CanJam at RMAF - Outstanding Product: The SPL Phonitor

SPL Phonitor
I've listened to just about every cross-feed circuit out there. I've twiddled and tweaked long hours to make it do it's thing and not get in the way of fidelity. It's not easy. But the SPL Phonitor just blew me away with its outstanding and fully adjustable cross-feed circuit.

Doing cross-feed well is very difficult. It tends to murk up the sound a bit. Any common bass information on both channels tends to build up when summed back on itself in the opposite channel, which results in a slight warming of the sound. Not so with the Phonitor. I spent a good deal of time playing with the cross-feed adjustments on this amp and I simply couldn't perceive any tonal changes while diddling.

Hearing cross-feed changes is difficult, the results are subtle and it takes a trained ear to tell what's going on. An easy way to listen to the effects of a cross-feed circuit is to remove either the left or right input to the amp and listen to the resulting display of what one channel is doing. When I did so on the Phonitor in order to be able to hear clearly what the device was doing I was immensely pleased to hear exactly what I would like to hear with such a circuit.

The Phonitor (~$2000) has adjustments for speaker angle (time delay), cross-feed level (how much signal is mixed into the opposite channel), and center image (L+R level). Having made a similar experimental unit when first playing around with cross-feed circuits, I can tell you that SPL must have done some very cool things inside their box in order to pull of this trick and still leave the music clean and clear.

There's a lot of contention in Headphonedom about whether cross-feed is needed or not to create a convincing audio image. Many claim that their head becomes accustom to the sound of headphones and have no problem hearing a solid image. I disagree. Having an inter-aural time difference (ITD is the arrival time difference from off-axis sounds at the ears) is fundamental to left-to-right aural localization (sometimes called lateralization). I think the lack of belief that cross-talk circuits are valuable for headphone listening is fundamentally due to the fact that all such circuits I've heard in the past either don't technically achieve the correct delay and mix of the cross-talk signal, or end up changing the tonality of the music.

I think the Phonitor may change all that. My experience was that I was able to adjust the cross-feed circuit in this amp easily, and was able to develop a very natural and organically correct listening experience without changing the tonality of the music one little bit. Just extraordinary.

I'd have to have some listening time here at home to comment on the clarity and articulation of this amp, but on first listen it seemed very good. I'd highly recommend that folks looking for better imaging on headphones have a listen to the Phonitor and it's excellent cross-feed circuit.

Edit: I should add that I don't consider the Smyth Realiser a cross-feed circuit. It's a very complicated, DSP-based, HRTF (head related transfer function) generator that does an even better job of producing a realistic headphone image presentation, but is far more ambitious and complicated a device.

mward's picture

I find crossfeed makes headphones significantly less fatiguing. The HeadRoom models I've used don't sound like speakers, per se, but they eliminate the extremely irritating feeling of hearing a sound in one ear but not the other, which I find feels almost as if someone is tickling my ear canal.

Seems like there are a lot of great headphone amps out there, but it's hard for me to contemplate owning most of them since it involves giving up crossfeed.

Good to see ongoing work in this area.

donunus's picture

I am a huge fan of crossfeed! I have been playing with a bunch of software plugins for a while and none of them get it quite right even with countless number of hours tweaking the settings on each one of them. They either make the sound too murky and lifeless(BS2B), Thin and sterile (Naive software crossfeed for foobar), have too subtle an impact on soundstage when setup to minimally color the sound(Headfit and Vnophones), Too phasey and too dramatic in coloring the tonality (Isone pro), and too dry sounding and colored (redline monitor). So far, I'm trying the stock crossfeed plugin of J River media center and its sounding good at the subtle setting but I have to listen more to be able to conclude that I have found the holy grail in a simple crossfeed setup.

I've also tried some hardware implementations like the one on my past headroom and meier amps but none of them are transparent enough for me.

The Phonitor has been an amp that I have been drooling over for a few years now. Too bad I don't have the moolah to buy one because if I did, this would have been the first amp in the higher price ranges that I would instantly buy.

13mh13's picture

Can't recall whether it was in one of HR's Stereophile famous 'info-ads' or a write-up from an old HR catalog ... but a major re-vamp (IIRC) of the orig. HR xfeed processor occurred (wild guess) 10 yrs back.

I recall the write-up mentioning you (Tyll)/HR brought back an orig. engr. (??) -- that dude locked himself up in a room for a few weeks (you fed him pizza boxes from under the door!) and came up with something new and improved. IIRC, the guy emerged from the room and noted something like ... okay ... it works ... and works better ... but I'm not really sure why! Who was that guy? What major changes to the xfeed occurred?

donunus's picture

Really? I have only heard the crossfeed of the headroom "little more power" that I had so I guess I haven't heard this newer crossfeed. I wish Tyll could have some programmers design a software version too :)

Tyll Hertsens's picture
It's been tweeked numerious times over the years, and the better the quality of components used (which changed over time) also seemed to improve the result.
13mh13's picture

In this site's forum, I posted on this general topic a few weeks back ... IAC ...
I've tried a lot of XFs as I really don't like the in-your-head fatigue that plagues nonXF amps. FWIW, HR's processing, to my ears, improves the fidelity of the acoustic (sounds richer, fuller; not talkin' 'bout HRTF imaging) -- pretty much the opposite of what "tradit." passive ckts do ... IME, anyway. E.g., Meier and the modified Linkwitz I use in some DIY digs.

What I'm not sure about is whether HR's non-XF ckt (aka std.) is 'unconventional' (so sounds worse than other non-XF, like Meier, etc.). This may 'splain the improvement in SQ (again I'm not talking about HRTF imaging) when switching to HR XF.

I haven't heard 'em all, and the new Phonitor has me Jonesin' ;) Hopefully, some DIYer will find a smaller, more portable, more-AFFORDABLE hacked sol'n.

Methinks active HRTF/XF is probably best, and the passive-is-better mentality is behind-the-times. I think ;)

donunus's picture

Now I am thinking that I may be behind the times but I think I have to get at least a headroom micro to see what the current state of the headroom crossfeed is.

Tyll Hertsens's picture
It's best in the "Ultra" versions.
donunus's picture

I did have a lot of interest in the ultra micro before but the product suddenly got discontinued.

donunus's picture

Did you mean it was better in the ultra because of the extra filter switches or just because of the better quality parts used?

Tyll Hertsens's picture
Better parts.
donunus's picture

By the way Tyll, have you tried the redline monitor vst plugin? I was wondering if the phonitors xfeed is much more transparent than this. It seems to have the same type of controls. Distance, angle, and center image.

Tyll Hertsens's picture
Haven't tried it, sorry.
jackhamm's picture

Hi Tyll, Wondering if you have any plans to do a review on the Smyth Research Realiser?

Tyll Hertsens's picture
I would love to. I'll work on it.
FSonicSmith's picture

Tyl; please tell me you got a chance to try the LCD-2s (Rev 2) with the Phonitor and good or bad, please let us know what you thought.

nunh's picture

If I had an extra 2k.... :)

mnaganov's picture

I've took a brief look on SPL's crossfeed frequency and phase response, and the characteristic of their the filter that is applied to the crossfeed channel are drastically different to what people usually do based on the description of XF provided by Bauer initially and then by Linkwitz (Chu Moi's circuit, Dr. Meier's amps, bs2b DSP plugin). In short, SPL's implementation introduces much less alterations to the phase and to the group delay, and it's a bit more complex in terms of frequency response alteration. I agree that the result sounds much more natural, I prefer SPL's XF to any other. I was also amazed how clean their analog filter is--my naive attempt of reproducing it in a DSP resulted in much bigger phase and group delay variations.