• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Bruno Putzeys ‘Life on the edge’

So can everyone now just let @amirm get on with the Class D amp reviews (including his IMD distortion tests), and kindly shut the f*** up about how Class D is allegedly "a toy" or "has a long way to go" simply because you have to band-limit THD sweeps when measuring them?
I assume you have realized that Bruno speaks about near-clipping twin tone 18.5+19.5kHz test, not about the 5W test as is used here. Try it, it is a BIG difference. Some remarks about “toys” that I like to use is not for the reason of THD sweeps at different frequencies, but for the reason of insufficient heat sinking (for example AIYIMA A07 and similar amps) and for the reason of frequency response highly depending on load complex impedance, which remains untested here. 4 vs. 8 ohm resistor load is not the test to reveal this. Proper measurement of output impedance vs. frequency would be the test.
 
So can everyone now just let @amirm get on with the Class D amp reviews (including his IMD distortion tests), and kindly shut the f*** up about how Class D is allegedly "a toy" or "has a long way to go" simply because you have to band-limit THD sweeps when measuring them?

There's a huge difference between 5W (used at ASR) and just under clipping.

Also, band limited THD sweeps showing THD figures that aren't even considering enough harmonics to generate a useful number are plain DECEPTIVE. Most people have no understanding of measurement bandwidth.

Edit: @pma said it first.
 
I want to bring this to attention again:
Sweeter still: active is cheaper: class D amplifier modules cost less than some inductors for passive crossovers. You can build something that runs rings around a top-flight separates rig and make it masquerade like a lifestyle product with a matching price tag.
Not a week goes by without some pundit asking, “Is high-end audio dying” and why we can’t get the younger cohort interested in high-performance audio.
How about giving them what they came into the shop for? A system that reliably sounds good and doesn’t present them with unnecessary choices. Preferably something that requires nothing but a power cable and a phone to play music. Give them all the user-friendliness of their cheap wireless speakers and soundbars but combine that with heart- stoppingly beautiful sound.

It’s not the fact that high-end amplification has become a commodity that should force manufacturers to rethink their strategy. It’s the survival of Hi-Fi itself that they need to secure. Class D isn’t the revolution, and neither will it trigger one. Class D is there to help the revolution along.
As much as I appreciate Mr. Putzeys contribution to the hifi world, we also need to remember the price of Hypex and Purifi. They are not cheap at all, making what he descibes above in regards to reaching the younger generation more or less impossible today. The latest and smaller Kii seven has a starting cost of $9113 including tax here in Norway! That's a price in the high-end segment and something even very few adults with a good salary would consider to spend on a hifi system.

And while the Hypex plate amps can bring the cost down some, they are still pricey compared to low cost AB plate amplifiers from Asia. None of the low cost Asian speaker manufacturers consider to use Hypex/Purif for this reason.

Both Hypex and Purifi are primarily aimed at the high-end market with their prices. Sure, I'm aware of some direct manufacturers selling some fairly low cost Hypex amps (though still very high for the younger generation), but with a very low building quality and non existing cooling system that will lead to an early death in many environments.

So who is he trying to fool here? His Class D needs a substantial reduction in cost in order to being able to reach a younger generation.
 
@Bjorn I agree. With Hypex and Purifi you are paying a lot more for IP than the hardware itself.

I pay for substance, build quality, support, spare parts, potential residual value, pride in ownership, reliability and overall enjoyment. That's why I'd buy the NAD M33 if I had to have a Purifi based product. At least I can get a full service manual and schematic should it ever need fixing.
 
the reason of frequency response highly depending on load complex impedance, which remains untested here. 4 vs. 8 ohm resistor load is not the test to reveal this.
Say what? The increased output impedance with frequency shows up just as well with 4 vs 8 ohm as it does with complex loads. If 4 and 8 ohm land on top of each other, then the output impedance is very low and therefore it is a non-problem.
 
I assume you have realized that Bruno speaks about near-clipping twin tone 18.5+19.5kHz test, not about the 5W test as is used here.
As Bruno says, and I have said many times, the 18.5+19.5 kHz tones do not remotely resemble real content:

"The solution is to blast the amplifier to near-clipping with two sine waves right at the end of the audio band and to inspect the resulting spectrum. You’ll agree that this is just about the worst possible test signal that still technically qualifies as “audio”. Nothing like it ever occurs in real music, so it’s a proper stress test."

He likes IMD tests because THD tests require wide bandwidth. Class D amps suffer here due to switching noise in ultrasonics being pulled in. I run such a test. He avoids it in his and rather see the IMD test that doesn't disadvantage class D.

In addition, harmonics of 1 kHz tone land where our hearing is most sensitive. The intermodulation from those high frequencies do not. Even in less than ideal implementations, IMD products don't rise above threshold of hearing there:

index.php


Notice how the products are all below -105 dB below 15 kHz and almost none in 3 to 5 kHz. And this is for this high amplitudue tones that don't exist. Reduce their levels and those spikes disappear quickly in noise floor.

Regardless, I run these tests as well, mostly to be able to compare them to Stereophile tests. Not because they have any interinsic value to show audibility of distortion.
 
The transparent aim was not to get younger folks interested in high-quality sound as such, but in tinkering with boxes. It won’t even work: when did kids ever get excited about the same thing their parents loved when they were young?

The older generation(s) were driven to tinkering with boxes in the quest for the best sound.
However modern audio hardware completely eclipses anything that was available up until the mid 80's (or possibly earlier) I would say, therefore there is no longer any need to 'tinker with boxes'. Even gear that doesn't measure that well will still sound good to younger generations.

How about giving them what they came into the shop for? A system that reliably sounds good and doesn’t present them with unnecessary choices. Preferably something that requires nothing but a power cable and a phone to play music.

This is a good comment. I also believe that multichannel audio has / is taking over from stereo, especially in countries like the UK where lounge space is limited. Most folk simply don't have space for a surround system and a stereo system, so the multichannel system is used for stereo as well (if / when required).
Also factor in that the younger generation typically listens to music via a phone & IEMs and they have little desire to have a HiFi system, and many more don't even particularly care about sound quality.

Times have changed.

Does anyone remember the video from Meridian when they introduced MQA? They were pulling people off the street and getting them to listen to MQA via a high quality system and headphones and capturing their reactions. Given how few folk have heard a decent system, many of the reactions came as no surprise.
 
His Class D needs a substantial reduction in cost in order to being able to reach a younger generation.

Low-cost Class D is owned by big IC companies -- board-level Class D specialists like Purifi and Hypex know they can't serve the big Bluetooth speaker and sound bar markets profitably. Chips like TI's now-classic TPA and TAS stuff may be popular with cost-conscious audio enthusiasts like some of the ASR crowd, but the marketing target for those parts is mass market products like the ones I mentioned. BTW, several of the key folks from TI's Class D team are now working with Bruno at Purifi, so they clearly knew what they were doing when they designed those chips.
 
Young people do care about audio.They just have other priorities and their time is limited.
A portion of them who does well (REALLY well) cause of jobs related with Internet stuff does buy into nice stuff.usually MC though as they come handy in gaming,etc.
Bruno is Bruno,affordability was never his concern and the misunderstanding about it probably started with ucd's who were mainly a contribution to his beloved DIY community.

Other than that he follows what the rest of western manufacturers do,follows the regulations and offers the high quality he can offer both in design and actual physical circuits.
He would be mad to put himself in the price race,he has no reasons to do so.And as a power lover he probably won't offer low power solutions either.
 
He is completely wrong about this. Class D had and continues to have a terrible reputation for "sound" quality.
To be fair though to Bruno - I doubt he wrote this article for an audience of ASR members.

I suspect the target is specifically those audiophiles with whom the rep of class d sound is poor. In which case pointing out again and again how and why class D sounds superb is the way to chip away at that reputuation.
 
Class d became more popular once people got exposed to board level, rather than chip level solutions. The availability of ucd in the diy market was a huge driver of this. I know a dozen guys with decent class d amps, none of them know this site.
 
But speakers like Grimm and Kii are priced way out of what's achievable for young people. So there's no revolution there and certainly nothing that will save hi-fi.
That's rather seen with studio monitors with class AB, but lacking in the user friendliness.
I have no idea what might rejuvenate the hi-fi industry I only appreciate Bruno’s designs.
Keith
 
Bruno discusses class-d evolution and much else,

Courtesy of hometheatrehifi
Keith
A long but interesting read.

Bruno is one of those rare beasts: An exceptionally talented engineer who is also able to communicate what he is doing in a way that is accessible for people with much less technical knowledge.
 
I have no idea what might rejuvenate the hi-fi industry I only appreciate Bruno’s designs.
Keith
I'm not sure there is much of a demand for HiFi any more. Do the younger generations really care?

Both of my sons were raised in a home where decent HiFi was always present. The eldest (mid 30's) was very keen on his HiFi until he settled down with his partner, and now although he still has a passing interest, he's nowhere near as enthusiastic as he used to be.

My youngest (early 30's) started out with a multichannel system, but that has now gone and he has been quite into Sonos products, although he did recently show a passing interest in a decent pair of bluetooth speakers.

My point is, here we have two people from the younger generation who, whilst they grew up with good HiFi, their interest has waned.

So I'm not sure how it's possible to convince the younger generations that they need separates with all the complexity that entails, when they can get quite reasonable sound quality from a mobile phone and a pair of IEMs, and if they want sound for everyone they either just play it through their phone speakers, or some cheap bluetooth speaker which seems satisfactory to many of them.

As I said in my previous post above, those who are interested in more typically have a multichannel system (or sound bar) which covers all bases.

Personally, I think that stereo is dead to all but the hard core enthusiasts as found here on ASR.
 
As a main system I would much rather have a "Steampunk" stack of streamer, preamp, power amp that does exactly what I want, gives me more power, and costs less than the M33.
A challenge for you! What pile of "Steampunk" boxes would match the sound quality of the M33 at a lower price? And how can that be if the pile of boxes by definition includes 3 cases (often a very costly item), 3 power supplies, 3 power cables and a bundle of interconnects. It just doesn't make sense, or the innards of that pile of boxes is seriously compromised as so much cost goes into what is shared in an all-in-one.

I suspect you've not had an M33 in your system. I have and chose it after comparing sound quality with numerous other amps. It really is very good. ;)
 
Last edited:
The Atma-Sphere monos are ones that I would love to see here. We know they won’t have the lowest distortion measurements but it would be valuable to add it to the database by quantifying the distortion versus frequency of the unit under the same test conditions that all of the other amps have been tested. Perhaps you can coordinate with Amir so that the turnaround time is minimal…

I buy kit not for how well it measures (I've had my fingers burnt twice after doing so), but for how good and lifelike the music sounds. To me, that's really all that matters and choosing an amplifier must be done with one's own speakers connected and installed in one's own listening room. Otherwise, I'm afraid measurement are an academic exercise that merely shows that a particular item "may" sound good - but that's far from a certainly in practice.

My Atma-Sphere amps are in the UK, but otherwise he'd be welcome to borrow them!
 
I think the only solution that can rejuvenate the hi-fi industry is blown away sound quality at a price many can afford.

Unfortunately that's not really possible today without a good acoustical room. At minimum a quite large room with good furnituring and low llevel of early reflections. This greatly rules out most of the younger generation. Advanced sound steering could help.

Personally I have never heard truly great sound at either hifi shows or audio stores. The focus is on wrong things.
 
And how can that be if the pile of boxes by definition includes 3 cases (often a very costly item), 3 power supplies, 3 power cables and a bundle of interconnects. It just doesn't make sense,

How well do you understand electronics and circuit design, or even the basic measurements typically done?
 
A challenge for you! What pile of "Steampumk" boxes would match the sound quality of the M33 at a lower price? And how can that be if the pile of boxes by definition includes 3 cases (often a very costly item), 3 power supplies, 3 power cables and a bundle of interconnects. It just doesn't make sense, or the innards of that pile of boxes is seriously compromised as so much cost goes into what is shared in an all-in-one.

I suspect you've not had an M33 in your system. I have and chose it after comparing sound quality with numerous other amps. It really is very good. ;)

Trouble is, the breathless proponents of the disparate "piles of boxes" only ever test their individual boxes in isolation and never as an entire system from end to end.

The M33 may well outperform that mythical pile of SOTA steampunk boxes, not only in absolute numbers, but also in the areas that actually matter- usability, integration, functionality and reliability.
 
Thread already becoming bizarre.

Wasn't audible noise and distortion in amplifiers solved in the 1970s? Why should we care if Hypex or Topping continue to push the inaudible boundaries? I'm paying for that now?

Can the amp drive the chosen speaker optimally? Yes or no. That's all that matters IMV. The Toppings struggle with low impedance. This N-Core Class D falters into two ohms despite costing nine grand:


As for pre-purchase sighted auditioning of amplifiers like they were musical instruments...
 
Back
Top Bottom