Marantz PM-11S3 Integrated Amplifier Review

[avengineer]

I happen to be friends with Henry Ott and I think he would find it amusing how you are misquoting his book. In fact, he's even written and peer reviewed many cable articles on this site. He and I had many discussions on the merits of fully differential designs in audio.

Nice name drop! Sure wish I could do that. Hmm...perhaps I can: Deane Jensen...we used to have long conversations about the benefits of multi-filar transformer windings and electrostatic shielding, and his modeling transformers on an early HP computer. Sadly, we all lost a Deane, a dear friend, in 1989. How'd I do? I really don't think name dropping has much value in a technical discussion, but you really don't want to get me started either, I'm an old guy who knows a lot of people. Can we leave that one alone now?

By the way, if you read my posts, I never quoted Ott's book, so I could hardly be misquoting either. I only referenced it, correctly I might add....but whatever.

Let me just quote a small excerpt from his book on page 116: "Balancing is an often overlooked-though in many cases cost-effective-noise reduction technique, which may be used in conjunction with shielding when noise must be reduced below the level obtainable with shielding alone."

So, getting to the nitty-gritty now....Gene, you are having issues with reading comprehension. Page 116 is referring to balanced interface, and it's CMRR (common-mode rejection ratio). And, for the bazillionth time, I do not disagree with that, in fact, that's been my entire point all along.

Perhaps you should have quoted the sentence that was just ahead of the one you mentioned above: "The purpose of balancing is to make noise pickup equal in both conductors, in which case it will be a common-mode signal which will then be cancelled out in the load." Bingo. Couldn't have said it better, though I tried many times. That's what I've been saying this entire time. And, he hasn't said word one about reduction of distortion. And he's not talking about internal amplifier topology, he's talking about external lines and interface, and common-mode rejection...and so am I. I'm surprised you haven't gotten that from your friend.

And check out Henry's little App note inspired by one of the articles we coauthored on this site many years ago:
http://www.hottconsultants.com/pdf_files/Audio Interconnections.pdf

Another fine paper...about interface! Geez, man, what's the problem? I agree with it all! We are in complete agreement here...except about that little distortion thing...ok, perhaps the possible noise penalty paid by a fully balanced amplifier (internal!!!!) topology. Details. It's a small point perhaps, but Ott makes no mention, does not even use the word "distortion" that article.

I honestly cannot believe we are still having this discussion, especially given your claimed background.

...and I honestly cannot believe we are still having this discussion, especially given your claimed background. But I think I may now understand why...reading comprehension. Just take a deep breath, shake off all that anger, and just read...slowly....my posts. You just may be surprised that we agree, and have been in agreement, after all...except ...possibly...about that little distortion and noise thing.

All I ever suggested (which was a positive suggestion that you could, if you chose, use to improve the value of your reviews) is to actually go ahead and test a balanced input for common-mode rejection vs frequency. You might even find it interesting how non-flat it can be.

The entire discussion about what goes on inside an amplifier built with balanced vs single ended topology is something you brought up, and is entirely tangential to this thread and review. I'm happy to drop it, if you are.

 

[ImcLoud]

Nice name drop! Sure wish I could do that. Hmm...perhaps I can: Deane Jensen...

I met Jenna Jameson at an mma event, we talked about leg work outs for 35 minutes... I also met adam sandler at a restaurant in NY, we talked about traffic and weather {weird}, then he asked if 1 sandwich was enough for a guy my size, and I said "nope, but neither is 2, so I may as well just have one", he laughed and said "then I should just have a bite"...

 

[RichB]

It's not what you know, it's who you know

- Rich

 

[gene]

Nice name drop! Sure wish I could do that. Hmm...perhaps I can: Deane Jensen...we used to have long conversations about the benefits of multi-filar transformer windings and electrostatic shielding, and his modeling transformers on an early HP computer. Sadly, we all lost a Deane, a dear friend, in 1989. How'd I do? I really don't think name dropping has much value in a technical discussion, but you really don't want to get me started either, I'm an old guy who knows a lot of people. Can we leave that one alone now?

By the way, if you read my posts, I never quoted Ott's book, so I could hardly be misquoting either. I only referenced it, correctly I might add....but whatever.



So, getting to the nitty-gritty now....Gene, you are having issues with reading comprehension. Page 116 is referring to balanced interface, and it's CMRR (common-mode rejection ratio). And, for the bazillionth time, I do not disagree with that, in fact, that's been my entire point all along.

Perhaps you should have quoted the sentence that was just ahead of the one you mentioned above: "The purpose of balancing is to make noise pickup equal in both conductors, in which case it will be a common-mode signal which will then be cancelled out in the load." Bingo. Couldn't have said it better, though I tried many times. That's what I've been saying this entire time. And, he hasn't said word one about reduction of distortion. And he's not talking about internal amplifier topology, he's talking about external lines and interface, and common-mode rejection...and so am I. I'm surprised you haven't gotten that from your friend.


Another fine paper...about interface! Geez, man, what's the problem? I agree with it all! We are in complete agreement here...except about that little distortion thing...ok, perhaps the possible noise penalty paid by a fully balanced amplifier (internal!!!!) topology. Details. It's a small point perhaps, but Ott makes no mention, does not even use the word "distortion" that article.



...and I honestly cannot believe we are still having this discussion, especially given your claimed background. But I think I may now understand why...reading comprehension. Just take a deep breath, shake off all that anger, and just read...slowly....my posts. You just may be surprised that we agree, and have been in agreement, after all...except ...possibly...about that little distortion and noise thing.

All I ever suggested (which was a positive suggestion that you could, if you chose, use to improve the value of your reviews) is to actually go ahead and test a balanced input for common-mode rejection vs frequency. You might even find it interesting how non-flat it can be.

The entire discussion about what goes on inside an amplifier built with balanced vs single ended topology is something you brought up, and is entirely tangential to this thread and review. I'm happy to drop it, if you are.

Geez man each response you write is a book. You must either be retired, independently wealthy, or have a cushion job to lurk here while you work

I'd be interested in knowing how balanced topology reduces distortion.

This is what you originally asked. I approached my response from a Systems standpoint while you seem to be focused solely on an interface and/or interconnect. I have CLEARLY stated since the beginning that I meant fully balanced from input to output meaning not only the interconnects and interfaces but all active components in between which usually involves amplification.

Yes you can do a balanced input the right way and the wrong way. Some companies simply do a phase splitter on the inputs and then convert to single ended in the amp stages. IMO this is lame, especially if the products are high $$$. One company I know for sure that DOES NOT DO this is Emotiva. When they say their product is fully balanced, it's fully balanced from input to output. Take their 2CH preamp for example (XSP-1 The Emotiva XSP-1 Differential Reference Preamp | Emotiva Audio | Emotiva Audio | High-end audio components for audiophiles and videophiles, spanning 2-channel music systems, as well as 5.1 and 7.1 home theaters. Products include multichannel amplifi), every circuit is fully differential in it. Kudos to them b/c so many of their far more expensive and prestigious competitors simply phase split the inputs and run everything single ended thereafter.

AGAIN I State for the LAST time, I prefer fully balanced designs FROM INPUT TO OUTPUT. It doesn't mean single ended, unbalanced is bad, it just means properly done differential/balanced is the very best you can get. It's not just my telcom experience but my experience designing audio systems for the Government as well as my 14+ years reviewing and evaluating high end consumer gear. We all have our preferences so if you prefer single ended designs then enjoy and move on.

The purpose of this thread was to support the product review which clearly hasn't happened since its been derailed.

 

[PENG]

Take their 2CH preamp for example (XSP-1 The Emotiva XSP-1 Differential Reference Preamp | Emotiva Audio | Emotiva Audio | High-end audio components for audiophiles and videophiles, spanning 2-channel music systems, as well as 5.1 and 7.1 home theaters. Products include multichannel amplifi), every circuit is fully differential in it. Kudos to them b/c so many of their far more expensive and prestigious competitors simply phase split the inputs and run everything single ended thereafter.

It makes me wonder how they can make a good profit. I guess they may have volume on their side as everyone seems to have at least one, except me, not yet anyway.

 

[avengineer]

Geez man each response you write is a book. You must either be retired, independently wealthy, or have a cushion job to lurk here while you work

You inspire me, man!

"I'd be interested in knowing how balanced topology reduces distortion."

This is what you originally asked. I approached my response from a Systems standpoint while you seem to be focused solely on an interface and/or interconnect. I have CLEARLY stated since the beginning that I meant fully balanced from input to output meaning not only the interconnects and interfaces but all active components in between which usually involves amplification.

Fully balanced systems, or just balanced interface, the real question is what advantage balancing provides. We both know common-mode noise rejection. You've said lower distortion...still waiting for backup on that one...sorry!

Yes you can do a balanced input the right way and the wrong way. Some companies simply do a phase splitter on the inputs and then convert to single ended in the amp stages. IMO this is lame, especially if the products are high $$$. One company I know for sure that DOES NOT DO this is Emotiva. When they say their product is fully balanced, it's fully balanced from input to output. Take their 2CH preamp for example (XSP-1 The Emotiva XSP-1 Differential Reference Preamp | Emotiva Audio | Emotiva Audio | High-end audio components for audiophiles and videophiles, spanning 2-channel music systems, as well as 5.1 and 7.1 home theaters. Products include multichannel amplifi), every circuit is fully differential in it. Kudos to them b/c so many of their far more expensive and prestigious competitors simply phase split the inputs and run everything single ended thereafter.

I don't disagree that there are benefits to a fully balanced system, it's just that there are also drawbacks, which should be "balanced" against the advantages, but seem to be glossed over, or incorrectly stated. There is also a lot of misinformation about what a fully-balanced system is, such as the Emotiva XPR-1 review where the reviewer made this unfortunate mis-step, "For multi-channel, the XPR-1 was connected up to my Denon AVP-A1HDCI processor for a fully balanced connection from source device to the speaker outputs!". Nope, sorry, internally the AVP-A1HDCI is fully unbalanced/single ended from just past the only balanced inputs (CD) to the balanced output drivers. And so what did we gain from all of that besides a warm-fuzzy balanced feeling? CMRR. And was that a real advantage? Only if there was actual common-mode noise to be rejected. Otherwise, we may easily have paid a noise penalty just by having more complex analog circuits require for balanced I/O.

AGAIN I State for the LAST time, I prefer fully balanced designs FROM INPUT TO OUTPUT. It doesn't mean single ended, unbalanced is bad, it just means properly done differential/balanced is the very best you can get. It's not just my telcom experience but my experience designing audio systems for the Government as well as my 14+ years reviewing and evaluating high end consumer gear. We all have our preferences so if you prefer single ended designs then enjoy and move on.

And I acknowledge your preference! I like balanced systems too, they've saved by butt countless times, and are completely the norm in pro audio and broadcast. The point is, if it is in fact balanced, then measure the CMRR so we know how good that is, be it a system or just an input. You have publicly stated preference, which is your right and privilege. You are also an influential force in the audio community, and people base their opinions on what you write, so it would be nice to back up the preference with real data, and advance truth in audio.

The purpose of this thread was to support the product review which clearly hasn't happened since its been derailed.

Hopefully all this will support a better review next time.

 

[PENG]

You inspire me, man!

Fully balanced systems, or just balanced interface, the real question is what advantage balancing provides. We both know common-mode noise rejection. You've said lower distortion...still waiting for backup on that one...sorry!

Base on the knowledge you have displayed so far (not trying to make you feel good but I am serious), why are you still waiting? I would think you know where to find the backup already, internet, library etc. IMO forums are not always the right place to explain things in details, more often than not people just post links to stuff that are fuly Googleable or Bingable. The problem is, not all the stuff on the internet are credible, but I am confident (again, I am quite sure you know your stuff) that someone like you can discern what's credidble and what's not).

I don't disagree that there are benefits to a fully balanced system, it's just that there are also drawbacks, which should be "balanced" against the advantages, but seem to be glossed over, or incorrectly stated.

This is what the man said:

"AGAIN I State for the LAST time, I prefer fully balanced designs FROM INPUT TO OUTPUT. It doesn't mean single ended, unbalanced is bad, it just means properly done differential/balanced is the very best you can get."

So it seems to me you are in agreement, no? I would think that if things are really done properly, those drawbacks you may be think will not be enough to "balanced" against the advantages. We can list examples of what properly done mean but I do not feel that is necessary.

Hopefully all this will support a better review next time.

Things can always get better, or worse. I think we can all agree, but it shouldn't need so much nit picking if that's the goal.

 

[avengineer]

Base on the knowledge you have displayed so far (not trying to make you feel good but I am serious), why are you still waiting? I would think you know where to find the backup already, internet, library etc. IMO forums are not always the right place to explain things in details, more often than not people just post links to stuff that are fuly Googleable or Bingable. The problem is, not all the stuff on the internet are credible, but I am confident (again, I am quite sure you know your stuff) that someone like you can discern what's credidble and what's not).

You can wait a very long time for something that doesn't actually exist.

This is what the man said:

"AGAIN I State for the LAST time, I prefer fully balanced designs FROM INPUT TO OUTPUT. It doesn't mean single ended, unbalanced is bad, it just means properly done differential/balanced is the very best you can get."

So it seems to me you are in agreement, no? I would think that if things are really done properly, those drawbacks you may be think will not be enough to "balanced" against the advantages. We can list examples of what properly done mean but I do not feel that is necessary.

Unfortunately, the concept of "properly" is not binary. It's not on or off, it's very much a question of degree. If a balanced input has 40dB of common mode rejection, and another has 80dB, which one is done "properly"? They may even have exactly the same circuit, but with minute actual component value differences, and the end results be as different as this example. Unless the parameter is tested specifically, who's to know if it's been done properly, and to what degree?

Things can always get better, or worse. I think we can all agree, but it shouldn't need so much nit picking if that's the goal.

Agreed.

 

[gene]

I don't disagree that there are benefits to a fully balanced system, it's just that there are also drawbacks, which should be "balanced" against the advantages, but seem to be glossed over, or incorrectly stated. There is also a lot of misinformation about what a fully-balanced system is, such as the Emotiva XPR-1 review where the reviewer made this unfortunate mis-step, "For multi-channel, the XPR-1 was connected up to my Denon AVP-A1HDCI processor for a fully balanced connection from source device to the speaker outputs!". Nope, sorry, internally the AVP-A1HDCI is fully unbalanced/single ended from just past the only balanced inputs (CD) to the balanced output drivers. And so what did we gain from all of that besides a warm-fuzzy balanced feeling? CMRR. And was that a real advantage? Only if there was actual common-mode noise to be rejected. Otherwise, we may easily have paid a noise penalty just by having more complex analog circuits require for balanced I/O.

Have you seen the internal schematics for the Denon processor? Everything they sent me indicated the entire signal chain from input to output is differential. Their 10CH amplifier when bridged also is fully differential. I've also never measured a lower noise floor or better FFT from any audio preamp regardless of price. The Denon AVP processor is/was the real deal. They will never make a product like that again.

http://www.audioholics.com/av-preamp-processor-reviews/denon-avp-a1hdci/Denonbalanced.jpg/image_view_fullscreen

 

[AcuDefTechGuy]

Have you seen the internal schematics for the Denon processor? Everything they sent me indicated the entire signal chain from input to output is differential. Their 10CH amplifier when bridged also is fully differential. I've also never measured a lower noise floor or better FFT from any audio preamp regardless of price. The Denon AVP processor is/was the real deal. They will never make a product like that again.

Denon-balanced.jpg Full Screen Image | Audioholics

I am surprised Denon gets the fully balanced AVP-A1, but Marantz does not. I expected Marantz to get all the top high-end fully balanced components.

 

[gene]

I am surprised Denon gets the fully balanced AVP-A1, but Marantz does not. I expected Marantz to get all the top high-end fully balanced components.

That won't be the case going forward. Denon is out of the separates business. The AVP is a rare animal and one that I plan on keeping for a very long time unless a similar replacement can be found with the enhanced bass management features I'm looking for. Working with Marantz engineers on this. Fingers crossed.

 

[AcuDefTechGuy]

The AVP is a rare animal and one that I plan on keeping for a very long time unless a similar replacement can be found with the enhanced bass management features I'm looking for.

I am utterly happy with my AVP.

Of course, I am not as demanding as you.

 

[PENG]

You can wait a very long time for something that doesn't actually exist.

Unfortunately, the concept of "properly" is not binary. It's not on or off, it's very much a question of degree. If a balanced input has 40dB of common mode rejection, and another has 80dB, which one is done "properly"? They may even have exactly the same circuit, but with minute actual component value differences, and the end results be as different as this example. Unless the parameter is tested specifically, who's to know if it's been done properly, and to what degree?

Agreed.

You don't have to wait, as I said they are Googleable but I did find the Googleable ones mostly refer to ADC or telecom applications using OP amps. The principles are the same, that the scheme reduces (not completely cancelled due to imperfect matching characteristics of components) even harmonic distortions. I am clear Gene was not talking about everyone should have a differential amp but if you want the best then that is the way to go.

By the way, Bryston.ca is one site where you can see the schematics of their amps, and you will see that they are not fully balanced/differential from in to out until you get to their flag ship models such as the 7B, 14B and 28B. Even the $5,000 4B SST2 is only fully differential input but not from input all the way to output. I don't think too many people would have problem with their 4B and 6B models but you just don't get the fully balanced feature unless you go for their best.

 

[PENG]

I am surprised Denon gets the fully balanced AVP-A1, but Marantz does not. I expected Marantz to get all the top high-end fully balanced components.

Are you sure all of the Marantz reference series amps are not?

 

[AcuDefTechGuy]

Are you sure all of the Marantz reference series amps are not?

I was just fishing.

Hoping to thank some engineering-minded individuals to tell me the answers.

It amazes me how a $5K Bryston is not fully balanced. And I believe it's like that with McIntosh, Parasound, etc.

It would be interesting to see a compiled table of the amps that are balanced & unbalanced.

But this $5K Marantz PM-11S3 is NOT fully balanced, which I agree is a disappointment.

It puzzles me how a $1600 Outlaw 7500 amp or $800 Emotiva XPA-1 amp can be fully balanced when much higher echelon amps are not.

 

[PENG]

It puzzles me how a $1600 Outlaw 7500 amp or $800 Emotiva XPA-1 amp can be fully balanced when much higher echelon amps are not.

Me too, that's why I wonder how they can make money. Fully balanced from one end to the other requires more components and they have to be of tigher tolerances than otherwise otherwise you may get the opposite results.

 

[avengineer]

Have you seen the internal schematics for the Denon processor? Everything they sent me indicated the entire signal chain from input to output is differential. Their 10CH amplifier when bridged also is fully differential. I've also never measured a lower noise floor or better FFT from any audio preamp regardless of price. The Denon AVP processor is/was the real deal. They will never make a product like that again.

Denon-balanced.jpg Full Screen Image | Audioholics

Apologies, I stand corrected. The schematics cover several models on the same prints...it's a bit confusing...the AVR version isn't fully balanced, the AVP version is. My mistake.

 

[PENG]

I was just fishing.

Hoping to thank some engineering-minded individuals to tell me the answers.

I am not "engineering-minded" but I can now tell you there is nothing on their website that screams fully balanced. In fact they don't seem to have separates in their reference series at the moment so the 11S3 is the best one can get. $5K is not that bad for a high quality integrated amp but in terms of value you can see why Gene gave it 3 stars only.

It amazes me how a $5K Bryston is not fully balanced. And I believe it's like that with McIntosh, Parasound, etc.

Again, with Bryston you can get their higher models if you are wiling to pay for the 20 years warranty that you most likely don't need. For Parasound, it would appear even their flag ship monoblock is not fully balanced (not 100% sure) but if you can believe all those rave reviews out there, for $6,000 a pair I think for sound quality their owners would have reached the end of the road. At that point I think "fully" or not, is just academic.

It would be interesting to see a compiled table of the amps that are balanced & unbalanced.

I think logically you only need the first table, and it will be a very small table.

But this $5K Marantz PM-11S3 is NOT fully balanced, which I agree is a disappointment.

For you and I yes, but for those who only wants integrated analog only amps and listen with their ears only, it may be a good choice.

 

[gene]

Apologies, I stand corrected. The schematics cover several models on the same prints...it's a bit confusing...the AVR version isn't fully balanced, the AVP version is. My mistake.

Yep its a very special product and was underrated unfortunately b/c many audiophiles couldn't handle the Denon nameplate. This preamp kicks the crap out of many of the so called "higher end" brands. If only Audyssey was better implemented and the bass management had a mode to allow LFE to simultaneously goto the main channels + sub channels, I'd never need to upgrade again

 

[RichB]

Yep its a very special product and was underrated unfortunately b/c many audiophiles couldn't handle the Denon nameplate. This preamp kicks the crap out of many of the so called "higher end" brands. If only Audyssey was better implemented and the bass management had a mode to allow LFE to simultaneously goto the main channels + sub channels, I'd never need to upgrade again

The Marantz Av8801 supports LFE+Main for digital/digitized sources.
it would be nice if this was available in the analog domain.

A well known reviewer described the AV8801 volume control as a cheaper design.
i have no idea if that is a problem.

I find the LPCM over HDMI sound to be lacking when compared to the analog outs of myBDP-105.

Even engaging stereo mode with all controls flat degrades the sound.
This is true of all the preamps/AVRs that I have owned so their is rook for improvement in handling the digital data stream.

Also, the sound quality is affected when in Pure Direct mode when changing the video processing setting.

The unit gets too hot, I have measure the temp using my infrared thermometer at the center back at 130f.
i can't say for sure, but I had a trigger output fail and had to get the unit repaired.

I would like to see better isolation of the video processing, changing uh video settings affects the sound with analog in a pure Direct mode.
the BDP-105 also has a pure direct mode but it is just about turning off displays but there is no change in sound quality.
The AV8801 affects the sound and that has nothing to do with the front panel display.

There is rook for improvement in separating the audio and video and shielding them for one another.
The copper plate is absurd when internally you have a Ethernet hub.

Don't get me wrong I like the AV8801 better than the ONKYO PR-SC5507, but it could be improved.


- Rich