How far have ss amps really come in the last twenty years??

[andreasmaaan]

Hopefully equipment that tests well makes you happy as well, if it doesn't, you may never scientifically know why.

I think the point others are making is that there is a very clear way to know why. If the differences you hear vanish under controlled conditions, then they can be attributed to factors other than how they sound.

 

[LTig]

A look at the development of Genelec speakers over the past 30 years gives clues to the question «what’s happened on the SS amplifier front in the last decades?». Genelec has made active speakers from day one since the company was established over 40 years ago.

These are data of the largest Genelec amplifier in the big 1234 model and its predecessor (called 1034):

Genelec anno 1989
1034a amplifier weight: 71 kg
1034a amplifier SNR (b, m, t): 101 dB, 105 dB, 106 dB
1034a THD: <= 0,05%

Genelec anno 1998-2000
1034a amplifier weight: 30 kg
1034a amplifier SNR (b, m, t): >=100 dB, >=100 dB, >= 100 dB
1034a THD: <= 0,05%

Genelec anno 2015
1234 amplfier weight: 11 kg
1234 amplifier SNR (b, m, t): >116 dB, >119 dB, >115 dB
1234 THD: <0,003%

So it seems like lots and lots of weight has been «lost» and specifications are a bit better than they used to be 30 years ago.

The specs are not just a bit better. THD is more than a factor of 10 better, SNR at least by a factor of 3. And all this with 15% of the weight ...

 

[LTig]

Remember, on this site, there are folks who only listen with their measuring gear and get confused if they happen to see anything with their eyes. One thing is certain, they do NOT like to listen with their ears, LMAO. Probably a good thing, as I seriously doubt that they have the ability to hear much anyway...at least according to what they report!!!

You know nothing, Jon Snow.

 

[Xulonn]

Pioneer SX-1980 please. - Some of those old receivers are absolutely gorgeous...

If you are interested in 1960's-1980's vintage audio gear, check out the excellent "Vintage Audio Addict" website - LINK. I find this to be an interesting topic, but it does predate the time period of "the last 20 years" noted in the title of this thread.

The extensive collection of projects at Vintage Audio Addict includes the restoration of a Pioneer SX-1980, and the website shows what is inside this and other classic components. One fascinating aspect of his presentation style is the opportunity to observe the evolution of audio circuits from the first one - point-to-point wiring - to including one or more simple printed circuit boards, and then moving on to more modern-style complex, high component density PCBs. If you look at some of his projects, you can also see some of the different circuit "styles" used by various manufacturers. The website's presentation style is not a step-by-step how-to, but rather the showing of a few interesting points in each project.

Below is a pic of some of the heat related damage in an SX-1980 after 35 years of use...

 

[dwkdnvr]

Until one goes into a test where the answer is known in advance, and provides the same answer, then what they think about accuracy of their hearing is just wrong.

I agree with this, but think that a big problem is that it is actually pretty difficult for lone individuals to perform ABX tests of reasonable quality on their own. This can leave 'us' in a position where the path forward is unclear if our subjective impressions are inserting themselves into the evaluation.

I'm kinda dealing with that right now. I have what I'd generally call an 'ASR Approved' system. Khadas Tone Board into JDS Atom into Neurochrome Mod-86; sounds absolutely fantastic driving my Kef Q150's in the desktop system, or driving the NHT Xd tops as part of a 2-way active system with RSS210 bass bins. The NHT Xds / Dayton RSS210 system in particular is utterly satisfying, and I'm trying to figure out how to make it our 'living room' system. (the JRiver based DSP setup is a non-starter for the living room due to the complexity of setup, switching inputs, dealing with Windows etc)

BUT

When I insert a MiniDsp Nano-digi in bypass mode into the system, it collapses into what I'd call 'generic mid-fi sound'. The subjective impression is that the sense of music emerging from a 'black background' is lost, and the texture/harmonic structure and rhythm get a bit muddled. All the NanoDigi is doing is applying SRC to 96KHz on an spdif input and outputting another spdif signal, and so the reasonably objective side of me feels that there's no way it could possibly be influencing things to this degree unless the engineering was surprisingly poor. But the impression has been consistent across 3 different system configurations that I've tried it on. (both in pure bypass and when reproducing the DSP chain implemented in JRiver)

So, what is one to do? I can't really ABX myself to any degree of reliability by myself, and my wife would shoot me on the spot if I tried to drag her into the process. (leaving aside the problem that with only 1 input the Tone Board requires physical re-configuration to switch setups which is suspect on it's own from an ABX perspective) I feel like I'm in Limbo. I REALLY want the Nano-Digi to be viable since it greatly simplifies the system design, but my subjective impression (which I fully recognize may not be accurate) is that it's not an acceptable level of fidelity.

So, ultimately this is just expressing a bit of sympathy for subjectivist perspective that ultimately we want to be happy listening to our systems, and at some point we might just have to take a flyer and accommodate our flawed perceptions if there isn't another practical answer due to the difficulty of validating our perceptions via testing/measurements.

 

[Blumlein 88]

I agree with this, but think that a big problem is that it is actually pretty difficult for lone individuals to perform ABX tests of reasonable quality on their own. This can leave 'us' in a position where the path forward is unclear if our subjective impressions are inserting themselves into the evaluation.

I'm kinda dealing with that right now. I have what I'd generally call an 'ASR Approved' system. Khadas Tone Board into JDS Atom into Neurochrome Mod-86; sounds absolutely fantastic driving my Kef Q150's in the desktop system, or driving the NHT Xd tops as part of a 2-way active system with RSS210 bass bins. The NHT Xds / Dayton RSS210 system in particular is utterly satisfying, and I'm trying to figure out how to make it our 'living room' system. (the JRiver based DSP setup is a non-starter for the living room due to the complexity of setup, switching inputs, dealing with Windows etc)

BUT

When I insert a MiniDsp Nano-digi in bypass mode into the system, it collapses into what I'd call 'generic mid-fi sound'. The subjective impression is that the sense of music emerging from a 'black background' is lost, and the texture/harmonic structure and rhythm get a bit muddled. All the NanoDigi is doing is applying SRC to 96KHz on an spdif input and outputting another spdif signal, and so the reasonably objective side of me feels that there's no way it could possibly be influencing things to this degree unless the engineering was surprisingly poor. But the impression has been consistent across 3 different system configurations that I've tried it on. (both in pure bypass and when reproducing the DSP chain implemented in JRiver)

So, what is one to do? I can't really ABX myself to any degree of reliability by myself, and my wife would shoot me on the spot if I tried to drag her into the process. (leaving aside the problem that with only 1 input the Tone Board requires physical re-configuration to switch setups which is suspect on it's own from an ABX perspective) I feel like I'm in Limbo. I REALLY want the Nano-Digi to be viable since it greatly simplifies the system design, but my subjective impression (which I fully recognize may not be accurate) is that it's not an acceptable level of fidelity.

So, ultimately this is just expressing a bit of sympathy for subjectivist perspective that ultimately we want to be happy listening to our systems, and at some point we might just have to take a flyer and accommodate our flawed perceptions if there isn't another practical answer due to the difficulty of validating our perceptions via testing/measurements.

OH, yes this is a problem. Do you have a way to record the output of the Tone board? One straight in and one via the Nano-digi. If the sound is different the signal has to be different.

 

[March Audio]

I agree with this, but think that a big problem is that it is actually pretty difficult for lone individuals to perform ABX tests of reasonable quality on their own. This can leave 'us' in a position where the path forward is unclear if our subjective impressions are inserting themselves into the evaluation.

I'm kinda dealing with that right now. I have what I'd generally call an 'ASR Approved' system. Khadas Tone Board into JDS Atom into Neurochrome Mod-86; sounds absolutely fantastic driving my Kef Q150's in the desktop system, or driving the NHT Xd tops as part of a 2-way active system with RSS210 bass bins. The NHT Xds / Dayton RSS210 system in particular is utterly satisfying, and I'm trying to figure out how to make it our 'living room' system. (the JRiver based DSP setup is a non-starter for the living room due to the complexity of setup, switching inputs, dealing with Windows etc)

BUT

When I insert a MiniDsp Nano-digi in bypass mode into the system, it collapses into what I'd call 'generic mid-fi sound'. The subjective impression is that the sense of music emerging from a 'black background' is lost, and the texture/harmonic structure and rhythm get a bit muddled. All the NanoDigi is doing is applying SRC to 96KHz on an spdif input and outputting another spdif signal, and so the reasonably objective side of me feels that there's no way it could possibly be influencing things to this degree unless the engineering was surprisingly poor. But the impression has been consistent across 3 different system configurations that I've tried it on. (both in pure bypass and when reproducing the DSP chain implemented in JRiver)

So, what is one to do? I can't really ABX myself to any degree of reliability by myself, and my wife would shoot me on the spot if I tried to drag her into the process. (leaving aside the problem that with only 1 input the Tone Board requires physical re-configuration to switch setups which is suspect on it's own from an ABX perspective) I feel like I'm in Limbo. I REALLY want the Nano-Digi to be viable since it greatly simplifies the system design, but my subjective impression (which I fully recognize may not be accurate) is that it's not an acceptable level of fidelity.

So, ultimately this is just expressing a bit of sympathy for subjectivist perspective that ultimately we want to be happy listening to our systems, and at some point we might just have to take a flyer and accommodate our flawed perceptions if there isn't another practical answer due to the difficulty of validating our perceptions via testing/measurements.

Have to say my experience with the minidsp boards I tried was the same. It just did something to kill the sound. On a technical basis I did establish their enforced re-sampling was not the best, but never got much further with investigation to what the overall problem might be.

 

[amirm]

I agree with this, but think that a big problem is that it is actually pretty difficult for lone individuals to perform ABX tests of reasonable quality on their own. This can leave 'us' in a position where the path forward is unclear if our subjective impressions are inserting themselves into the evaluation.

If you can't do your own controlled testing, then trust the advice of the people who do run them!

That is why many of us are here. We run the tests you cannot.

 

[dwkdnvr]

OH, yes this is a problem. Do you have a way to record the output of the Tone board? One straight in and one via the Nano-digi. If the sound is different the signal has to be different.

At the moment my highest resolution device is a Focusrite Scarlett 6i6, and I'm not sure it's up to the task of revealing something like this. I'm expecting/planning to get a better interface, but only when my system design crystallizes which is something of a circular problem.

I do have an Emotiva Bas-X A-100 that is unallocated at the moment and makes a frequent appearance in the 'amps we'd like to see tested' thread, so maybe I should send that to Amir and throw the NanoDigi into the box to see whether the AP tests reveal anything obvious.

 

[dwkdnvr]

If you can't do your own controlled testing, then trust the advice of the people who do run them!

That is why many of us are here. We run the tests you cannot.

Sure - that's how I ended up with the Khadas/JDS/Neurochrome setup. $650 all-in which is just crazy value. But you can't test everything, unfortunately.

 

[dwkdnvr]

Have to say my experience with the minidsp boards I tried was the same. It just did something to kill the sound. On a technical basis I did establish their enforced re-sampling was not the best, but never got much further with investigation to what the overall problem might be.

Interesting that you found this as well. Given that the SHD unit did very well in Amir's tests, the one flaw was jitter spikes that sure looked to me like they were potentially related to the SRC process. It didn't seem to compromise SINAD/distortion numbers though, so I was left unsure whether they improved the implementation in the SHD vs earlier units or whether my guess that SRC might be a problem was a red herring.

 

[Juhazi]

More OT - Minidsp boards have sort of different generations or species. The first 2x4 definitely has low performance, the Nanodigi and 4x10HD are better. 2x4HD is even better and the latest SHD yet better. But dirtect comparison is not possible because many units are systems with several modules. I am happy with my 4x10HD and 2x4HD boxed units and I can't hear any degradation of sound with them.

Neurochrome's Tom has measured the 4x10HD https://www.neurochrome.com/minidsp-4x10-hd/

 

[Blumlein 88]

At the moment my highest resolution device is a Focusrite Scarlett 6i6, and I'm not sure it's up to the task of revealing something like this. I'm expecting/planning to get a better interface, but only when my system design crystallizes which is something of a circular problem.

I do have an Emotiva Bas-X A-100 that is unallocated at the moment and makes a frequent appearance in the 'amps we'd like to see tested' thread, so maybe I should send that to Amir and throw the NanoDigi into the box to see whether the AP tests reveal anything obvious.

Well, I say run the test with what you have. If the 6i6 shows nothing that at least limits how large the differences are. Might want to use pkane's software too.

https://deltaw.org/

Free download of a very useful nulling tool with many other functions as well. pkane is a member here and he can help if needed or some of us who have used it will help if you need it.

 

[Hypnotoad]

Below is a pic of some of the heat related damage in an SX-1980 after 35 years of use...

Discolouration from hot voltage regulators? Some of these are over 50 years old and the electrolytics would fail long before the regulators. My sons recent expensive Yamaha receiver has to have a snubber capacitor replaced every few years, from poor circuit design. A lot of new gear would not last anywhere near as long as some of these vintage units have.

 

[andreasmaaan]

At the moment my highest resolution device is a Focusrite Scarlett 6i6, and I'm not sure it's up to the task of revealing something like this. I'm expecting/planning to get a better interface, but only when my system design crystallizes which is something of a circular problem.

I do have an Emotiva Bas-X A-100 that is unallocated at the moment and makes a frequent appearance in the 'amps we'd like to see tested' thread, so maybe I should send that to Amir and throw the NanoDigi into the box to see whether the AP tests reveal anything obvious.

The Scarlett has good performance (see here, specifically the mic input measurement). It should be fine for recording line levels signals with/without the NanoDigi in line so that you can compare the outputs using a software comparator. Please try! You've got me interested

Interesting that you found this as well. Given that the SHD unit did very well in Amir's tests, the one flaw was jitter spikes that sure looked to me like they were potentially related to the SRC process. It didn't seem to compromise SINAD/distortion numbers though, so I was left unsure whether they improved the implementation in the SHD vs earlier units or whether my guess that SRC might be a problem was a red herring.

The SHD unit is their top of the line, while the NanoDigi is one of their budget items. Not sure you can draw much of a conclusion from Amir's measurements.

 

[Wombat]

Hah! Nearly twenty years ago, I was the moderator of the three AudioAsylun tube audio forums after JackG left because of his nastiness towards objectivists. Romy was - and is apparently still is - a real "anal orifice."

Lucky you. You missed the last decade, or more, of wacky AA contributions(SET/HiEfficiency) by TubeWrangler(mfr) and drlowmu(aka,elsewhere, as lowohms/dowto1000), an acolyte).

 

[anmpr1]

I'll need to look into those papers. Please provide some links when you have some time.

In the meantime, I'd like to say I think you're vastly underestimating the power of human hearing. Hearing a well-regarded tube amp vs. some new ncore module-based amp is like night and day to me. And no, whether or not people prefer one or another is not the question I'm trying to tackle. I'm siding with Stereophile's claim that A-B testing (as the ones I have read up until now) is flawed in several areas.

With a tube amp, into a low impedance wildly reactive load, you might be able to hear something consistently. On the other hand, David Clark's experiment using NYAL Futterman OTL-1 amps, machines notorious for causing problems with loads under 8 ohms, showed no difference in audibility into Magnaplanar speakers when compared to other amplifiers. I posted his results in the thread.

You might be able to tell amplifiers apart in a controlled test. But you are the only person that can do it. So you are a pretty unique guy.

PS: funny we don't buy eye glasses like we buy audio gear. When you go to an optometrist you are given an AB test to determine which lens is better. But when it comes to hearing audio gear, it's always non-rigorous. Unless you go to an audiologist. They perform rigorous hearing tests, with good reliability.

Again, this doesn't mean that quality engineered gear has no purpose. I own both Benchmark and a 20 year old Yamaha amp. And a couple of tube units. I may not be able to hear a difference blind. But I enjoy the Benchmark because I know what goes in it.

PSS: Since the tests you read are 'flawed in several areas' (your words), I wonder what part of the Nousaine/Maki/Zipser test was not done well? Zipser sold high end gear for a living. He claimed to hear big differences in electronics. The test was done in Zips living room with his own equipment and his own source material. He had as much time as he needed to listen. And he couldn't tell his super duper class A Nelson Pass amps from a then 10 year old Yamaha integrated. So don't think I'm being churlish or a crank. That's not my intent. Just curious, what makes you think your ears are/were better than Zips (may he rest in audio peace)?

 

[Frank Dernie]

With a tube amp, into a low impedance wildly reactive load, you might be able to hear something consistently. On the other hand, David Clark's experiment using NYAL Futterman OTL-1 amps, machines notorious for causing problems with loads under 8 ohms, showed no difference in audibility into Magnaplanar speakers when compared to other amplifiers. I posted his results in the thread.

I think the Futterman was "lucky" with the Maggies since the thing which causes changes in the frequency response is the high output impedance into a complex load and the Maggie is not a complex load.
All amps with a high output impedance change the sound, the question is - how much?
Pretty well every valve amp I know of has a high enough output impedance to sound different on the same speakers to a SS amp with low output impedance.
Quite a few SS amps do also to a lesser extent, particularly those with low feedback.
Given the propensity to equate price with quality my theory is that expensive amps which sound a bit different are judged to be better because they are expensive.
That means all valve amps, to varying degrees and things like Dartzeel which are SS but high output impedance for SS.

 

[anmpr1]

I think the Futterman was "lucky" with the Maggies since the thing which causes changes in the frequency response is the high output impedance into a complex load and the Maggie is not a complex load.
All amps with a high output impedance change the sound, the question is - how much?
Pretty well every valve amp I know of has a high enough output impedance to sound different on the same speakers to a SS amp with low output impedance.
Quite a few SS amps do also to a lesser extent, particularly those with low feedback.
Given the propensity to equate price with quality my theory is that expensive amps which sound a bit different are judged to be better because they are expensive.
That means all valve amps, to varying degrees and things like Dartzeel which are SS but high output impedance for SS.

I know this is 'subjective', but I remember buying my Acoustats. The day I was in the store another customer was demoing the speakers (tall 2+2 panels), using his own tube amp. It was something by EAR (Esoteric Audio Research). I didn't (and don't) know much about them, other than they were expensive. I was using the large Acoustat amp, a pretty beefy MOSFET solid state amp designed with those particular speakers in mind. Anyhow, the EAR tube gear sounded distinctly distorted to my ears. My guess was that the speakers were asking too much of the amplifier. But, again, it was a short term listening experience, at a dealer, and not my living room.

I'll be honest. Whenever I casually listen to amps I always hear 'differences.' But I'm not doing it level matched, so I'm not making any claims about amplifier difference in audibility, or that my hearing is 'golden', which I know it's not. I'm an older guy, and older guys' hearing is never the same as a teenager. I just think that it's up to those who claim they hear big differences in these things to explain why they can't do it, once brands are hidden, and the levels matched.

 

[anmpr1]

Boy, that one takes me back. It was a real comeuppance for Zipser.

... I was happy to do some blind testing and accept any result, because I was just as curious if it could be my imagination as well. The tests even being guided by some of the more hard-nosed "objectivists." I easily discerned between the CD players/DAC. The results weren't accepted by the objectivists in question - e.g. Arny K - the general reply being "something MUST have gone wrong somewhere."

Whereas other blind tests I've performed have left me unable to tell audible differences between certain components.

If you demonstrated it under controlled conditions, then that is that. I remember how the rec audio thing would get pretty nasty. Those were the days!

Next, the question becomes, what was it about the device in question that allowed you to hear the differences? Everything I've read on the subject (I admit to not being a scholar on the topic) suggests that FR changes would likely be the culprit.

One of my first experiences in an audio store, where I 'heard' a difference, was between a Yamaha B1 amp and an SAE (don't remember the model, probably post Jim Bongiorno, who I think had moved on to Ampzilla, or Sumo, or GAS, or whatever). Associated gear was the Sao Win/Ira Gale Lucite DD turntable. I remember that. Not sure the arm or cartridge. The Yamaha sounded bright and crisp, the SAE dark and murky. I mean, how can anyone explain this stuff? Hallucinations? LOL