Experiences with Sonarworks Reference 4 Studio

[RichardT]

Sonarworks had some good Black Friday reductions so I took my chance and upgraded from Reference 4 ‘headphone’ edition to ‘studio edition with measurement microphone’.

For those that aren’t familiar with it, this product lets you measure your listening environment and construct a frequency curve that can be applied to your monitors to flatten their response. Martin strongly recommended it.

I had some interesting findings. I ran the process twice, because first time I hadn’t switched out the existing frequency adjustments in my KEF LS50Ws. The results were remarkably consistent between the two runs, taking into account the clear effects of the shelving introduced by the KEFs in the first run.

I had some significant troughs and peaks (6dB) in the bass region, and big discrepancies between the L and R channels resulting from one speaker being much closer to a wall (unavoidable at the moment). Fixing this has made a big difference to low frequency evenness, clarity and imaging, and this reaches right up into the midrange.

On the other hand, I had a very even tilt downwards in the higher frequencies, reaching around 3-4dB at 10k and above. This was a bit of a shock! When I applied the adjustment, everything sounded very bright, too bright for comfort.

I decided to partially cancel out the HF adjustment using the tilt settings inside the Sonarworks software, and I’m getting used to the effect now. I may switch out the tilt in the future as I get used to things more.

Do other people monitor completely flat or do you prefer a slight HF cut?

[manwilde]

I´ve been using it at home for the past month or so, and for over a year at the project studio. It never made sense to me to tilt the frequency response, as it seems to counteract what the calibration is supposed to fix, to a point. But then, if you get used to the way it sounds "tilted" and your mixes translate well, why not?. I mean, as long as you know what you´re doing...

[Martin Walker]

Hi Rich!

Glad to hear that you're already revealing better stereo imaging using Sonarworks Reference 4. Like you I already have acoustic treatment, and the measurements my biggest 'before' excursions are no more than 6dB, and mostly significantly less than that - here's my 'before' frequency response:

Judging by all the comments I've read concerning Sonarworks Reference 4 on forums around the world, improving from a starting point of perhaps +/-6dB has a mostly successful outcome, whereas those who try it with no acoustic treatment and who are therefore attempting to correct significantly larger peaks and troughs find that the software starts struggling to remedy those limitations (hardly surprising really).

I've accepted the corrected response in my tiny studio 'as is', and mix to that, and both frequency response and particularly imaging are significantly better.

The only thing I can think of that might result in exaggerated high frequencies is if you're not using one of Sonarworks' individually calibrated mics - it can be frightening to see just how much some cheap measurement mics vary in their frequency response, particularly above a few kilohertz.

On the other hand, unless all your reference tracks sound overly bright as well as your own creations, it might be worth persevering with the flat Sonarworks setting - you'll mix with less top end as a result and might like what you hear elsewhere - as manwilde says, it's all about translation.

Martin

[PeterD9988]

Nope - don’t get it. I’ve calibrated my room with Adam 7X monitors, Cubase 10.5 and have the resulting Sonarworks Ref 4 EQ as the last plugin on my master bus.
But why am I to disable/ bypass that EQ when rendering ? What was the point of all of that if not rendering with the EQ included? Confused.

[muzines]

Because you are using the software to compensate for what you *hear* in your room, so you can make effective mix decisions to make a mix that translates properly.

You don't want to print that compensation into the mix, unless your audience all intend to listen to your song in your room, too.

[The Elf]

But then, if you get used to the way it sounds "tilted" and your mixes translate well, why not?

Because when you work in another studio that doesn't have it you're stuck.

[Aled Hughes]

Nope - don’t get it. I’ve calibrated my room with Adam 7X monitors, Cubase 10.5 and have the resulting Sonarworks Ref 4 EQ as the last plugin on my master bus.
But why am I to disable/ bypass that EQ when rendering ? What was the point of all of that if not rendering with the EQ included? Confused.

As Desmond says, the plugin is intended to improve the listening environment specifically in your room, so that you can hear things more accurately. You don’t want that EQ as part of your actual songs!

Think of it as part of your monitoring, not a part of your mixing.

[Hugh Robjohns]

But why am I to disable/ bypass that EQ when rendering ? What was the point of all of that if not rendering with the EQ included? Confused.

The EQ is to sort out the speakers > room > ears part of the signal chain -- and that is a unique correction for listening only in your room. When listening in a corrected space, you'll make more acuraye and consistent mix and processing decisions, so your finished tracks should translate more reliably for others listening on different sustems elsewhere.

But the correction for your room obviously doesn't apply to anyone else, listening anywhere else.... Which is why, if the plugin is in your master bus, it needs to be bypassed before rendering.

A better option, if your DAW allows it, is to place the room correction plugin into a separate monitoring output path only, and that has two benefits:

1. You won't screw up the render if you forget to bypass the plugin
2. The output master metering won't be erroneously showing the room corrected signal peaks!

[RichardT]

Nope - don’t get it. I’ve calibrated my room with Adam 7X monitors, Cubase 10.5 and have the resulting Sonarworks Ref 4 EQ as the last plugin on my master bus.
But why am I to disable/ bypass that EQ when rendering ? What was the point of all of that if not rendering with the EQ included? Confused.

Because that EQ is compensating for your room and speakers. Other rooms and speakers are different and so it’s best to export without any room compensation at all.

[The Elf]

Nope - don’t get it. I’ve calibrated my room with Adam 7X monitors, Cubase 10.5 and have the resulting Sonarworks Ref 4 EQ as the last plugin on my master bus.

Wrong place. Switch on Cubase's 'Control Room' and put the plug-in in the insert slots there.

This way you will never need to bypass it.

[muzines]

Wrong place. Switch on Cubase's 'Control Room' and put the plug-in in the insert slots there.

That's a nice feature - does that let you put plugins basically *after* the mix render bus?
This is a bit of faff in Logic to implement.

[The Elf]

Wrong place. Switch on Cubase's 'Control Room' and put the plug-in in the insert slots there.

That's a nice feature - does that let you put plugins basically *after* the mix render bus?

Yep.

It's one of Cubase's best features and it dismays me that I see it switched off for the majority of Cubase users I help. Even after I've shown them the joys of Control Room I invariably find it switched off again the next time I'm there.

[manwilde]

But then, if you get used to the way it sounds "tilted" and your mixes translate well, why not?

Because when you work in another studio that doesn't have it you're stuck.

But that applies whether you're using Sonarworks or not, ain't it?. If you mix in a different room, the sound is gonna be different...

[The Elf]

But then, if you get used to the way it sounds "tilted" and your mixes translate well, why not?

Because when you work in another studio that doesn't have it you're stuck.

But that applies whether you're using Sonarworks or not, ain't it?. If you mix in a different room, the sound is gonna be different...

Not if you take your headphones with you, and you are used to how they sound without Sonarworks to prop them up.

[Martin Walker]

Wrong place. Switch on Cubase's 'Control Room' and put the plug-in in the insert slots there.

That's a nice feature - does that let you put plugins basically *after* the mix render bus?
This is a bit of faff in Logic to implement.

Reaper has a very similar feature named 'Monitoring FX', which is where I place Sonarworks Reference 4 and ADPTR MetricAB while evaluating mixes.

Martin

[manwilde]

Not if you take your headphones with you, and you are used to how they sound without Sonarworks to prop them up.

Oh, well, I hadn´t thought of that scenario. Fair enough!.

[Hugh Robjohns]

If you mix in a different room, the sound is gonna be different...

And that's why professional rooms have properly designed acoustics to ensure the flattest response possible, and the best consistency between different rooms using different monitors... A level of consistency and accuracy which can be approximated well by using sonarworks in Bedroom /project studios like yours and mine and many others.

[Sam Spoons]

The purpose of Sonarworks, as I understand it, is to make your speakers, in your room, sound as near to flat as possible, if you go to a different room with different speakers also corrected with Sonarworks they should also sound as flat as possible in that room. If both rooms are sufficiently well treated you should be able to take a mix from one to another and not experience significant differences in balance. Using the slope feature to change your speakers in your room to something other than "as flat as possible" means that will no longer be the case?

[The Elf]

The purpose of Sonarworks, as I understand it, is to make your speakers, in your room, sound as near to flat as possible...

Speakers and headphones.

[Sam Spoons]

[PeterD9988]

Ok, that seems to make sense - SW Ref4 EQ is compensating and making the necessary corrections for my room/ monitors.
So I render my mix with SW Ref4 EQ bypassed, but the next person listening to my music hears the 'uncorrected' version. Since I cant obviously generate an EQ for every living person to correct their systems, I cannot predict what they will hear surely.
It seems a better approach mould be for SW to generate an EQ which should be included in the render and would improve the sound for the average listener.

[PeterD9988]

Also another point: to conserve memory due to large orchestral templates in Cubase 10.5 I freeze tracks periodically (render in place). So even if I were to bypass the SW Ref 4 EQ on the master bus when exporting, the correction EQ curve is already printed on previously frozen tracks. Do these tracks need to be 're-rendered' individually? getting v. messy.

[RichardT]

Also another point: to conserve memory due to large orchestral templates in Cubase 10.5 I freeze tracks periodically (render in place). So even if I were to bypass the SW Ref 4 EQ on the master bus when exporting, the correction EQ curve is already printed on previously frozen tracks. Do these tracks need to be 're-rendered' individually? getting v. messy.

If you rendered them with full master bus processing, then yes.

[RichardT]

Ok, that seems to make sense - SW Ref4 EQ is compensating and making the necessary corrections for my room/ monitors.
So I render my mix with SW Ref4 EQ bypassed, but the next person listening to my music hears the 'uncorrected' version. Since I cant obviously generate an EQ for every living person to correct their systems, I cannot predict what they will hear surely.
It seems a better approach mould be for SW to generate an EQ which should be included in the render and would improve the sound for the average listener.

There isn’t such an EQ setting - you’d make things worse for most people, not better!

[Hugh Robjohns]

Ok, that seems to make sense - SW Ref4 EQ is compensating and making the necessary corrections for my room/ monitors.

Correct.

So I render my mix with SW Ref4 EQ bypassed, but the next person listening to my music hears the 'uncorrected' version.

It's not 'uncorrected', it's a 'flat' mix that will sound perfect in any room (or car or headphones) that has a reasonably flat response too.

Since I cant obviously generate an EQ for every living person to correct their systems, I cannot predict what they will hear surely.

Precisely. So the best we can do is assume that, as serious listeners, they will have taken the trouble to buy decent gear and, ideally, to have done whatever they could to get the flattest response from it.

It seems a better approach would be for SW to generate an EQ which should be included in the render and would improve the sound for the average listener.

But what is 'the average listener'? You can't Pre-EQ for every different headphone, or earphone, or every hifi speaker, or every car, or every pocket radio, or every phone, or every tablet.

Instead, we trust the manufacturers to get as close as they can to the flattest possible response, and we trust serious listeners to take whatever actions they can to make their rooms as flat as possible.mand we make our mixes sound good when heard on flat monitoring systems.