It's amazing how often AutoEQ settings get discussed without context; this is exactly why I find them nearly useless in practice, at least without further corrections. I don't have a good enough ear to EQ without guidance, but whenever I'm testing new cans I like to collect multiple EQ profiles from all the usual sources and swap between them. Eventually I'll settle on whatever sounds most pleasing to my ear and typically make a few minor tweaks to taste (usually slight adjustments to bass and a softer hand with treble adjustments). The AutoEQ profiles are fascinating as they almost always sound insane, but sometimes produce entertaining effects, like exaggerated dead spots alongside intense treble spikes. Makes life a bit spicy!That's not Oratory's EQ, that's AutoEQ's purely algorithmically generated EQ, using Oratory's measurements, which, as can be seen from the 'Equalized' curve on the graph in your link, in this case has produced an EQ that erroneously boosts the 8-10 kHz region by around 5 dB. This is Oratory's actual EQ for the HD560S. I would only use AutoEQ's EQs if he hasn't made his own one for a particular headphone, due to the potential for such robo-errors as seen in this case.
I'm pretty experienced with EQing headphones and have A/Bed numerous PEQ presets with 10 or more headphones and eight IEMs or so.It's amazing how often AutoEQ settings get discussed without context; this is exactly why I find them nearly useless in practice, at least without further corrections. I don't have a good enough ear to EQ without guidance, but whenever I'm testing new cans I like to collect multiple EQ profiles from all the usual sources and swap between them. Eventually I'll settle on whatever sounds most pleasing to my ear and typically make a few minor tweaks to taste (usually slight adjustments to bass and a softer hand with treble adjustments). The AutoEQ profiles are fascinating as they almost always sound insane, but sometimes produce entertaining effects, like exaggerated dead spots alongside intense treble spikes. Makes life a bit spicy!
But it's distressing that these profiles are often shared without that necessary context. The Qudelix app (which is otherwise phenomenally well designed and feature rich) now has them searchable by default. At first blush, it seems amazing - saves the effort of having to manually enter profiles and it's a wealth of information at your fingertips... until you actually listen to the presets.
I suppose some of the problems with not liking certain Oratory EQ's is unit to unit variation (some models of headphones good other bad in this regard), because then it makes the EQ less valid and less accurate, but of course that's the same with EQ's from any source. But I'd say Oratory makes good decisions re how he EQ's his measurement, whereas with AutoEQ in the case that GaryH brought up in the following link the 8-10kHz notch was boosted, which is a mistake on the part of AutoEQ: https://www.audiosciencereview.com/...er-hd560s-review-headphone.29603/post-1040898I'm pretty experienced with EQing headphones and have A/Bed numerous PEQ presets with 10 or more headphones and eight IEMs or so.
In my experience, oratory's presets that he creates himself sound comically bad about as often as any preset you may find on AutoEQ.
For me at least, the only reliable way to find that one PEQ which genuinely improves sound quality is to A/B every preset and measurement from reliable sources that you can find.
This is where I personally think we need a standardized database of presets, i.e.: Harman research utilizing their measurements of various headphones and providing a PEQ based on their measurements. The reality of it is if the majority of users are using any form of EQ to conform to Harman Curve, then there should be a standard set of PEQ' they should be using to minimize uncertainty and variances based on testing methodologies.I'm pretty experienced with EQing headphones and have A/Bed numerous PEQ presets with 10 or more headphones and eight IEMs or so.
In my experience, oratory's presets that he creates himself sound comically bad about as often as any preset you may find on AutoEQ.
For me at least, the only reliable way to find that one PEQ which genuinely improves sound quality is to A/B every preset and measurement from reliable sources that you can find.
How is that a mistake on part of AutoEQ? The raw response that oratory provided to Jaakko was (after smoothing HF) deficient in that area compared to the target, so the python script boosted it.which is a mistake on the part of AutoEQ
Taking a look at the graph showing the raw & equalised, the following graph, if I'm not mistaken:How is that a mistake on part of AutoEQ? The raw response that oratory provided to Jaakko was (after smoothing HF) deficient in that area compared to the target, so the python script boosted it.
You can read about why AutoEQ uses heavy smoothing above 6kHz here, which will result in slight over-/undershoot of the target: https://github.com/jaakkopasanen/AutoEq#technical-challengesI think it was a mistake to boost 8kHz so that it now peaks there above the Harman Curve
That's you criticizing the Harman target curve, not AutoEQ.I think it was wrong to boost that 10000Hz notch [...] as there's supposed to be a dip there
Yes. AutoEQ uses a modified version of the AE/OE2018 target with 2dB less bass, akin to the Harman target from 2013.Not to mention the fact that the blue target line is not the Harman Curve, it's underrepresented in the bass
I see those more of the limitations of the AutoEQ system being used. Smoothing is already applied in the Oratory measurement, by definition that the published measurement is an average of very many different measurements, both within one headphone itself being measured, but also of multiple headphone units if more than one example of that model is measured, so the published measurement aims to have peaks & dips showing which are consistent from unit to unit & on-head placement position to on-head placement position - so best not to ignore peaks & troughs in the published measurement (apart from the 10kHz notch of course which is what we discussed & above 10kHz it's unreliable and you should use either very broad peak filters (say Q1) or High Shelf Filters). Further to the 10kHz point, blindly EQ'ing up the 10kHz notch and then saying that's a "Harman Problem" is naive - most people who know about headphone measurements know that there is a natural dip at 10kHz that should not be EQ'd up, so if you or Auto EQ ignores that bit of wisdom then that's really not optimal. I'm sure that AutoEQ can produce some decent results for some headphones, but the HD560s example here is certainly not optimal.You can read about why AutoEQ uses heavy smoothing above 6kHz here, which will result in slight over-/undershoot of the target: https://github.com/jaakkopasanen/AutoEq#technical-challenges
That's you criticizing the Harman target curve, not AutoEQ.
Yes. AutoEQ uses a modified version of the AE/OE2018 target with 2dB less bass, akin to the Harman target from 2013.
People like Andrew Park aka Resolve Reviews argue that the vanilla Harman target favors preference across a diverse target group over neutrality and that a 2013-style bass response is a closer fit for audiophiles looking for faithful reproduction.
I think i'ts at a point that instead of comparing sound quality, you decide on features. Especially when using equalizer. I wanted daily driver headphones that have good ventilation and not get hot so for me AKG K371 was not consideration because they are closed and HE400se seems heavy and has probably worse ventilation than HD560S due to planar drivers. All of them have good performance.Looks alright but how does it really compare in this class? There are so many headphones in this price tier now.
You'd be surprised at the ventilation of the HE400se, the HD560s pads started to cause sweat on my earlobesI think i'ts at a point that instead of comparing sound quality, you decide on features. Especially when using equalizer. I wanted daily driver headphones that have good ventilation and not get hot so for me AKG K371 was not consideration because they are closed and HE400se seems heavy and has probably worse ventilation than HD560S due to planar drivers. All of them have good performance.
It's actually a good sounding pair of headphones, EQ'd or not, I was surprised, the angled drivers seem to open it up nicely.Looks alright but how does it really compare in this class? There are so many headphones in this price tier now.
4 whole days. Give the man a break.@amirm no review since past many days ? I wait eagerly.
Sheesh man, let the man enjoy the holiday' a bit.@amirm no review since past many days ? I wait eagerly.
HD560s certainly can compete with HD600, certainly for me, HD560s is a lot better once EQ'd, mainly soundstage related, but also re bass clarity. Also, you don't "get what you've paid for" when it comes to headphones, there's not a correlation between price & performance. I might make the leap of faith though that more expensive headphones probably on average have better driver matching & less unit to unit variation, which is an important consideration if you're using EQ's published on the net by people like Oratory, etc.Nice to see so many reviews telling the same - it seems obvious to me that despite measurement differences 560S sound just can't compete with 600/650/6xx same as AKG 361/371 is obviously a lower grade compared to K7** lineup (yes, I dare to compare closed vs open). Again despite all 7** problems (which are perfectly EQable, I've done this). No miracle, you get what you've paid for in these cases.
I second this, the HD560s can compete with the 600s. This isn't a scenario of 'you get what you pay for', Sennheiser did a good job of tuning these to almost reference tuning, which many of their upper grade headphones don't have. These are still a good recommendation to have for entry level audiophile headphones before leaping to the upper echelons.HD560s certainly can compete with HD600, certainly for me, HD560s is a lot better once EQ'd, mainly soundstage related, but also re bass clarity. Also, you don't "get what you've paid for" when it comes to headphones, there's not a correlation between price & performance. I might make the leap of faith though that more expensive headphones probably on average have better driver matching & less unit to unit variation, which is an important consideration if you're using EQ's published on the net by people like Oratory, etc.