Marantz AV 10 installed: - Early Review and Impressions.

[PaulBe]

I think PaulBe's point is, things like level/volume, frequency response can be assessed quite effectively (yes, always to a point) by subjective measurements. The reality is, amplification do not change frequency response significant enough for it to mean anything audible, once it gets pass the point of diminishing return such as +/- 1 dB 20-20,000 Hz with the amp operating well below clipping point. Volume/level can be matched to within 0.5 dB without too much difficulty when doing AB comparisons, yet people (I mean forum posts) who reported the different "sound" their amps produce never (almost...) seem to both such level matching and FR verification by measurements), to them subjective measurements is everything because it is their ears.

Processing is different, for obvious reasons, DTS, DD, and their variants, and upmixing etc., do sound different enough for the die hard objectivists, but it is very difficult to compare in terms of apples to apples for person HT use.

One thing I find very difficult to understand is, why don't enthusiast who believe in amps are tuned to different sound think logically, that if that is true, then amps measured on the bench will not the typical 20-20000 Hz +/- 0.5 to 1 dB, yet they do, whether we are comparing bench test results of a Benchmark AHB2, McIntosh's flagship amps, Passlab's, or a <$1000 buckeyeamp (class D), Fosi's <$100 class D chip amp.

Also, Peter Walker, the EE who designed @TLS Guy 's favorite Quad amps clearly stated they didn't tune their designs by ears, and that they only conduct listening test to make sure there's nothing unusual there in the end.

I posted the link long time ago, the link to the Geocities article no longer works though:

Interview with Quad's Peter Walker in 1978

I did copied/pasted the contents at the time in the post below:

Amplifier distortions - what, and how much are audible | Page 2 | Audioholics Home Theater Forums

I highly recommend that Peter Walker interview to those who is willing to apply logic to the claim that well design amps (based on specs and measurements) do sound audibly different when the compared amps are operating at well within their input and output limits.

Again, @TLS Guy has made it clear, the much improved sound quality he referred to so far about the AV7705/6 Vs AV10 is about DSP sound, such as Atmos, he has not said anything about audibly better sound quality for non processed, such as in pure direct mode.

My main point about subjective measurement is the measurements are a set of empirical data with which to make objective conclusions. Math is developed from the subjective measurements, and then tested for repeatability. It’s science – inductive reasoning. The science stands until someone or something knocks it down. That someone is usually someone who stands on the shoulders of the someone(s) who made the previous standards.

I wasn’t thinking of tiny differences in overall volume.

Concerning amplifiers –

Amplifiers are voltage amplifiers with various real world current limits. The favorite load of Any voltage amplifier in the real world, regardless of the cost and quality, is a resistor. Most speakers are a load of complex impedance – also consider this sentence an amusing euphemism. The biggest variance in frequency response of amplifiers occurs from a variability to drive a complex load. Sound quality is another matter. Most of the sound quality in an amp is derived in the ‘first watt’.

As Peter Walker wrote in your post of his article, and thanks for posting his article – “The peripheral effects are what get people into trouble. You can see why you find these differences in amplifiers. You can always find them. If people test two amplifiers and say, "These sound different," there's no magic in it. Spend two days, maybe a whole week in the lab, and you find out exactly why they're different and you can write the whole thing down in purely practical, physical terms. This is why these two sound different, and the cause is usually peripheral effects. It is not really a case of good or bad amplifiers, it's that the termination impedances are wrong, or something of that sort.” https://forums.audioholics.com/forums/threads/amplifier-distortions-what-and-how-much-are-audible.121482/page-2#post-1467053

The terminations in consumer audio are often Wrong. Using the right terminations would put a big kink in the subjective reviews, minimize the ‘circle of confusion’, and interfere with industry profits – can’t do THAT /s.

IIRC, Peter Walker made sure his amplifiers would be stable and complementary driving his electrostatic speakers – a highly capacitive load at high frequencies. Some class D amps may have greater issues driving capacitive loads while doing great at handling back EMF.

Side notes –

I went to the artofsound link and looked at the first page of comments – entertaining.

I did the Klippel test you mentioned, some time ago. My ability to distinguish distortion was limited to -18dB on one try. Ouch! It may have been the next level down (-21dB?). My memory on this is dim. Not so good in either case.

Thanks for your well-reasoned approach to this field.

So far, I hear the same audio improvements with all sources using the AV10. The improvements are not limited to processing, though it appears more dramatic with ATMOS discs and perhaps because of the number of channels. All my sources are disc or file. I’m not going to test why I hear improvements. I’m just going to enjoy the sound of recordings, and flawless (knock on wood) function of the processor.

* What if some of the sound and noise improvement in the AV10 is as simple as proper terminating impedances? Speculative example - The HDAM4 modules are cascodes. Cascodes linearize the signal and also have a higher output impedance. Why would Marantz engineer a higher output impedance into what is essentially a buffer?

 

[PENG]

My main point about subjective measurement is the measurements are a set of empirical data with which to make objective conclusions. Math is developed from the subjective measurements, and then tested for repeatability. It’s science – inductive reasoning. The science stands until someone or something knocks it down. That someone is usually someone who stands on the shoulders of the someone(s) who made the previous standards.

I never thought it wasn't science, of course it is. Just that it is not a reliable way of determining, or used as proof that two amps with almost the specs in multiple metrics, and measurements including the usual suspects such as THD, DR/SNR, IMD, crosstalk, frequency response, damping factor/output impedance etc.

I wasn’t thinking of tiny differences in overall volume.

Me neither, but volume matching when comparing using something like a REW sweep 20-20,000 Hz, that's what I used in the past such as the following:

I have plenty of such comparisons, this one and all the others too, basically showed no significant differences in FR even with 1/24 smoothing (this one happens to be 1/12) and the minor differences could well be mostly due to repeatability related, considering the not super accurate U-mik-1 mic and room noises.

Concerning amplifiers –

Amplifiers are voltage amplifiers with various real world current limits. The favorite load of Any voltage amplifier in the real world, regardless of the cost and quality, is a resistor. Most speakers are a load of complex impedance – also consider this sentence an amusing euphemism. The biggest variance in frequency response of amplifiers occurs from a variability to drive a complex load. Sound quality is another matter. Most of the sound quality in an amp is derived in the ‘first watt’.

I don't disagree with what you stated, but I do feel the point of using resistor vs reactive load that most speakers are, seem not relevant if and when tests are done at well below the DUTs output limits, such as those found on ASR and Audioholics.com. For example, even an AVR such as the mid range Marantz Cinema 40 can do a very transparent job if test and measured at output up to their published rated 125 W, so if compared with any other integrated amps or separates at say the typical 5 W that ASR uses, and up to say 50 W, then who it won't matter if the load is a resistor or a simulated reactive load such as one used by the Stereophile.

Here we are looking at a mid range AVR, the matching power amps for the AV10 and AV20, the amp10 and amp20, obviously should and I suspect would measure even better when connected to reactive, even highly reactive loads.

So far, I hear the same audio improvements with all sources using the AV10. The improvements are not limited to processing, though it appears more dramatic with ATMOS discs and perhaps because of the number of channels. All my sources are disc or file. I’m not going to test why I hear improvements. I’m just going to enjoy the sound of recordings, and flawless (knock on wood) function of the processor.

I believe you and TLSGuy, and that makes me thinking of eventually selling my AVM70 and grab the AV20 or keep the AVM but also grab the presumably more affordable AV30 when it launches later this year (use either one as backup, otherwise in one of my two channel system).

* What if some of the sound and noise improvement in the AV10 is as simple as proper terminating impedances? Speculative example - The HDAM4 modules are cascodes. Cascodes linearize the signal and also have a higher output impedance. Why would Marantz engineer a higher output impedance into what is essentially a buffer?

Can's say I disagree but I do doubt that, besides according to Gene and I think some of Masimo's marketing info, the AV10/20 uses HDAM version SA3, not HDAM4.

As to their higher output impedance, its a question of what "higher" is, if it is as high as even 1 kOhm, it still wouldn't make enough difference when used with the vast majority of power amps that typically have input impedance not lower than 15 kOhm, more often in the >20 kOhm range

Besides, Marantz marketing highlighted the claimed advantage of their HDAMs are that they are discrete, that offers higher slew rate than the corresponding OPAs. That, to anyone who can think logically, should know that is probably BS, if they bother checking into the details found in service manuals, that there are other mediocre IC opas up and down stream, as well other ICs such as the volume IC, switches and of course the DAC IC, so the higher slew rate of the HDAMs, assuming it really is superior, won't make a difference, simple bottleneck analysis.

 

[PaulBe]

I never thought it wasn't science, of course it is. Just that it is not a reliable way of determining, or used as proof that two amps with almost the specs in multiple metrics, and measurements including the usual suspects such as THD, DR/SNR, IMD, crosstalk, frequency response, damping factor/output impedance etc.



Me neither, but volume matching when comparing using something like a REW sweep 20-20,000 Hz, that's what I used in the past such as the following:

I have plenty of such comparisons, this one and all the others too, basically showed no significant differences in FR even with 1/24 smoothing (this one happens to be 1/12) and the minor differences could well be mostly due to repeatability related, considering the not super accurate U-mik-1 mic and room noises.

View attachment 73788



I don't disagree with what you stated, but I do feel the point of using resistor vs reactive load that most speakers are, seem not relevant if and when tests are done at well below the DUTs output limits, such as those found on ASR and Audioholics.com. For example, even an AVR such as the mid range Marantz Cinema 40 can do a very transparent job if test and measured at output up to their published rated 125 W, so if compared with any other integrated amps or separates at say the typical 5 W that ASR uses, and up to say 50 W, then who it won't matter if the load is a resistor or a simulated reactive load such as one used by the Stereophile.

Here we are looking at a mid range AVR, the matching power amps for the AV10 and AV20, the amp10 and amp20, obviously should and I suspect would measure even better when connected to reactive, even highly reactive loads.

View attachment 73789




View attachment 73790





I believe you and TLSGuy, and that makes me thinking of eventually selling my AVM70 and grab the AV20 or keep the AVM but also grab the presumably more affordable AV30 when it launches later this year (use either one as backup, otherwise in one of my two channel system).



Can's say I disagree but I do doubt that, besides according to Gene and I think some of Masimo's marketing info, the AV10/20 uses HDAM version SA3, not HDAM4.

As to their higher output impedance, its a question of what "higher" is, if it is as high as even 1 kOhm, it still wouldn't make enough difference when used with the vast majority of power amps that typically have input impedance not lower than 15 kOhm, more often in the >20 kOhm range

Besides, Marantz marketing highlighted the claimed advantage of their HDAMs are that they are discrete, that offers higher slew rate than the corresponding OPAs. That, to anyone who can think logically, should know that is probably BS, if they bother checking into the details found in service manuals, that there are other mediocre IC opas up and down stream, as well other ICs such as the volume IC, switches and of course the DAC IC, so the higher slew rate of the HDAMs, assuming it really is superior, won't make a difference, simple bottleneck analysis.

Oops! I didn’t remember the HDAM number correctly.

Regarding charts and smoothing – I think 6th, 3rd, and half octave smoothing relate more to what people actually perceive. There was a time when 3rd octave GEQ was used to suppress feedback at live events and it didn’t interfere much with sound at the events. A Q of 4.4 is a 3rd octave filter. Today’s filter choices in DSP can be much higher Q. 1/24 smoothing gives more detail. Is it important to the analysis?

Thanks for the reactive load test charts. Either I haven’t seen them or didn’t pay attention to them in reading articles. I’d like to see the bars relate to not only the 5 simulated reactance conditions, but with real loads that change impedance across the audio spectrum. The results would be a different curve with regards to different amplifiers driving different speakers. The curves would look a bit like using small amounts of EQ. While absolute level requires a relatively large change to perceive a volume difference – at lease .5dB, small changes – even .1dB - at as low as 1 octave fractions of the audio spectrum act as tone controls. I think this is what the subjectivists hear most of all with different amplifiers. It can change the character and interpretation of a piece of music – even pushing forward or laying back particular singers or musicians in a mix.

I said ‘higher impedance’ for a cascode, a relative term, and was speculating on its use for proper termination to a load, not its ability to drive the load. It was just a thought. Audio is a lump sum condition but proper or closer transmission line termination can still make an audible difference in practice. If you look up pro audio cables, they often come with characteristic impedance specs, and the specs are all over the place. Experimentation applies. Does HDMI have a termination standard or range?

A personal example of the paragraph above is when I terminated the inputs to an amplifier driving my acoustats. I was having problems with the local taxi cab CB linear amps interfering with and over driving the inputs to my amp. It was pretty scary. The amp had a 47.5Kohm input impedance. I added Y connectors at the inputs and made up 3 sets of parallel terminators with 10Kohm, 4.7Kohm, and 2.2Kohm resistors in RCA plugs. I tried each set. The 10K set lowered the CB interference. The 4.7K and 2.2K sets eliminated the interference, lowered the audible noise floor, and improved the sound for any condition. This is when I realized boutique audio cables are BStique. From then on, all the RCA connected line cables I used sounded good and about the same.

The discrete vs opamp argument in the Marantz marketing seems like more audiophoolery. But, as long as what they implement works, I don’t care. I think most of the marketing buzz that any of the manufacturers use is amusing.

My favorite opamp that I used from older days was a Burr-Brown OPA671. A single 8 pin dip, ran class A without a pull up resistor, slew rate of 100v/usec, 50 ma output current, $7.45 a piece in small quantities of 25 in 1991. I don’t recall one audio manufacturer that used it.

 

[TLS Guy]

I think PaulBe's point is, things like level/volume, frequency response can be assessed quite effectively (yes, always to a point) by subjective measurements. The reality is, amplification do not change frequency response significant enough for it to mean anything audible, once it gets pass the point of diminishing return such as +/- 1 dB 20-20,000 Hz with the amp operating well below clipping point. Volume/level can be matched to within 0.5 dB without too much difficulty when doing AB comparisons, yet people (I mean forum posts) who reported the different "sound" their amps produce never (almost...) seem to both such level matching and FR verification by measurements), to them subjective measurements is everything because it is their ears.

Processing is different, for obvious reasons, DTS, DD, and their variants, and upmixing etc., do sound different enough for the die hard objectivists, but it is very difficult to compare in terms of apples to apples for person HT use.

One thing I find very difficult to understand is, why don't enthusiast who believe in amps are tuned to different sound think logically, that if that is true, then amps measured on the bench will not the typical 20-20000 Hz +/- 0.5 to 1 dB, yet they do, whether we are comparing bench test results of a Benchmark AHB2, McIntosh's flagship amps, Passlab's, or a <$1000 buckeyeamp (class D), Fosi's <$100 class D chip amp.

Also, Peter Walker, the EE who designed @TLS Guy 's favorite Quad amps clearly stated they didn't tune their designs by ears, and that they only conduct listening test to make sure there's nothing unusual there in the end.

I posted the link long time ago, the link to the Geocities article no longer works though:

Interview with Quad's Peter Walker in 1978

I did copied/pasted the contents at the time in the post below:

Amplifier distortions - what, and how much are audible | Page 2 | Audioholics Home Theater Forums

I highly recommend that Peter Walker interview to those who is willing to apply logic to the claim that well design amps (based on specs and measurements) do sound audibly different when the compared amps are operating at well within their input and output limits.

Again, @TLS Guy has made it clear, the much improved sound quality he referred to so far about the AV7705/6 Vs AV10 is about DSP sound, such as Atmos, he has not said anything about audibly better sound quality for non processed, such as in pure direct mode.

That is true, although I think the improved SNR is a factor. I think the deleterious effect of noise is something that is underestimated. It takes little noise added to the program even if the program seems to mask it. I have always waged war on noise, which is why I am obsessional about the ground plane. So i do think the absence of noise from the AV 10 is a factor.

Undoubtedly though, the biggest issue is the improved Atmos performance with improved localization and much better preservation of the natural ambience of the venue.

My major issue is that I am highly trouble averse. Equipment failures are a huge PITA. I have found that in generally a high build quality tends to correlate with improved reliability. The AV 10 has not been around long enough to get a picture of its long term reliability. However, its build quality inspires a lot of confidence. The common denominator in my equipment that has been with a long time, is superb build quality with an orderly internal layout. If it looks like thrown together junk, then it probably is. Just one example, I have had a number of NAD units brought to me over the years. They are difficult to work on, because of layout/construction issues. People have, I think correctly said that NAD stands for Not Always Dependable. In my experience that has been a deserved moniker.

 

[Squishman]

I think you said once that NAD looks like a birds nest under the cover. I have a NAD cassette player that seems to be a pretty good one though. Mid-line for sure. Not a three head, but seems very solid after all these years.