How far can EQ really go towards truly equalizing headphones? - Page 3

Meh - they're sitting on the shelf with their dust-cover on. Maybe later, when I'm not listening to ASOT. Or if it cools off a bit and I stop craving the perfection of velour pads.

My apologies, not trying to add to the confusion.

 

Edited post above.

 

My two cents are in that I don't think EQ can make all headphones sound the same or even make a bad characteristic of a pair of headphones go away completely. There maybe a set of headphones designed in the future which could respond to EQ that way, but I don't think they exist today. Headphones can only reproduce the character of what they are, both good and bad, as there is no perfect set out there.

 

That said, I think complete systems are built around tuning each headphone to respond in a desired way. I look at each piece of gear as adding a small EQ to the signal. I would rather use power-cords, interconnects and different sources to EQ a signal to my liking. I don't use EQ and am not sure of why? 

 

I posted the word "Wow" as I found it related to my own personal believes about the sound being affected by the room in relation to the sound being affected by aspects of the headphone design resonance "colorations" which I feel can not be EQed out in the end.

I would guess I have heard super flat and uber expensive, completely "perfect" rigs..............VPI analogue tables with perfect disks, Einstein Phono-preamps and all powering Sony R-10s and yes it was clear and flat. I maybe had to listen longer than I did to come to grips with how "perfect" everything was. It just wasn't that much fun. I will go for  the rose colored glasses on the musical world any day.

Edited by Redcarmoose - 6/5/12 at 11:35pm

Goldenears mentions a point that bass should be boosted with headphones more so than speakers.

 

Similar to what is acheived with the LCD-2, headphone.com (I think) mentions that treble should be rolled off - something to do with the fact that high frequencies in a listening room are absorbed more when listening with speakers and having the high frequencies shot directly into your ears by headphones.

 

Besides the above the headphones should be fairly flat.
 

Edited by Matt head 777 - 6/5/12 at 11:27pm

I measure my response from my listening position and try to make it flat at that precise spot. So it isn't high frequencies dissipating on their way to the listener.

I'm guessing that even if a headphone puts out a perfecatly flat response, shooting it directly into the ear magnifies some frequencies like a horn. So you slant the EQ to favor the bass and lower the treble to compensate. The end result is the same... Flat response.

I do think that it's possible to EQ reasonably good headphones to perfect perceived flat, and that they will all sound pretty much the same. I think most manufacturers make pretty much the same acoustic tradeoffs. They differentiate their brand's signature sound by EQ coloration, which is easily reversable.

You would have better luck equalizing speakers by stacking up dirty laundry in random piles around your house than by trying to equalize with wires. They just can't do that. Even if they could, you could never make precise enough adjustments that way. The same goes for headphones. An equalizer is the tool for the job.
Edited by bigshot - 6/6/12 at 12:22am

Redcarmoose: Ah, I understand now. I just saw my post repeated and "Wow" added at the end (and again, I'm used to Huddler having hick-ups sometimes, so I figured it was me; when BigShot was confused as well, I realized that something had run amuck - maybe ask Currawong or Joe about it, unless you know what happened?). I agree.

BigShot: Again, there's a limit to what EQ can do. You cannot fix resonance. It just isn't possible. To give you a room treatment example - if you have a room that has a mode at a given frequency, you cannot eliminate that mode with EQ unless you have the most aggressive notch-filter known to man. Because *any* output at that frequency will excite the node. Reducing output doesn't make it go away. The same is true of headphones. So if you look at something like a Grado, Ultrasone, or newer Beyerdynamic model, that makes use of resonance for (or as a result of) their overall design. You can't fundamentally EQ them to sound like a Koss ESP, STAX, or Sennheiser model that has very little resonance. That resonance will impact the overall sound as well - FR is not the only thing we look at when dealing with a system design.

Here's Tyll's recent article about this:
http://www.innerfidelity.com/content/presto-change-o-measurement-transformations

Sure, with something like an ESP or STAX, you can EQ them and change their flavor. But you aren't making them not a planar anymore; you can't make it sound like an S-LOGIC or Auranomic model. That changes the overall SQ and response.

Finally, you'll never get "exactly flat" for a variety of reasons (the easiest is the simplest: there is no perfect).

Resonances usually go hand in hand with peaks in the frequency response. By cutting problematic frequency ranges you can definitely improve sound quality.

Notch filters aka narrow band-stop filters, as the name suggests, attenuate to the limits of precision (usually 32- or 64-bit float). Guess you're talking about narrow peaking EQ / bell filters.

Lastly, how do you know that you can't EQ headphone A to sound like headphone B, have you tried it? Unless we compare two different headphones (in terms of resonances) with equally EQ'd frequency response I don't see how this statement can be supported, at all.

 

 

How do you know that? Most ultrasones (s-logic) have a very "special" frequency response and just because one usually doesn't aim for such erratic FR doesn't mean it's impossible to EQ a headphone to sound similarly.

Edited by xnor - 6/6/12 at 8:21am

You cannot make this:

into this:


With EQ alone. Nor can you "fix" the later one with just EQ. Sure, you can decrease the energy going those frequencies (not to the exact frequency that's being aggravated most likely, but you can pull it down nonetheless) but it will still ring. You also can't change the radiation pattern of the HD 650 to that of the S-LOGIC equipped PRO2900 with EQ. The PRO2900's drivers are still placed low and forward and still have their ported buffer board, and the HD 650 will never have that without physical modification.

Now, you can EQ the HD 650 to be brighter, sure, but it still won't re-create the PRO2900. From FR alone, they actually look relatively similar apart from the treble peaks on the PRO2900.

http://en.goldenears.net/index.php?mid=GR_Headphones&search_target=title&search_keyword=pro2900&document_srl=10258
http://en.goldenears.net/index.php?mid=GR_Headphones&search_target=title&search_keyword=hd+650&document_srl=11616

The point is not that EQ does nothing, but that EQ is not the "transmorphing magic" that will turn water into wine and KPH7s into SR-009s that many want it to be. It will not correct errors or otherwise act in the time domain. It will always be limited by the downstream components (things like the amplifier and transducers). More robust DSP solutions can act in the time domain, although I've not seen measurement results for the really good ones. I have seen measurement results for cheaper DSP solutions though, like this:
http://www.geocities.jp/ryumatsuba/mdr-ds1000.html

And indeed, you can see it acting in the time domain. But that isn't EQ.

It did in practice in my room. The volume had to be very high before the walls shook. It wouldn't bump into the point often, but when it did, it was irritating. Maybe it wasn't a resonance. Maybe I was overdriving the room. My Sunfire subwoofer is easily able to blow the roof off. It took a while to balance it into the system. Perhaps the EQ I applied just muted it a bit from one point all the way to the bottom. It was a very low frequency.

It would certainly be able to match the response. It's possible to do extremely precise corrections with pro grade equalizers. The response is a much bigger part of the overall sound than the differences in time. You couldn't change the physical layout of the drivers though. You would probably have to correct by ear to account for that. But I bet you could get pretty darn close.

In terms of frequency response you could. It's questionable how audible a slightly slower decay time is, especially considering auditory masking.

 

Which manifests itself in changes in frequency response which can be EQ'd. Not solely, of course. Radition pattern, in a headphone? You mean that different placement on the ear will result in different frequency response? That's kinda irrelevant.

 

Imho, they are not similar.

 

Nobody said it's magic nor that cheap/bad headphones can be transformed prefectly into expensive/great ones. EQs do act in the time domain. In fact, most work in the time domain. You can correct both magnitude and phase response.

Edited by xnor - 6/6/12 at 1:18pm

I believe if you don't use eq with headphones your missing out. Maybe some people are put off, or don't realise what a good parametric eq can do cos all they've tried is a graphic eq. I'm using the LCD-2 and it's nearly amazing to how you can tailor the sound for better considering so many people seem to shy away from it and would rather spend money on expensive cables or even amps to tailor the sound.

 

Can make the bass tought without spoiling the high's add some low bottom end to suit all genres, tweak the mids and highs to make things like acoustic instruments sound so real. The 'real' effect seems to be easily spoiled. What makes people think that a headphone manufacturer can make a headphone sound so real and right with all the constraints, manufacturing limitations etc, even personal taste makes people seem a bit dumb.
 

I think the problem is a lot of people don't have an engineering technical background and don't realise how hard it is too make things work. Some people on these forums seem to have an attitude of expectation (not realising the technical constraints in designing a driver etc) that if you spend enough money it should just work. Hence the cables instead of eq.

 

It's the mentality either that or head-fi is actually secretly using sponsors (probably based in Asia based on all the manga that's around ) to pose as average consumers in the forums.

Edited by Matt head 777 - 6/7/12 at 12:46am

 

Cheap cans typically also have cheap drivers. Some cannot handle "higher" volumes well. Also, bass sometimes is severely rolled-off (due to the overall construction, not necessarily the drivers only) so you'd need to boost the bass a lot adding even more distortion. The pads might be out of cheap material that'll fall apart soon, seal very badly etc. Then there's intermodulation distortion, which I've even seen expensive headphones "fail" at. And of course there's comfort, which imho, is very important with headphones.