Before going back to work tomorrow, I want to give a final update on today's listening tests, begging your pardon. I'm not super-comfortable with this audiophile stuff, and I'm a little embarrassed by it all. But I realize you guys are paying me a generous salary to write about it, and I don't want find myself in breach of contract.
First thing is, I've heard the sound of the A3 change during warm-up so many times now that I believe it's real and not just my imagination. I didn't figure this out until today, so many of my earlier comments were based on listening to a cold or lukewarm amp. I will wait until I get the new power supply built, then start again with a fresh review and a properly warmed-up amplifier.
Second thing is, I've been trying to evaluate the A3 in comparison to the A2 and it's been really frustrating because I come away with different impressions every time. I thought it was just me being a bad listener, but I'm not so sure now. All I can say is the A2, to my ears, has improved dramatically over the weekend. If it sounded bright or overly-analytical before, I can't hear that at all now. In fact, I feel it's right up there with the A3. This is extremely inconvenient because it doesn't fit the tidy narrative about complex circuits and high-feedback amps versus simple, low-feedback ones. Plus, I have no credentials as an audio designer. It makes no sense.
When I replaced the attenuator in the A2 I was discouraged by the sound and stopped listening to it. In fact, I've been busy and preoccupied and have barely turned the amp on for weeks. At the time, Mark told me to reserve judgment on the attenuator until it had broken in. And I was like, yeah, sure Mark, whatever you say. I can only speculate that Mark was right. Or maybe the capacitors in the A2 got a case of the sadz during their time off, and needed some time to wake up again. Or I'm imagining everything. Dunno.
This is a weird hobby. Sometimes you have experiences that totally surprise you. This is one of those moments.
I'm not going to try to explain this right now. Jam, you're right about the Twilight Zone. Ridiculous; I shoulda taken the blue pill.
BTW, I start a new job Tuesday, so don't be surprised if the entertainment stops for a while.
First thing is, I've heard the sound of the A3 change during warm-up so many times now that I believe it's real and not just my imagination. I didn't figure this out until today, so many of my earlier comments were based on listening to a cold or lukewarm amp. I will wait until I get the new power supply built, then start again with a fresh review and a properly warmed-up amplifier.
Second thing is, I've been trying to evaluate the A3 in comparison to the A2 and it's been really frustrating because I come away with different impressions every time. I thought it was just me being a bad listener, but I'm not so sure now. All I can say is the A2, to my ears, has improved dramatically over the weekend. If it sounded bright or overly-analytical before, I can't hear that at all now. In fact, I feel it's right up there with the A3. This is extremely inconvenient because it doesn't fit the tidy narrative about complex circuits and high-feedback amps versus simple, low-feedback ones. Plus, I have no credentials as an audio designer. It makes no sense.
When I replaced the attenuator in the A2 I was discouraged by the sound and stopped listening to it. In fact, I've been busy and preoccupied and have barely turned the amp on for weeks. At the time, Mark told me to reserve judgment on the attenuator until it had broken in. And I was like, yeah, sure Mark, whatever you say. I can only speculate that Mark was right. Or maybe the capacitors in the A2 got a case of the sadz during their time off, and needed some time to wake up again. Or I'm imagining everything. Dunno.
This is a weird hobby. Sometimes you have experiences that totally surprise you. This is one of those moments.
I'm not going to try to explain this right now. Jam, you're right about the Twilight Zone. Ridiculous; I shoulda taken the blue pill.
BTW, I start a new job Tuesday, so don't be surprised if the entertainment stops for a while.
A couple of people have asked about my A3 PC board files. I've decided to split off a separate A3 build thread and I will start by publishing my signal board Gerbers, schematic, and BOM there. I'm going to hold off on the Kicad files because I'm satisfied with the layout and I don't want to encourage a dozen variants to pop up in the field. Once I design the power supply board, I will publish Gerbers for that as well. I will get this going by tonight, hopefully.
I'm want to do this because I think the HPA-1 clone is a classic design and, with a little encouragement, has the potential to attract more builders.
I will reserve this thread for any future projects I come up with, to support people building the A2, and as a place for my yammering.
About my trying to do subjective reviews of my amps, I guess I still have a lot to learn. I'm not prepared to go full audiophool, but it seems things aren't as simple as I thought. I try to post information I can back up with facts and reason. So I'm going to limit the subjective posts until I feel I have a better handle on all of this.
I'm want to do this because I think the HPA-1 clone is a classic design and, with a little encouragement, has the potential to attract more builders.
I will reserve this thread for any future projects I come up with, to support people building the A2, and as a place for my yammering.
About my trying to do subjective reviews of my amps, I guess I still have a lot to learn. I'm not prepared to go full audiophool, but it seems things aren't as simple as I thought. I try to post information I can back up with facts and reason. So I'm going to limit the subjective posts until I feel I have a better handle on all of this.
Tonight's project was to build a 12V auxiliary power supply to run the muting relay. That's done now, so no more being chained to the workbench and the HP power supply.
Next step is to put a bi-color LED on the front panel and hook it up to the timer connector. Can't live without that LED. The problem with these breadboards is I can't keep the cat from walking in the circuitry. :-(
I decided it would be best to ask the Pass Labs folks if there is any concern about IP before I release board files. I don't believe there's a strict legal issue, but think it's best to be polite and inquire anyway. I posted over there and didn't get a response, but I'll wait a little longer. I think instead of starting a build thread right now, I will just give out the files privately on demand and then reconsider a build thread once the power supply board is done.
Next step is to put a bi-color LED on the front panel and hook it up to the timer connector. Can't live without that LED. The problem with these breadboards is I can't keep the cat from walking in the circuitry. :-(
I decided it would be best to ask the Pass Labs folks if there is any concern about IP before I release board files. I don't believe there's a strict legal issue, but think it's best to be polite and inquire anyway. I posted over there and didn't get a response, but I'll wait a little longer. I think instead of starting a build thread right now, I will just give out the files privately on demand and then reconsider a build thread once the power supply board is done.
Yeah, ok, quick listening impression. I didn't play music at all today until about a half hour ago. So something I guess I learned yesterday is that listener fatigue is a thing, and it helps to rest your ears or everything starts to sound the same. A2 still sounds very nice this evening, but the A3 is more... vivid... I would call it. I won't say any more for now.
I received no reply to my posting on the Pass Labs forum. Given NP's historical support for hobbyists building copies of his products, I assume no one objects to my releasing the A3 design files. As I said earlier, for now I will keep the distribution private. Send me a PM for a link.
The Dropbox folder contains the Gerbers, the schematic, and a copy of my Mouser order form. The latter does not include the audio transistors. Name-brand BC550C and BC560C are discontinued or out of stock at the major distributors, but seemingly reliable sources can be found on eBay. The other devices can still be had, but you will need to order from several suppliers. An indispensable resource is octopart.com. Check it out if you haven't already.
The 500 Ohm trimpot is mostly out of stock everywhere. RS components had 39 in stock when I checked earlier this evening. The JFETs come from the diyaudio store.
The 2-pin, 0.1" Molex receptacle housing is out of stock everywhere. TE Connectivity has a similar connector series that could be used instead.
The schematic shows an Omron relay part number, but I used the Kemet EA2-12NJ. This is in limited supply and marked as end-of-life. The substitute is EA2-12NU, which is the same relay with shorter pins, IIRC. The latter part is unavailable. Any other quality relay in the same form factor will work as well.
I used the tiny Vishay CPF2 parts in a 2W rating for the power resistors. They are more than adequate in this application. Be mindful of the small footprint; standard 3W parts will not fit.
Pads for the two "clean" ground wires are covered by resistors on top. The wires are intended to be soldered from underneath the board.
I believe the information in the files to be correct, but have not triple checked. I recommend that you verify everything carefully before placing orders for boards and parts.
If anyone actually gets around to building this thing, we can talk later about some more technical details.
-----
I prefer not to explain what prompted me to consider the IP rights for this project. This design is intended for personal use only. Please respect the folks at Pass Labs and do not redistribute the files or go into business selling kits or finished clone products. I mean that seriously.
The Dropbox folder contains the Gerbers, the schematic, and a copy of my Mouser order form. The latter does not include the audio transistors. Name-brand BC550C and BC560C are discontinued or out of stock at the major distributors, but seemingly reliable sources can be found on eBay. The other devices can still be had, but you will need to order from several suppliers. An indispensable resource is octopart.com. Check it out if you haven't already.
The 500 Ohm trimpot is mostly out of stock everywhere. RS components had 39 in stock when I checked earlier this evening. The JFETs come from the diyaudio store.
The 2-pin, 0.1" Molex receptacle housing is out of stock everywhere. TE Connectivity has a similar connector series that could be used instead.
The schematic shows an Omron relay part number, but I used the Kemet EA2-12NJ. This is in limited supply and marked as end-of-life. The substitute is EA2-12NU, which is the same relay with shorter pins, IIRC. The latter part is unavailable. Any other quality relay in the same form factor will work as well.
I used the tiny Vishay CPF2 parts in a 2W rating for the power resistors. They are more than adequate in this application. Be mindful of the small footprint; standard 3W parts will not fit.
Pads for the two "clean" ground wires are covered by resistors on top. The wires are intended to be soldered from underneath the board.
I believe the information in the files to be correct, but have not triple checked. I recommend that you verify everything carefully before placing orders for boards and parts.
If anyone actually gets around to building this thing, we can talk later about some more technical details.
-----
I prefer not to explain what prompted me to consider the IP rights for this project. This design is intended for personal use only. Please respect the folks at Pass Labs and do not redistribute the files or go into business selling kits or finished clone products. I mean that seriously.
I started my new job this week and it looks like I'm going to be scrambling for a while to get up to speed. I've been doing some listening, but taking a break, out of necessity, from 24x7 amplifier obsession. I need to rethink how I evaluate amplifiers, as the full-frontal approach wasn't very helpful. Here are some short comments while I wait to build the new A3 power supply.
I'm still impressed by how fun the A3 is to listen to. It's got great "PRAT." The most impressive qualities are how natural and unstressed the high frequencies are, the ambience, and the size of the soundstage. The downsides are a slight coloration and some sloppiness in the midbass.
The A30 Pro, in contrast, seems to do practically every individual thing right, but it sounds like a collection of parts that don't quite fit together. The highs are a little unnatural and I swear I hear some congestion or grain sometimes. But I would never think to criticize it if I had nothing else to compare it to.
The A2 is kind of in the middle. TBH, I'm impressed with it. You may remember I was alarmed when I first listened to it because it sounded soft or rolled-off on top. The A3 is like this, too. I'm starting to think that the softness is actually a lack of high-frequency artifacts. The A2 doesn't rock out with the same reckless abandon as the A3. Ok, I'm biased. But I like it. I'm listening to the A2 right now with my K702s and having a good time.
The A3 has been cooking in the other room all afternoon (to keep the cat from walking in it and chewing on live wires). I was listening earlier and I'm going to go listen some more now. I've got a different DAC (AMB Gamma2) in there, so it's not a fair comparison, but we'll see.
Anyway, I think I can recommend both the A2 and the A3 as good projects to build. I'm not sure which is my favorite between the two, or if it matters. I haven't listened to the A1 recently, so no comment. Last time I played it, I was happy too.
It's easy to exaggerate these differences, so don't take my word for it. "It's just entertainment."
Edit: Yup, I just did the comparison with the A3 after listening to the A2. I'm listening to Gabor Szabo's "Bacchanal" album. This is not the cleanest recording, but it's got a nice acoustic sound with lots of reverb. The first think you hear is that the images are bigger on the A3, instruments are placed farther "out there" and have more of that uncanny realism. But also the midbass exaggeration. Now, I'm back on the A2, and it's great, too, but read what I said above.
I'm still impressed by how fun the A3 is to listen to. It's got great "PRAT." The most impressive qualities are how natural and unstressed the high frequencies are, the ambience, and the size of the soundstage. The downsides are a slight coloration and some sloppiness in the midbass.
The A30 Pro, in contrast, seems to do practically every individual thing right, but it sounds like a collection of parts that don't quite fit together. The highs are a little unnatural and I swear I hear some congestion or grain sometimes. But I would never think to criticize it if I had nothing else to compare it to.
The A2 is kind of in the middle. TBH, I'm impressed with it. You may remember I was alarmed when I first listened to it because it sounded soft or rolled-off on top. The A3 is like this, too. I'm starting to think that the softness is actually a lack of high-frequency artifacts. The A2 doesn't rock out with the same reckless abandon as the A3. Ok, I'm biased. But I like it. I'm listening to the A2 right now with my K702s and having a good time.
The A3 has been cooking in the other room all afternoon (to keep the cat from walking in it and chewing on live wires). I was listening earlier and I'm going to go listen some more now. I've got a different DAC (AMB Gamma2) in there, so it's not a fair comparison, but we'll see.
Anyway, I think I can recommend both the A2 and the A3 as good projects to build. I'm not sure which is my favorite between the two, or if it matters. I haven't listened to the A1 recently, so no comment. Last time I played it, I was happy too.
It's easy to exaggerate these differences, so don't take my word for it. "It's just entertainment."
Edit: Yup, I just did the comparison with the A3 after listening to the A2. I'm listening to Gabor Szabo's "Bacchanal" album. This is not the cleanest recording, but it's got a nice acoustic sound with lots of reverb. The first think you hear is that the images are bigger on the A3, instruments are placed farther "out there" and have more of that uncanny realism. But also the midbass exaggeration. Now, I'm back on the A2, and it's great, too, but read what I said above.
Last edited:
I wouldn't know how to begin answering that question. The only way I know for sure to determine if an amplifier is any good is to Send It To Amir For Testing.
I'm about to throw in the towel on listening comparisons. It's just impossible, and no amount of analysis is going to get to the bottom of it. Sometimes things sound great, sometimes they sound like garbage, there's no rhyme or reason to it.
I've been listening to the A3 for a few hours now at background level, so no chance of hurting my ears, but definitely plenty of time for them to adjust. If the A3 is a little fat on the bottom, it helps it sound good when playing quietly. I'm in Roon Radio mode, so it's just playing a lot of different music, and I'm not paying much attention. But Steely Dan came on, and I like Steely Dan, so I turned it back up. Wow, what is this awesome sound?
My first instinct is to unplug the A3 and hook up the A2 to compare. But I'm not going to do that. I've tried playing that game and it didn't get me the answers I sought. Enough is enough.
I have a theory. First of all, I think I believe Jam: this A3 needs to be on ALL the time. Two hours is like the minimum warm up, but it sure seems to improve even after that. I would leave it on all night, but the kitty likes to walk on it, and chew the wires.
But here's the thing. I constantly have the experience that music starts out sounding bad, and after a while it just gets better and better. People describe this as warm-up. Could it actually be the brain warming up? Or, more accurately, tuning in?
I am starting to think that during the day, when I'm doing my job, that I may be physically incapable of listening to music properly. I know I can't write computer software with headphones playing. It's too distracting. Perhaps the reverse is true, and intense analytic thinking and task-oriented activity interfere with auditory signal processing in the brain. I say this because when I kick back for a while and just listen, I start to hear better. A lot better.
I know it sounds fruity, but the brain is an incredibly complex machine, and we barely understand how it works. In all seriousness, the ability to hear differences between audio components may have less to do with the ears, and more to do with attention and focus. I seem to remember that early theories of perception and consciousness modeled the brain as a kind of movie screen onto which our senses project images of the outside world. But now it's understood to be a lot more complex and arbitrary than that. This may explain my trouble figuring out how these amps sound, and why doubling down trying to analyze it doesn't help.
Just saying. It may be nonsense. But tell me, seriously, where else can you read stuff like this? ;-)
"And I liked the A3 so much, I bought one for myself..."
I've been listening to the A3 for a few hours now at background level, so no chance of hurting my ears, but definitely plenty of time for them to adjust. If the A3 is a little fat on the bottom, it helps it sound good when playing quietly. I'm in Roon Radio mode, so it's just playing a lot of different music, and I'm not paying much attention. But Steely Dan came on, and I like Steely Dan, so I turned it back up. Wow, what is this awesome sound?
My first instinct is to unplug the A3 and hook up the A2 to compare. But I'm not going to do that. I've tried playing that game and it didn't get me the answers I sought. Enough is enough.
I have a theory. First of all, I think I believe Jam: this A3 needs to be on ALL the time. Two hours is like the minimum warm up, but it sure seems to improve even after that. I would leave it on all night, but the kitty likes to walk on it, and chew the wires.
But here's the thing. I constantly have the experience that music starts out sounding bad, and after a while it just gets better and better. People describe this as warm-up. Could it actually be the brain warming up? Or, more accurately, tuning in?
I am starting to think that during the day, when I'm doing my job, that I may be physically incapable of listening to music properly. I know I can't write computer software with headphones playing. It's too distracting. Perhaps the reverse is true, and intense analytic thinking and task-oriented activity interfere with auditory signal processing in the brain. I say this because when I kick back for a while and just listen, I start to hear better. A lot better.
I know it sounds fruity, but the brain is an incredibly complex machine, and we barely understand how it works. In all seriousness, the ability to hear differences between audio components may have less to do with the ears, and more to do with attention and focus. I seem to remember that early theories of perception and consciousness modeled the brain as a kind of movie screen onto which our senses project images of the outside world. But now it's understood to be a lot more complex and arbitrary than that. This may explain my trouble figuring out how these amps sound, and why doubling down trying to analyze it doesn't help.
Just saying. It may be nonsense. But tell me, seriously, where else can you read stuff like this? ;-)
"And I liked the A3 so much, I bought one for myself..."
You finally found it! Yet I bet you won't be able to really let go .
Read Dolby C theory! It's the best thing you can do !l
Read Dolby C papers once again!
You don need to remember anything , just read Ray Dolby's papers.There's a reason why Dolby's patents dominate the audio industry for 50 years.
I know. Hence the smiley. TBH, much as I acknowledge and respect your and Jam's subjectivist leanings, I've never taken this golden-eared stuff all that seriously, and until recently had decided it didn't really matter. I'm still not convinced, but since some people seem to find my rants amusing, I thought I would share my growing befuddlement. It's clear the methodology is flawed, since my results fail to converge. To the extent that you are trying to turn me into a true believer, you should be encouraged by my anti-progress.
I had an interesting experience a few years ago. I took my son to a show called "Star Wars in Concert." There was a live orchestra there, and they played the film score to clips from the movies synchronized to the music. I was amused to find that I could hear, but not understand the music as long as I looked at the screens. I could only make sense of what I was hearing if I looked away or closed my eyes.
I don't read subjectivist forums because 99% of it is nonsense. I do read hi-fi reviews and there's a lot more straight-out assertions about sound quality nowadays, and a lot less of the kind of thoughtful framing I seem to remember from, say, TAS back when the high-end was a new concept and its proponents were trying to build a framework to justify the methodology. Now that the concept of high-end is mainstream, a lot of charlatans have moved in to take advantage.
I'm a stubborn individualist. You can tell me what to do, but I won't believe it until I hear it and explain it for myself. I don't expect anyone to believe what I say, and I barely believe it myself. I am starting to think most of these differences are due to perceptual shifts, and have much less to do with the actual stimuli. The one hope seems to be that maybe the differences are real, but only perceptible if the listener is properly trained and in the right frame of mind. This may be a double-edged sword, because it seems equally possible that getting in the right frame of mind is akin to religious indoctrination.
I'm mainly doing this A3 project because you put me up to it, Mark. It's a big investment in time and money and it will be a waste if I can't get some positive results. Well, not really a waste, but definitely a disappointment. I'm very wary, though, of talking myself into believing in baloney.
Edit: Actually, I'm a little uncomfortable with the direction all this is heading, and I blame you and Jam, hah. My original intent was to keep things technical.
Last edited:
Further thought: I'm not very interested in any of this subjectivist stuff unless I can relate it to some concrete physical mechanisms. I have recently been prodded by someone I respect to consider that I may be too focused on materialism (in the philosophical sense) and should be more open to metaphysics. I am, like, ok, it's not that I am closed to metaphysics, but I don't know what to make of it. The only conclusion I've come to so far is that it may be human nature to have faith in things we can't prove. In fact, there seems to be a lot of that going around lately, if you read the news.
I haven't done a deep dive on psychoacoustics, but most of what I've read seems focused on laboratory measurements of listener sensitivity to low-level stimuli, double-blind tests, and so on. I am starting to understand why most "serious" audiophiles are so hostile to DBTs. I think the only path forward is through a general theory of human cognition. You need to model the neuropsychological perception of "beauty," and good luck with that.
The ONLY reason I'm indulging in this is because I can't justify building more amplifiers if they all sound the same. I've already built a couple of "perfect" amplifiers. It's not that interesting anymore otherwise.
I haven't done a deep dive on psychoacoustics, but most of what I've read seems focused on laboratory measurements of listener sensitivity to low-level stimuli, double-blind tests, and so on. I am starting to understand why most "serious" audiophiles are so hostile to DBTs. I think the only path forward is through a general theory of human cognition. You need to model the neuropsychological perception of "beauty," and good luck with that.
The ONLY reason I'm indulging in this is because I can't justify building more amplifiers if they all sound the same. I've already built a couple of "perfect" amplifiers. It's not that interesting anymore otherwise.
There are lots of things that could be said. Let's take DBT hostility for one example. Typical DBT is harder than it seems like it should be. PMA, one of our more measurement oriented forum members, used to put on listening tests where members could download audio files to test their blind listening skills using Foobar ABX. Eventually, he found that he could pass ABX and thus confirm his sighted listening impressions. However, he said it was indeed much harder to learn how to pass ABX than it seems like it should be. It takes a lot of practice and extended concentration. His finding is consistent with the complaints of people who are DBT hostile, and consistent with the findings of perceptual testing experts in other areas such as food taste research for the food industry. ABX has comparatively poor sensitivity as compared to some other blind protocols, such as, for example, A/B. Why? Apparently due to something about that mystery of exactly how brain processing works. The field of "Auditory Scene Analysis' is where the research action is today. I would not give too much weight to some of old psychoacoustics
Last edited: