing.

Something wasn't right so a couple of weeks later I did it all again despite the fact that my back was already hurting at the thought of it, but this time I devoted more attention to the measuring points. Using a tape measure, a laser pointer and blue tape on the floor I pedantically made sure there was 450mm (18 inches) between each measurement, which probably somewhat mimicked the wide sofa setting. The result was well worth it. No more ultra dry bass and the presentation or sound stage was much improved over my initial experience with Dirac.

I guess Dirac or any room correction software can only be as good as the quality of the data it has to work with, or as we say down here...sh*t in/sh*t out.

 

[Worth Davis]

I remove the back of my main chair for Dirac and Audyssey - mostly voodoo, I dont have a clear science reason for it

 

[Chromatischism]

I remove the back of my main chair for Dirac and Audyssey - mostly voodoo, I dont have a clear science reason for it

And you also listen without a back and sit up and cross your legs? Not my preferred way to listen, but we're all different.

 

[Coach_Kaarlo]

Again subjective comment disclaimer applies. Although I did make many many measurements to arrive at this opinion.

After 6 months of experimenting and measuring and listening I noticed a few trends. The ideal vertical distance between measurement points seems to be a function of ceiling height - meaning unless you have very high ceilings 250-400mm works best. The horizontal distance between measurement points is really important, nothing less than 500mm for some sense of spaciousness when listening, I have tried 500mm upto 1000mm. The focused setting in DIRAC and corresponding measurement grid tended to give me a more headphonic listening experience (meaning less image and less spaciousness - less of the room the recording was made in). Using the full DIRAC Wide Lounge 19 measurement points gave the best concert hall or outdoor rock arena type sound. Lots of space and very good positioning of instruments within the soundscape. The real live feeling / sound.

I also remove furniture from the listening area, but leave the edge of room furniture as the boundary effects seem to matter. However anything around the microphone area is clear to allow the reflections to be measured well.

The system is sensitive, I have often spent an hour measuring with DIRAC, and after finishing discovered the result is crap. Playing my test tracks and taking some REW measurements shows a poor frequency response and or the sound is crap. After another set of measurements at identical points I often get a better frequency response and far better listening experience. Background noise, in-room interference, software anomalies - who knows.

I must say once a decent set of DIRAC measurements have generated a good correction, the result is very very good - well worth the effort.

 

[Chromatischism]

I also remove furniture from the listening area

Only for the measurements?

 

[Coach_Kaarlo]

Only for the measurements?

indeed.