What's new

Should I trust Sonarworks?

JT

Senior Member
I got Sonarworks Reference last month. I'm using it with Senn HD600. Everything is working out except some high frequencies, around 2.5 Khz. I have a Glockenspiel line that sounds perfect in the cans with Sonarworks. It's quiet and delicate, perfect for this spot. But when I A/B this with my Neumann Monitors, this Nell part is barely audible.

In the past, I've always trusted my monitors. But Sonarworks is giving me something different.

Should I trust Sonarworks and keep my levels as is, or trust my monitors?
 
Is the balances fine with your headphones with sonarworks off? Senn hd600 is a good pair of cans and I'd mix with those any day, so in terms of fine balances I'd trust the cans. But for energy and overall presentation the monitor mix should feel correct aside from small balance changes. If they're not comparable and translating you probably have 2 possible things happening. Either your mix isn't clean yet, there's resonances and things that are masking on other systems that are not apparent in your corrected headphone monitoring, or your monitorin setup is not accurate enough and you're getting things like standing waves and other room artefacts that are mucking up your balances majorly. Take it to few other places and try it there, try earbuds for example. You should be able to tell if the mix is actually alright or needs more work after trying it in few different systems.
 
I agree with the importance of calibrating the monitors. I trust mine implicitly post-correction. Headphones are always tricky and I tend to use mine uncorrected. 600s are my choice, too, and I never found the Sonarworks calibration on them to be particularly helpful in my situation. But I can't recommend the speaker calibration (properly done) highly enough.

As always, YMMV.
 
I don't understand the logic behind sonarworks and headphones. To some degree it can work if the headphones are clearly messed up, but on good headphones, it seems to make the frequency response even more unbalanced past 20% or so.

For example on HD800S (tried it with them), the dips and bumps are there for a reason because headphones need them to feel flat. Of course some frequency response bumps are technically undesirable but it seems that sonarworks is trying to simply apply an opposite curve as if they were speakers. Past 20% it resulted in obvious weird resonances (actual ringing in the mids) that I didn't hear on my studio monitors or with sonarworks off.

I'm not sure what technique they are using to come up with these target curves but it just doesn't sound right. 100% is an absolute mess, 10-20% can sometimes sound better depending on the headphone model from what I heard.
 
I don't understand the logic behind sonarworks and headphones. To some degree it can work if the headphones are clearly messed up, but on good headphones, it seems to make the frequency response even more unbalanced past 20% or so.

For example on HD800S (tried it with them), the dips and bumps are there for a reason because headphones need them to feel flat. Of course some frequency response bumps are technically undesirable but it seems that sonarworks is trying to simply apply an opposite curve as if they were speakers. Past 20% it resulted in obvious weird resonances (actual ringing in the mids) that I didn't hear on my studio monitors or with sonarworks off.

I'm not sure what technique they are using to come up with these target curves but it just doesn't sound right. 100% is an absolute mess, 10-20% can sometimes sound better depending on the headphone model from what I heard.
This is interesting to hear. I've been tempted to get sonarworks for my headphones but the reason I haven't done so is that the sound / frequency response changes quite a bit just by putting a bit of pressure on them with my hands (i.e. squeezing them against my head a little bit). If that changes the sound so much, it makes me wonder how one calibration curve can be "correct" for those headphones given they will fit slightly differently on everyone. I've heard good things about Sonarworks so I'd be interested to hear other's thoughts on this.
 
Last edited:
Guys. I think that all your problems come from the fact that you are using "averaged" curve for any given headphones model.
I have pre-calibrated pairs (Sonarworks did the measurement and provided custom headphones profile) and the sound and its shape is so different from "average" curve that I ended up using it (~80% wet) on another pair of same model (without Sonarworks calibration). In fact, "average" curve was so bad and so off that I didn't like it all.

Also I really recommend you to listen your favourite music (and thus calibrating your ears) both through speakers and headphones with Sonarworks enabled. And then Sonarworks will become an invaluable part of your workflow - especially if you are using Cubase where setting 2 instances of Sonarworks (for speakers and headphones) via Control Room is a breeze.
 
Guys. I think that all your problems come from the fact that you are using "averaged" curve for any given headphones model.
I have pre-calibrated pairs (Sonarworks did the measurement and provided custom headphones profile) and the sound and its shape is so different from "average" curve that I ended up using it (~80% wet) on another pair of same model (without Sonarworks calibration). In fact, "average" curve was so bad and so off that I didn't like it all.

Also I really recommend you to listen your favourite music (and thus calibrating your ears) both through speakers and headphones with Sonarworks enabled. And then Sonarworks will become an invaluable part of your workflow - especially if you are using Cubase where setting 2 instances of Sonarworks (for speakers and headphones) via Control Room is a breeze.

That's possible but for the HD800S for example, the manufacturer's tolerance for variation is very low (if the drivers don't respond like they should they discard them) so it still shouldn't sound as bad as it does at 50% I think. The averaged curve should be very close.. I'm sure it sounds better for some headphones than others though.
 
I don't understand the logic behind sonarworks and headphones. To some degree it can work if the headphones are clearly messed up, but on good headphones, it seems to make the frequency response even more unbalanced past 20% or so.

For example on HD800S (tried it with them), the dips and bumps are there for a reason because headphones need them to feel flat. Of course some frequency response bumps are technically undesirable but it seems that sonarworks is trying to simply apply an opposite curve as if they were speakers. Past 20% it resulted in obvious weird resonances (actual ringing in the mids) that I didn't hear on my studio monitors or with sonarworks off.

I'm not sure what technique they are using to come up with these target curves but it just doesn't sound right. 100% is an absolute mess, 10-20% can sometimes sound better depending on the headphone model from what I heard.
i really like using my audio technica ath-m50x + sonarworks, it tamed harsh frequency and bass, also midrange frequency are more clear
 
As I've mentioned before in other posts, the perception of how headphones "sound" has a much wider range than speaker comparison. While two listeners might listen to speakers and have different preferences, rarely will one fall in love and the other find it completely unlistenable. But two people can listen to the same phones (with the same source material) and have a MUCH greater gap in their opinions. I've opined that the decreased distance between drivers and ear allows the shape of the canal to play a much greater role in the perceived sound of the phones, so much so that I place very little stock in published frequency response results. Others feel that other factors weigh in heavily. I'll leave the argument as to "why" to better and more experienced minds than mine.
Find phones that match how you feel known source material should sound uncorrected and go from there. If the correction works for you, great. If it doesn't, perhaps try a curve from other models and find one that suits your combination better. At worst, you have phones that already sounded right to you.

But speakers - Sonarworks has been a godsend to me.

As always, YMMV.
 
I don't understand the logic behind sonarworks and headphones. To some degree it can work if the headphones are clearly messed up, but on good headphones, it seems to make the frequency response even more unbalanced past 20% or so.

For example on HD800S (tried it with them), the dips and bumps are there for a reason because headphones need them to feel flat. Of course some frequency response bumps are technically undesirable but it seems that sonarworks is trying to simply apply an opposite curve as if they were speakers. Past 20% it resulted in obvious weird resonances (actual ringing in the mids) that I didn't hear on my studio monitors or with sonarworks off.

I'm not sure what technique they are using to come up with these target curves but it just doesn't sound right. 100% is an absolute mess, 10-20% can sometimes sound better depending on the headphone model from what I heard.

Now that there is a 20% off for a few days, I'm thinking about this too.
I used the trial period to activate SW Ref 4 on my DT880 Pro and I don't like the outcome. It certainly tames the high and enhances the low nicely but the upper mid is ringing loud.

I really doubt that average correction profile. If someone has a custom correction curve for DT880 Pro, I'd like to try. I may try the dry/wet setting as well but I'm afraid it'll bring up some phase issues.

So, I'm not sure about headphone correction BUT I'm pretty sure I'd have splendid results on my CMS 40 monitors in my small music room.
 
the sound / frequency response changes quite a bit just by putting a bit of pressure on them with my hands (i.e. squeezing them against my head a little bit). If that changes the sound so much, it makes me wonder how one calibration curve can be "correct" for those headphones given they will fit slightly differently on everyone
Yes, this! It is very dependent on pressure, how you position the cans on your ears, how big your ears are, how big your head is, the angle of the membrane compared to your ear, etc etc. I don't use sonarworks anymore. If you have good cans to start with, and learn how they translate you should be fine.
 
Now that there is a 20% off for a few days, I'm thinking about this too.
I used the trial period to activate SW Ref 4 on my DT880 Pro and I don't like the outcome. It certainly tames the high and enhances the low nicely but the upper mid is ringing loud.

I really doubt that average correction profile. If someone has a custom correction curve for DT880 Pro, I'd like to try. I may try the dry/wet setting as well but I'm afraid it'll bring up some phase issues.

So, I'm not sure about headphone correction BUT I'm pretty sure I'd have splendid results on my CMS 40 monitors in my small music room.

The average profile for the DT880 pro was really out in comparison a the calibrated file. When I questioned them on this they said berdynamic had really changed their DT880 pro models response in the bass on recent releases and probably needed to update that particular averaged profile.
 
... If someone has a custom correction curve for DT880 Pro, I'd like to try ...
"Custom correction curve" refers to a curve made by measuring one sample of a given make and model of headphones. "Average curve" refers to a curve made by testing several (I don't know how many) samples of the same make and model. If someone sends you the custom curve for their DT800 Pro's, it's not any more relevant to your DT800 Pro than the average curve. To get a custom curve, you have to send your own headphones to Sonarworks. They measure them and send them back with a curve for that specific pair.
 
I picked up sonarworks a while ago..I got a behringer measurement mic, but haven't got around to(or have been putting off..) doing the measurement..Does it make that much of a difference?..Does the volume you're mixing at matter,?,..as in, can you vary the volume you're working at with it?
 
It does make a big difference. Not sure how accurate that Behringer mic will be though.

Volume doesn't matter, crank it up
 
Every set of headphones or monitors + room combo is going to have a bit of a learning curve, regardless of whether you're using room/headphone correction software or not. So the real question becomes how do your mixes translate into the "real world" outside of your studio and does using this correction software help you get a better translating mix quicker?

For me the answer has been yes using Sonarworks for my monitors and room. I could definitely hear into the mix better and my imaging tightened up, especially the center. I also ordered a pair of Sennheiser HD600 pre-calibrated with a custom curve from Sonarworks and found it difficult to mix with the curve... it just threw me off too much, even when using a much lower mix blend on Sonarworks. So I don't use the curve on the HD600, but I do on my monitors and room.

I have found that his combo works well for me and overall, mixes have been translating a lot better outside of the studio with much less effort. So I would say give yourself some time to acclimatize to using a new setup with Sonarworks but also go with what feels right for you and especially what allows your mixes to translate better with less hassle and effort.
 
I could definitely hear into the mix better and my imaging tightened up, especially the center.
Same here. I didn't even realize my image leaned slightly to one side until I compared the corrrected an uncorrected room. The overall imaging is also more balanced, and the impression of 'height' improved a bit as well.

I actually like the phone correction but as some have pointed out headphones are one of those things where people have widely varying preferences. I've always hated mixing in cans. Even at low levels my ears fatigue so I found SW made them much more tolerable for extended periods of time... If you like your cans as is you may very well not like the result.

For me the fact that my room and cans sound closer to one another is what matters most... But again, YMMV depending on what you're used to...
 
Last edited:
In terms of how my monitors and headphones sound the difference has been day and late afternoon. In terms of translation the difference is day and night and that is what matters the most.
It has saved me many trips to the car and other speaker systems for testing.

Last month I had to move my studio to home due to lockdown and I was delivering final masters to clients in less than a day, normally it took me at least a week to learn a new room.
 
"Custom correction curve" refers to a curve made by measuring one sample of a given make and model of headphones. "Average curve" refers to a curve made by testing several (I don't know how many) samples of the same make and model. If someone sends you the custom curve for their DT800 Pro's, it's not any more relevant to your DT800 Pro than the average curve. To get a custom curve, you have to send your own headphones to Sonarworks. They measure them and send them back with a curve for that specific pair.
Thanks for your reply.
I reckon now that they are providing this service. I was just curious to know whether or not a particular profiles deviates a lot from the average one.
I think I found a compromise by applying half of the correction only.

Anyway, I don’t doubt it could bring a lot to correct my room.
 
Top Bottom