I have a few to try on my list in a years time. Would of loved to of got hold of the Sega as I think it be a cracking amp but looks highly unlikely so the Benchmark and Belles are a few that be on my list

Now I realise that on a civilised forum like this one nobody shrieks "shill" like a banshee, but both John and Chris appear to have had a dizzy spell earlier in this thread and needed their punka-wallah's to increase their work rate to keep them cool on fear of "promotion on behalf of a manufacturer".  As noted elsewhere, the idea that Benchmark, or the amplifiers designers and patent holders - THX Corporation, part of the George Lucas group are

1) interested in promoting their product on an,  ... whats the word - elite and select (but not large) HiFi forum
2) Connected to me

are as amusing as Bunsen Honeydrew picking up a role in the next star wars movie.  Oh, ok, I'll come clean - George Lucas is my mate really, and I wrote half the Star Wars films with him over a pint here in Yalding.  (It's about as credible)

Back on thread, assuming we are interested in "new stuff" - Martin - regarding your interest in SMPS, I am merely retyping from page 37 of the AHB2 instruction manual...

"
Resonant Switch-Mode Power Supply - Most high-end power amplifiers use unregulated linear power supplies.  The conventional wisdom is that linear supplies are electrically quieter than switching supplies.  Benchmark's research shows that this is a myth.  Linear supplies have transformers that operate at AC mains frequency.  The problem with these 50 to 60 Hz transformers is that they must be very large in order to supply the required power.  These large low-frequency transformers generate large stray magnetic fields that can interfere with other nearby devices, and with the amplifier itself.  The noise performance of many traditional power amplifiers is limited by the amplitude of the AC line-related hum.  This hum is often magnetically induced and is very hard to eliminate.  To solve these problems Benchmark now uses switching power supplies in all of its newer products.  Switching frequencies are set well above the audio band.  More importantly, the high switching frequency greatly reduces the size of the magnetic components and the magnitude of the stray magnetic fields.  Resonant switching is used in AHB2 to minimise switching noise.  Resonant supplies switch at low voltage and low current and are much quieter than traditional supplies"

It goes on to say who this enables them to use a regulated supply, and as a result output voltage is not a function of AC line voltage, prior audio content, or speaker loading.

Now - those are just the manufacturers comments in their own manual (which appears not to be copyright - but if it is, all rights acknowledged).  However,  it is near certain Benchmark will be able to back those statements with evidence.

So SMPS is serious audiophile Martin.  (Even though this amp was primarily designed for the pro market - for use in studios)

Whilst those specs are fantastic they are not unique.... Just saying
SMPS and 50Hz ones can be equally quiet. It's all down to implementation and, of course, the amount of money thrown at the problem!
It has become cheaper nowadays, in mass production anyway, to make SMPS's as there is less copper and iron involved. They are also much lighter and so there is the saving in transport costs. I still prefer a traditional psu myself but it's purely on grounds of reliability and ease of repair if it does go wrong. At much more budget price levels, SMPS (and also class D sometimes) has rendered a lot of affordable equipment as disposable i.e. it's both cheaper than ever before but at the same time more difficult and expensive to repair than ever before, hence if it goes wrong, skip it and buy a new one. This has now rendered much of the latest equipment (hifi and group/PA as well as electronics in general) with a purchase cost much below £500 as land fill waiting to happen.... and is making the job of repairing equipment for a living less and less tenable. It's pretty much only vintage gear and high end(ish) gear that is economical to repair in the times we live in...

I'm interested Jez.  Which other power amps achieve that sort of performance? -132db A weighted stereo ?  I wouldn't pretend to have exhaustively researched the subject - just took Benchmarks statements, and all subsequent press affirmation of that statement at face value.  But I haven't managed to catch them out yet

You obviously have something in mind - what is it?

Yes.  That would appear to be sensible - in my case it is limited by Pip on any input I use which only has about -98db at best.  It is of course designed for direct feed from DAC's which achieve those sort of figures - and the design spec was to match the performance of the DACs - see the launch blurb

Benchmark launch - stereophile

However - it seems that the result has proved good to listen to.  It's anathema to Marco and many others - an amp designed with measurements to a set of specification targets.  It's designer didn't listen to it  until he got sent the one I have, and he didn't have a working pair of speakers until he came here.  The pro market isn't fussed by such a statement

Frankly, whilst I knew he was working on it, I wasnt interested in it because I had expected (and he had managed that expectation) it was for pro use only, and he could do better focussing on a home audio design specifically.  It was only when it started to pick up so many reviews and all of them were so positive, that I got interested and asked for a chance to hear it.

Somewhere (I can't find where) Benchmark or John Siau says that "its  job is to render a near perfect high-power replica of the incoming line-level signal. It does this very well." - that's pretty much a wire with gain.  Maybe that's all it needs for an amp to sound really good.

I don't have any reference points to tell how good it is compared with what else is out there.  But , on still fairly limited audition, its as good as anything I can remember, and good enough to have started a domestic dispute about whether we can afford one when we are supposed to be conserving all our spare cash for "unexpected costs" on the house move.

But as an engineering exercise - as intended for its pro market - it has unique measurements (claimed better by at least 15db) - and the result doesn't sound too shabby at all.  Especially paired with its designers best pre-amp  - recently refurbished by the man himself (after I set fire to it trying to build one) 

ONE unique measurement.... Which could no doubt be equalled. S/N ratio in power amps has never really been an issue or something that much design effort is put into.

I'm afraid I found it flat as a pancake, the most disappointing hunk of Japanese muscle I think I've ever heard.

As for what equipment and methods magazines use, it varies a lot and so does my respect for their writings.

<cough> Devialet <cough>

'Shortest signal path in the world (less than 4 inches between the DAC and the speakers output connectors): lossless by design.
Unequalled Signal/Noise ratio (> 130 dB): abyssal silence, emotional power, the music emerges from nothingness.
Harmonic distortion barely measurable (0,001%): absolute fidelity of instrumental timbres and sonorities.
Intermodulation distortion equally barely measurable (0,001%): seizing realism of soundstage (separation and placing of instruments).
Thermal distortion non measurable: very natural fast variations of dynamics (impacts, resonances).'

Hell Pinky a designer that doesn't listen to his product until the manufacturer makes one? Sorry but that is not the way to design audio. If it measures good it doesn't necessarily sound good. That is the problem with a lot of speakers these days. Also I am sceptical about the use of switch mode power supplies as they dump a load of nasties into the mains. I have heard the latest version of the Chord and it sounded Ok but not a revelation. I certainly wouldn't dump £8K on one and I think £3K is rather excessive for a pro mkt product. People use switch mode PS to save space/weight, what other reason not to use a proper toroidal?