Comparison of DRCs: Dirac Live for Studio, IK Multimedia ARC System 3 and Sonarworks Reference 4 Studio edition

[vavan]

Been using manual peq/convolution based on Rew generated filters for years (see one of results - https://www.audiosciencereview.com/...l-f208-tower-speaker-review.13192/post-401549) but couldn't resist discounted dirac offer few weeks ago and though wouldn't call it night and day difference in general I like it

 

[wjc]

Hello everyone,

Excellent work dominikz! My question concerns the miniDSP DDRC-24 that I have recently purchased. Would it be better if I connect my miniDSP between the preamp and amp? The correction filter will then be applied after the DAC where there will be less degradation to the signal. At the output stage of the miniDSP, I have also imported my EQ curve from REW. This will help flatten some of the peaks in the room.

Cheers.

 

[Snarfie]

Great comparison i was waiting for. Would be interesting to see how Mathaudio Room EQ would compare as a no cost freeware solution.

 

[dominikz]

Thousand thanks and thump up share details and hard work, first i comment subjective feelings resulted from listening to your shared FLAC files, on Dirac FLAC as natural as possible transient for symbols is killed there probably because phase goes north (delayed) as frequency increase and feel that is strange and the wrong way because any tranducers sadly have inductance L(e) and that parameter also delay or distort time domain signal as frequence goes up plus probably also most DACs have excess phase up there, think for ARC3 and Reference 4 that correction fall apart as track material goes complex by the end, so all in all for best natural sounds i really like hear your room how weird that opinion can sound ..

Comment on objective part is because of Amir's detailed acostic reviews and data sharing thousand thanks, then have done some observations that never thought of before and one is below say 1000Hz we mostly have full space coverage that means room sound power will easy be dominated by that full space area down there so if there is any hot peaks in that area that push sound power curve higher relative to ideal and will probably dominate system sound because its full space area verse half space area say above 1000Hz where a certain directivity pattern set in.

Animation bekow is what i mean, in right side graph have marked half space curve (90º) black colour so as to help see the better ballance of sound power, if you like to try that correction its two PEQ's -1,0dB/Q2,81/@94,5Hz & -3,7dB/Q1,23/@128Hz..

View attachment 100747

A funny observation talking for you should try above simple correction is if we allign your Dirac/ARC3/Reference4 correction curves with above simple correction there is coherence .. (Dirac Live=blue, ARC3=green, Reference4=red, above two PEQ's=black)
View attachment 100745

Should the two above PEQ's work good have added four more PEQ's in below string that target smooth power response minus XO region directivity..

View attachment 100750

Animation of modeled prediction for the 6 times PEQ..

View attachment 100753

Thanks a lot! Didn't really have a lot of time to compare, but I did a quick listen and basic in-room measurement of your proposed EQ today.
Definitely the 1st and 2nd band PEQs reduce the boominess that otherwise happens at louder SPLs - so great job! I would need to spend a bit more time listening and comparing to be able to make any judgement, though.

Here's an in room measurement comparison between your 6-band EQ version, Dirac Live correction and no EQ - done at only one position (MLP), average of both speaker responses for each:

Obviously this PEQ correction doesn't fill in the dips, but it does reduce the peaks similar to Dirac. Great work!

 

[Jazmanaut]

Thank you for comparisation. This is on line with my tests from few years ago. My conclusion was, that ARC and Dirac are very very close sounding. I liked ARC a hair better tho. This was V2. tho. I like V3 even more. And i do like to correct full spectrum in my monitoring setup.

 

[MediumRare]

Great comparison i was waiting for. Would be interesting to see how Mathaudio Room EQ would compare as a no cost freeware solution.

Is there a free version? I see there's a download, but I thought to use it past a trial period you need to pay. No?

 

[Snarfie]

Is there a free version? I see there's a download, but I thought to use it past a trial period you need to pay. No?

Incombination with foobar2000 it is free .
As systemwide solution it is around 100,- usd. Mathaudio.com Foobar is my main audio media player solution.

 

[dominikz]

Hello everyone,

Excellent work dominikz! My question concerns the miniDSP DDRC-24 that I have recently purchased. Would it be better if I connect my miniDSP between the preamp and amp? The correction filter will then be applied after the DAC where there will be less degradation to the signal. At the output stage of the miniDSP, I have also imported my EQ curve from REW. This will help flatten some of the peaks in the room.

Cheers.

Unfortunately I don't own any of the miniDSP devices so maybe I'm not best placed to answer this question.
That being said, with the DDRC-24 I would probably first try use its digital input to avoid doing unnecessary DA/AD conversation before hitting the DAC.

The setup in that case would be:
source digital out (USB, coax or optical) => digital in - [miniDSP DDRC-24] - analog out => analog in of external power amplifier => speakers

So effectively if you did that you would be using DDRC as your DAC.
Note that your system's SINAD will anyway be limited by its weakest link, and unfortunately DDRC-24 doesn't provide any digital output.

Hope it helps!

 

[PeteL]

Thanks for the writeup.
For Sonarworks, where you using zero latency, mixed or linear phase filters?

 

[dominikz]

Thanks for the writeup.
For Sonarworks, where you using zero latency, mixed or linear phase filters?

I believe I used mixed phase most of the time - though I tried all three options.

Frequency response differences in magnitude between them are not large - but they do differ in phase:

EDIT: These are soundcard direct loopback tests with Sonarworks Reference 4 systemwide correction active, with 100% wet setting.

 

[PeteL]

I believe I used mixed phase most of the time - though I tried all three options.

Frequency response differences in magnitude between them are not large - but they do differ in phase:
View attachment 101042
View attachment 101043View attachment 101044
EDIT: These are soundcard direct loopback tests with Sonarworks Reference 4 systemwide correction active, with 100% wet setting.

Is it correct that it is the only one to offer true linear phase FIR filters out of the 3 options?
I have to say I don't have yet dove into DRC for my home system but I'm a bit more familiar with Sonarworks. Interestingly they are more rooted in the professional music industry, it's been in recording and mastering studios for a while now, where Dirac was more in the market of home theater from the start. I have heard some quite good result with Sonarworks, but haven't been been able to do a direct comparison like you did, so hard to know what's going on, it did sound good to me in the environments that I experienced it. I was more naturally leaning to it as my choice but i'll dig a bit more now, + a hardware solution might be more appropriate for me in my use case. Recording Studios normally have already treated room tough (often too much). Sonar works where I heard it was for relatively minor corrections.

 

[dominikz]

Is it correct that it is the only one to offer true linear phase FIR filters out of the 3 options?
I have to say I don't have yet dove into DRC for my home system but I'm a bit more familiar with Sonarworks. Interestingly they are more rooted in the professional music industry, it's been in recording and mastering studios for a while now, where Dirac was more in the market of home theater from the start. I have heard some quite good result with Sonarworks, but haven't been been able to do a direct comparison like you did, so hard to know what's going on, it did sound good to me in the environments that I experienced it. I was more naturally leaning to it as my choice but i'll dig a bit more now, + a hardware solution might be more appropriate for me in my use case. Recording Studios normally have already treated room tough (often too much). Sonar works where I heard it was for relatively minor corrections.

ARC3 also has linear phase option - you can see an example of that in the first post.

I can see Reference 4 being used in studios though - it works a lot better in nearfield than it does for farfield IMHO (see my post on nearfield use on previous page) and in addition treated rooms will have smother bass response so its somewhat blunt filters might not be a problem.

Suggest to give the trial a go and see how you like it (if you have some measurement mic)

 

[trl]

Many thanks for this great review and comparison!

• IK Multimedia ARC System 3 - €99.99 SW only or €182.99 with mic

I decided to give it a try to the ARC 3, but then I realised that the prices from above are for the "crossgrade" version, while the prices for new joiners to the IK world are: €237.99 €178.49 software only (on sale right now) or €304.99 mic included. I stopped any purchase for now and I'll see about this in the future, perhaps.

Based on IK FAQ: "A Crossgrade is simply a discounted price for users who own another qualifying IK product already. Its not an upgrade, as it does not require a previous version of the same earlier product version, as normally one would find with an upgrade offer. To qualify for a Crossgrade to another product, you must be registered with a paid version of another IK Multimedia product over $/€ 99 or higher. This allows for any IK user to expand into our wide range of music creation tools".

 

[dominikz]

Many thanks for this great review and comparison!


I decided to give it a try to the ARC 3, but then I realised that the prices from above are for the "crossgrade" version, while the prices for new joiners to the IK world are: €237.99 €178.49 software only (on sale right now) or €304.99 mic included. I stopped any purchase for now and I'll see about this in the future, perhaps.

Based on IK FAQ: "A Crossgrade is simply a discounted price for users who own another qualifying IK product already. Its not an upgrade, as it does not require a previous version of the same earlier product version, as normally one would find with an upgrade offer. To qualify for a Crossgrade to another product, you must be registered with a paid version of another IK Multimedia product over $/€ 99 or higher. This allows for any IK user to expand into our wide range of music creation tools".

Thanks, I missed that! I've now updated the original post with both types of ARC3 pricing). FYI they also offer a 7-day full-featured free trial if you just want to give it a try.

 

[trl]

That was what I initially meant, to download and try it out, but when I saw the price I said to myself: "What if I will like this software?"

I simply don't think I should spend right now such amount of money on a software, but I'll think again on this next year for sure.

Thank you!

 

[Snarfie]

That was what I initially meant, to download and try it out, but when I saw the price I said to myself: "What if I will like this software?"

I simply don't think I should spend right now such amount of money on a software, but I'll think again on this next year for sure.

Thank you!

Why not download for free mathaudio room eq for free in combinatie with foobar2000 i realy have excelent results.

 

[tecnogadget]

mitchco Very nice write up on the subject!

I've been following your articles about Audiolense and Acourate and I'm planning to add DRC to my system in a near future. The problem is that besides being a stereo system for music, it is also multichannel for movies and I am tired of waiting for Dirac to make its integration simpler (multiple delays and beta testers programs, the launch of Bass Control on limited and expensive hardware, lack of integration with the multichannel option for PC, etc), and of MiniDSP to re-launch a device capable of manipulating multichannel signals digitally (many of us don't need extra DAC's in our chain).

As far as I know, Audiolense and Acourate depend on a PC to process the signal. I would like for a matter of simplicity and compatibility to continue playing Blu-Ray discs through a dedicated player, therefore the audio output options are HDMI and S/PDIF.
So the first limitation I see is to go from HDMI to SPDIF for simplicity, and on the way, we lose lossless tracks like DTS HD MA & Dolby True HD and we are left with standard DTS & Dolby Digital in 5.1 like DVD audio (I don't see it as a bad price to pay in exchange for the benefit of DRC)

I would like to know if you know any hardware way to make the PC receive the digital stream from the player (maybe the easiest way is in LPCM) through S/PDIF, process the signal with Audiolense/Acourate, and then send it through S/PDIF again to an AVR? I imagine that on the way we will run into DRM problems...but maybe there is some kind of extractor or sound interface similar to those used by recording studios, I don't know.

 

[mitchco]

@tecnogadget Thanks. There are many approaches to this, including what you describe. But It's complicated and a worthy topic for another thread. There might already be one on ASR... Here is one "hybrid" way: https://yabb.jriver.com/interact/index.php/topic,127712.msg886045.html#msg886045

 

[turbotuff]

Thanks for the comparison. I personally went with Sonarworks for its seamless integration with its "systemwide" feature. I have been very pleased since day one with it. The difference is clear, music sounds so muffled with it off! I can't go back!

 

[Robbo99999]

As promised, here are my thoughts on using these DRCs in a nearfield setting.

General comments
My work desk / nearfield setup is arranged around JBL LSR305 (so older MKI) powered monitors connected to RME Babyface silver edition sound card. Speakers are spread ~70cm, close to the back wall and not set symmetrically within the room (which is also untreated). At the listening spot I have a really nasty suckout of energy in the ~60-~90Hz range and a huge peak at ~125Hz.

A note on target curves - in the nearfield I find I'm OK with a relatively flat target curve with less bass boost (I measured with flat bass, but while listening I still preferred 2-3dB boost below 100Hz). This is in contrast to the typical Harmanesque target curves with ~6dB bass boost and constant downward slope I prefer in our living room (which could be considered far field with speakers at around 2,2m distance from the listening position). This I assume is because in the nearfield much more of the perceived sound energy comes from loudspeaker's on-axis radiation.
In any case, flatter target curves play nicer with most of these DRCs so I found in general that fullrange correction worked better in the nearfield for me than it did in the farfield.

Here's the regular 3-band PEQ I use normally in the soundcard's DSP (based on some old REW measurements):
View attachment 100722

Dirac Live for Studio full-range response used for the measurements:
View attachment 100719

IK Multimedia ARC System 3 full-range response used for the measurements:
View attachment 100720

Sonarworks Reference 4 Studio edition full-range response used for the measurements:
View attachment 100721

In-room measurements
View attachment 100735
Note: variable smoothing shown.
I'd say we see similar results as before - Dirac and ARC3 seem close, 3-band PEQ is not too far-off either , and Reference4 is a bit less sharp so doesn't fully address the resonances.
View attachment 100725
This time I'm showing step response (thanks @mitchco for the suggestion!). We see that Dirac is doing its time-domain magic here, Reference4 exhibits some pre-ringing and the rest don't seem to care much about the time-domain

Subjective thoughts and summary
The order of preference for me still hasn't changed between the three (Dirac > ARC3 > Reference4), but I noticed a few things:

• Reference 4 gave quite solid results in the nearfield - perhaps because the target curves it provides work better with on-axis / nearfield responses. Could be confirmation bias from seeing the measurements, but I thought bass wasn't as smooth as the others - all in all still very workable and much better than either 3-band PEQ or listening without correction (which many times sounds like someone blowing in a bottle really loudly )
• I thought ARC3 and Dirac full-range correction sounded nice and non destructive
• Limiting the correction of Dirac and ARC3 to below ~700Hz also sounded great, but I might even prefer the full-range correction. I'd need to do more listening to decide.
• The 3-band PEQ in comparison to all three DRCs sounds hollow - for sure more than 3 filter bands are required to get comparable results - but that is a task for another day (or another person )

That is a massive suck out between 50-100Hz, I don't think that can really be cured by any EQ....I think the solution would involve room sound treatments or rearranging speaker & listening position in the room.