[Alleyman]

Quote:

Originally Posted by Emon /img/forum/go_quote.gif Uhh...what? Foobar2000's default crossfeed is little more than a slight channel mixing with delay. In any case try the Bauer stereophonic-to-binaural DSP component for foobar2000.

Seconded! I find it essential while listening with any kind of headphones...

 

[TheSloth]

Quote:

Originally Posted by cotdt /img/forum/go_quote.gif Something is definately wrong with common crossfeed algorithms. It's very difficult to stimulate the natural distance effect that speakers take advantage of. Most just downmix to mono below 700Hz but this introduces a lot of complicated problems and messes up the sound.

We know something is wrong in that they aren't perfect in themselves, and aren't tailored to one's ears. But the idea that using a high quality crossfeed does the things to 'female vocals' that you describe is very misleading. I have played my headphone system (avec crossfeed) for many fellow professional musicians, and no-one including myself has ever had such a reaction to any musical instrument or timbre. There may be a particular interaction with your ear geometery at work here, but there are many other reasons why crossfeed might not be working optimally in your system. As a quick example, if I insert my QUAD QC-24 tube preamp between my source and headphone amp, and use crossfeed, the interaction of the 'tonality' of the QC-24 and the slight FR shift of the crossfeed algorithm does create a slightly strange spatial effect that is worse than dual mono (dual mono seems to be a decent way of referring to the headphone effect). The fact that inserting that extra step reduces channel seperation in the first place doesn't help the crossfeed either.

Unless you describe your system, and what 'hardware' crossfeed you are actually referring to, I'm not sure this thread can serve any purpose other than rehashing the same old wives tales about crossfeed. To address one of them as it has already come up again, the K1000 has more inherent natural crossfeed than any other headphone design, but it is still a negligable amount compared to speaker monitoring, and in terms of which frequencies are acoustically crossfed, the balance is not even close. There is no LF crossfeed whatsoever with the K1000 - it's not very difficult to approximately test the amount of crossfeed in the K1000 by simply measuring the distance between your ears, placing the K1000 that distance away from you on the same horizontal plane and listening to the FR and volume level.

This downmixing to Mono below 700hz is something I have never heard of before. Of the two commercially implimented hardware crossfeed designs by HeadRoom and Meier Audio, neither do anything remotely as stupid as this. 700hz is still relatively directional and any system designed to mix those frequencies to mono is a result of very stupid design. For those that haven't considered it and think that 700hz is 'low', it would be a little above the middle an operatic soprano's range (of course they can go a fair bit higher, but usually not in most melodies).

 

[Uncle Erik]

Quote:

Originally Posted by m00hk00h /img/forum/go_quote.gif This little black box can do wonders.

Yeah, I use one of them, as well. I've listened to a couple different crossfeed implementations, and I prefer Dr. Meier's approach. I use it all the time, with a couple of exceptions. One, I don't like it with the K-1000. The natural crossfeed is fine and is one of the many reasons these are my favorite headphones.

Second, I don't like crossfeed with vinyl for some reason. I'm not entirely sure, but I think the stylus gives a little crosstalk between channels and it acts like crossfeed, and is why the artificial crossfeed doesn't sound right. But maybe that's just me, and I really should read up on how channel separation works with vinyl.

As for the K-1000, you may not be able to simply order them any more, but AKG did make around 12,000 pairs. They come up for sale pretty often. If you keep looking, you will get a pair. They are absolutely worth it, too, even with the amp madness they inspire.

 

[Davesrose]

I've just gotten used to the headphone sound myself: While the soundstage might not be as engaging as a speaker system , it makes up for some nice intimate detail. It might sound dumb, but if you have any classical CDs....start listening to those first! Oddly mixed rock CDs can have those horrible hard pans, but music that's recorded live in an auditorium sounds more natural at least.

 

[GreatDane]

Quote:

Originally Posted by cotdt /img/forum/go_quote.gif I've tried Crossfeed, both hardware and foobar versions, and while it fixes the problem, the medicine did kill the patient. The sound becomes thick and muddy, and female vocals sound like barbarian males howling.

I've never used any software versions but I do use a Meier Corda Cross-1 in my main home rig. I don't hear any mud in the music. On the low setting it's sometimes hard to detect. Using the bass & treble trims can even change any discernible effects that the crossfeed circuit might cause.

I also use a portable amp built by Jan Meier that has the same pleasant crossfeed results. Another portable amp that I use(Xin SuperMini IV) also has crossfeed which has a slightly more dramatic effect and it never causes the sound to become thick or muddy, these options simply eliminate the extreme stereo effect that causes listening fatigue for most all music, certain classical music being the exception.

Yet another portable amp that I use (HeadRoom Micro Amp) which has crossfeed actually does add some very noticable coloration which I'd refer to as muddy with certain types of music. Some types of music actually benefit from this coloration but only with certain headphones that I use.

Which hardware crossfeed options have you tried?

 

[cotdt]

Thanks guys, I've studied everyone's suggestions and have found a solution. I will build my own hardware crossfeed, specifically the hardware version of the Bauer stereophonic-to-binaural DSP, which is the only crossfeed I've come across that doesn't muddy up the sound. It's a completely passive network so I can also use it to add bassboost to my AKG K701's.

 

[Carl]

Quote:

Originally Posted by cotdt /img/forum/go_quote.gif It's well known that recordings are made for speakers, so each of the stereo channels are heard by both ears, but for headphones, each channel is only heard by one ear so you get a fatiguing ultrastereo effect. It crowds the instruments together and messes up the soundstage, and gets fatiguing easily if you're used to speakers. I started out with headphones btw, then I went for high-end speakers, and now I'm having a hard time coming back to headphones. But I must, since neighbors complain about my speakers. I've tried Crossfeed, both hardware and foobar versions, and while it fixes the problem, the medicine did kill the patient. The sound becomes thick and muddy, and female vocals sound like barbarian males howling. Something is definately wrong with common crossfeed algorithms. It's very difficult to stimulate the natural distance effect that speakers take advantage of. Most just downmix to mono below 700Hz but this introduces a lot of complicated problems and messes up the sound. So sometimes I just listen with the channels completely downmixed to Mono, and it is much more pleasing to listen to than stereo. Stereo is more fun at first, but fatiguing and not as emotionally involving. But there's no illusions that I'm definately missing something with Mono. It's the best I've got at the moment, but I want a headphone rig that can match speakers in spatial presentation, or at least come close. So what have you guys done to combat this annoying ultrastereo effect?

I also find the problems with playing back stereophonic recordings with headphones to be unpleasent. My solution is to use Dolby Headphone, which is basically a stereo->binaural converter.

 

[smeggy]

I think I've tried the majority of software crossfeeds and Dolby/dsp progs and haven't found any of them worth using. I did build a passive crossfeed circuit myself and that works better than any of the software ones but still isn't perfect. As a result I listen almost exclusively in plain old stereo.

 

[Carl]

Quote:

Originally Posted by smeggy /img/forum/go_quote.gif I think I've tried the majority of software crossfeeds and Dolby/dsp progs and haven't found any of them worth using. I did build a passive crossfeed circuit myself and that works better than any of the software ones but still isn't perfect. As a result I listen almost exclusively in plain old stereo.

Stuff like that always works better in hardware.

 

[ferraro25]

Quote:

Originally Posted by Emon /img/forum/go_quote.gif Uhh...what? Foobar2000's default crossfeed is little more than a slight channel mixing with delay. In any case try the Bauer stereophonic-to-binaural DSP component for foobar2000.

Wow, using that component right now; it is GREAT! Thank you for helping me to hear the K701 in a new light. Hearing things out in space improves listening so much, and I am pretty sure detail is easier to pick out now (and, of course, instrument separation has been improved).

 

[Emon]

Quote:

Originally Posted by Carl /img/forum/go_quote.gif Stuff like that always works better in hardware.

Not really...analog hardware creates loss, however small. DSP in software is lossless.

 

[PiccoloNamek]

I am not sure what to think of that plugin. On some songs (especially songs with instruments that are hard-panned) it sounds great, but on others, it completely destroys the reason I like to listen on headphones in the first place. When listening to "Canta Per Me" from the Noir soundtrack, the effect of having the nearly heavenly female vocals drawn out of my head and placed in front of me ruined the experience.

There is also a fair amount of cognitive dissonance involved; the sound is expansive and out in the open, and yet there is a pair very snugly fitting headphones on my head; the sound I'm hearing shouldn't be possible. And yet, there it is. It makes it difficult to concentrate.

 

[Carl]

Quote:

Originally Posted by Emon /img/forum/go_quote.gif Not really...analog hardware creates loss, however small. DSP in software is lossless.

Digital isn't even remotely lossless. Take a look at how impulse response filters actually work, don't treat them as a black box.

Signal processing of any type will degrade your signal quality, it's just a question of whether the positives will outweight the negitives.

 

[countach]

This is one of the greatest threads i've seen in head-fi. I just used to think that I loved that ultrastereo effect, but trying the binaural bauer plugin, I discover a great improve of the listening experience in my ER4 !

Men, I'm wondering about explaining all this to the bunch of crapywhite-apple-earbuds people around... I'm sure they will look me/us as if we were insane...