This review page is supported in part by the sponsor whose ad is displayed above
In response to an online poster's query on "why NuForce has audible hissing noise when the amplifier is idling (no input signal)?", the company's Jason replied: "Our NuForce amplifier has a S/N ratio of 120dB @ 100 watts and so is exceptionally quiet when playing music. NuForce's feedback and control system runs at 1MHz and actively performs noise cancellation in real time at every switching cycle (comparing noise with music signal) .But when there is no input signal, it merely amplifies noise in the system and occasionally you can hear a swash/hissing noise if you are very close to the speaker. When the source device (preamp or CD player) is turned off, it acts as a noise receiver/transmitter and could generate even more noise for the NuForce amplifier during idling time. Because our switching amplifier uses hundreds of components in its circuit, minor deviations in parts values may mean that some units have slightly more idling noise than others."


In response to "why is there a small DC offset between the two +/- speaker terminals", Jason explained that "the op amp has variations and will result in 0.05 to 0.13 volts on the speaker terminals. 0.13V is equivalent to 2mW into an 8-ohm load. This is hardly a concern for your speakers." With regards to why some units exhibit turn on/off noise, Jason explained that "the low level turn on/off noise (pop), even though sounding unpleasant, is no voltage spike that could harm your speaker."

With extremely efficient speakers, I was in an unusually heightened position to check on these matters. I'd be able to determine readily whether the poster was hyper reactive to minuscule operational noise floor -- only audible with one's ear glued to a tweeter -- or whether there was anything of more practical concern. The amp does indeed have a bit of a turn-on noise. However, it's nothing like a thump. It's more of a strange digital fart. It sounds a bit like rustling or tearing cellophane paper. Truly, it's nothing to worry about and over very quickly. If you're allergic to -- non-odiferous -- farts, leave the amps permanently on. Regarding noise (no signal, preamp not muted, volume at average listening levels, source turned on but not playing), I heard nothing with my ears as close to the Zu Definition drivers as humanly possible. And I mean nothing as in 0.000%.


For a 26dB 150-watt amplifier into 101dB speakers, that's excellent -- and quite unexpected -- performance. It's also better than most any 8wpc tube amp I've ever had through. And not just on one Ref 9 but both. No fluke then, this silence. As far as I'm concerned, when there's proper grounding, noise with these NuForce monos is a complete and utter non-issue (and I'm quite a stickler on the subject)! As is en vogue, the power LED is blue, quite bright but, unless at eye level, partially obscured by sitting inside the cosmetic recess.


With some ICEpower-based amps that came through here a lot
earlier, I had noticed a strange phenomenon. It was inner-ear pressure like ultrasonic ringing. It was somewhat similar to the high-pitched 16kHz squeal you heard when an older television got turned on - except this I couldn't hear. I only felt it as pressure sensation on the ear membrane. This would subside the moment the amps were turned off. I blamed it on insufficient filtering of that particular implementation at the time. Remembering, I paid close attention to the NuForce monos and noticed no such thing. While on the subject, a reminder seems in order. The moniker digital amps is a complete misnomer. This would occur to anyone asking how come they have analog inputs and if so, how come they don't use A/D and D/A converters internally. With all the bashing 'perfect sound forever' has bequeathed on the terms 'digital', associating the latter with amps conjures up nasties like brittle, bright, flat and sterile. It would be best to disassociate the two -- amps and digital -- which, I believe, is why NuForce insists on calling theirs an "analog switching amp". That is the correct term, albeit not just for NuForce but for Tripath and ICEpower and Hypex and others of the same ilk. (TacT and Sharp's OneBit might properly be called digital but we'll let the engineers worry about that.) With those cobwebs cleared out of the attic, let's address a few more that relate to NuForce's website statements as of September 20. I've talked to Casey at NuForce about what is to follow. I suggested they revise certain statements to become dead-on accurate rather than "pushing the envelope" by incorrect implications. To wit.


In their promotional lit, NuForce prides itself on exploded bandwidths. Their amps are deliberately bandwidth-limited to 50kHz. That's an octave higher than RedBook cut-off and smack in the heart of SACD extension. That's laudable but certainly not unique. Most Pass Labs amps are down 3dB @ 100kHz, for example. Goldmund amps reach to 3MHz at full power. That's ultra-wide bandwidth. 50kHz is not. NuForce extols their high damping factor with "the higher, the better". Categorically not. High-efficiency light-cone big-magnet drivers such as Lowthers want to see high output impedances in order to not suffer overdamping and premature bass roll-off to the tune of up to 15dB. That's why the FirstWatt current-source amps have Z-outs of up to 50 ohms. Granted, such drivers are in the vast minority on the market. Still, categorical statements deserve closer inspection. Citing a damping factor of 4000 as though this automatically implied superiority regardless of speaker load is simply not factual. Overdamping is just as objectionable as underdamping and much in this regard will depend on drivers and cabinet loading, i.e. sealed, vented, resistively vented, transmission-line, back-horns and their various permutations. (Incidentally, talking to an ICEpower manufacturer, I learned that his amp shows a 4 milliohm output impedance at the amp's speaker terminals at 100Hz, making for a similarly astronomical damping factor as NuForce. However, once you add the load characteristics of an average 8' speaker cable, you should expect a significant lowering of such damping factors. That's not to deny their existence but to be realistic about what the speaker actually sees through the connecting cable.)


NuForce also claims that all amplifiers use output filters and induce phase shift. Again, not true. Class D amplifiers must use filters but traditional linear amplifiers do not. Well-designed linear amplifiers don't suffer phase shifts but ICEpower for example does inject -70 degrees @ 20kHz/4-ohm for one of their boards as imposed by the output filter. The NuForce claims thus do apply to their direct competition -- so-called digital Class D amplifiers -- but not to traditional 'linear' amplifiers.


NuForce also makes much of their switching power supply and how it renders traditional linear power supplies obsolete or at least outdated. SMPSs are by design regulated whereas
linear power supplies can be but don't have to be. That is an advantage of SMPSs. They're also far more efficient. However, they do generate massive HF switching noise which needs to be filtered. In the end, there's no free lunch.


The Reference 9 was authored to provide more current reserves into lower impedances than the Reference 8 whose power supply is current-limited to 100 watts. The Reference 9 is voltage-limited and thus provides far higher current delivery into the 3 to 6-ohm loads most speakers present in the bass. That's despite its 8-ohm power output which, being very similar to the Reference 8, could compel potential shoppers to wonder what the Ref 9 offers extra. More raw grunt into he-man speakers, that's what. Lastly, the custom Stealth power cord for the Ref 8 can cause the appearance of a rising HF response if used on the Ref 9. Similar tonal shifts could occur as a function of power cords with floating shields. If you notice undue treble bite or energy, Casey at NuForce recommends you swap power cords and begin with the included black cord. It includes a ferrite core for EMI shielding and will give you the "factory default" sound from the Ref 9.


Lastly, NuForce has an upgrade for the connectors - WBT NextGen RCAs [$55 per amp] and Eichmann binding posts [$25 per amp]. These are low-mass connectors and Casey feels they can subjectively mitigate HF brightness not by causing any rolloff at all but by removing a certain "smearing" that could translate as treble hardness in certain systems. With regards to the balanced and single-ended inputs, the Ref 9 is a fully balanced circuit and the single-ended inputs are run hyper-phase to convert to balanced inside. The only measurable difference NuForce can detect between XLR vs. RCA feeds is lower THD on the former, likely a function of the raw connector. Whether that's audible remains to be discovered. Casey did not make any statement of balanced-input superiority. Unless listening tells me otherwise, prospective customers probably shouldn't worry. Instead, merely consider the XLR/RCA option a convenience feature. It accomodates
preamps of either persuasion without requiring cable adaptors. No need to "go balanced" in the fear you'd miss ultimate performance - or so it appears at present. Listening impressions to follow, both with the stock unit and then the connector option which will entail shipping the amp back to California for a quick mod.