New Benchmark - on loan (pending domestic negotiations)

[Oldpinkman]

I had a visit from a chum this weekend, and mercifully the weather was so good we didn't want to spend time indoors playing with HiFi. So he lent me this to give me time to let him know what I think of it

http://i1315.photobucket.com/albums/t582/oldpinkman/DSC_4081-1_zpsj6l7il5f.jpg (http://s1315.photobucket.com/user/oldpinkman/media/DSC_4081-1_zpsj6l7il5f.jpg.html)

I rather like it so far :cool:

[Marco]

Hi Richard,

What exactly is 'this'? I'm presuming that it's an amplifier? If so, how does it sound compared to what you were using before? :)

Marco.

P.S I've heard of Benchmark DACs, but not amps. Is the amp you've got from the same manufacturer?

[Oldpinkman]

Hi Marco

Yes its an amplifier and yes its from the same manufacturer - Benchmark, a great big American company selling pro audio equipment that I have nothing to do with. The amplifier was designed for them under contract by THX (owned by George Lucas) using their patents. The lead engineer on the assignment (basically the amplifiers designer) was my friend Owen - who designed the Pip. I watched him designing this in his laboratory at his home and have been eagerly awaiting the "birth". I could take it out and post better pictures but they won't be as good as the dozens of professional photos on the web, including those in the Hifi News review I have put a link in for below

He was sent this amplifier a few months back, but hasn't really got round to listening to it and so bought it down this weekend to listen to here. It was too lovely out to sit in and listen to HiFi, so he has left it with me to "let him know what I think" for a few weeks.

The more I listen, the more I think it is very very special. So does everyone else on the internet who has reviewed it (mostly pro audio reviews - and it was designed for the pro market). But it is rapidly making a big impression in the audiophile market

https://cdn.shopify.com/s/files/1/0321/7609/files/HFN_Mar_Benchmark_AHB2_Reprint.pdf?100480359912993 87752

It is quite extraordinary - as in not at all ordinary, and therefore I think interesting for the revolution it amounts to, as much as how good it sounds

To quote the last paragraph of the user manual

"In summary, Benchmark has taken a radical approach to audio amplification, has leveraged patented technologies (licenced from THX Ltd) and has reversed many common practices. The result is an amplifier that is significantly quieter, cleaner, smaller, and more efficient than any traditional design. There is nothing ordinary about the Benchmark AHB2"

It arrived with my serviced Pip 2, which since that uses RCA's (Pip 1 is Din connections) that has meant all new interconnects. And the inputs to the Benchmark are balanced, so that connection is a custom cable Owen made for me. And we "hairdryered" my ESL 63s to tighten them up a bit, and I reset the anti-skate accurately - so its all sounding a bit different and I am taking my time getting used to it.

It is extraordinary. I fear another "ratchet" component (which will be the second Mr Jones is responsible for)

[Marco]

Nice one. Glad you're enjoying it. I rate the Benchmark DACs, so I can imagine that their amps are pretty good, too :)

Btw, as an aside, those Quads you've got can also be modified to sound rather special. Two friends of mine use them (one of them Tubehunter, who posts here) and they sound STUNNINGLY GOOD - probably the best Class A solid-state amps I've ever heard.

I'm not sure exactly what the mods were that were done, but if you asked Duncan (Tubehunter), I'm sure he'd outline them in detail. All I can say is that if anyone here has a pair of 405s, and wants to hear what they're truly capable of, they should have them modified thus (for not much dosh), as the results are positively breathtaking.

Marco.

[Oldpinkman]

Not sure which Quads you are talking about. The Esl63's are speakers. I have one Quad 405-2 power amp

And that was modified by - Owen! (who designed the Benchmark) - so it's already quite good.

I think the modifications to make a 405-2 Class A would be fairly substantial - akin to the NASA space program.

However - back on topic - the Benchmark does resemble the Quad in that it lets the power get delivered by rough and ready class AB devices, and "corrects" the final output. But the Benchmark does it much much better than the standard Quad, or even my modified one. It does it so well that it is being consistently compared with "Class A" - because there is zero cross-over distortion. I don't want to swamp you in my enthusiasm with stuff that may be of no interest to others, but there are lots of details of the innovations refered to by John Siau I could post about if anyone is interested.

The sound has been likened to a valve amplifier in the "Everything Audio Network" review "The AHB2’s sonic character is dynamic, open, with quick, taut bass — yet with the silky ease of the best tube amp. But tube amps don’t have the energy and speed in the transients, nor bass, that the Benchmark possesses."

http://everythingaudionetwork.blogspot.co.uk/2014/10/audiophile-review-benchmark-ahb2-stereo.html

No bottles, no heat, but valve sounds. (That now makes 2 Owen designs that sound more like a valve amp than a valve amp does. You may recall I described the Pip in those terms.)

[Marco]

I'll reply more later, as I'm just popping out. However...


Not sure which Quads you are talking about. The Esl63's are speakers. I have one Quad 405-2 power amp

And that was modified by - Owen! (who designed the Benchmark) - so it's already quite good.

I think the modifications to make a 405-2 Class A would be fairly substantial - akin to the NASA space program.


This is (sort of) what I'm referring to: http://www.dadaelectronics.eu/ForumRetrieve.aspx?ForumID=317&TopicID=97322&NoTemplate=False

Also see here (from post #42 onwards): http://theartofsound.net/forum/showthread.php?34047-Quad-405-servicing-modding-advice

I thought you already had two, but yes, it's about having a pair of 405s and converting them into monoblocks. Duncan tells me that they are (or become) Class A. Don't ask me how... However, what I can state is that the resulting sound quality is stunning - easily the best I've ever heard from ANY solid-state amps, at any price.

Anyway, apologies for the thread drift. Laters! :cool:

Marco.

[Oldpinkman]

No worries - as I say, in a funny sort of way its not thread drift. I confess I didn't read everything in careful detail but I think I am (broadly) familiar with those mods. There were some variations like using the chassis as a monoblok , with benefits for separation and dedicated power supplies, but most was the standard well known mods to the Quad 405

These are

Improve the power supply (dual mono supplies are available if not going the whole hog for monoblocks)
Sort out the birds nest wiring with the mains and signal cables stuffed together
"component upgrades" - in particular opamps
lower the input sensitivity (which improves the noise performance)
Take electrolytics out of the signal path (a lot get this wrong)
rebuild the bridge

Owen - who designed the Benchmark has done most of these for me - although he devised his own solutions. Certainly those sort of improvements take the amp well beyond its honest but not exciting standard performance. They don't make it class A. Maybe Duncan means "as good as class A" - which is what the Benchmark does, but it is not going to achieve the same level of performance

The Quad is a current dumping design. It has a class A/B amplifier to deliver most of the output, and this is then sampled and corrected by the bridge, which uses a much higher quality amplifier for correcting the "roughness" of the AB amp.

The Benchmark does the same, but using feedforward and is able to achieve zero cross-over distortion - effectively the same as class A but without the heat and power waste.

It's a bit special ;)

[Marco]

Owen - who designed the Benchmark has done most of these for me - although he devised his own solutions. Certainly those sort of improvements take the amp well beyond its honest but not exciting standard performance. They don't make it class A. Maybe Duncan means "as good as class A" - which is what the Benchmark does, but it is not going to achieve the same level of performance


I'll ask Dunc to comment and clarify what he meant by the Class A thing. Anyway, enjoy the Benchmark, which clearly you're already doing :)

Marco.

[Firebottle]

The 405 uses a feed forward design as well.

The first few hundred miliwatts of output (not sure of the exact figure) are class A as they effectively come from the class A driver stage, before the current dumpers come in to effect to give greater output.

So there :ner:

:)

[Marco]

Can't comment on that, Alan. I just remember Dunc saying something about Class A. I've asked him to pop in and comment. Nice meeting you at the show yesterday, btw! :)

Marco.

[DSJR]

The driver stage was definitely Class A, as Quad made a big thing of it at the time. the feed forward bit is, I believe, an update on the thing Crown had been doing for some time before (Class AB + B). This latter sounds incredibly crude in description, but closely matched transistors/resistors where necessary enabled it to work well for decades ;)

I still rate the 405-2, although I believe the circuit was improved in the 306 onwards when the input op-amp was taken out of the signal path I believe..

The Benchmark amp looks interesting, but it's very expensive for the output and heaven help a service engineer in years/decades to come if the thing ever breaks down ;) I'd like to hear if it's really any better WITH MUSIC than say a £400 Crown Drive-Core design as Jerry had and liked for a while.

[tubehunter]

Let the man explain it

The Current Dumping Amplifier

http://hifisonix.com/wordpress/wp-content/uploads/2012/12/Quad_405_Power_Amplifier_Wireless-World-Article1.pdf

[Roy S]

£400 Crown Drive-Core design as Jerry had and liked for a while.

I've got Jerry's Crown amp now & run it with another in bridged mode and the results (to me) are amazing.

[Marco]

Let the man explain it

The Current Dumping Amplifier

http://hifisonix.com/wordpress/wp-content/uploads/2012/12/Quad_405_Power_Amplifier_Wireless-World-Article1.pdf

'Fraid all that techy stuff flies right over my head... The only bit I'm interested in, which I was telling Pinky about, is that you called them Class A amps. Are they Class A or not, dude? That's all I'd like to know :)

Marco.

[Oldpinkman]

Basically it says it is not pure class a, but done properly it can perform and sound like class a. What's good about class a is that the devices, (valves, or transistors of any flavour) are working in a range where they are fairly linear. If working means going up and down in voltage, they do that smoothly and accurately. That linear range is always going to be above the switch on area. They are not very linear in the range zero to not-very-much.
The nastiness in a class b amplifier is that it operates in that close to zero range as the sine wave signal goes from positive to negative. And the non linearity in that range gives a "cross-over distortion" as the signal crosses over from positive to negative and the device on the positive half runs down into that nonlinear region before it turns off, and the device on the negative half of the cycle switches on and runs through the rough patch until it gets up into its own fairly linear region.
Class a is about amplifying the signal without cross-over distortion - which is done by biasing a device (valve or transistor) so that the middle of the sine wave signal is the middle of the devices linear operating range, and the swings the signal makes stay in that linear range.

To keep this not too long, ignore the differences between current and voltage amplification, in essence what both the quad current dumping and benchmark feed forward technologies do is let a class b (strictly ab) part of the amplifier deliver the heavy current (power and heat) part of the output and "correct" the cross over distortion using a very accurate but low power class a amplifier.

The quad never operates as pure class a, with all of the output from a device biased into the middle of its operating range, but it uses class a amplifiers and some clever jiggery-pokery to get the same effect.

Now, we get into measurement I'm afraid. If the problem with not class a is cross-over distortion - then successful jiggery-pokery is going to mean low distortion. The quad was OK. The best of the mods (including owens on mine) are really good, the benchmark amplifier is mind-boggling good. As the blurb says "vanishingly low levels of distortion"

Edit : Actually - whilst giving the gist for someone who doesn't understand the technical paper - that needs footnotes - see next post

[Oldpinkman]

So - footnotes

The principle remains the same - but , the benchmark uses class H and class AB, and feed-forward circuitry, which THX have patented as the AAA(TM) Achromatic Audio Amplifier "to achieve high efficiency and vanishingly low distortion". This patented technology has been integrated with a Resonant-Switch Mode Power Supply. This has implications both for power consumption and error correction accuracy.

This bit from the blurb might help too Marco

"Low-Bias Class-AB Main Amplifier - Class-AB output stages always produce "crossover distortion" whenever the push-pull output drivers change state. Precise bias control is usually required to minimise this crossover distortion. Traditional feedback networks have difficulty correcting the transients caused by push-pull crossover transitions. However, the THX patented feed-forward error-correction system has the speed and precision necessary to virtually eliminate all traces of crossover distortion. With this feed-forward system, bias currents have no significant impact on the distortion at the output of the amplifier. This topology provides the freedom to conserve power by setting unusually low bias currents. The AHB2 uses very low bias currents but achieves vanishingly low distortion - even under heavy loading.

Class-H Tracking Rails - Multiple power supply rails (Class G), and tracking power supply rails (Class H), can improve the efficiency of a class-AB amplifier. The AHB2 uses a lower fixed rail and an upper class-H tracking rail. The improved efficiency provided by this topology usually comes at the cost of increased distortion. For this reason, there are very few amplifiers that use multiple supply rails. In a traditional feedback design, every rail switch point adds another layer of crossover distortion. In contrast, the feed-forward error correction used in AHB2 is so effective that multiple rails can be added without incurring any distortion penalty"

Note 1 - that last comment is true - but not as easy as it sounds. Knowing how much design time was spent getting those multiple rails to produce low distortion, its not as easy as buying the patent licence and "away you go". Hundreds of hours.

Note 2 - semantics. There are those in sad parts of the HiFi world who look to jump on words to find hidden nuances of secret conspiracies and cover-ups. So "no significant impact" on distortion is taken as "but still has an impact". Forget subtle readings - lets use numpty old measurements of "vanishingly low distortion" (and while we are at it "low noise")

Compare these numbers with any other power amplifier design

Signal to noise ratio A weighted stereo mode (the amp can also operate at the flick of a switch in bridged mono) -132db. (Yes you read that right)
Distortion THD 1Khz, 20kHz LPF at full rated output into ANY rated load - 0.00011% (Stereo Mode) (Count the zero's)

That's what "no significant impact" means ;)

DSJR - you know a lot of kit. Can anything get within 15db of those figures (Marco - 3db is just audible. 10db is "twice as loud")

[Marco]

Thanks for that, lol!

All I know is that the buggers sound superb (essentially they don't sound, 'hard', grey and grainy, like so many SS amps do), and so I would heartily recommend anyone on a tight budget, who wants serious SPPV from their amplification, to go down the monoblocked Quad 405 route :)

Marco.

[Oldpinkman]

and heaven help a service engineer in years/decades to come if the thing ever breaks down ;)

You sound like my former father-in law every time I bought a car more modern , or with "more things to go wrong" than an Austin Allegro. By all means stick with points ignition Dave, but realise the world is going to sail past you in something smoother and more reliable.

As for "I'd like to hear if it's any better with MUSIC..."

did you mean personally - or in the opinion of anyone else trying it? I posted a link to a HFN review in an earlier post. I assume google has broken on your PC - cos there are a few dozen others out there. Let me help you with just one

http://everythingaudionetwork.blogspot.co.uk/2014/10/audiophile-review-benchmark-ahb2-stereo.html

;)

[NRG]

This looks like a very interesting amplifier and very well thought out, however, there are lies, dammed lies and statistics! The performance figures are good but need to be taken in context, this review gives a better Lab report than just the headline numbers Pinkie has posted: http://www.thx.com/files/2014/12/BenchmarkPDF1-v3.pdf

[Oldpinkman]

Your link doesn't work Neal. There are a bunch of graphs for the manufacturers measurements. I do know Mr Jones commented that the HFN measurements were flawed. The last link I posted has some consistent and worthwile measurements on it.

It sounds pretty good too. Not impressive. It's not one of those "wow" amps. Indeed the first couple of times we swapped the Quad back in - the Quad was the more "impressive". But in a "boom tizz" way. And whilst the Quad could be obviously richer and brighter, by comparison

the bass is boomy on the Quad
the bass on the Benchmark is tight and fuller
the Benchmark is wider in its soundstage and sorts out separating instruments better
the Benchmark has better detail presentation - very natural, very clear. (for example, background vocals sound like a person, or different people singing, rather than like a synthesiser BV channel)
the Benchmark is very relaxing. It is lovely to listen and listen and realises there is none of the stridency

These comments are all relative. The Quad is a damn good amp, and the sounds are more alike than different. But they are subtle and important differences, and like all such differences grow on you. Also my current choice of input (vinyl) is hardly exploring the full range of the device. But it sounds good even so.

[NRG]

It did work! Somebody has pulled it plus the other link Google found! Luckily I have a cached version... Paul Miller is well respected in the industry and has tested hundreds of Amps in his time using the same in-depth test methods....they can't all be flawed ;)

[NRG]

No, it's still here: http://www.iglooaudio.co.uk/get_manual_file/3732.pdf

Get it while you can before it gets pulled ;)

[Oldpinkman]

The link works now. From the time it took to load I knew it was going to be the HFN review. I forget now what the problem was, but a clue if you are technical probably lies in the fact everyone else measuring it got similar results close to Benchmarks published figures. I could ask...

:cool:

[Ali Tait]

It should be good given the asking price...

[Firebottle]

Also my current choice of input (vinyl) is hardly exploring the full range of the device.

At least the amp isn't a bottleneck :D

Like the design approach with this amp, would like to hear one.

:cool: Alan

[Oldpinkman]

Very good Alan

I expect to be bringing one to France with me when we move out permanently (that should fund the purchase!!) so maybe we can all get together sometime. I'll probably hang on to the Quad, so since this is small light and robust I could even post it as a loaner (once I have one of my own). I fear this one is not going to be staying long - which is unusual with Owen, but he's keen to get it back.

Your signature reminded me of something - and since its my thread I feel entitled to drift it

What do you think of my mate Sam? (21 last month)


https://www.youtube.com/watch?v=cixo3etePCo

[Firebottle]

Whoaaa.....I enjoyed that :rock:

One of my favourite tracks as well, that boy knows how to swing his axe. Presumably a tribute band?

:)

[Oldpinkman]

They have just got together. First gig is in ë weeks time. First half of the set is general blues stuff and the second half is hendrix :)

[Marco]

What do you think of my mate Sam? (21 last month)


https://www.youtube.com/watch?v=cixo3etePCo

Freakin' awesome, dude - that boy has SOUL! :guitar: :hairmetal:

That track reminds me a bit of one of my favourite ones from a band called Nine Below Zero... I give you Sugar Mama:

HaqVZW2k8aU


Sam should try covering that track for his 'general blues' stuff. I think it would go down well :)

Marco.

[Oldpinkman]

He'd have to play the harmonica - they are a trio for now. Mind you - he probably would only need a day to pick it up. He plays keyboards and drums at least as well as he plays his "first" instrument - guitar.

[Joe]

Signal to noise ratio A weighted stereo mode (the amp can also operate at the flick of a switch in bridged mono) -132db. (Yes you read that right)
Distortion THD 1Khz, 20kHz LPF at full rated output into ANY rated load - 0.00011% (Stereo Mode) (Count the zero's)

That's what "no significant impact" means ;)

DSJR - you know a lot of kit. Can anything get within 15db of those figures (Marco - 3db is just audible. 10db is "twice as loud")

Yes. The Devialet amps have the following specs:

THD+N (Tonal/Harmonic Distortion) at full power: 0,001% IMD (Intermodulation Distortion) SMTPE: 0,001%
SNR (Signal-to-Noise Ratio): 130dB unweighted
Thermal distortion: not measurable
Output impedance: <0,001Ω
Bandwidth through digital inputs at 2Ω to 8Ω load: DC–87kHz (-3dB),
DC–60kHz (-1dB), DC–30kHz (-0,1dB), same on analog inputs except 0,1Hz instead of DC

They also have a preamp, and a built-in DAC, and the more expensive models have configurable MC phono stages and wireless connectivity for streaming.

[Oldpinkman]

No. Ha!

My new 40% lead solder for naughty boys. My word is it a bloody sight easier than lead free. Even making up cables for the Benchmark I don't seem to be able to get the human barbeque going like I normally do

:D

[BTH K10A]

I've just handed back a Yamamoto A-08S that was loaned to me for a week. Amazing!!

[Oldpinkman]

I've just handed back a Yamamoto A-08S that was loaned to me for a week. Amazing!!

Not good then. I never like having to hand back "amazing" stuff. I was hoping that with this one I would hand it back before going on holiday, have 2 weeks with no HiFi, come back to the Quad, and be perfectly happy with it. Even that simple plan has been buggered by not getting round to returning it before going on holiday

I took it round to Sovereigns place to compare with his SECA, and realised that the cables I was using at his house were not the same as the ones he was using (Klotz MC5000) which were the cables I was using at my house, so intended to make up a 2m set of the short cables, so I could compare the "short" cables with the 2m Klotz ones.

Actually I got side-tracked into making a completely different set of cables, so I'm not sure what this is going to prove, regardless of how these 2 sets of cable sound.

Still - done now, and no burned flesh! :cool:

[BTH K10A]

I'm not sure I understand why handing something back an amplifier a kind work colleague had lent it to me is "not good",as it wasn't mine to keep. :scratch:

Without the loan I would probably not have chanced purchasing one on reviews alone. Apart from only being suitable for very efficient speakers the Yamamoto was capable of levels of detail, nuance and delicacy that , coupled with a deep wide soundstage and wonderful tonal colours that I've never encountered before in any amplifiers I've owned or auditioned. :)

[Oldpinkman]

Mine, my precious

All Mine

http://i1315.photobucket.com/albums/t582/oldpinkman/ahb2_zps6gsmy5z6.jpg (http://s1315.photobucket.com/user/oldpinkman/media/ahb2_zps6gsmy5z6.jpg.html)

[CageyH]

I think you need to revisit your HiFi rack.
And there you were the other week criticising my glass shelf. :eyebrows:

[Oldpinkman]

I think you need to revisit your HiFi rack.
And there you were the other week criticising my glass shelf. :eyebrows:

I think it can stay in its wrappings given it may be travelling to France in a couple of weeks. Although I suppose maybe it makes sense to plug it in and confirm it works. I've had a good day for work so far, so maybe I can have a play tonight. The funny thing is not a peep out of the better half - and she is adamant we spend what it takes to keep the Quads. Very funny
:cool:

[misterdog]

I'm toying with the idea of trying the AHB2 with my Quad 989 ESL's has anyone else tried these with ESL's ?

[CageyH]

Richard uses bis to drive ESL63s.
It works quite well.

[hifinutt]

I'm toying with the idea of trying the AHB2 with my Quad 989 ESL's has anyone else tried these with ESL's ?

what attracts you to them ? afraid the hi fi critic review was enough to put me off but be interested in your thoughts

https://pinkfishmedia.net/forum/threads/benchmark-media-ahb2-power-amp.239617/

https://www.audioshark.org/amplifiers-8/thoughts-benchmark-ahb2-sound-18278.html

[misterdog]

what attracts you to them ?

Clinical accuracy, though i currently use Neurochrome amps with 'vanishingly low distortion'.
I think the AHB2 may offer more especially in the bass department.

Adjustable input level, allowing me to play my SMSL M400 direct.

The M400 outputs 5V with lowest THD (0.000075 % ).

My Quad 989 ESL's have a swing of impedance from around 3.5 to 21 Ohms. The AHB2 may control this better.

But mostly curiosity.

[hifinutt]

Let us know how you get on