Sonarworks Reference 4 vs SoundID: Same sound? Same algorithm? I don't think so.

VST, AU, etc. plug-in Virtual Effects discussion
200 posts since 18 Mar, 2007 from London

Post Fri Nov 26, 2021 11:59 am

After upgrading from Sonarworks Reference 4 to Sound ID Reference I imported my room calibration file (I made with Reference 4) into Sound ID. Straight I've heard some difference in the sound. :nutter:
It made me curious, I checked levels etc but it sounded a little bit different. Than I checked the Sonarworks official reference 4 - SoundID comparison chart (see attached pic), but no mention about sound/algorithm improvement. I thought: "Something is not right, what's going on"? :bang:
I decided to make a null test.

Here is what I have done in Bitwig Studio.

1. Copied the same wav file on 2 audio tracks.

2. I verified all settings are the same (panning, levels etc)

3. I put Reference 4 on one track and Sound ID on the other one, with the same calibration file, same levels, exact same settings.

4. I flipped the polarity on the first track, expecting I'll get silence, expecting the two tracks will null.

5. BUT they did not, I was left with a signal that moved between -65 to -40 dB.

6. Than I tried the same test with a sine sweep. This test showed me that the signal difference is bigger in the low frequencies.

7. Than I have done the same test in Reaper, exact same results.

8. Than I tried the same test with my HD800 calibration profile, both in Bitwig and Reaper. Same result, they do not null.

9. I also tried to play with levels, thinking perhaps I can match the two signals closer but no, as soon as I changed levels, the sum signal got louder indicating they were closest to nulling when level matched.

10. Finally I checked whether Reference 4 on both tracks would give me a null, and it did. Than the same perfect null when using SoundID on both tracks.

Could someone please do a similar test to cross check my results?

It would be important because if my test is correct, than room measurement profiles made in Sonarworks Reference 4 are completely useless in Sound ID.

If my test was correct, than everyone who upgrades from Reference 4 to Sound ID has to make a new measurement (a new calibration file) in Sound ID, and no one should import measurements made in earlier versions. :dog:

I just found a similar post on SoundOnSound.
I copy that message here:
"I bought Sound ID reference and compared it Reference 4. I did hear a difference - more clarity, a little less 'niceness' (which is a pity on my system!).
I thought I'd actually test if there is a difference.
So I created a Cubase with two tracks, both containing the same audio, but one with the phase set to 180 degrees.
I checked that they nulled, and that with the Reference 4 plug in on both, they also nulled. They did.
But with Reference 4 on one and Sound ID Reference on the other, with the same settings, they do not null. Depending on the material, the difference signal is in the range -65 to -75 dB. The differences are spread pretty evenly across the audible frequency band.
I've checked the settings religiously, and I think they are the same, and I've also tried with different settings between the two, and the difference signal is much bigger when they are out of line.
Would anyone be interested in doing the same thing to cross check my results?
I used 24/96 audio, linear phase, extended treble but other limits at defaults, listening position set to on, and the same profile."


01 Sonarworks Reference vs Sound ID comparison chart.png
You do not have the required permissions to view the files attached to this post.

User avatar
699 posts since 19 Jan, 2008

Post Sat Nov 27, 2021 9:13 am

Thank you for this information, recalibrated it again. Most likely and I hope this is the reason, they did some improvements on the calibration part.

3111 posts since 2 Oct, 2004

Post Sun Nov 28, 2021 2:15 am

It's best to hit their support channels. But their CEO had previously said nothing has changed on the GS forum.
Orion Platinum, Muzys 2

Return to “Effects”