• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Purifi class-D Review and Observations

Status
Not open for further replies.

GUTB

Member
Joined
Dec 29, 2020
Messages
8
Likes
1
I decided to register here to discuss the Purifi modules. I recently purchased a stock Purifi stereo amp from VTV. It was noticeably better than every other class-D I've heard before. However it suffered from a really bad mid-range: blurry and artificial-sounding. Just awful. I almost returned it, but I was talked into buying a pair of VTV custom input boards with Sparkos opamps. The upgrade is on the way and I'll update the thread with my findings. A few other peculiarities:

  1. Burn-in time. It seemed like it needed more than the usual amount of 200 hours, maybe closer to 300 hours. I wasn't counting. But burn-in was essential, out of the box the amp was atrocious: narrow soundstage, no dynamics, lacking in bass control (something I expected it to be good at out of the gate). Unnatural sounding.
  2. Doesn't care about mains quality. I tested my reference amp mains: Lessloss Level 1 filtering cable + Lessloss Firewall 64X into a Furutech GTX-D(r) NCF outlet against a bog standard black (14ga I believe) mains cable into a regular outlet in my listening room. I could not reliably tell the difference between them. Perhaps there was a minor difference but the change is so small it wouldn't overcome audio memory effects. In fact, I think all the class Ds I've tried have had this issue. Perhaps something to do with switching mode power supplies?
  3. The Purifi seems to have been engineered to bring back even-order harmonics -- or at least that what it sounds like to me. The sense I get that in terms of harmonic overtones it's truer to life than what I've experience in all other class Ds. I disagree with the assessment that it's "clinical" or "dry". However this could be because of my all-tube preamp, a Yaqin B-T2 with all NOS and Herbie damping rings. I just got cheap-o NOS tubes as I didn't want to spend big $ on high end NOS if I didn't want to keep it long-term.
  4. Lack of engagement. Compared to my current reference, an Odyssey Kismet, the Purifi is seriously un-engaging. However this may be completely the fault of the compromised input stage. Possibly the ugly midrange is what is killing it for me.
 

ahofer

Major Contributor
Forum Donor
Joined
Jun 3, 2019
Messages
2,532
Likes
4,331
Location
New York City
0E811FC1-48B1-4B18-A39E-FE61B0FCE537.jpeg
 

PaulD

Senior Member
Joined
Jul 22, 2018
Messages
452
Likes
1,316
Location
Other
I decided to register here to discuss the Purifi modules. I recently purchased a stock Purifi stereo amp from VTV. It was noticeably better than every other class-D I've heard before. However it suffered from a really bad mid-range: blurry and artificial-sounding. Just awful. I almost returned it, but I was talked into buying a pair of VTV custom input boards with Sparkos opamps. The upgrade is on the way and I'll update the thread with my findings. A few other peculiarities:

  1. Burn-in time. It seemed like it needed more than the usual amount of 200 hours, maybe closer to 300 hours. I wasn't counting. But burn-in was essential, out of the box the amp was atrocious: narrow soundstage, no dynamics, lacking in bass control (something I expected it to be good at out of the gate). Unnatural sounding.
  2. Doesn't care about mains quality. I tested my reference amp mains: Lessloss Level 1 filtering cable + Lessloss Firewall 64X into a Furutech GTX-D(r) NCF outlet against a bog standard black (14ga I believe) mains cable into a regular outlet in my listening room. I could not reliably tell the difference between them. Perhaps there was a minor difference but the change is so small it wouldn't overcome audio memory effects. In fact, I think all the class Ds I've tried have had this issue. Perhaps something to do with switching mode power supplies?
  3. The Purifi seems to have been engineered to bring back even-order harmonics -- or at least that what it sounds like to me. The sense I get that in terms of harmonic overtones it's truer to life than what I've experience in all other class Ds. I disagree with the assessment that it's "clinical" or "dry". However this could be because of my all-tube preamp, a Yaqin B-T2 with all NOS and Herbie damping rings. I just got cheap-o NOS tubes as I didn't want to spend big $ on high end NOS if I didn't want to keep it long-term.
  4. Lack of engagement. Compared to my current reference, an Odyssey Kismet, the Purifi is seriously un-engaging. However this may be completely the fault of the compromised input stage. Possibly the ugly midrange is what is killing it for me.
Forgive some of the preceding comments! You have probably arrived in the wrong place for that sort of post...

You are amongst friends, but no one here will believe such a post, and it will be heavily scrutinised.

Here is a post you probably should read before going on:
https://www.audiosciencereview.com/...iles-posting-at-asr-for-the-first-time.17598/

ASR is about the science of audio, everything that can be heard can be measured. We can measure to well beyond the limits of our hearing. Controlled listening tests are needed to remove the biases that we all have. They are difficult to do, but worth the effort. I have found them VERY educational, for myself and others.

Here is another post that will help, particularly to do with amplifier "sound": https://www.audiosciencereview.com/...-hypex-nc400-diy-amp.5907/page-64#post-585358 Here is some of that post:

I think the conclusion that you should draw from this is that a double-blind test is required to verify that these effects actually exist rather than being the effects of bias. This is the standard of evidence ASR promotes, not casual listening tests.

I would point you to this post for reference https://www.audiosciencereview.com/...aphic-dac-available.17062/page-17#post-556223
When people do a blind test suddenly all of the "obvious differences" between the amplifiers disappear. It is a very sobering experience to be part of and I HIGHLY recommend taking the time to do it. Also worth reading is the PDF from that post https://linearaudio.nl/sites/linearaudio.net/files/Valves versus Transistors DCD.pdf

This has been proved so many times I am sorry to keep repeating that double-blind and level matched tests are needed to verify audible differences. Casual listening tests, particularly with amplifiers (and DACs), tell us basically nothing. What audible differences actually exist between amplifiers can be readily explained by their measurements (frequency response, distortion, output impedance, and so on).

For more information on testing, try SIY's article on LA here:
https://linearaudionet.solide-ict.nl/sites/linearaudio.net/files/LA Vol 2 Yaniger(1).pdf
And Sean Olive's blog on the dishonesty of sighted listening here https://seanolive.blogspot.com/2009/04/dishonesty-of-sighted-audio-product.html

And directly to the point, here is a review of the Purifiy amplifier that was carefully done by SIY https://audioxpress.com/article/fre...d-audio-bosc-and-purifi-audio-eigentakt-eval1
Here is a quote from that review: "For both of these amplifiers, regardless of where they were inserted in my system, I could discern no difference in the sound, nor was there any difference compared to some pretty fine conventional linear amplifiers that I’ve used recently. No matter what sort of music I played — from acoustic jazz to Americana-folk to electronic progressive rock, whether my own minimally miked uncompressed recordings or highly produced loud studio rock — the amplifiers did exactly what they were supposed to, noiselessly, without stress or strain, and in a perfectly transparent way."

This is not to say that you did not hear the characteristics that you did, with all of the conviction of an eye witness (we've all done that!), but that was not what anyone else has heard when they have carefully examined this unit. So that would suggest you should reexamine the testing and results.
 
Last edited:
OP
G

GUTB

Member
Joined
Dec 29, 2020
Messages
8
Likes
1
oil the capstans on your cassette deck. That will do the trick.
I have three sources.
  1. CD transport from AliExpress. Claims to be a clone of a Wadia circuit. Uses a spring-loaded drive tray for vibration control.
  2. Custom audio PC. Linear ATX power supply, Paul Pang V3 USB card on a 5V USB battery pack, SATA power line isolated to a separate 5V battery pack. Supra USB cable. Running Roon Core, local music library, Fidelizer Pro. Also has HQPlayer but currently not in use.
  3. Triangle Art Concerto + upgraded platter. TA Zeus cart, Reed 2A tonearm. TA speed controller. Chinese all-tube phono stage (clone of an old unit, Marantz I think). DIY built SUT using Hashimoto transformers (off eBay, only way to get Hashimoto transformers in a SUT).
So far I've only played via the transport and audio PC. Spun a record just once. Once the new input boards and opamps come in I'll listen to some more records.
 

Crane

Active Member
Forum Donor
Joined
Dec 16, 2018
Messages
145
Likes
163
This forum is a science believer forum, therefore the responses you see although it shouldn't be the case. People here need to remember most people believe mumbo jumbo until they learn and understand which this forum has done for many.

As measurements and objective testing is what we do here we look for equipment made to recreate the original recordings or sound as intended meaning:

- There is no such thing as burn-in, the only thing relative to this temperature affecting the performance or in physics a material being expanded/collapsed due to the same affect (typically never seen in equipment).

-DAC/Amps/Cables do not affect performance unless seriously built wrong or intended to color the sound by filtration or other means that may cause distortion or noise (tubes do this, so recordings aren't being played as intended).

- Since you have used some subjective terminology, here is one you probably hear a lot "Synergy". This is basically each equipment down the line affecting the other, this only happens with if equipment is badly designed.

- Testing should be done blind, matched, and with multiple trials to ensure what you're hearing is correct.

- Speakers are difficult to design to be flat-on playing the full range, so if you have everything beforehand clean and designed to perfection that is the only thing you need worry from a "sound signature" perspective.

The only advise we can give is that you attempt testing with clean equipment ones designed to avoid distortion and noise. Also, attempt to read more from the forum you can find some helpful guides on many things and look at the measurements performed here to see how each equipment performs. If you do this you'll find out that the purifi amps are as clean and perfect as possible (unless VTV messed up the connections).
 
OP
G

GUTB

Member
Joined
Dec 29, 2020
Messages
8
Likes
1
I guess I'll discuss my history with class D. I've always been interested in the promise of highly efficient amps that do the job linear ones (A and A/B).

My first experiment with the class in my system (ie, not a home theatre setup) was an inexpensive integrated / all-in-one Teac (forget the model) that used an older ICEPower module. Awful, worst amp I've ever heard. Didn't burn it in though so take that with a grain of salt.

Later I tried a D-Sonic stereo amp, the version with the newest (at the time) Pascal module -- M3-800 or something like that. I really liked the dynamics and driver grip thanks to its super high damping power, and it was very resolving / low noise. Ultimately lacked musicality, seemed too "artificial" in tone, and soundstage was struggling, ie an image would "wobble", never experienced that phenomena before.

I gave up on class D for a while from these experiences but then I learned about Technics' new GaNFET modules, something which I thought might go a long way to solve class D's main issue: the lack of switching bandwidth. I picked up a Technics SU-G30 which is a network-enabled "smart" style of all-in-one integrated. Burned it in for over a month: left it in my walk-in closet running its' built-in internet radio feature into a set of little bookshelves I had lying around. I honestly don't know if the burn-in did anything to be honest I don't recall there being a large change. Perhaps it just didn't have enough muscle to handle my speakers in an enjoyable manner, but even 50W a channel should have been enough. I found it sterile and boring, not enjoyable at all. Didn't have the terrific dynamics and grip of the D-Sonic. Ended up selling it.

After this disappointment I gave up on class D yet again -- until I was talked into trying the Purifi module. I read about how the lack of switching bandwidth was eliminated as an issue by a clever application of giant amount of feedback, creating a "self-oscillating amplifier". I have no idea what a self-oscilating amplifier is, but if it could overcome this major limitation as I saw it then I figured it was worth giving a shot. VTV has a money-back guarantee so the only thing it'll cost me is time and maybe return shipping.

EDIT: Forgot about the Crown amp I trailed. Based on the DriveCore2 modules if I recall. Picked it up from Amazon based on recommendations. Very strong dynamics, but also very "blurry", not resolving, as if it suffered from heavy distortion. It was recommended that I use SpeakON connectors for this amp but that was too much effort for me at time to figure out (my speakers didn't use SpeakONs). I think the idea behind that recommendation was that speaker terminals used by the Crown were pretty cheesy. Going by memory of course.
 
Last edited:

VintageFlanker

Major Contributor
Forum Donor
Joined
Sep 20, 2018
Messages
3,357
Likes
11,723
Location
Paris
I decided to register here to discuss the Purifi modules. I recently purchased a stock Purifi stereo amp from VTV. It was noticeably better than every other class-D I've heard before. However it suffered from a really bad mid-range: blurry and artificial-sounding. Just awful. I almost returned it, but I was talked into buying a pair of VTV custom input boards with Sparkos opamps. The upgrade is on the way and I'll update the thread with my findings. A few other peculiarities:

  1. Burn-in time. It seemed like it needed more than the usual amount of 200 hours, maybe closer to 300 hours. I wasn't counting. But burn-in was essential, out of the box the amp was atrocious: narrow soundstage, no dynamics, lacking in bass control (something I expected it to be good at out of the gate). Unnatural sounding.
  2. Doesn't care about mains quality. I tested my reference amp mains: Lessloss Level 1 filtering cable + Lessloss Firewall 64X into a Furutech GTX-D(r) NCF outlet against a bog standard black (14ga I believe) mains cable into a regular outlet in my listening room. I could not reliably tell the difference between them. Perhaps there was a minor difference but the change is so small it wouldn't overcome audio memory effects. In fact, I think all the class Ds I've tried have had this issue. Perhaps something to do with switching mode power supplies?
  3. The Purifi seems to have been engineered to bring back even-order harmonics -- or at least that what it sounds like to me. The sense I get that in terms of harmonic overtones it's truer to life than what I've experience in all other class Ds. I disagree with the assessment that it's "clinical" or "dry". However this could be because of my all-tube preamp, a Yaqin B-T2 with all NOS and Herbie damping rings. I just got cheap-o NOS tubes as I didn't want to spend big $ on high end NOS if I didn't want to keep it long-term.
  4. Lack of engagement. Compared to my current reference, an Odyssey Kismet, the Purifi is seriously un-engaging. However this may be completely the fault of the compromised input stage. Possibly the ugly midrange is what is killing it for me.
Welcome there! With such a first post, you obviously have no clue what ASR is all about (if you do, this would be trolling, but I'll assume it's not the case). Few things then:

- You obviously cannot, in any shape or form, find an amp sounding different after 100, 150, 200 or 300h of burnin whatsoever. Your hearing memory doesn't last that long and there is unlikely not any difference measurable (nor audible) of burnin effect with electronics.
- Of course, your amp doesn't care about mains quality. None should.
- Before trying to evaluate such a transparent amp, I would first suggest you to pair it with an equally transparent preamp/DAC. Your Yaqin is nowhere near this requirement. It is obviously colored and full of harmonic distortion and doesn't do it justice.
-And so on...

Not that big of a deal for sure. Live and learn!;)
 
Last edited:
OP
G

GUTB

Member
Joined
Dec 29, 2020
Messages
8
Likes
1
Welcome there. With such first post, you obviously have no clue what ASR is all about:

- You obviously cannot, in any shape or form, find an amp sounding different after 100, 150, 200 or 300h of burnin whatsoever. Your hearing memory doesn't last that long and there is unlikely not any difference measurable (nor audible) of burnin effect of electronics.
- Of course, your amp doesn't care about mains quality. None should.
- Before trying to evaluate such a transparent amp, I would first suggest you to pair it with an equally transparent preamp/DAC. Your Yaqin is nowhere near this requirement. It is obviously colored and full of harmonic distortion and doesn't do it justice.
-And so on...

Not that big of a deal for sure. Live and learn!;)
Well, I do have a Freya. I could put that into the review rotation. I have a lot of DACs and several amps but only a couple of preamps, and of those only one (Freya) is balanced.

Mains quality helped the Odyssey amps significantly. My first one was a Stratos but I managed to break it by messing around cables on the preamp while it was on. THAT was an expensive mistake, besides taking out the Stratos it also blew on of my ProAc D30R's ribbons which was a $500 repair. The Stratos is pretty old as I had bought it used. I tried to fix it myself, had some success then killed off one of the channels for good, maybe a short on the back of the PCB...something which would be a huge PITA to get at. Found a good used price for a Kismet and just got that. Ultimately though the Odysseys are fillers until I can afford to get an amp I really want -- Mark Levinson No.585, Boulder 866, perhaps an Esoteric, maybe even a T+A if I can find a great used price. Unless, of course, I can get something as good out of a Purifi at a fraction of the cost. That's the dream.
 
Status
Not open for further replies.
Top Bottom