OLLO S5X Headphone Review

[Zensō]

1dB in a wide Q is audible.

I’d argue that’s true only if we’re talking a 1 dB discrepancy between L/R channels. If we’re talking about a 1 dB discrepancy from their target, which is what they appear to be saying (see below), it’s highly unlikely anyone would hear it.

To achieve even flatter sound, we had to approach hardware and software tuning as a unified whole while respecting different needs and use cases.

 

[GaryH]

Waves Audio themselves don't seem to think using 'pre-chewed' 10-band EQ to correct studio headphones to the Harman target generated through science and industry standard measurements by a professional acoustic engineer is a bad idea:

 

[Robbo99999]

Waves Audio themselves don't seem to think using 'pre-chewed' 10-band EQ to correct studio headphones to the Harman target generated through science and industry standard measurements by a professional acoustic engineer is a bad idea:

View attachment 303097

So am I right in thinking that Ollo users could use the USC (Unit Specific Calibration) aspect, which is the calibration from Ollo, and then use Waves NX to either choose one of the Ollo Target Curves or an Oratory EQ to Harman, and this would all be done through Waves NX (including the USC aspect)?

EDIT: Oratory hasn't measured the S5X yet though, but he has measured S4R and S4R 1.1 - whatever they are!

 

[bodhi]

Waves Audio themselves don't seem to think using 'pre-chewed' 10-band EQ to correct studio headphones to the Harman target generated through science and industry standard measurements by a professional acoustic engineer is a bad idea:

I also think it's a good idea and works great in theory.

I don't believe that many of those filters do anything in practice (those with mild gain and high Q). And this is only my opinion based on my own experiences.

I would of course be interesting to have controlled blind testing where EQ tweakers are put into test by having to assign their preference correctly when the controller is messing with the filters first slightly and with increasing deviations. And the test must be to pick the preferred profile, not to discern when something changed.

 

[Robbo99999]

I also think it's a good idea and works great in theory.

I don't believe that many of those filters do anything in practice (those with mild gain and high Q). And this is only my opinion based on my own experiences.

I would of course be interesting to have controlled blind testing where EQ tweakers are put into test by having to assign their preference correctly when the controller is messing with the filters first slightly and with increasing deviations. And the test must be to pick the preferred profile, not to discern when something changed.

If you look at Oratory's EQ's then there's normally only one or two filters per headphone that could be described as mild gain and high Q, so it's not true what you say that "many of those filters don't do anything in practice" - plus those filters can help add up to an accumulative effect if two of those filters are working in the same direction. But intrinsically that doesn't make 10-band EQ bad, even if those filters you mentioned don't do much they're not detrimental. 10 band EQ is a sensible number to choose - doesn't take too long to input into your device when you're setting up the EQ and it's certainly enough to fix headphones. To adapt your opening phrase: It's still a good idea, and it works great in theory as well as in practice.

 

[GaryH]

So am I right in thinking that Ollo users could use the USC (Unit Specific Calibration) aspect, which is the calibration from Ollo, and then use Waves NX to either choose one of the Ollo Target Curves or an Oratory EQ to Harman, and this would all be done through Waves NX (including the USC aspect)?

I don't think so, as the 'USC' EQs to their made-up target(s) anyway.

 

[bodhi]

If you look at Oratory's EQ's then there's normally only one or two filters per headphone that could be described as mild gain and high Q, so it's not true what you say that "many of those filters don't do anything in practice" - plus those filters can help add up to an accumulative effect if two of those filters are working in the same direction. But intrinsically that doesn't make 10-band EQ bad, even if those filters you mentioned don't do much they're not detrimental. 10 band EQ is a sensible number to choose - doesn't take too long to input into your device when you're setting up the EQ and it's certainly enough to fix headphones. To adapt your opening phrase: It's still a good idea, and it works great in theory as well as in practice.

Just to clarify, you yourself would be pretty confident that after you have chosen certain 10 band profile, maybe modified it to your own taste, you could consistently identify it in blind test when the test controller starts meddling with the filters? Let's say for example you take one hour break from listening and then get presented X and Y where you have to choose preferred one and where from other a +1.5dB Q4 filter at 6kHz is removed.

 

[Robbo99999]

I don't think so, as the 'USC' EQs to their made-up target(s) anyway.

That would be a bit disappointing if the implementation of the individualised unit calibration was restricted in that way, because theoretically it could be added to the headphone to remove unit variation before then applying an EQ of choice on top of it (eg an Oratory EQ). I mean it could still be done, if the USC thing EQ's it to their Ollo Target then you're removing the unit to unit variation, then it just needs to allow for the Oratory EQ to be put on top of that, but that's only valid if the Ollo Target is an average of many units of S5X that have been measured, but if the Ollo Target sits consistently away from the S5X in some way then you couldn't put an Oratory EQ on top of it, as Oratory's EQ are based on an average of whatever numbers of S5X he would measure, so the individual unit calibration is only gonna work if Ollo were to calibrate all their units to their average S5X, which would still remove unit to unit variation for them, maybe they could think about implementing it in that way to give users more flexibility to put different EQ's on top of their unit calibration too.

 

[Robbo99999]

Just to clarify, you yourself would be pretty confident that after you have chosen certain 10 band profile, maybe modified it to your own taste, you could consistently identify it in blind test when the test controller starts meddling with the filters? Let's say for example you take one hour break from listening and then get presented X and Y where you have to choose preferred one and where from other a +1.5dB Q4 filter at 6kHz is removed.

I'm not sure about that, it's a moot point anyway. 10 band EQ is a good idea, it's not detrimental, and it doesn't matter if one or two of those filters are hard to spot the difference - it doesn't invalidate the whole endeavour, it doesn't invalidate the benefits of EQ. What exactly is your beef with EQ, are we on an anti-EQ crusade here?! (It's not going too well.)

 

[GaryH]

So my question to you is whether what I understood and outlined above is true

No.

does the claim made in the article that you can correct resonances by applying filters still holds true?

Yes.

One of the defining characteristics of resonance is non-linearity and strong level dependence if I am not mistaken

You are. This will not necessarily be the case. An example:

There are several resonances in the mids and treble here, but none of them show nonlinear level-dependence. Only the bass does, and only a small amount, and only at very high levels, so not really a problem anyway. And of course, this bass nonlinearity, here manifesting as dynamics compression, will also show up as distortion. So just as @Robbo99999 has correctly been saying, frequency response and distortion are sufficient really to characterize the audible performance of a headphone.

 

[bodhi]

I'm not sure about that, it's a moot point anyway. 10 band EQ is a good idea, it's not detrimental, and it doesn't matter if one or two of those filters are hard to spot the difference - it doesn't invalidate the whole endeavour, it doesn't invalidate the benefits of EQ. What exactly is your beef with EQ, are we on an anti-EQ crusade here?! (It's not going too well.)

No, I'm using EQ with most of my headphones. What I'm getting is that this scientific optimizing, while great in theory, might not be all that useful in practice. With a bit exaggeration one might wonder if some crude method, let's say piece of felt, could get pretty darn close, preference wise, to the result from a set of filters where something like tone correction is required.

We are using EQ to improve our listening experience, right? Not to achieve some theoretically optimum FR based on measurement which may or may not be accurate, fixture which might or might not match our head and ears and a target curve which might or might not match our personal preferences.

 

[Zensō]

I may be missing something here, but my understanding is that the USC calibrations are meant to run inside of Waves NX plugins, so you’re stuck using their control room sims to implement your headphone corrections. Because of what I feel is the gimmicky nature of the NX sims and how much they color the sound, this would be a non-starter for me for mixing, but even more so for general listening to music. If someone knows otherwise I’m all ears.

 

[GaryH]

 

[Zensō]

View attachment 303183

Thanks. That was unclear in the product description, at least to me.

 

[markanini]

I don't believe that many of those filters do anything in practice (those with mild gain and high Q). And this is only my opinion based on my own experiences.

I think I see your point. With Oratorys1990 presets there's not a large difference in post EQ score, between the full, and RME ADI presets with limited bands. I once asked Oratory1990 to make a PDF of a octave band preset I had made manually in REW. Even he was surprised how close the post EQ score was to his full version preset.

As much as I like Oratory1990 contributions, I wish there was more to encourage EQ users to use a more limited set of filters to feasibly personalize their EQ.

 

[IAtaman]

There are several resonances in the mids and treble here, but none of them show nonlinear level-dependence. Only the bass does, and only a small amount, and only at very high levels, so not really a problem anyway. And of course, this bass nonlinearity, here manifesting as dynamics compression, will also show up as distortion

Thank you for the response.

Yes, that graph is pretty conclusive, headphone resonances are mostly linear it looks like. Where is it from?

I went back to the review to figure out why I thought they would be non-linear. I think it is THD vs Frequency graph that show levels as dB and THD as % is the culprit. If you convert THD% to dB, looks like they would align linearly indeed.

This is the graph I am talking about. Level is logarithmic, and so is the x-axis, not sure why the y-axis is linear.

frequency response and distortion are sufficient really to characterize the audible performance of a headphone.

That may be true if the FR and distortion you are referring to is the FR and distortion we perceive. Surely it is NOT true for the 1/12 octave smoothed, rig specific, unreliable measurement data we have though. Let's not pretend they are the same.

 

[bodhi]

I think I see your point. With Oratorys1990 presets there's not a large difference in post EQ score, between the full, and RME ADI presets with limited bands. I once asked Oratory1990 to make a PDF of a octave band preset I had made manually in REW. Even he was surprised how close the post EQ score was to his full version preset.

As much as I like Oratory1990 contributions, I wish there was more to encourage EQ users to use a more limited set of filters to feasibly personalize their EQ.

Yes. It seems that one of the main things here in ASR is educating people to either stop buying equipment or doing tweaking that doesn't actually do anything OR owning it that these changes are for non-audio quality reasons. Buying better DAC for the best SINAD or tweaking to perfect FR even though you can't hear the difference.

Amir himself seems to rather use less than more filters and many times has concluded that he couldn't really be sure if some filter helped or not.

And still, just yesterday I tested his EQ profile with AKG k371 and boy was the improvement subtle to my (cloth?) ears. I would bet exactly zero euros to be able to notice that in blind test let alone randomly tell after a few songs if the profile was enabled or not. And some of the tweaking I'm referring to is a lot milder than that.

 

[Mulder]

Yes. It seems that one of the main things here in ASR is educating people to either stop buying equipment or doing tweaking that doesn't actually do anything OR owning it that these changes are for non-audio quality reasons. Buying better DAC for the best SINAD or tweaking to perfect FR even though you can't hear the difference.

Amir himself seems to rather use less than more filters and many times has concluded that he couldn't really be sure if some filter helped or not.

And still, just yesterday I tested his EQ profile with AKG k371 and boy was the improvement subtle to my (cloth?) ears. I would bet exactly zero euros to be able to notice that in blind test let alone randomly tell after a few songs if the profile was enabled or not. And some of the tweaking I'm referring to is a lot milder than that.

Yesterday evening I tried amirm’s eq setting again, but this time just the 213 Hz setting, omitting the others. When the full setting is applied, ie the four different settings, then the difference in soundcharchter is not subtle. It is very obvious. Better? not to me, but others may differ in their verdict.

Anyway. amirm stated this: ”I was amazed how much difference that filter at 213 Hz made. I think the harmonics of that distortion were polluting the spectrum above above.”

So I tried just this setting. I am not sure I could hear any difference at all. I don’t know at what volume amrim did his listening test, and I didn’t puch the volume to very high levels. Anyway. What I hear and what amrim describes do not correnspond at all, when just this 213 adjustment is done. Actually. To me this adjustment alone seems to be completly meaningsless. Placebo?

 

[bodhi]

So I tried just this setting. I am not sure I could hear any difference at all. I don’t know at what volume amrim did his listening test, and I didn’t puch the volume to very high levels. Anyway. What I hear and what amrim describes do not correnspond at all, when just this 213 adjustment is done. Actually. To me this adjustment alone seems to be completly meaningsless. Placebo?

If I remove the low self then the rest don't make any kind of meaningful difference (if any) to me. I tried playing about twice as hard as I usually listen. Also tried really push the Sinxer SA-1 so that I wouldn't dare to have the earcups next to my ears.

But I think this horse is about well done by now and maybe some of it should be moved to general EQ discussion. I'm happy to admit that for me personally careful EQ tweaks are wasted, even though I like general, tone altering modifications.

 

[Robbo99999]

No, I'm using EQ with most of my headphones. What I'm getting is that this scientific optimizing, while great in theory, might not be all that useful in practice. With a bit exaggeration one might wonder if some crude method, let's say piece of felt, could get pretty darn close, preference wise, to the result from a set of filters where something like tone correction is required.

We are using EQ to improve our listening experience, right? Not to achieve some theoretically optimum FR based on measurement which may or may not be accurate, fixture which might or might not match our head and ears and a target curve which might or might not match our personal preferences.

We're not gonna go through all that again, we've debated the pros & cons of EQ vs insert, so I don't have anymore to say on that. I also pointed out earlier that you were getting confused by bringing in the whole topic of "EQ'ing a headphone from a measurement you find on the internet" whilst we were actually talking about insert vs EQ that replicates that insert's effect. (Going back to this post: https://www.audiosciencereview.com/.../ollo-s5x-headphone-review.46166/post-1675154 ). So it's not relevant to be going on about the pros & cons of "EQ'ing a headphone from a measurement you find on the internet" here in this thread - do it if you want to, don't do it you don't want to. I made an (in brackets) reply to you when you were commenting off track from the discussion on your opinion of 10-band EQ, and then we also had a quick subsequent discussion on that even though it's not relevant. So I think we're dead on this topic now. (You can find many discussions here on ASR on pros & cons of Harman Curve EQ's (using 10 band EQ for instance because Oratory's EQ are the most common), so I'm not going through all the pro's & cons of all that as it's not relevant.