• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Dirac for nearfield 2.0?

MarcR

Member
Forum Donor
Joined
Apr 12, 2021
Messages
10
Likes
11
I just acquired a pair of Kali IN-8v2 and a miniDSP Flex TRS w/ Dirac. The physical setup is pretty good - small carpeted office with speakers on audio isolation foam pads toed in from the corners of a 60x34" desk centered on a wall.

I'm new to nearfield and am blown away by the out-of-the-boxes sound quality.

I wonder though if I'm missing out on what Dirac can do. It's my first time using it. I configured it for a "tightly focused" setup with nine measurements, designed a Harman-like target curve, and loaded it to the Flex. When I A/B Dirac Live OFF/ON, I hear no difference.

I noticed that the Flex preset with the Dirac filters has a 0.02 ms delay for Dirac 1 whether the Dirac Live button is ON or OFF, so I also A/B'd between presets. The several seconds of dead air while the preset switches make judgments difficult, but any possible difference there is subtle.

So I have a couple of questions:

1. Does Dirac Live generally offer much to nearfield 2.0 setups with close-to-optimal physical configuration? Maybe it's just not necessary?
2. When I load filters from the Dirac Live app to a Flex preset, does it change the sound of the preset even when Dirac Live is OFF?
 

usersky

Active Member
Forum Donor
Joined
Jan 20, 2020
Messages
183
Likes
261
Hi! I use Dirac with Minidsp Flex in my office in semi near field (1.2m between ears and ELAC DBR6.2). For my setup it is truly beneficial. And for the second question: turn Dirac off and it it won't affect the sound at all. You can play however you want with the 4 presets (I use 2 of them for 2 positions in my office, chair and sofa and the other two with Dirac off, with eq for Senn HD600 and Hifiman HE4xx).
 

DjBonoBobo

Addicted to Fun and Learning
Forum Donor
Joined
Jan 21, 2020
Messages
702
Likes
1,276
Location
any germ
If it is a small room, I would be surprised if there would be not even a boomy room mode in bass anywhere without Dirac.
You could test if everything works correctly by trying a new filter with a very "crazy" target curve that could be identified clearly if applied.
 
OP
MarcR

MarcR

Member
Forum Donor
Joined
Apr 12, 2021
Messages
10
Likes
11
Thanks for the feedback and advice. I remeasured with more care regarding mic placement. I also accepted the default target curve instead of building a custom one. This time I do hear a difference between OFF and ON. Even better, I prefer ON. :)

The speakers both have a deep and wide low-frequency dip from about +3 dB at 45 Hz to -15 dB at 55 Hz and then back up to +7 dB at 103 Hz. The correction limits the double dip from +0.5 db at 45 Hz to -4 db at 55 Hz. The frequency response curves of the two speakers cross and thereby support one another as they climb back up to 0 dB at 75 Hz. Looks better, sounds better.

Should I expect to be able to improve the LF response curve with a single well-placed subwoofer crossed at 80 Hz, or will I need two to improve on Dirac's management of the Kali IN-8 pair's combined LF curve?

The Dirac 1 input has a 0.04 ms delay, and the Dirac 2 input has a -0.3 dB gain adjustment. I was careful to position the UMIK-2 precisely and am surprised by Dirac's adjustments. Does anyone know (or care to speculate about) what's up with that?
 
OP
MarcR

MarcR

Member
Forum Donor
Joined
Apr 12, 2021
Messages
10
Likes
11
I'm satisfied with my Dirac Live configuration and overall setup now. I don't think a subwoofer (or two!) will improve this office desk setup enough to justify the expense, space, and complexity.

A key insight was to move the default control point target curve down 5 dB to give Dirac room to correct more fully the giant null in the bass response at my listening position - the before and after pictures are below. I used trial and error to settle on a 5 dB reduction - less leaves improvement on the table, and more does no good.

I wish there were an instantaneous gain-matched way to compare Dirac ON vs OFF with a gain-shifted target curve, but I see no way to do that. To make comparisons with the other Flex presets easier, I added 5 dB to the two output channels to compensate for the shift down in the target curve.

Related, while switching presets back and forth, I discovered a Flex bug - when I switch from one preset to another while Dirac is ON, it appears to be ON in the DDR-24 app and on the Flex display, but it is actually OFF. I have to cycle it OFF and then back ON to enable it. The -5 dB target curve shift makes this really obvious.

The Dirac ON/OFF comparisons for the default curves seem to be gain matched - I tested with a sound meter app - and I do prefer ON in sighted comparison. With the preset switch delay and the need to turn Dirac OFF and then ON, I'm not able to discern a difference between the default and gain-shifted target curves - they both sound awesome to me. I'm just relying on Dirac's before/after pictures in preferring the gain-shifted curve.

I also noticed a "Disable delay/gain" menu item in the Dirac Live app. Choosing that and overwriting the Flex preset allowed me to ditch the irritating small delay and gain that I presume is due to measurement error rather than anything real.

Dirac default sloped target curve.png
Dirac default sloped target curve -5 dB.png
 
Top Bottom