Integrated Amp discussion...

[GrimSurfer]

That was negative feedback they referred to. Benchmark uses feed forward, implying they don't use any amount of feedback but I could be wrong. They are low in distortions period, whether they are THD (including crossover distortions),IMD, or TIM distortions.

Benchmark did provide a lot of details about their feed forward error correction approach.

https://benchmarkmedia.com/blogs/application_notes/feed-forward-error-correction

Back to bandwidth, for class AB amps, I believe high slew rate (hence bandwidth potentially) is one of the indicator of less likely to have bad TIM distortions. Another review did show a slew rate consistent with 100 kHz bandwidth, that's not that unusual, in fact seems average for amps at the $3000 price point, but for a 100 W rated amp its more than high enough aside from TIMs that we have measurements for, not surprisingly.

Here's the review if you are interested:
http://www.avmentor.net/reviews/2016/benchmark_dac2hgc_ahb2_2.shtml
and it says the following, about bandwidth/slew rate:
"Finally, the square wave response shows a very good transient behavior, without any particular problem. Slew rate was estimated at 16V/uS and rise time at 3.5mS, a value that corresponds to a conventionally calculated bandwidth of 100kHz."

Thanks for the link. I agree that Benchmark puts a lot of performance and engineering data forward. It certainly adds confidence, especially when confirmed independently.

When I read specs like -130 dBA for IMD and 0.1 dB for L/R volume differentials, with no significant change at 4 ohm loads, one can't help but be mightily impressed.

My old 90s AB amp has a slew rate of 80V/us and a rise time of 1.8 ms. So I guess it doesn't suck too bad, though it's likely no longer performing exactly to spec because of its age (kind of like me).

Still, it's spec THD is in the -60 dB range and the stereo separation is 78 dB, so could very well be audible when pounding out the tunes... keeping in mind that performance may (LOL) have drifted over 20 years.

Horse and buggy performance compared with the Benchmark, as just about everything is other that the lovely Bryston I listened to.

That Benchmark pair is a killer, man. Wish I could justify the additional cost of the DAC. An impressive combo!

 

[<eargiant]

That was negative feedback they referred to. Benchmark uses feed forward, implying they don't use any amount of feedback but I could be wrong. They are low in distortions period, whether they are THD (including crossover distortions),IMD, or TIM distortions.

Benchmark did provide a lot of details about their feed forward error correction approach.

https://benchmarkmedia.com/blogs/application_notes/feed-forward-error-correction

Back to bandwidth, for class AB amps, I believe high slew rate (hence bandwidth potentially) is one of the indicator of less likely to have bad TIM distortions. Another review did show a slew rate consistent with 100 kHz bandwidth, that's not that unusual, in fact seems average for amps at the $3000 price point, but for a 100 W rated amp its more than high enough aside from TIMs that we have measurements for, not surprisingly.

Here's the review if you are interested:
http://www.avmentor.net/reviews/2016/benchmark_dac2hgc_ahb2_2.shtml
and it says the following, about bandwidth/slew rate:
"Finally, the square wave response shows a very good transient behavior, without any particular problem. Slew rate was estimated at 16V/uS and rise time at 3.5mS, a value that corresponds to a conventionally calculated bandwidth of 100kHz."

Feedforward is usually used in combination with NFB. They work together. Benchmark uses both in the AHB2. This is really nothing new, Sansui was already applying both these techniques in the late '70s early '80s.

Download AES paper here:
http://www.aes.org/e-lib/browse.cfm?elib=3934

Download FF Brochure here:
https://www.hifiengine.com/manual_library/sansui/super-feedforward-system.shtml

 

[GrimSurfer]

Feedforward is usually used in combination with NFB. They work together. Benchmark uses both. This is really nothing new, Sansui was already applying both these techniques in the late '70s early '80s.

Perhaps that's how they get the performance levels specified.

The early Sansui stuff was groundbreaking. Too bad they were produced at a time when solid state was perceived as crap (lots of it was, but some of it was still lght years ahead of Class A stuff).

 

[PENG]

My old 90s AB amp has a slew rate of 80V/us and a rise time of 1.8 ms. So I guess it doesn't suck too bad, though it's likely no longer performing exactly to spec because of its age (kind of like me).

It depends on the maximum rated output too, that's just math... My A21's (John Curl design) spec says 130 V/µsecond, better than my Bryston's 60 V/µsecond and the Anthem STR's 30 V/µsecond. The A21's spec for TIM is: "unmeasurable" if you can believe them. Passlab's class A and class AB amps are all rated 50 V/µsecond.

So yours 80 V/µsecond does not suck at all, it is actually excellent, as long as it is not rated for the likes of 1 kW or more super high output. Based on published specs only, it looks like Parasound's John Curl designed Halo amps all have very high slew rate, and actually specified TIM, that is one spec that is hard to find, but when found, they seem to typically say unmeasurable or virtually non existent etc.

None of those were measured though, and we are not even speculating whether the different specs would result in audible differences. I am just talking about difference in specs so far.

To be clear, I have no intention to participate in any debate, but I find it amazing to read about how so many people had heard night and day differences while others have hard time telling the differences even between very different designs. Roger Russell (someone who presumable wanted to tell the truth about audio..) cited a relatively recent case in 2007, when people had hard time telling the difference (sound quality of course) between a 35 W classic tube amp that was just updated, and a >200 W Denon solid state amp. He noted that the tube amp has moderate negative feedback. I was wondering why they didn't pick a so called warm sounding Marantz SS amp to have a better fighting chance.

http://www.roger-russell.com/truth/truth.htm

Sorry I digressed, and I just want to share some interesting reads on perhaps remotely related topics.

 

[everettT]

I'm quite happy with under 30lbs and AB if available. Published specs are good, and the units previous THX specs havent been changed

https://www.qsc.com/cinema/products/power-amplifiers/dca-series/dca-1622/

I own a couple of crowns class d for subs, no issues measurably.

My neighbor has an active system with NCore modules and it fantastic.

 

[GrimSurfer]

It depends on the maximum rated output too, that's just math... My A21's (John Curl design) spec says 130 V/µsecond, better than my Bryston's 60 V/µsecond and the Anthem STR's 30 V/µsecond. The A21's spec for TIM is: "unmeasurable" if you can believe them. Passlab's class A and class AB amps are all rated 50 V/µsecond.

So yours 80 V/µsecond does not suck at all, it is actually excellent, as long as it is not rated for the likes of 1 kW or more super high output. Based on published specs only, it looks like Parasound's John Curl designed Halo amps all have very high slew rate, and actually specified TIM, that is one spec that is hard to find, but when found, they seem to typically say unmeasurable or virtually non existent etc.

None of those were measured though, and we are not even speculating whether the different specs would result in audible differences. I am just talking about difference in specs so far.

To be clear, I have no intention to participate in any debate, but I find it amazing to read about how so many people had heard night and day differences while others have hard time telling the differences even between very different designs. Roger Russell (someone who presumable wanted to tell the truth about audio..) cited a relatively recent case in 2007, when people had hard time telling the difference (sound quality of course) between a 35 W classic tube amp that was just updated, and a >200 W Denon solid state amp. He noted that the tube amp has moderate negative feedback. I was wondering why they didn't pick a so called warm sounding Marantz SS amp to have a better fighting chance.

http://www.roger-russell.com/truth/truth.htm

Sorry I digressed, and I just want to share some interesting reads on perhaps remotely related topics.

My amp is a modest 30W @ 8 ohm but the specs also say it has a peak current delivery of 30A.

WRT TIM specs, I'd feel a whole lot more comfortable seeing "below 0.00001%" than " unmeasurable", which could mean the head of the marketing department told the engineers that the food truck was in the parking lot so that they didn't capture anything that would be too difficult to sell.

But when it doesn't appear anywhere on the spec sheet of a design prone to such things, my spidey senses start tingling.

Roger Russell articles are excellent. Probably would notice the difference between my amp and one from the era with 200W output provided the spl was the same. I can pick up some differences between tube and solid state but only on the extremes of the audio spectrum... certainly before the former has reached operating temp. No blindfolds though...

I usually don't listen much past 85-90 dBA from my seating position, which doesn't require much power to achieve. Probably less than 5W given my speakers' sensitivity and distance of my listening position. So clean and transparent with good headroom is my thing, not heavy or modulated. Even when gettin the Led out!

 

[GrimSurfer]

This is an interesting article on TIM from Matt Otala. It's old (1977) but discusses TIM on a fundamental level. So, while some of the means to suppress TIM may have changed, they don't change the physics that drives TIM susceptibility in high feedback circuits.

https://pdfs.semanticscholar.org/b3c0/a892a982ebde91f83f228905dac30186f827.pdf

My eyes started to glaze over somewhere less than halfway into the Greek alphabet (between eta and theta LOL).

Anyone crazy enough to push thru the main body will eventually reach and appreciate the more digestible statements in the last two sections of the paper. There's also a table that shows calculated slew rates for a 100W amp operating at various power levels and treble boost levels (which reinforces PENG's point about the appropriateness of a given slew rate being dependent on power output -- good one, man).

 

[JerryLove]

Limited impact at low listening levels. Some might call this a lack of muscularity. The limited muscularity goes away at completely at moderate listening levels. At higher sound levels, higher freqs don't have the same crispness. The frequency is there (as well as muscularity),along with a trace of sibilence that I do not hear in the ABs I've listed. It's almost Naim-ish if you get my drift, though not quite as shrill.

Can I ask what the speaker and amps in question were? It would be nice to make sure there's not some obvious other issue we can use to account for the discrepancy.

The only word in there I can put specific behavior to is "sibilance". That should be something we can test for.

So the issue is sonic characteristics. Perhaps it's a price range issue, but I haven't yet heard a D in that general range that measures up to my short list. And it's not as if I'm comparing a $8k Mc with a $2k D either.

I am comparing a McIntosh to a $500 Class H in my own case. There was a pair of speakers I could hear the difference on; but I'm reasonably certain that the cause was not the topology but the performance envelop. The ClassH in question isn't really intended to power lower than 4Ohm, and the area I was having issues with was where (or near to; as I didn't do enough investigation to verify my impression) the woofer dipped to 1-2Ohm. On everything else here I've done that A/B comparison with, I could not hear a difference.

Am I listening to things in the same space, thru the same speakers? Nope... Not always (did with the Bryston though). But I do bring the same music and even my DAC to auditions to get as close to fair as I can. I also avoid the classic mistake of auditioning amps through different classes/types of speakers. (Standmounted, towers, 2 or 3 driver, bass reflex, acoustic suspension etc.)

I was at the audio show earlier this month. Several of the speaker manufacturers drove me from the room with the unpleasantness of their HF behavior (it hurt my ears).

I do understand that, with enough sample cases, you could start to say "hey, every time I hear class D there's a silibance problem and when I don't there isn't"; but at this point in the conversation I am concerned that there are too many variables in play.

My A/B testing, though not blind, was identical except for amplifier; and while my amplifier comparisons weren't comprehensive (indeed: I had only one brand of ClassD/H involved) they do seem more apples-to-apples than yours.

Of course, it's also possible that I simply don't notice/hear the difference... but I'm skeptical of that.

All of these are light years ahead of a friend's PS Audio M700 monoblock set up (which is the best Class D I've heard in the $3000 range) in qualitative terms, though the M700 crushes them on power.

Have you done a drop-in, level-balanced A/B comparison with it? I'm very concerned that there's a great deal more in the chain, particularly the speakers and the room, involved.

I'd like to know why this is. I don't think I'm being biased (a hybrid H/AB is on my list, I listened to G class etc.). I was hearing this stuff before I did a lot of research into Class Ds, so I don't think this is an issue of confirmatuon bias.

I just can't ignore what is, and has been written, about the issues that affect Class D amps. Just as I spend an equal amount of time obsessing/looking for types of AB amps that minimize their inherent weaknesses (save power consumption which, frankly, lies somewhere between whale poop and the seabed on my list).

Obviously: eliminating confirmation bias requires blind testing... but right now I'm more interested in simply removing other variables... especially speakers and rooms, though there are many other potential areas (is your friend running room correction? Is it part of the issue? Conversely, are you running it on "not your friends", is it making it better).

In my listening experience, over the last couple of decades, The speaker, the room, and the mastering of the source are the variables that actually make a difference. Rarely, as in my example above, I've run into "insufficient amp", and sometimes I've run into overly-lossy compression, and (of course) there can be actual problems like ground loops or bad contacts or defective caps; but I've not found differences in (functional and sufficient) readers, DAC, switches, amps, or wires that could be heard by myself or anyone else I personally witnessed (that I had the opportunity to take to single-blind A/B testing if they claimed to hear it).

 

[GrimSurfer]

Can I ask what the speaker and amps in question were? It would be nice to make sure there's not some obvious other issue we can use to account for the discrepancy.

The only word in there I can put specific behavior to is "sibilance". That should be something we can test for.


I am comparing a McIntosh to a $500 Class H in my own case. There was a pair of speakers I could hear the difference on; but I'm reasonably certain that the cause was not the topology but the performance envelop. The ClassH in question isn't really intended to power lower than 4Ohm, and the area I was having issues with was where (or near to; as I didn't do enough investigation to verify my impression) the woofer dipped to 1-2Ohm. On everything else here I've done that A/B comparison with, I could not hear a difference.


I was at the audio show earlier this month. Several of the speaker manufacturers drove me from the room with the unpleasantness of their HF behavior (it hurt my ears).

I do understand that, with enough sample cases, you could start to say "hey, every time I hear class D there's a silibance problem and when I don't there isn't"; but at this point in the conversation I am concerned that there are too many variables in play.

My A/B testing, though not blind, was identical except for amplifier; and while my amplifier comparisons weren't comprehensive (indeed: I had only one brand of ClassD/H involved) they do seem more apples-to-apples than yours.

Of course, it's also possible that I simply don't notice/hear the difference... but I'm skeptical of that.


Have you done a drop-in, level-balanced A/B comparison with it? I'm very concerned that there's a great deal more in the chain, particularly the speakers and the room, involved.


Obviously: eliminating confirmation bias requires blind testing... but right now I'm more interested in simply removing other variables... especially speakers and rooms, though there are many other potential areas (is your friend running room correction? Is it part of the issue? Conversely, are you running it on "not your friends", is it making it better).

In my listening experience, over the last couple of decades, The speaker, the room, and the mastering of the source are the variables that actually make a difference. Rarely, as in my example above, I've run into "insufficient amp", and sometimes I've run into overly-lossy compression, and (of course) there can be actual problems like ground loops or bad contacts or defective caps; but I've not found differences in (functional and sufficient) readers, DAC, switches, amps, or wires that could be heard by myself or anyone else I personally witnessed (that I had the opportunity to take to single-blind A/B testing if they claimed to hear it).

If I gave you a specific answer, other than my friend's PS Audio M700s, I'd be guessing beyond the Peachtree 300 that was so unremarkable that I dropped it faster than grammar class. But I'll dig around the study to see if I kept any brochures or notes on the others...

I tried to get close on the levels, but was using the NIOSH sound level meter app on my tablet -- not the best instrument but likely better than saying " yeah, that sounds about right". I use it more than my RS spl meter and it always seems to be around when I need it.

I definitely recall the speakers when auditioning the Bryston and other D... B&W 705s. The reason why I recall this arcane fact is because they were moderately sized bookshelf models, one level up from the base 600s, without the externally diamond tweeters. I'm a stand mounted/bookshelf guy. That was easy to look up and verify. But the sub? Not a clue.

My friend is a semi Luddite. He has no problem with Class Ds but isn't into the whole ARC scene (if he had it, wouldn't use it). I'm the other way but feel that it's best to tame the room and position speakers etc. before even going with audio correction (which I don't have but appreciate it on a technical level, though within sensible limits).

Like you, I'm very sensitive to variables. They're so difficult to eliminate, especially when hopping from store to store on different days. Sometimes "marking" them is the best one can do.

That's the way it is for the average consumer though. The only thing we can often do is conduct clustered sessions, manage distances and basic sound levels, bring along the same home equipment to take that variable out of the equation, and call-the-ball.

That's where specs can help keep us (and manufacturers and sales staff) honest when everyone's lab coat has been lost at the cleaners.

 

[AcuDefTechGuy]

Is the OP @Mangoman even reading any of this after about the 2nd or 3rd page?

What are we talking about here?

Still thinking about buying an integrated amp?

As PENG and I say, integrated amps are AVR-derived - they are nothing more than AVRs without the radio tuner and disguised with an amp chassis.

So might as well just buy an AVR.

If you want separates MCH, then get a pre-pro + amp.

If you want separates 2CH, then get an analog stereo preamp + amp.

 

[GrimSurfer]

Is the OP @Mangoman even reading any of this after about the 2nd or 3rd page?

As PENG and I say, integrated amps are AVR-derived - they are nothing more than AVRs without the radio tuner and disguised with an amp chassis.

So might as well just buy an AVR.

LOL.

That's a generalization. The integrated amps I'm looking at have much more in common sonically with great separates. Compare a B60 with an NAD 165 preamp and 275 power amp. Personally, I'd take the Bryston. I'd wager 90% of people at that audition would too.

The NAD and the Moon 250i are even competitors, but you'll save a few hundred on the Moon and gain stronger resale value when it's time to sell.

But I'll grant you that not many integrateds exist in the $3k range that can hold their own with separates at the same price. But they are out there.

 

[AcuDefTechGuy]

That's a generalization. The integrated amps I'm looking at have much more in common sonically with great separates.

And that's your own generalization with absolutely no proof, except a bunch of hearsay.

It's all in your head. It really is. But it is up to you to find the truth, which a lot of people can't handle.

 

[everettT]

What constitutes great separates? If one were to remove the amps in the denon x4500, would that be a great preamp?

 

[Irvrobinson]

That Benchmark pair is a killer, man. Wish I could justify the additional cost of the DAC. An impressive combo!

The Benchmark DAC3 is very impressive; I have the DAC3L version. I previously had the DAC1-HDR, and for the music I listen to there isn't any audible difference between the two that I could discern. I used the qualification about "music I listen to", because Benchmark claims the DAC3 processes digital recordings that include max signal levels better (so-called "inter-sample overs" in BS recording industry jargon),but I apparently don't own any recordings like that, probably because max level instances are most prevalent on pop music mastered for highest relative loudness. The reason I traded up to the DAC3 is that Benchmark dramatically improved the volume control circuit in the DAC3, and the vastly improved precision of the remote volume control over that in the DAC1-HDR.

In my system the Benchmark DACs have an advantage over most of the cheaper units in that their balanced output stages are so robust. I use the DAC3 as a pre-amp, I split the output signal twice, and the ATI amps have somewhat below average sensitivity on their balanced inputs, so significant volts and low output impedance are good things in my system. Using the DAC3 just as a line-level DAC seems like a bit of a waste of money; in that mode I'm not sure there would any audible advantage to a Benchmark DAC at all (unless what they say about their inter-sample overs advantage is real).

 

[PENG]

What constitutes great separates? If one were to remove the amps in the denon x4500, would that be a great preamp?

Good point, never thought of it that way. In that case even the x3500h would be good. I know that because it held it's own against my A21. Back on topic, integrated amps that don't use LSI ICs of course are available but are likely the higher end models of manufacturers who make AVRs or boutigue brands that don't make receivers at all. The question is, do they "sound better"? I would say again, it depends.

 

[PENG]

The Benchmark DAC3 is very impressive; I have the DAC3L version. I previously had the DAC1-HDR, and for the music I listen to there isn't any audible difference between the two that I could discern. I used the qualification about "music I listen to", because Benchmark claims the DAC3 processes digital recordings that include max signal levels better (so-called "inter-sample overs" in BS recording industry jargon),but I apparently don't own any recordings like that, probably because max level instances are most prevalent on pop music mastered for highest relative loudness. The reason I traded up to the DAC3 is that Benchmark dramatically improved the volume control circuit in the DAC3, and the vastly improved precision of the remote volume control over that in the DAC1-HDR.

In my system the Benchmark DACs have an advantage over most of the cheaper units in that their balanced output stages are so robust. I use the DAC3 as a pre-amp, I split the output signal twice, and the ATI amps have somewhat below average sensitivity on their balanced inputs, so significant volts and low output impedance are good things in my system. Using the DAC3 just as a line-level DAC seems like a bit of a waste of money; in that mode I'm not sure there would any audible advantage to a Benchmark DAC at all (unless what they say about their inter-sample overs advantage is real).

Have you ever tried an Oppo one such as the Sonica or the class A HA-1? I don't have perfect hearing, and I can't imagine the Oppos at less than half the price of Benchmark's could be the bottleneck in mine or any system.

 

[everettT]

Good point, never thought of it that way. In that case even the x3500h would be good. I know that because it held it's own against my A21. Back on topic, integrated amps that don't use LSI ICs of course are available but are likely the higher end models of manufacturers who make AVRs or boutigue brands that don't make receivers at all. The question is, do they "sound better"? I would say again, it depends.

The 3400 I'm using hasn't introduced any issues to the sound the Salks are producing and I'm running full range with in room low end at 20hz and high end out to 40kHz. Incredible for $599 price.

 

[everettT]

Have you ever tried an Oppo one such as the Sonica or the class A HA-1? I don't have perfect hearing, and I can't imagine the Oppos at less than half the price of Benchmark's could be the bottleneck in mine or any system.

I've heard the HA1 in a demo room, seemed like very capable product. I'm getting past the point of diminishing returns with electronics for the most part.

 

[Irvrobinson]

Have you ever tried an Oppo one such as the Sonica or the class A HA-1? I don't have perfect hearing, and I can't imagine the Oppos at less than half the price of Benchmark's could be the bottleneck in mine or any system.

I tried the Sonica and made some comments in this thread:

https://forums.audioholics.com/forums/threads/external-dac-pick-one.107601/page-3#post-1239280

I ended up sending it back due to an unacceptable turn-off thump which I couldn't get rid of, and I found the phone app to have very poor responsiveness for use as a remote volume control. The SQ of the Sonica when everything was set properly and in stasis seemed to be the equal of the Benchmark. I was disappointed that the Sonica was unacceptable, seeing as it was only US$799. I considered leaving it powered up all the time to negate the thump problem, but that isn't a good idea in an area that gets some rather impressive thunderstorms. The Sonica did have an IR sensor, so a programmable IR remote could be used, but in the end the Sonica's weaknesses frustrated me, so I sent it back, and ordered the DAC3-L, which is flawless for my use in every way. (And well it should be, at nearly three times the price.)

 

[AcuDefTechGuy]

What constitutes great separates? If one were to remove the amps in the denon x4500, would that be a great preamp?

The main culprits are the Tuner and the chassis itself.

The Tuner will make it sound sonically inferior in every way and the AVR-looking chassis will also make it sound sonically inferior.

If you want something to sound sonically superior regardless of price, you must place it into an amp chassis (not AVR) and you must remove the Tuner at all cost!

Screw the speakers and subs. It's all about amp chassis without tuners.

Come to think of it, all you have to do is put the insides of an AVR into an Amp chassis and it will sound sonically superior.

It must be the superior metal that conducts the audio signals better through the cerebral cortex.