Columns Retired Columns & Blogs |
An excellent review- KM.
Always good to read and see a new AYRE product.
Happy Listening!
Analog sources: VPI Avenger Direct turntable with VPI 12" Fatboy Gimbal tonearm and Hana Umami Blue MC cartridge.
Digital sources: HoloAudio May DAC, Sonore opticalRendu, Roon Nucleus+, Small Green Computer power supply, TRENDnet switch, Apple iPad mini.
Preamplifiers: Sugden LA-4, Ayre K-5xeMP line preamplifiers; Manley Labs Chinook, Goldnote PH-5 phono stages.
Power amplifier: Pass Labs XA-25.
Loudspeakers: Volti Audio Razz SE.
Cables: Interconnects: AudioQuest Pegasus Zero-Tech, Triode Wire Labs Spirit II (RCA). Speaker: Analysis Plus Silver Apex Speaker. AC: Triode Wire Labs Obsession NCF.
Accessories: IsoTek EVO3 Aquarius line conditioner, Salamander five-tier Archetype rack (2), IKEA Aptitlig bamboo chopping boards. Hi-fi set up on short wall firing into 10' × 12' room, wood slat on plaster walls.Ken Micallef
An excellent review- KM.
Always good to read and see a new AYRE product.
Happy Listening!
Love your reviews. Your music selections are eclectic and always introduce me to new and excellent discoveries. Thank you!
Isn't it great to see the actual real square wave under test coming out of an amp, without the need of the dreaded AUX-0025 Class-D filter cloaking device that makes a Class-D's squares wave look reasonable if not real and what disgusting noise really comes out of them and onto your speakers.
PS: Love the high input impedance, means NOTHING has an excuse not being able to drive into it, every thing should be that high, might be noisy if switched on with nothing attached, but who does that anyway.
Cheers George
Didn't matter when that filter wasn't use on the 175w CI and Belcanto amps way back, showed the switching noise and if the was any instability ringing that also would show up, so you get to see it all without the AP filter. And a bonus is it shows how effective the switching noise filter in the amp/s is also.
https://www.stereophile.com/images/archivesart/805CIAfig2.jpg
Same also with the Belcanto design Black
https://www.stereophile.com/images/615BCBfig13.jpg
The Halcro Logic had quite a bit of switching noise on it output, which you also commented on being quite large, delicate efficient tweeter coils could get warm with this on them 24/7. I had to replace a pair of Wilson/watt 7 speaker teeter diaphragm/coils because of slight distortion from both, when I opened them up both coils were very "blued" compared to the new replacements, nothing had been blown up and he's not a loud listener, he had big mono DM Halcros not the Logic here, and Nu-Force Reference 9 v2 mono's, one or the other "blued" the voice coils.
https://www.stereophile.com/images/archivesart/406Halfig01.jpg
Cheers George
Charley Hansen's "way of thinking and design" has resulted in a "100 watt" power amp that overheats and shuts down when asked to output a mere 20 watts? Or, is this part of Ariel Brown's "core design thinking?"
Far better heatsinking (for thousands less) with the Parasound Halo A 21+.
https://www.stereophile.com/content/parasound-halo-21-power-amplifier
I was concerned/confused about the heat issue as well. But, it can produce a square wave w/o a special filter, what more do you want? Haha
Ayre's 8-series products are "modular designs". The amp section of the VX-8 is very similar to that of the EX-8 2.0 integrated hub. Both are rated at 100 watts into 8 ohms. Both suffered overheating issues when measuring THD+N in bench testing (as summarized next). When JA measured the THD+noise of the Ayre EX-8 2.0 he did so at 10 watts into 8 ohms. At 40 watts into 2 ohms he could only test one channel at a time. Driving both channels at 40 watts into 2 ohms resulted in overheat shutdown. Notably, there was no overheat at 10 watts into either 8 or 20 watts into 4 ohms when measuring THD+N for the EX-8 2.0 integrated hub. However, the VX-8 THD+N was measured at 20 watts into 8 ohms (twice the dissipation) and suffered shutdown after a few minutes operation. As mentioned in the review-- after cooling and reset, measurements could be completed, presumably before hitting the overheat time limit. Which leads to the question: why didn't JA test the VX-8 at 10 watts into 8 ohms, as he did with the EX-8 2.0 integrated, after experiencing VX-8 overheat at 20 watts? Would the VX-8 have no issues, like its integrated sibling? Further no THD+N measurements were attempted on the VX-8 into 2 ohms. Perhaps at 10 watts power into 8 ohms (40 watts into 2 ohms), this could have been accomplished--even if one channel at-a-time. Yes this is indeed nitpicking. But consistency between reviews is important to the reader for their interpretation of the results. Looking on Ayre's website: neither the EX-8 2.0 or VX-8 have 2 ohm power ratings in their specifications. So why check at 2 ohms on the EX-8 2.0, and not the VX-8? Don't check 2 ohms, because the VX-8 had issues at 20 watts into 8 ohms? Again, consistency would suggest halving the power to 10 watts/8-ohm a rechecking THD+N. To assess how the VX-8 performed against its identically rated EX-8 2.0 integrated in this bench test. Nitpicking indeed! The bottom-line is yes, the 8-series amp sections would appear to benefit from increased heat sink area in this static bench test. However, under dynamic signals of music playback no overheat issues were reported for either product in their respective reviews. And yes there are other less costly amplifiers that don't overheat at a fraction of their max rated power in THD+N tests, but they probably also do not have the same "sound" as Ayre products (as described in the VX-8 review). But for these two products reviewed, I believe a little more consistency in measurements would be of benefit.
... subjective reviews, either, since the Parasound amp was evaluated by a different reviewer than the one for this Ayre amp.
However, KR characterized the Parasound amp as "impossible to fault." Further, he states that it "is remarkably powerful, transparent and quiet" and it "imposes no constraints on one's desire to listen through it to the recorded performance or to experience the size, weight, and dynamics of the original event."
Your argument makes good sense.
You could ask some other questions along those lines:
Why measure distortion at 50 Hz? That should be pretty much the easiest part of the audio with regard to distortion. That's especially true for amplifiers using global feedback.
There's great reason to measure distortion at 1 watt output. But, why not also at 5 or 10 watts? Unless you have very inefficient loudspeakers, output levels beyond 10 or so watts are going to be distorting your hearing system. Higher power output levels are interesting, of course, but not as relevant to your living room.
Along the same lines, why not perform multitone tests at some normal listening level power? That would be more representative of actual usage. Audio Precision test systems have this function built-in. Some of the excellent free software packages do this, too. (It would make sense to use "pink" weighting on this test, so that the power rolls off with increasing frequency, like music.)
I get that there's an entire legacy of reviews that should be respected. But, maybe it all could be better today.
Understood. But, let me explain my reasoning for suggesting different testing.
For several years, we used a Wavelength Audio Proton DAC. Its output voltage when operated as recommended by Wavelength and the Stereophile review is just about 900 mV for a 0 dBFS signal. In other words, that's the limit. The Proton was connected to a preamp that had a maximum voltage gain of unity. The power amplifier had a voltage gain of just over 18 (~25 dB). Do out the math, and you find that the maximum voltage output from the power amp was ~16 Vrms. That's for 0 dBFS source material.
The electronics all fed a pair of Vandersteen 2Ci's, in a room about 20 feet by 15 feet, with openings into stairs and other rooms near the back.
Any music that was recorded close to 0 dBFS was impossible to listen to with the volume control at maximum on that system. Or, even close to it. It was painful and not something we would want to do here. It definitely violated any reasonable hearing health standards. How Manny Noriega stayed in that hotel, I will never know.
The Vandersteen loudspeakers are certainly not what would be consided high sensitivity. The room is not tiny. It may not be worst case in terms of power requirements for a home system, but I'd bet it's in the top quartile of that group. 16 Vrms into 8 Ohms is 32 watts. That was the maximum for the system, and that was too much power.
So, I decided that there was no reason to have a power amplifier that produced more than 32 watts. That was even stretching the limit. We plain could not use even 32 watts. There are certainly reasons why power amplifiers that could make more power might sound better, but it wouldn't be because we needed or could use more power than that. A perfect 32 watt amplifier would do the job better than a less perfect amplifier that could produce hundreds of watts.
Later, I used this method to more accurately determine the power that we could use:
https://www.diyaudio.com/community/threads/a-test-how-much-voltage-power-do-your-speakers-need.204857/
My results were that 15 watts would do the job. That was for loudspeakers that are on the lower end of the sensitivity scale in a room probably as large in volume as used by most people for music listening.
So, my point is this: Stressing an amplifier to the corner cases might be interesting for testing and certainly makes for lively discussion on the internet, but it doesn't really say much about how the amplifiers would perform in actual use. It's far more important to see how the amplifier performs in the range of a watt or less up to maybe 15 or 20 watts.
In addition, it's probably especially important to test in frequency bands where the human ear is most sensitive for both the fundamental tones as well as distortion products.
Did I explain that well?
The tests traditionally performed aren't wrong, per se; they just might be improved upon. We've upgraded our system quite a bit since the days with the Proton DAC. I think testing could be upgraded, too.
End note: It could also be that testing doesn't tell much about the perceived sound quality, so my observations above are just bunk.
... all of the program material to which you listen will have been recorded/mastered with an average RMS value no less than -12dBFS.
As the FAQ for the test points out, classical and some jazz recordings might be mastered at -22dB to -24dB. In that case, your maximum 15-watt power requirement increases to about 250 watts.
Perhaps you've seen the video of a Harbeth demo regarding peak power requirements. The Monitor 40 speakers are driven by CH Precision power amps, which have peak reading power meters. The peak power displayed during the demo was, at least, 600 watts. The program material was Laptevinmeri/Laptev Sea by Pan Sonic.
Demo video: https://www.youtube.com/watch?v=Y3WpRY-EtX8
Laptev Sea: https://www.youtube.com/watch?v=XhuJxdaU87I
I think you may be misinterpreting things a bit.
What you do for this test, or, at least, what I did, is to find the volume setting that is the loudest you ever play anything with. (I ignored a couple albums that I know were mastered at well below 0 dBFS - easy to observe in an audio editing package. These albums need more system gain to be played at a reasonable level, but there you just increase the system gain to make the maximum output level the same.)
Once you've established that volume control level, you play the -12 dBFS tone and measure the voltage. That's it.
As described in the thread, -12 dBFS was chosen because it's low enough in level to not damage any of the system components, including your ears. It's also high enough that it's easy to measure on an RMS voltmeter. The tone could have just as easily been at -22 dBFS, -32 dBFS, or -60 dBFS. It has nothing to with how the recordings are mastered. It's just a pragmatic level to deal with. It's 12 dB - 1/4 the voltage - below the maximum level your system can play with a recording mastered with its peak at 0 dBFS. You can't go any louder.
Assuming that the loudest we ever listen, and then some, is the loudest we'd ever want to listen, 12 dB above that (250 versus 15 watts) wouldn't be enjoyable at all.
In any case, as I first mentioned, our system at the time could only get to about 32 watts maximum, even with the preamp gain set to 11, I mean full volume. We (I keep saying "we" - we being my wife and I) never once found that the volume was too low with any recording.
One other reference point: https://www.stereophile.com/content/ayre-acoustics-ax-5-integrated-amplifier
"As Hansen puts it, "Since most preamps are used anywhere between –10dB and –40dB for an average listening level,"
This is consistent with what I found. Since Ayre has always preferred balanced interconnections, the typical DAC used within an Ayre system has about 7 dB more output voltage than the Proton I was using at the time I made my measurements. Note that the maximum gain of Ayre preamps is nominally 0 dB.
You may have noticed in the poll associated with the DIYaudio test that a very large percentage of the people who performed that test had very low voltage requirements from their power amps. I guess I'm not alone in this.
But, whatever people want to do with regard to any and all of this is entirely good by me. It's a hobby for most of us. I just wanted to highlight why at least some of us might be very interested in an amplifier's performance at some level between 1 and 20 watts output. It seems like there's little interest in that after all, so I'll just let it drop here.
Sometimes when a founder passes there’s no one to pass the torch off to and the company fades away.
Sounds like Ayre is in good hands. Long live High End Audio!!!
In considering this Ayre Acoustics VX-8 ($6.8k) two channel stereo power amplifier, compare it to the Benchmark Media Systems AHB2 ($3.5k).
https://www.stereophile.com/content/benchmark-media-systems-ahb2-power-amplifier
https://www.stereophile.com/content/benchmark-media-systems-ahb2-power-amplifier-measurements
Benchmark AHB2 is approximately half of the price of the Ayre VX-8, while AHB2 exhibits similar bandwidth and similar power output, but at lower noise and lower nonlinear distortion, is much more efficient, and exhibits much cooler operating temperatures.
The price difference is more than enough to also buy a Benchmark LA4 Line Amplifier ($3.0k) control preamplifier.
https://www.stereophile.com/content/benchmark-la4-line-preamplifier
https://www.stereophile.com/content/benchmark-la4-line-preamplifier-measurements
In his review Kalman Rubinson observed and concluded, "The LA4 is probably the most transparent and revealing audio component I've ever used. It does not seem to leave any fingerprints on the sound. It has a useful array of inputs, outputs, and functions, and the use of sealed relays for volume and input selection assures a long and noise-free life. Benchmark might improve on the display and remote control, but I do not have any criticism of the LA4's sonic performance."
John Atkinson summed up his measurements with the comment, "Benchmark's LA4 is the widest-bandwidth, widest-dynamic-range, lowest-noise, lowest-distortion preamplifier I have encountered."
Or, instead of the LA4, add another $200 to the price difference to buy the Benchmark HPA4 ($3.5k) which is similar to the LA4, but adds a very good headphone amplifier.
I, too, read the reviews for equipment evaluations as much as music recommendations.
Suggestion: Would it be possible to begin including a separate page similar to the Specifications page listing the tracks and sources used in the review? This would make it easier to cut and past the track titles into Roon, Tidal, Qobuz, etc.
Just a thought. Anyone else think this would be useful?
Come'on man...get your sh$% together Ayre. Take this as harsh criticism from an actual fan and previous owner of several of your model lines; all of which I loved.
I can think of several Amps that I have owned in the past at a fraction of the cost of this one that you could damn near fry an egg on the top plate and they never shut themselves down trying to output a fraction of their advertised rating. And this is after 6-8hrs straight of taking a pure beating over 100db at college parties, house parties and just for the hell of it at times (Adcom, Pass, Classe, Rotel to name a few examples)
How are these kinds of issues allowed to leave the factory on their way to the consumer without being noticed? These Audio companies must know what kind of testing they will go thru when submitted to Stereophile.
The embarrassment is avoidable of having an audio rag discover the equipment's short comings. Just test that sh*& till your blue in the face before sending it out the door. If it cant take a real world beating then send it back to the drawing board. Simple as that. This goes for every audio company, not just Ayre.
I personally applaud a company that designs its products for its customers rather than the press.
I anxiously await reports from customers experiencing excessive high temperature shut downs.
I wonder how an amplifier with distortion profile like this one, dominated by class AB output stage crossover distortion not corrected by FB action, can get quite positive rating by J.A. ("The Ayre Acoustics VX-8 offers respectable measured performance,")
Crossover distortion? Feedback?
A very reputable designer of equipment who uses lots of feedback in his designs seems to disagree with you on the value of feedback when it comes to crossover distortion.
https://benchmarkmedia.com/blogs/application_notes/131424519-amplifier-crossover-distortion
Besides, I'm not seeing any real evidence of crossover distortion in the shown plots. I just see harmonic distortion. I'm certainly missing something - please show me what I'm not seeing.
It is evident from the distortion spectra that the output stage is in highly biased class AB. At the moment when the output stage leaves class A (I = 2 x idle current) and moves to class AB, you can see the crossover distortion resulting from infamous Gm doubling (higher Gm in class A) and transition from A to AB. See the knees in Fig.4. e.g. Low loopgain does not correct it and the level of distortion is high, with high order harmonics forest spread to infinity. The famous designer spoke about totally different circuit.
OK, but I'm not quite following why you think it is crossover distortion, per se. Yes - the amplifier appears to be highly biased class AB. I don't think that's a secret and that should be evident to anybody looking at the box. There cannot possibly be enough heat sinking capability for it to operate as class A even at more than a few watts.
Perhaps it's because I never considered Gm doubling to be crossover distortion. After all, it doesn't actually happen at the crossover point, only on each side. Semantics, perhaps.
My understanding is that the VX-8 is completely open loop in its design. So, low loop gain would not be relevant. That could be wrong on my part, and Ayre could be lying on their web page. No idea. I've never even seen a VX-8.
My point about crossover distortion and John Siau is that he explained why crossover distortion often cannot be fixed with feedback because of the bandwidth of the crossover distortion relative to the bandwidth of the feedback system. But, that's not relevant here, either, if the open loop architecture description is true.
I have not heard the amp reviewed here, but I have owned a KX-5 Twenty preamp and VX-5 Twenty amp for a year now. I used the VX-5 Twenty to drive a pair of Magnepan 3.7i speakers, without a single issue. I'm now using it with Wilson Audio Deutte speakers. I don't care how any of this stuff measures, whether it shuts down during some lab test, etc. What I do care about is that it's the most musical solid state equipment I've ever heard. I am no expert and I have certainly not heard everything. What I have heard and still own are full tube Joule Electra OTL amp / preamp, Pass Labs XA, Coda, Levinson... I forget what else. There is something about the "Rightness" and Musicality of the Ayre gear that makes me think it is going to be hard to replace / do without. Any sort of consideration along those lines would be in terms of making sure I could sell the latest contestant without losing money, and the Ayre gear would be waiting to show it the door (likely).
With the Ayre gear, I want for absolutely nothing. I'm just happy I was able to afford it at used pricing. Prior to owning the Ayre gear I was a "tube guy"... any tube lover looking for a viable replacement for tube musicality would be well advised to hear this stuff. There is something special and different going on... and I'm glad Ariel is carrying the ball forward.