Rossini direct to tubes?

Hi. I’m thinking of moving to mono blocks and wanted to ask some advice:

(A) is anyone on the forum running the Rossini direct to tubes, i.e., no pre-amp?
(B) what monos are you running?
(C) what other brands/amps did you consider?
(D) any thoughts on C.A.T.?

Thank you in advance!

Just a couple of points to bear in mind:

Few tube amps have balanced inputs so you will need to use Rossini’s RCA outputs which do not sound as good as the XLR balanced ones. However I have no information for the new
XLR v. RCA outputs on the Apex version.

Not using a balanced connection may have a consequence regarding noise pickup especially if the cable run from Rossini to power amp is long.

Thx Pete. Understood re: XLR/RCA. It’s a short run.

@James @Phil I would be interested in hearing comments on this one as I (and several others here) use our Rossini to front our Naim systems. As such I connect with a Chord RCA > DIN cable.

Gregg, you may have forgotten but you asked this a couple of years ago and James responded:

As he points out the preference for using the XLR output is covered in the FAQs.

Of course with a Naim amplification system you either have to use RCA/RCA or use a converter cable. However with a converter you lose the advantage of CMRR noise rejection as although the result provides an electrical connection it is no longer balanced. As I said at the time such a cable is nevertheless preferable to adaptor plugs.

As just about everything in this hobby it is a case of fitting theory to the practical reality of what we actually have so compromises often need to be made.


Ah…good sleuthing Pete. I forgot about that as it was many years ago.

With that said I wonder if things are different now that I am running

  1. a Rossini
  2. which is now APEXED?

I see @James reply is back from 2019 and perhaps refers just to Bartok….and pre APEX?

Certainly your posting some years ago was pre-Apex. Of course you wrote with your forthcoming Rossini in mind.

The point about loss of noise rejection without a balanced connection stays true but the inherent sound quality of Apex RCA v. XLR is for me, as above:

Although I have a Vivaldi Apex here unfortunately I can no longer easily carry out a comparison so I look forward to anyone else’s opinion.

I run my Vivaldi stack directly via XLRs into a Tenor OTL Hybrid; pure Tube input stage and a MOSFET output stage in a fully balanced differential design.

Prior to that I had the Vivaldis direct into a pair of Wavac MD-805m SETs; obviously a single-ended design which was via RCAs.

And prior to my Vivaldis, I had the dCS Scarlatti stack directly into a VTL S-400; via XLRs as it’s a fully balanced design. That was also pretty glorious sounding too.

I’ve never heard the CATs, but IMHO dCS direct tends to be so transparent and dynamic that if you really like how the CAT Triodes sound, then don’t sweat that it’s non-differential, the RCAs out of the dCS units are still pretty class-leading. So, go for it :wink:

By the way, I googled “CAT Amp Rossini”; looks like The Audio Beat did a CAT Amp review with the Rossini as a source that you might be interested in reading;

1 Like

Hi Guys,

There’s not really an easy or simple answer to this question…

The first point to make is that balanced connections are all about reducing the amount of noise picked up between the two units. In that sense, whether it’s balanced or not is immaterial to the sound quality where the cables lengths are modest, the physical environment isn’t harsh and the amount of noise picked up should be low.

Given that, one could surmise that a balanced connection is going to be detrimental to the sound quality as the balancing/unbalancing might be done by a transformer, which is definitely not an audio neutral device, or else if it’s an electronically balanced circuit, there are twice the number of amplifiers in the output stage and they have to be carefully matched. There may also be some cancellation of even harmonics generated in the output stages (which could be deemed as being positive or negative) if the balanced audio output isn’t given the attention it deserves - but we feel that we do a very good balanced output which should be up there with the very best.

In our implementations, the balanced output impedance is much lower than that of the unbalanced. In THEORY this means that cable (etc.) capacitance will tend to roll off the frequency response sooner with the unbalanced than the balanced output stage however in practice even an implausibly capacitive cable would only start to roll off things above 1MHz, so it’s hard to imagine that having much effect in reality given the general response of amplifiers, speakers and ears in the MHz+ range.

The final factor could be the input stage of the destination equipment which is something we have no control over and which can be expected to vary a lot from one piece to the next. Implementing an unbalancing stage with valves is quite involved. Sometimes this is skipped and the XLR “hot” connection is simply connected to an unbalanced input (grounding or leaving open the unused leg). Some may use a transformer and transformers are typically not “audio neutral” components (as above). They may well add distortion at low frequencies and roll off high frequencies, the former of these can definitely have euphonic properties so it could well be possible that some listeners who express a preference the “sound” of the balanced output simply prefer the audio signature of the balanced input of whatever they’re driving.

So, to sum up there’s no easy answer and lots of variables, many of which are beyond our control but there is no intrinsic reason that one connection will “sound” better than the other…

Myself, I use the RCA outs on my dCS kit here at home into the unbalanced (but XLR) inputs on one of my NAP300’s and I’m really happy with the sound apart from I wish it went louder, I do have a NAP500 but I prefer the sound of my NAP300.

(Yes, Naim use XLR’s on the NAP250, NAP300 and NAP500 but for UNBALANCED audio, not balanced audio.)

I hope this helps…



This may be thread drift however may I ask what kind of cable you are using between your dCS and 300? If it is of your own making can someone like Chord fashion one on order? If you recall I am in the States - 552/500.


Hi Greg,

Here’s where I suddenly sound like a total heathen - mine are “home made” and are simply a set of RCA to RCA cables (I’m afraid I don’t know the brand but highly unlikely to be anything “high end”) that I had sitting around that I whipped the ends off and added a set of Neutrik XLRs to.

They definitely could make up a set for you if you wanted (but they’re not difficult for anyone that knows which end of a soldering iron to hold). Not sure what the deal would be with shipping though so you’d need to check with them.


Thread hijacked in less than 12 hours!

Anyone else beside @Anupc running tubes or similar direct from Rossini?

I’d love your thoughts.



This post from an amplifier manufacturer may be useful. They describe how they implement balanced/unbalanced in their amps and pre amps and how CMRR is archived. Whether tubes or solid state I believe the issues are the same. From post 2.

don’t know if this helps but I just skipped my pre-amp and connected the Rossini Apex directly to my poweramp (pre/power both Nagra). Even with the Apex (a huge improvement in SQ) the preamp in my chain sounds better. Better means more details, more flow, more soundstage, more “3D” impression… My guess: if you have an entry- or good midrange -priced preamp then skip this and go direct to poweramp. But this is really a matter of taste of course…

1 Like

this is a nice one :slight_smile: Innuos Statement + Bartok + D’Ago + Magico (skipped preamp obviously)



Sent you a message as i have my old silver cables i used when i had my naim 500 directly to my dcs

That system just didn’t sound great at the weekend, it was quite flat sounding.
Asked if it was an apex bartock, and the guy didn’t have a

wow- never thought that! With this price tag it should sounds like live Disco!

I’ve shared my experiences with running my Bartok direct to my power amp in other threads - here is a quick summary of my subjective observations. While my amp is solid state, many of the considerations are the same for tube users.

  • I’ve had the best results running direct between -10dB and 0dB - using either .2V or .6V out into my amp. Many Tube amps need higher voltage to reach rated output so one may be able to use the 2 or 6V setting depending on hardware and listening conditions
  • Sonically speaking, it is a mixed bag - running direct is more transparent. Warm recordings sound warm but clear, bright recordings sound even brighter. Bad recordings have nowhere to hide
  • Low bass and micro-dynamics are simply better running direct. Soundstage is narrower but deeper. Highs on great recordings sound so natural
    Subjectively speaking… on some well recorded - DSD transferred analog material; running direct sounds more like reel-to-reel than digital
  • The dCS digital volume control is simply wonderful. It tracks perfectly and left right balance is rock solid. My preamp has a digital controlled analogue circuit that measures very well, but the dCS feels more secure
  • Running direct via balanced, is much quieter with audibly less hum and hiss - this is extremely important in my room which has a 25 to 27dB ambient noise floor.
  • The down side is that my pre amp seems to add a bit of mid bass warmth, and what some may call ‘tonal saturation’, that makes less than favourable recordings more tolerable.
  • Running direct means no turntable at the moment