CROSSFEED: software vs. hardware
Jan 19, 2004 at 10:06 PM Thread Starter Post #1 of 18

slunk007

100+ Head-Fier
Joined
Dec 17, 2003
Posts
307
Likes
10
Hey everybody,

As I am possibly investing in a PIMETA, I am wondering if there are any benefits from using a software crossfeed (such as foobar's) compared to a hardware crossfeed (built-in to the amp's circuitry), or vice-versa.

I use my PC as my only source for music listening while at home. Recently I switched from using Winamp2 to Foobar 2000. I must say that I notice a slightly better SQ using foobar. It's cool and I've customized it and it looks pretty sweet now. Then I heard about crossfeed (available for both winamp and foobar) and I've read about what people like about it and all that. I've read about how some people simply cannot listen to headphones without it, whereas some people fall into the "i don't even notice that it is on" category. So I tried it out on foobar2k. With my musical tastes, I honestly found nothing spectacular about the effect whatsoever. I could barely hear a difference with it on or off. I would definitely say that this effect is subtle.

So is there a qualitative difference between crossfeed dsp (on foobar2k) compared to the crossfeed built into an amp? Or are the differences so subtle that either option would be fine to use?

Any help on this subject would be greatly appreciated.
 
Jan 19, 2004 at 10:11 PM Post #2 of 18
Listen to it a bit longer and try not to focus on sonic differences but more-so the differences in how natural the sound is being presented.

While it's very subtle, I think that's the way crossfeed should be.
It should really only make things sound a bit more natural.

I don't have problems with headaches, but I do appreciate what crossfeed does.
 
Jan 19, 2004 at 10:28 PM Post #3 of 18
Doing signal processing in digital domain has it's advantages but foobar's crossfeed isn't really customizable. Some find if too much and other find it doesn't do anything. Just play it by ear and see if you like it.
 
Jan 19, 2004 at 10:31 PM Post #4 of 18
I am a big fan of software crossfeed, becouse:

1) it's computed with 64bit precision and shoudn't have any nasties of analog implementation like quality of caps, signal way lenght, many solder joints etc. in fact you cannot build hardware crossfeed to exactly meet the computed specs, you just get close to them..
2) it's for free
3) you can turn it on and off very simply in the most convinient way

the effect is subtle or no on some material, but may be huge on other.. try for example some older Beatles or Rolling Stones recordings
smily_headphones1.gif
to me the effect is like spreading bass into the whole space, it's not comming from left and right can anymore.. very pleasant effect, resembling speaker listening session.. if it doesn't change the sound significantly it's only good as you won't loose any detail, you get just the benefits of it
wink.gif
 
Jan 19, 2004 at 10:43 PM Post #5 of 18
@Ian:
Do you know of any better software solutions? I would be interested.

@Glassman:
Yes. Free is good. Also I did hear that hardware crossfeeds attenuate the signal slightly. But is it really noticable?
 
Jan 19, 2004 at 10:47 PM Post #6 of 18
yes it will definitely attenuate the signal, but hey, you're using it before amplifier
wink.gif


if you want to tweak the crossfeed plugin, there's nothing easier than asking it's developer for some presets or something like that, but I'm quite happy with it's current state..
 
Jan 19, 2004 at 11:02 PM Post #7 of 18
I am not aware of any other crossfeed plugins. I havn't looked though as I don't need crossfeed.
 
Jan 19, 2004 at 11:07 PM Post #8 of 18
Okay... I think I now have a better idea as to what crossfeed sounds like. I put on some Beatles (Come Together) and without the crossfeed, I can certainly hear the bass coming from my left ear and the drums coming from the right, and the vocals coming at me from the front. When listening, the bass and the drums would make my eyes flinch with the impact. But it felt like someone (ringo) was playing my right eardrum as a drum. I turn on the crossfeed and give it another listen. This time, the sounds from extreme left/right are still there, but the impact of the playing is subdued and my eyes don't do that flinching thing. The sound does seem a little more "full". Now I more clearly see the difference between crossfeed and no-crossfeed.

I don't think I would use it all the time (most of today's recordings have a bit of crossfeed in them to begin with from what I've read) since I don't listen to many recordings that would require having crossfeed turned on.

One thing I would like to hear about still is some more opinions regarding hardware versus software crossfeed.

Thanks guys... keep 'em coming!
 
Jan 20, 2004 at 2:50 AM Post #9 of 18
Slunk007:

Allow me to give my opinons. I too have the latest incarnation of Foobar2000 and I have indeed used the software cross feed option in conjunction with my laptop. While my laptop does not nearly sound as good as my audio rig, I did notice the differences between the .wav / .mp3 files being played straight vanilla (i.e., no cross feed) and with it enabled. That's the key distinction -- different. Different does not necessarily mean "better" in my humble opinion because I define good sound as not necessarily being neutral / transparant / reference nor colored but my affinity for that sound. It's all about how well does it allow me to "get my fix" of music! You're right about one thing: most rock / pop recorded / mixed today doesn't sound like it needs cross feed be it software / hardware. Perhaps older recordings do but that's where I draw my line in the sand with respect to headphones vs. speakers: a headphone is supposed to sound like a headphone (i.e., detail, stereo separation, intimacy, portability, closed design, etc.) while speakers should sound like speakers (i.e., soundstage, bass impact, a cohesive sound, room filling, etc.). Hardware cross feed as implemented by HeadRoom Corporation doesn't sound too good; it muddies up the midrange by introducing blurred imaging and lifts the lowest bass notes up by 1/4 - 1/2 an octave depending on the album while artificially pooling together the sound in the middle of your brain. With all due respect to myself, I found this appealing in my inchoate stages of headphone Hi-Fi but I realized that there was better to be had. Part of the headphone Hi-Fi journey leaves indelible marks in how you prefer / perceive sound later down the road. If you are still reading this, then you get my drift: I do not like cross feed anymore. It makes the sound sound unnatural. I'm not trying to belittle anyone but I think that my brain is smart enough to know where to image sounds in my mind and what a headphone should sound like as compared to what a speaker should sound like too. Lastly, cross feed falls under one of those sophisticated tweaks that has a small but audible affect on how you perceive sound but it is not the most important criteria by which to judge whether your next headphone amplifier will sound good. Design, engineering, parts quality, and final build quality are the true benchmarks by which to judge any headphone amplifier. These are my opinions.
 
Jan 20, 2004 at 4:52 AM Post #10 of 18
I personally use hardware (low Linkwitz) crossfeed since the foobar crossfeed is a bit too extreme for me. But hardware crossfeed is a good addition if you're going to be using the amp portably, since very few portable devices come with a crossfeed option.
 
Jan 20, 2004 at 5:09 AM Post #11 of 18
Thanks Welly Wu. I appreciate your fresh perspective on things. I think for me, the only reason I would opt for a hardware crossfeed would be potentially to increase its re-sale value if I so chose to get something better in the future, it wouldnt completely kill the wallet in one fell swoop.

I have to admit. I'm not blown away by crossfeed (on foobar anyway). If people are saying that foobar's crossfeed is a little too strong, I definitely don't see a point in getting a hardware crossfeed. Given my current (and past) musical tastes, I don't think I would find any huge benefit from having a hardware crossfeed.

This is not to say that I don't hear a difference. I'm just saying that I don't listen to enough music that needs crossfeed to use it.
When I used it today, I definitely saw it's purpose... and it does a good job for that niche, IMO.

If I ever feel the need to use it, I'll pop open foobar.

Thanks guys.
 
Jan 20, 2004 at 12:15 PM Post #13 of 18
Quote:

Originally posted by slunk007
Do you know of any better software solutions? I would be interested.


Here I've posted a comparison of crossfeed-like winamp/foobar plugins I know. You might want to read ASDFer's posts. He has coded a HRTF plugin for foobar2000. It uses impulse responses to generate the effect (it's meant to work with freely available head related transfer function measurements aka HRTF/HRIR). By using custom impulse responses you can modify the effect as you want, you can even copy the crossfeed effect of other software or hardware. Because of this I don't see a chance that anything else can sound better.

My approach on this: Everytime I listen to binaural recordings I'm amazed by the way sound sources are located in the sound stage. When closing my eyes it feels like I could reach out and touch the sound source. I expect from crossfeed-like DSPs to do something similar: move the sound sources outside of my head in front of me to defined positions in soundstage - similar to listening with very good speakers or even live.

Unfortunately ASDFer's plugin is still somewhat limited (to impulse responses with a length of 512 samples), so I wasn't able to try with customized longer responses (simulating room echos etc.), but chosing the right HRTF set, for me the result is already better than anything else I heard so far.
 
Jan 20, 2004 at 12:25 PM Post #14 of 18
Quote:

Originally posted by blessingx
Best software crossfeed is the Canz3D VST plugin, but it's unfortunately OS X only.


In case you still have/use it: Could you please create an impulse response from it with your favourite settings?
Just in case - how to do it:
Generate a uncompressed stereo audio file (was it aiff for mac?) containing silence, something like 2 seconds. Put a single sample in the middle of the file in *one channel only* to max value (32767). Run the result through the plugin and record the output.
 
Jan 20, 2004 at 12:55 PM Post #15 of 18
hi tigre, does this HRTF plugin do something different than Convolver? aren't they about the same?
 

Users who are viewing this thread

Back
Top