Next level headphone above Sennheiser HD 650 and Beyerdynamics DT 1990

[Garrincha]

Try EQing your Beyers to Harman and see how big of a difference it makes to your ears. You might find that it doesn’t matter much for most albums.

Well I did. The difference is actually quite high, but is is difficult to A/B test as without EQ the volume is lower so I have to crank it up, which takes some time and thus memory is lost to a great extend.

 

[Garrincha]

If you could go slightly over 2000 on used market, go with the Focal Utopia. Otherwise, Fostex TH909 is a good and vibrant candidate.

Are there any objectivist test of the Fostex TH909? Crinacle didn't like them all that much.

 

[mrbungle]

Well I did. The difference is actually quite high, but is is difficult to A/B test as without EQ the volume is lower so I have to crank it up, which takes some time and thus memory is lost to a great extend.

Not saying there is no difference, but it might just be a slightly different flavor and you are not clearly and consistently favoring one over the other. Especially if you are adding just 2-3db here and there. Not that you shouldn’t look for perfection, but not having EQ on your phone or something might not be a big deal.

 

[Jose Hidalgo]

I agree with the most you are saying. But still, since it was already mentioned that even relatively cheap headphones manage to stay close to the Harman curve and the FR of it is widely accepted to be some kind of norm, it is really not easy to understand why many or even most manufacturers come up with different tonality. Why? Being close to the Harman curve for a headphone should be the same goal as having lowest possible SINAD for a DAC or amp. I for example am now that I discovered how to use EQ on my Macbook inclined to buy the Sennheiser HD 800S, principally because of the supposed huge soundstage everybody is talking about. But for example on my iPhone I will have difficulties using the perfect EQing, so I think it is really a drawback to being almost forced to use an EQ to have the best sound.

The Harman curve is very well known, but it's definitely not a 'norm', as in 'absolute reference'. Manufacturers are free to provide other frequency responses (even for high-end headphones) and still have their fans. Suffice to talk about warm headphones (some come to mind: Denon 5200/7200/9200, Nighthawk, etc.) which are far from the Harman curve but still have a solid fan base of people who wouldn't change their default FR for the world. Like I said, you can't please everybody, even with Harman. Some like it warm! (or piercing high for the famous 'Beyer sound', etc).

If Harman is your absolute reference (it is for me most of the time, but that's just my personal case), then a HD600/650 is obviously a great choice. Then the HD800S seems like a logical step-up, but like said it may need EQ to please you, and you may lose its soundstage (if soundstage is so important to you... it isn't necessarily for me, with free apps like HeSuVi that can dramatically play with it on a song-per-song basis, for that nice 'out of your head' sound).

Well I did. The difference is actually quite high, but is is difficult to A/B test as without EQ the volume is lower so I have to crank it up, which takes some time and thus memory is lost to a great extend.

My free EQ app (see my signature) would allow you to instantly A/B test with previously matched volumes: default sound vs. any target curve you can imagine, Harman being of course the most popular. It also supports HeSuVi. Unfortunately it only works on Windows.

 

[tomtrp]

Try EQing your Beyers to Harman and see how big of a difference it makes to your ears. You might find that it doesn’t matter much for most albums.

beyers, especially older beyers like dt880 dt990 T1 have very high q peaks and dips from 6-10k, which makes these headphones terrible for eq. if you use higher Q compensations for those peaks, you have to be quite precise at the frequency, if you use low q filters, to compensate our the bright/ringingness, you will still end up with narraw peaks or dips (see amir review on dt990, there is a measurement after eq). so quite annoying experience and time consuming to get the balance right. I will never spend my time with any of the high q peaks beyers again (I experiement peq on dt880, dt990 and t1 gen 1, all has some problems even after many many peq profiles while I spent little time to get Hd800s and he6sev2 right.)

 

[Garrincha]

beyers, especially older beyers like dt880 dt990 T1 have very high q peaks and dips from 6-10k, which makes these headphones terrible for eq. if you use higher Q compensations for those peaks, you have to be quite precise at the frequency, if you use low q filters, to compensate our the bright/ringingness, you will still end up with narraw peaks or dips (see amir review on dt990, there is a measurement after eq). so quite annoying experience and time consuming to get the balance right. I will never spend my time with any of the high q peaks beyers again (I experiement peq on dt880, dt990 and t1 gen 1, all has some problems even after many many peq profiles while I spent little time to get Hd800s and he6sev2 right.)

Well I wasn't really considering any Beyerdynamics actually. Although I am quite happy with my DT 1990 Pro and it is even not the top of the line, from what I have read and heard, I wouldn't expect any better performance from T1, T5 or any other model, for that matter.
Concerning EQ or PEQ, shouldn't a competent software with built in profiles (like SoundSource on Mac) do the job easily?

 

[MayaTlab]

Concerning EQ or PEQ, shouldn't a competent software with built in profiles (like SoundSource on Mac) do the job easily?

I already know you won't like that answer, but unfortunately at higher frequencies you're facing sample variation (including pad wear issues), and inter (between different people) and intra (for the same person, seating the headphones in different ways over one' head) individual coupling variations that make EQing headphones quite difficult (without even considering HRTF differences possibly requiring the ideal target to be slightly different between listeners).

What is often done is to perform spatial averaging on ear simulator fixtures to try to identify the invariant peaks / dips from the ones that may vary because of positional / coupling variation. Baring gross sample variation issues, these invariant peaks can be quite safely EQed down with somewhat narrow Q filters, as it is reasonable to presume that most people will experience them at the same frequency (ex : Focal headphones' 6kHz or so peak).
That's the basis behind presenting the data in this way with confidence interval, here from Oratory1990 :

https://twitter.com/i/web/status/1516350384650002433

Peaks / dips that vary with fit over the fixture may turn out to present themselves in a significantly different way once they're on your head, and using ear simulator measurements to EQ them might not bring ideal results then.
In the example above it's probably a good idea to leave some room for individual tweaks to the preset above 1.5kHz.

Presets such as the ones you'll find in SoundSource (which are derived from the autoEQ project) are a good starting point, but won't "easily" address some of the issues you may experience with Beyerdynamic headphones, particularly in the treble range.

I'm not quite certain that you'll be interested but I can send you the PDF of some (paywalled) relevant articles on the subject. This one is free (but unfortunately doesn't feature more modern HP designs) : https://vbn.aau.dk/ws/portalfiles/portal/227875204/1995_M_ller_et_al_AES_Journal_b.pdf