[Joe Bloggs]

Quote:

Originally Posted by Maverickmonk /img/forum/go_quote.gif
 
More practically: would anyone care to share their EQ methods and techniques?
 
What defines "forwardness" other than a bump in the mid spectrum EQ?

 
My guide to EQing all my headphones to my ideal frequency response starts here:
 
Quote:

 
It takes a lot of steps and I don't have the energy right now to write a complete guide so I'll try to walk you through step by step.  First you need to install Virtual Audio Cable and VSTHost on your computer
http://software.muzychenko.net/eng/vac.htm
http://www.hermannseib.com/english/vsthost.htm
 
And get it working as per this review http://www.head-fi.org/products/beyerdynamic-dt-770-pro-closed-studio-headphones-250-ohms/reviews/5928 (but replace SAVIHost with VSTHost) so that you can equalize system sounds using VST plugins.  And instead of his Marvel GEQ you need a parametric EQ like Electri-Q
http://www.aixcoustic.com/index.php/posihfopit_edition/30/0/
which is what I use.  The paid version may be less bugg though:
http://www.aixcoustic.com/index.php/Electri-Q-FULL/13/0/
(I know how to work around the bugs in the free version but it may be easier to work with the paid version)
 
Come back to me when you've got this setup working so you can hear system sounds (such as stuff playing on youtube or spotify) being changed when you play with Electri-Q in VSTHost, or if you have problems setting this up and we'll discuss what to do from there.

 

[RPGWiZaRD]

So it actually works to get a system wide VST EQ? That sounds highly interesting. What's with this "Virtual Audio Cable" though I wonder, I'm very dubious to if the extra software "layers" behind it will lead to any noticable sound quality impact though, I've tested "My Ears" HRTF manipulation algorithms in the past which operated on somewhat similar grounds (added an own output device) but the sound quality through that was very poor at best. I don't know anything about how Virtual Cable Works though so I'm just shooting in the dark here.
 
Basicly what I'd love to get is system wide VST EQ possibilities with WASAPI playback support and otherwise no audible sound quality loss versus just using your soundcard drivers like normal, in getting to that point. If this was a possibility I could at least give up the need of soundcard as my source as I really want minimum a 10-band EQ (due to it being system wide and usually better working than what's found on media players anyway) but VST EQs like electri-Q for example works a lot better than media player EQs though so that would also be good enough as a replacement for a 10-band EQ on a soundcard.

 

[Joe Bloggs]

It sounds all good to me, when it's not glitching.  There are numerous technical issues that may come up but I think I've licked them all, for my desktop system at least.

 

[RPGWiZaRD]

I just gave this a try and I was very positively suprised how great it works. Not hearing any loss in sound quality whatsoever when setting input wave device as MME: Virtual Cable and Output as "MME: Speakers" and WASAPI output in foobar2000 in my case in VSThost versus using WASAPI in foobar2000 with the Realtek onboard chip, more like the opposite almost (guess it's due to my quick EQ setting used to bring the lower-mids I've always wanted to put ever slightly more forward with my config). Guess I just need full version now to remove the annoying "trial" announcer ever few secs haha.
 
Thanks a lot for this tip Joe Bloggs, the playback has no compressed sound to it to my ears, sounds like using WASAPI or something.

 

[Joe Bloggs]

That's cool.  If you want bit perfect output you can play with getting ASIO output on VSTHost via something like ASIO4ALL.  I didn't get it working but others have.  One real sound issue that may crop up with this config though, I only encountered when I started using Sinegen to test the frequency response of my ear-earphone combo.  I found that sine tones produced by Sinegen sounded impure when played through VAC+VSTHost if not set up properly; it's most evident when the sine tone goes above my audible frequency range (~14kHz-22kHz), there was this aliased tone that was audibly going back and forth that shouldn't be there at all.  If I recall correctly, using ASIO was one way to solve this problem (but it introduced other problems I couldn't solve), but playing with the sampling rate settings in different parts of the setup solved the problem in my case too without using ASIO (in my case, I just set everything to 44.1kHz everywhere).  After this, I did not perceive any degradation going through VAC+VSTHost compared to bypassing them, except for random sound hiccups, which I recently finally eliminated with the Audio Services Priority Change tweak.
 
If you decide to further tweak the sound through VAC+VSTHost using Sinegen or want to know more about how I equalize headphones using this setup, post in this thread
www.head-fi.org/t/615417/how-to-equalize-your-headphones-advanced-tutorial-in-progress
and I will continue my tutorial and provide any technical support I can to problems that come up.

 

[RPGWiZaRD]

Thanks but my preferred EQ method is EQing by ears towards my personal ideal sound curve where noone but my own ears can tell what that is. You happen to know of any nice "3D sound enhancing" / HRTF tweaking VST plugins that may be interesting to use for gaming in mind.

 

[Joe Bloggs]

I am EQing by ear though, no measurement instruments involved.  Just good ole Sinegen playing sine tones at different volumes, going through equalizers in VSTHost.  If nothing else smoothing out those resonance peaks in the treble can make for much smoother, refined highs without substantially altering the signature of the headphones.
 
3D sound... Isone Pro?  VAC+VSTHost adds a small delay to system sounds though, which can be compensated for when watching movies (by changing the audio video sync) but could play havoc with performance gaming?  You've been warned...  test this before playing competitively.  The latency can be minimized with the right settings but that increases CPU utilization.  I also find that the delay could increase to an unreasonable level after leaving the computer on for days, at which point closing and restarting VSTHost cures the problem.

 

[RPGWiZaRD]

I prefer EQing to my favorite tracks, I can't enjoy sinetones so. I'm not playing competitively atm but neither did I feel that latency would be an issue if using MME @ buffer set to 525 samples, no noteworthy delay to speak of. My CPU is an i7-860 @ 4.0GHz which is rather slow for my standards as I'm a computer hardware & overclocking freak too which should be using an 2500K/3570K @ 4.5GHz by now if I kept updating with my normal upgrade cycles (about once a year) but that hobby has more faded towards headphones these days. ^^

 

[Joe Bloggs]

Cool

Even so it may be worth using Sinegen for a bit to check if the impure sine tone issue comes up.  Can't imagine it'd be much good for music if that happens--but for a time I had that issue but couldn't really detect how it affected sound when, say, watching a movie.

 

[kiteki]

Yeah, they have different impulse response, I was asking about the FR, for the third time.
 
Anyway, I am honestly asking questions, and you're making up imaginary agendas to project some kind of issue you have with the questions.

 
And we've been telling you not to draw sweeping conclusions from just one type of graph...

 
The point is these headphones look too similar on _ALL_ graphs to justify the differences, or they really do sound that close to eachother.
 
- especially from 30Hz to 11kHz (in other words listening to MP3 96 kbps, minus the 20Hz bass frequencies you can't detect in a HP in the first place, or using cut-offs).
 
 
If you have more graphs you can show them, but differences alone aren't justification.  I'm trying to match the data to exactly what it tells, so the data is more exact / it's a more exacting science.
 
If two headphones measure identical and still sound different, that is evidence that the data is lacking.
 
Likewise, if they measure identical and sound identical, that's leaning towards that the data is rather complete.
 
That's the point of 'cherry picking'.

 

[kiteki]

So kiteki how's the software installation and config coming along?

 
It's fine, as per PM, just need more time to play with it.
 
I don't have much hope I can turn every IEM into an ER-4S though, if that's what you're hoping , and my position is there are better sounding IEM's than the ER-4S, like the JH Audio JH11, irrespective to equalizer settings.
 
Frequency response is... SPL.  The JH11 has for example realism you can't find nor emulate in lesser IEM's, IME.  On that note UE's 'perfect FR' looks different than Etymotics.
 
Edit:  Ok, I am open to the idea you can turn an ER-4S into a JH11/13/16, or UM Miracle, with some amazing software, but that would be quite shocking to me.  That is my current stance anyway.

 

[rhythmdevils]

Quote:

 
If two headphones measure identical and still sound different, that is evidence that the data is lacking.
 
Likewise, if they measure identical and sound identical, that's leaning towards that the data is rather complete.
 
That's the point of 'cherry picking'.

 
Yes, the first 2 lines I agree with.  But that's not the point of cherry picking.  In fact, cherry picking is causing you to miss the fact that you have to look at more than one headphone.  Yes, if you can sit there for 6 hours scouring over measurements and find one example of two headphones that sound different but measure similarly, then there is clearly something missing.  But it doesn't mean what the measurements are showing is invalid.  Which is what you seem to have been arguing everywhere else.  Incomplete is not the same as invalid.  No one thinks that Tyll's measurements represent the entire sonic picture.  I sure don't.  I don't even care much about square waves.  But they are still meaningful for what they represent. 
 
What you seem to keep doing is taking small bits of measurements and then making sweeping generalizations about measurements as a whole because you disprove something they do not claim in the first place. 
 
Would be like a mechanic deciding that a tire pressure gauge is useless because his car still doesn't run even though the tire pressure gauge shows perfect air pressure in the tires. 

 

[kiteki]

I never said the measurements are invalid in any way shape or form, or anything else you're making up, you can project your issues somewhere else.

 

[Joe Bloggs]

So what you need now is for someone to sit down with the MMM and LCD-3 and tell you whether they sound all that different?
 
Going by your own description though, unless you find 96kbps mp3s sound very similar to lossless, that *is* a pretty big difference, no?
 
Quote:

It's fine, as per PM, just need more time to play with it.
 
I don't have much hope I can turn every IEM into an ER-4S though, if that's what you're hoping , and my position is there are better sounding IEM's than the ER-4S, like the JH Audio JH11, irrespective to equalizer settings.
 
Frequency response is... SPL.  The JH11 has for example realism you can't find nor emulate in lesser IEM's, IME.  On that note UE's 'perfect FR' looks different than Etymotics.
 
Edit:  Ok, I am open to the idea you can turn an ER-4S into a JH11/13/16, or UM Miracle, with some amazing software, but that would be quite shocking to me.  That is my current stance anyway.

 
Okay.  I've posted a guide to equalizing headphones to match an ideal loudspeaker response.
http://www.head-fi.org/t/615417/how-to-equalize-your-headphones-advanced-tutorial-in-progress#post_8491380
 
If you have an "ideal" IEM, you can pretty much follow this guide, except using your reference IEM as your reference speakers.
 
If you can, follow the steps in that post up to step 9 and post a screenshot of the equalizer settings you've arrived at so far, then I can talk you through the rest of the process. (of transforming another pair of phones to sound 99% like the JH11 or *gasp* make another pair of phones or the JH11 themselves sound better than the JH11)
 
The remark about room nodes in step 9 doesn't apply to headphones, of course.  OTOH, if you find sharp peaks and dips above 2kHz (probably much higher than 2kHz though, for in-ear customs), those may be ear canal resonances.

 

[kiteki]

So what you need now is for someone to sit down with the MMM and LCD-3 and tell you whether they sound all that different?

 
Um, it would help if someone in this thread had heard both, or if someone sat down and tried to make the MMM sound like the LCD-3 with an equalizer, yes.
 

 
Going by your own description though, unless you find 96kbps mp3s sound very similar to lossless, that *is* a pretty big difference, no?

 
If the MMM and LCD-3 sound pretty much the same with 96 kbps MP3 (or lossless with 30Hz and 11kHz cut-off) then the LCD-3 is a complete joke, or the MMM is a very overlooked headphone.
 
If the difference becomes more evident switching from 96 kbps MP3 to 192 kbps or lossless, that is definitely a suble difference, since that is already pretty subtle.  From what I can tell most people, including those in this thread, expect the differences in sound vastly more different than that, like two completely different sounding headphones, when playing 96 MP3, with any eq, well that's just my take on what most people think.