Marantz AV7705 Home Theater Processor Review

[Masza]

A perfect example of how Audyssey (without the app) sounds terrible. Even after all these years, they wrongly try to produce a flat in-room response. It boggles my mind.

Limiting the correction range is possible with the mobile app.

 

[rynberg]

Limiting the correction range is possible with the mobile app.

Yep, it's the only thing that makes Audyssey usable to me.

 

[peng]

And how would I easily up-sample all my content (music and movies)? I shouldn't have to spend extra money or effort to bypass Marantz's poor decisions.

Agreed, I am just wondering if you are actually hearing the effects of the gentle roll off from 10 kHz up, that is, just a test.. Marantz should have do the right thing and make the filters selectable like many cheap dacs do.

 

[peng]

The AV8801 is a really nice unit IMO.

Did you notice that Denon returned to the TI PCM1795 DAC IC that is used in the AV8801 for their 110th anniversary DCD-A110. The move away from the current output PCM1795 in the AV8802 and AV8805 was likely based on cost savings, not quality, since a current output DAC costs more to implement properly than the AKM DAC IC. With AKM's inability to deliver product perhaps we'll see the PCM1795 again! (Doubtful)

No I don't, but did they use the 1795 for the AVR-A110 also? I also if that AKM factory fire would force them to switch back to the TI chip for their 2021 models and if they do, may be they would give up on the slow roll off filters too, hopefully..

My AV8801 doesn't have that same roll off that the AV8802A, and as we all know the AV8805 do. That's -0.25 dB vs -2.12 dB at 20 kHz according to S&V's:

From soundandvision.com:

 

[Sal1950]

If I use the App/Rat, I could have flatten the 20-120 Hz easily to almost flat, say about +/- 1.5 dB no smoothing, or +/- 1 dB with 1/12 smoothing.

Yep, those who continue to bash Audyssey haven't use the App or Rat or just choose to ignore it and enjoy beating a dead horse. The OTB Audyssey was built for the JoeSixpack HT user and designed with ease of use in mind. It works and sounds just fine under those conditions for those folks. The more demanding user has two more stages of complexity to scale if he so chooses. At those levels it's as good as any other "mid-level" DRC system (DiracLive, ARC, Yamahahaha). Later if your a real audiophile you'll get a serious DRC piece of software like AcourateDRC, Sonarworks, Raw files in REW, Dirac Full, and the rest. Bottom line is Audyssey has matured into a very good piece of software for the user willing to go under the hood.

 

[rynberg]

Yep, those who continue to bash Audyssey haven't use the App or Rat or just choose to ignore it and enjoy beating a dead horse. The OTB Audyssey was built for the JoeSixpack HT user and designed with ease of use in mind. It works and sounds just fine under those conditions for those folks. The more demanding user has two more stages of complexity to scale if he so chooses. At those levels it's as good as any other "mid-level" DRC system (DiracLive, ARC, Yamahahaha). Later if your a real audiophile you'll get a serious DRC piece of software like AcourateDRC, Sonarworks, Raw files in REW, Dirac Full, and the rest. Bottom line is Audyssey has matured into a very good piece of software for the user willing to go under the hood.

The problem is that Audyssey plain-and-simple uses the wrong target. A flat in-room response is black-and-white WRONG, not simply a matter of preference. As I said above, the App is the only thing that makes Audyssey acceptable -- most "JoeSixpack" users just turn up the sub channel by several dB as a rough fix.

At least ARC and Dirac use a default target that is much more appropriate, even without user intervention.

Audyssey can sound just fine if the App is used to fix the inherent flaws, although I've always had to manually play with the subwoofer distance setting to get better integration at the crossover.

 

[Sal1950]

The problem is that Audyssey plain-and-simple uses the wrong target. A flat in-room response is black-and-white WRONG, not simply a matter of preference.

Preference - no, opinion - yes. There are some well educated and knowledgeable people working at Audyssey making the decisions, not a bunch of dummies. Your entitled to yours, them to theirs.

I've always had to manually play with the subwoofer distance setting to get better integration at the crossover.

Works just fine for me

Audyssey can sound just fine if the App is used to fix the inherent flaws

But what did you do to fix the missing dual subwoofer feature, not available in Dirac till just this year, at an extra cost?

 

[rynberg]

Preference - no, opinion - yes. There are some well educated and knowledgeable people working at Audyssey making the decisions, not a bunch of dummies. Your entitled to yours, them to theirs.

It is not opinion. If you take an anechoically flat loudspeaker and place it into a room, it will have a downward trending response. That is fact. A "correction" system producing a flat in-room response from a good speaker means either elevating the high frequencies or decreasing the low frequencies. That is fact.

A flat in-room response is factually wrong, not a matter of opinion. Why do 99+% of Audyssey users jack up the subwoofer channel by several dB? It's not because the resultant spectral response just has less bass than they are used to (or at least, not just because), it's because the resulting response is not how real speakers sound in real rooms.

And well-educated people can be knowledgeable in some areas and ignorant in others or simply make poor decisions. Yamaha designed the NS-10 to have a flat power response instead of a flat anechoic response, which is why they are horrific-sounding speakers. How could a company like Yamaha make such a colossal error?

Works just fine for me

That's great, it doesn't for everyone, and by a long shot. It's a valid criticism.

But what did you do to fix the missing dual subwoofer feature, not available in Dirac till just this year, at an extra cost?

I didn't. I don't have a Dirac device and I have one subwoofer. Your comment is a valid criticism of both Dirac and the current version of ARC (although this is being addressed with the new release of ARC).

Your responses seem to be coming from a place of personal affront...my criticisms of the system aren't personal criticisms of you, they are factually valid flaws in the Audyssey room correction algorithm.

 

[Vasr]

The problem is that Audyssey plain-and-simple uses the wrong target. A flat in-room response is black-and-white WRONG, not simply a matter of preference. As I said above, the App is the only thing that makes Audyssey acceptable -- most "JoeSixpack" users just turn up the sub channel by several dB as a rough fix.

At least ARC and Dirac use a default target that is much more appropriate, even without user intervention.

Preference - no, opinion - yes. There are some well educated and knowledgeable people working at Audyssey making the decisions, not a bunch of dummies. Your entitled to yours, them to theirs.

I mostly agree with what @rynberg is saying here having used all the three - Audyssey, ARC, Dirac.

But it is not a question of correctness, rather design philosophy.

First of all Audyssey allows for both flat and sloping curves (called flat and reference). I believe it uses flat as default fro music mode and reference for Cinema/Movie mode. This is without having to use the app. It does seem like it should be the other way around since HT can use some bright treble while music is fatiguing with a flat target (unless you are pretty much deaf above 10k or so in which case it doesn't really matter).

So, it does lead to people feeling the Audyssey auto correction sounding too thin or not enough bass.

ARC sounds much better with the auto EQ because whatever heuristics they use to arrive at the automatic setting is the best out of the three systems. Their automatic target philosophy is more of follow the natural response of the speakers and smooth it out rather than follow some set target design. So, if your speakers sounded good without room eq (except for room modes), using ARC auto eq makes it sound much cleaner and crisper without introducing anything much artificial relative to the capability of the speakers. I think this is a smarter approach and have used this very successively in setting my targets using REW in cases where I did not have one of these EQ systems. Of course, you can change the target curve using the full ARC Genesis adjust parameters mode.

You can use the Audyssey app to do the same as ARC above for the target and it will likely get as good a result as ARC, just that ARC auto eq is so much easier out of the box. But their decision to have that separate flat and reference sections and mapped to music/movie mode is likely driven by product management decisions than technical ones.

Dirac is like Audyssey in the sense that the auto eq is not always satisfactory because of the flatter target it chooses. But can be adjusted in the Dirac Live software (although the user interface is horrendous unless you are faily well conversant in toom eq concepts). But the effect from Dirac after you get it to your satisfaction has some "magic" in it that seems to make speakers blend much better (especially noticeable in multi-channel settings) than the other two. Technically, it has some unique factors but I am not an expert in this to say whether they play a role or not in the sound differentiation.

 

[Sal1950]

Your responses seem to be coming from a place of personal affront...my criticisms of the system aren't personal criticisms of you, they are factually valid flaws in the Audyssey room correction algorithm.

Not personal at all, I simply acknowledge that the designers have much more knowledge in the area than I and most likely you, although I see you don't agree.

Why do 99+% of Audyssey users jack up the subwoofer channel by several dB?

Don't know where your 99+% number came from, did you do a survey of a large sample of Audyssey users settings? But in any case, yea, when given an option a large number of users will crank up the bass a bit, because they like it, not because of any component or speaker designers err in design. That truth goes back to the earliest days of HiFi with tone controls and loudness buttons.

And well-educated people can be knowledgeable in some areas and ignorant in others or simply make poor decisions. Yamaha designed the NS-10 to have a flat power response instead of a flat anechoic response, which is why they are horrific-sounding speakers. How could a company like Yamaha make such a colossal error?

Humm, the NS-10 is probably the most widely used mixing console speaker in the world, I guess the mixing engineers have no idea what they're doing either? What you find "horrific" they find revealing and accurate.
When it comes to these midlevel semi-auto DRC systems, they all have their weak and strong points and the users all have their preferences in sound. When left to do their own thing you will end up with the decisions made by the designers. You then have the choice to live with it or modify it to your preference depending on the tuning tools given.

 

[oupee]

Audyssey works pretty well for me. Personally, I think the biggest bug in Denon / Marantz is Bass Sync. Not much is known or written about because most users add extreme bass and are after trouble. Practically at the beginning of every film, it should be experienced what value is correct (0ms, 5ms, 10ms). Hans Zimmer in Prague 0ms, Blade 4K (1998) 5ms, Doctor Sleep (2019) 10ms even with the bass down, it starts to play deep, dynamic bass.

http://manuals.denon.com/AVRX4700H/NA/EN/WBSPSYvsbphziy.php

 

[peng]

The problem is that Audyssey plain-and-simple uses the wrong target. A flat in-room response is black-and-white WRONG, not simply a matter of preference. As I said above, the App is the only thing that makes Audyssey acceptable -- most "JoeSixpack" users just turn up the sub channel by several dB as a rough fix.

At least ARC and Dirac use a default target that is much more appropriate, even without user intervention.

Audyssey can sound just fine if the App is used to fix the inherent flaws, although I've always had to manually play with the subwoofer distance setting to get better integration at the crossover.

According to Audyssey:

Audyssey LabsDecember 31, 2009 18:31
Contrary to popular belief, a target curve that is flat from 20 Hz to 20 kHz is not always the one that will produce the correct sound. There are several reasons for this including the fact that loudspeakers are much more directional at high frequencies than they are at low frequencies. This means that the balance of direct and room sound is very different at the high and low ends of the frequency spectrum.

The Audyssey Reference target curve setting (also called Movie in some products) makes the appropriate correction at high frequencies to alleviate this problem. A slight roll-off is introduced that restores the balance between direct and reflected sound.

So on the high frequency side, I am not sure (that is, you may right..) Audyssey uses the wrong target, and if they did, probably not intentionally.

On the LF side, I would agree with Sal1950 that flat or slope is a matter of personal preference/opinions. It is not a "fact" that flat 20 to say 300 Hz is wrong, in fact in a large well designed concert hall or outdoor, the flatness would be similar to measurements in an anechoic chamber too. Just because studies found most people prefer the sound of a sloped response towards the low end does not make it a "right" as such. Some may prefer a more steep slope than others too, so it is hard to call it right, or wrong or in between.

Without the App, then obviously then AARC and Dirac are better in the sense that they allow the user to customize their target curves. With the App, Audyssey is fine imo as long as one also use Ratbuddyssey for ease of use, and for better precision too if that's important to the user.

 

[rynberg]

Not personal at all, I simply acknowledge that the designers have much more knowledge in the area than I and most likely you, although I see you don't agree.

As I said before, this isn't a personal opinion. If you put a flat loudspeaker in a typical listening room, the response is not flat, it trends downward with increasing frequency. Go and look at any speaker reviews on this website -- if you see any flat speakers that have a flat in-room response, let me know. Myself, I am dumbfounded that Audyssey will not acknowledge they were wrong 15 years ago and fix it. The engineering behind Audyssey is solid -- they are simply using the wrong target. This happens all the time across many industries.

Besides you don't have to agree with me...physics doesn't care what you believe. Perhaps Dr. Toole meets your qualifications?

Don't know where your 99+% number came from, did you do a survey of a large sample of Audyssey users settings? But in any case, yea, when given an option a large number of users will crank up the bass a bit, because they like it, not because of any component or speaker designers err in design. That truth goes back to the earliest days of HiFi with tone controls and loudness buttons.

No, I have not personally performed a survey, I've just been on A/V forums for more than 15 years. It's a pretty much default Audyssey tip for any newbies. Tone controls and loudness buttons have a very real purpose - correction of poorly mixed/mastered content and listening at a loudness level below that which the content was mastered.

And of course, there is some level of personal preference, which I alluded to before.

Humm, the NS-10 is probably the most widely used mixing console speaker in the world, I guess the mixing engineers have no idea what they're doing either? What you find "horrific" they find revealing and accurate.

You misunderstand how the NS-10 was/is used. No knowledgeable professional ever used an NS-10 to develop their mix, other than to make sure it translated to a really crappy playback system -- that's right, the NS-10 was used to see how their mix sounded on a boombox or cheap headphones. Sure, a lot of ignorant amateurs saw them being used in professional studios and adopted them in their project studios without understanding this fact. That doesn't change the reality that it is an excellently engineered product with completely the wrong target (just like Audyssey).

Genelec and Neumann are used for revealing and accurate.

When it comes to these midlevel semi-auto DRC systems, they all have their weak and strong points and the users all have their preferences in sound. When left to do their own thing you will end up with the decisions made by the designers. You then have the choice to live with it or modify it to your preference depending on the tuning tools given.

Yes, except for the fact that the Audyssey target is wrong, as compared to others. Why is the app or Rat so popular if not?

 

[rynberg]

On the LF side, I would agree with Sal1950 that flat or slope is a matter of personal preference/opinions.

I'm not talking about preference, I'm talking about the real physical response of a speaker in a listening room.

It is not a "fact" that flat 20 to say 300 Hz is wrong, in fact in a large well designed concert hall or outdoor, the flatness would be similar to measurements in an anechoic chamber too.

Well, we are talking about small room acoustics, not concert halls or outdoors. Music is not mixed in these environments, it is mixed in rooms much closer to home listening rooms (size-wise).

Just because studies found most people prefer the sound of a sloped response towards the low end does not make it a "right" as such. Some may prefer a more steep slope than others too, so it is hard to call it right, or wrong or in between.

As I said, it's how a speaker behaves in a real room. That's why most people prefer the sound!

You are 100% correct that there will be some individual preference in the degree of slope. Still doesn't change the fact that Audyssey should not aim for a flat response from 20-3,000 Hz.

 

[peng]

I'm not talking about preference, I'm talking about the real physical response of a speaker in a listening room.

Well, we are talking about small room acoustics, not concert halls or outdoors. Music is not mixed in these environments, it is mixed in rooms much closer to home listening rooms (size-wise).

So we are just talking about different things then, not that we are disagreeing.

You are 100% correct that there will be some individual preference in the degree of slope. Still doesn't change the fact that Audyssey should not aim for a flat response from 20-3,000 Hz.

Agreed on that too, but I think they realized that eventually, hence the launch of the App.

 

[giuppo77]

Just my two cents.

Audyssey aims at a flat response between 20 and 3 kHz (and even further) even because of the DEQ (Dynamic EQ) feature. If the response were not flat, then DEQ would not work properly. Just enable DEQ and the curve is not flat any more. Sure, we could discuss about the issues DEQ brings into the game; at the same time Audyssey is flat only if you do not use DEQ (with or without the app).

 

[Vasr]

Just my two cents.

Audyssey aims at a flat response between 20 and 3 kHz (and even further) even because of the DEQ (Dynamic EQ) feature. If the response were not flat, then DEQ would not work properly. Just enable DEQ and the curve is not flat any more. Sure, we could discuss about the issues DEQ brings into the game; at the same time Audyssey is flat only if you do not use DEQ (with or without the app).

No, this is not correct. DEQ does a tonal adjustment (hence why the tone controls are disabled when using DEQ) to compensate for volume changes and independent of what the target curve was. It can be selected whether you have the Reference target (rolled off) or the Flat target in XT32.

DEQ is basically adding a variable frequency-dependent delta to the spectrum to increase or decrease the relative levels of the response to compensate for difference in audibility at different volumes. An automatic tone control so to speak.

 

[Sal1950]

Just my two cents.

Audyssey aims at a flat response between 20 and 3 kHz (and even further) even because of the DEQ (Dynamic EQ) feature. If the response were not flat, then DEQ would not work properly. Just enable DEQ and the curve is not flat any more. Sure, we could discuss about the issues DEQ brings into the game; at the same time Audyssey is flat only if you do not use DEQ (with or without the app).

Problem with DEQ is how it messes up front/rear balance, now there's a sure mistake.

 

[giuppo77]

No, this is not correct. DEQ does a tonal adjustment (hence why the tone controls are disabled when using DEQ) to compensate for volume changes and independent of what the target curve was. It can be selected whether you have the Reference target (rolled off) or the Flat target in XT32.

DEQ is basically adding a variable frequency-dependent delta to the spectrum to increase or decrease the relative levels of the response to compensate for difference in audibility at different volumes. An automatic tone control so to speak.

Here is the snapshot of a REW measurement I took some time ago with DEQ enabled. I do not remember the level of the master volume but the DEQ offset level was 0.

It looks correct to me that with DEQ enabled the frequency response is not flat at all.

 

[giuppo77]

Problem with DEQ is how it messes up front/rear balance, now there's a sure mistake.

I agree, they tried to do too much. They argue that psychoacoustics can justify that.