Headphone Reviews

The Orpheus was, however, unusually effective in keeping the soundstage from clustering inside my head. It wasn't completely successful in this—no headphones in my experience have been—but the soundstage seemed less of the clothesline-between-the-ears variety than is typical, extending instead from my left ear, across an arc just in front of my nose, then around to my right ear. This is more a matter of spectral balance than of any signal manipulation. The diffuse-field compensation referred to in Sennheiser's specifications refers to the design of the earpieces themselves—no electronic compensation is performed in the PEV 90 amplifier.

Stacked against the Stax
I've already noted that the Orpheus is several times the cost of the next-most-expensive headphones I know of—the $2000 Stax SR-Lambda Pro Signature. Like the Sennheiser, the Pro Signature comes with a tube amplifier whose design appears far less elaborate than the Sennheiser's—though this isn't necessarily a point against it. Fortunately, we had a sample of Stax's flagship headphones available for direct comparison with the Sennheisers. How did they "stax" up?

Overall, I found that the Orpheus, while a bit more tipped-up in the top octaves, sounded sweeter. The Lambda Signature was more forward in the low to mid-treble. On difficult or overbright recordings this difference benefited the Orpheus. On Leo Kottke's My Father's Face (Private Music 2050-2-P), a clean, highly detailed, but also rather etched-sounding recording, I was shocked to hear how much more listenable the Sennheiser was, without stinting in any way on clarity. The Sennheiser clearly won out in lack of grain and overall smoothness. The Stax's more pronounced mid-treble gave it a more smeared, less pristine quality. In contrast, the Sennheiser better separated the musical threads, clarifying detail in a way in which the Stax did not. The Stax had a bit more sonic weight—noticeable primarily on vocal tracks—but this didn't come close to counterbalancing the Orpheus's advantage through the top end on this recording.

On better, more naturally miked recordings, however, the verdict was much less obvious. On Mokave, Volume 2, I heard generally the same differences in balance, though this time the game was much closer. The Stax sounded brighter with slightly less top-end air but a noticeably (though not dramatically) deeper, tighter bass. Here I preferred the overall balance of the Stax, but only by the narrowest of margins. The same was true on a succession of good recordings. The Sennheiser's top end, despite a bit of emphasis, sounded notably more airy and silky. The Stax was livelier in the low to mid treble and a bit more immediate, which did give it a somewhat more analytical, but not aggressive, quality. Its added forwardness also brought the sound a bit more into my head, with a slightly less natural perspective than the Orpheus.

But the Stax's bottom end remained superior. The bottom was there in the Sennheiser, but the bass from the Pro Signature was more convincingly real. Vocals were more naturally fleshed out with the Stax—not by much (and certainly not to the degree of the Koss ESP/950 sampled earlier, which, for my money, remains the champ in vocal reproduction), but just enough to give the voice a more solid foundation.

An interesting comparison: lesser 'phones might envy even the weaknesses of the Orpheus and the Lambda Signature. I still long for a headset combining the best of both—the Stax's lower bass extension and tightness combined with the Orpheus's clearly sweeter, airier top end. But even Class A (where the Orpheus clearly belongs) does not denote perfection.

So far I haven't addressed the D/A processor built into the Orpheus's HEV 90 amplifier. All of the above observations, as already noted, were made by driving the HEV 90 at its analog inputs from a Krell Reference 64. I admit that I find puzzling the presence of an internal D/A processor which can only be used with the headphones. It has no analog outputs separate from the five-pin headphone jacks (which is why you'll see no measurements here), so it can't be used with, say, the rest of a system for driving loudspeakers. The only possible explanation for this is that Sennheiser believes the Orpheus will form the heart of an exclusively headphone-based system.

If you wish to also use loudspeakers, you'll need an outboard processor or all-in-one player. If you have such an outboard processor or player—and it seems unlikely that any Orpheus buyer will cheap-out on the latter—then why would you need the built-in DAC unless the latter is a knockout, performancewise? And if it is, why not give the user other options for its use?

Just how good is the internal converter? I drove it from the same Krell DT-10 transport which had been used with the Reference 64 processor above, but this time from the DT-10's S/PDIF digital outputs using an ART coaxial digital cable. Since the Reference 64 was driven from the DT-10 using ST optical cables, it was possible to continue driving the Reference 64 simultaneously with the Sennheiser's internal processor, both from the DT-10, keeping the Krell D/A's output plugged into the Orpheus's analog inputs. This made a direct comparison of external and internal processors a simple matter of throwing a toggle switch on the rear of the HEV 90. The only problem with this arrangement was the slightly higher output of the Krell processor; level matching had to be done on the fly, and with less accuracy than I normally prefer. There's no convenient means of metered level matching with such a setup.

In any event, no one is likely to buy or not buy the Orpheus on the basis of the quality of its internal D/A processor. Still, it's a very good one. It has the same weaknesses I've heard on other occasions from Bitstream converters—a slight graininess at the very top end and a subtle lack of life and liquidity compared with the best multi-bit D/As, including the Krell. But I do mean slight and subtle. I did prefer the sound of the Orpheus driven by the Krell processor, but in no way did the difference in performance come close to reflecting the difference in cost. I haven't heard a Bitstream D/A yet that sounds any better than the one in the Orpheus. That doesn't change my opinion of its ultimate usefulness. It certainly increases—perhaps substantially—the Orpheus's price, but since you can't buy the package without it, consider it a bonus.

Conclusions
The Orpheus is in a very different price category from the Stax headphones, and that, indeed, is my most serious reservation about it. It's clearly Class A, but it doesn't redefine Class A in all respects—something you'd almost expect considering its cost. Still, for those of you who fall in love with its pristine, airy sound, withdrawal will be painful unless your checking account balance is such that a $12,900 debit is just another routine entry. But for the rest of us, Sennheiser has available a much less expensive model (at under $2000, competitive with the flagship Stax) which makes use of some of the same technology. Something clearly had to go to bring the price down—the prime deletion is that luscious-looking tube amplifier. Still, we're hoping for a review sample, and may even have received one as you read this. Stay tuned.