NAD T778 Audio/Video Receiver (AVR) Review

[peng]

QC is of bigger concern for me. Though I think that NAD has shortchanged their amps, using Dirac to make up for their deficiencies, ultimately...if someone’s happy with the product, so be it. I would not purchase a product with this type of engineering ethic, but that’s me. My hope is their next gen. will at least live up to their own published spec.
One thing ARS does is call out companies that exaggerate their specs for marketing purposes. Consumers deserve honesty and consistent across the board apple to apple capacity for comparison from one brand to another. Not only is spec dishonesty endemic to much of the industry, but it allows for “just good enough“ engineering. As a consumer, I want more than a just good enough sexy box with useless vu meters, under performing amps, noisy fan and broken coaxial inputs, I want the best possible components, build, specs., support and QC I can get for my 3k. No way is this it. For those of you who have purchased this receiver, I hope it remains healthy, that NAD adequately supports it, and hardware upgrades are reasonably priced. Given NAD’s history of QC et al...it would be too risky for me. I really hope I’m wrong. They need prove themselves.


Dirac, along with all else it does, corrects frequency response in the room. If the amp doesn’t have a flat frequency response, which the NAD does not, it will be correcting for the amp deficiency as well.

I think it is a moot point either way because amps FR are practically flat from 20-20,000 Hz analog or 20-15,000 kHz digital if it is a recent year Marantz model because of the DAC filter choice but even then it is mainly the case for sampling freq 44.1 kHz.. Though I tend to agree with you that Dirac does not know who is the culprit for the dips and boost in the FR so it will try to do its "room correction thing" based on the data collected by the mic.

 

[Matyam]

Oh and the original nad pe 2200 snr is 116 and it still loses ,just saying.

 

[Dmitri]

I think it is a moot point either way because amps FR are practically flat from 20-20,000 Hz analog or 20-15,000 kHz digital if it is a recent year Marantz model because of the DAC filter choice but even then it is mainly the case for sampling freq 44.1 kHz.. Though I tend to agree with you that Dirac does not know who is the culprit for the dips and boost in the FR so it will try to do its "room correction thing" based on the data collected by the mic.

The NAD analog FR is pretty bad. I use the DAC in my Oppo for two channel listening. So, in that realm, the NAD is not adequate for my purposes. For digital sources, much less of an issue.

 

[peng]

How about without Dirac?

The thing is, no one can agree or disagree with someone who says device A sounds better than device B. The only thing for sure is, in a blinded listening test, we can expect no consensus because the "sound better.." thing is highly subjective. Apparently some people even think a heavier dosage of 2nd and 3rd harmonic could sound better than a lighter dosage of the same, so to some people, 2nd and 3rd, (may be 4th and 5th too) could be like spice/seasoning, if not too excessive obviously.. Same with REQ systems, years ago, Harmon's study showed even the worse REQ used in the comparison test were preferred by some, just fewer people. That worse was guessed to be Audyssey MultEQ, probably not even the XT versiion. If Dr. Sean Olive were to conduct another such comparison listening sessions, I would bet Dirac Live, Anthem ARC, and XT32's scores will be a lot closer, especially if "correction" is limited to below the room transition frequency, because all such products have probably been improved to, or near the point of diminishing return already. It would like comparing SINAD 90 to 100 to 110 dB in a room with noise floor at > 25 dB over the full audio band.

 

[peng]

The NAD analog FR is pretty bad. I use the DAC in my Oppo for two channel listening. So, in that realm, the NAD is not adequate for my purposes. For digital sources, much less of an issue.

Oh yes, I forgot, it did look bad, for an amp but it could also be the reason why some thought it sounded good, subjectively, and seriously..

 

[Vasr]

Of course Dirac is not supposed to fix poor frequency response in the amp section. That’s exactly what I’m saying. Dirac is for room equalization only. I’m well aware of that, and am not confusing it’s purpose. But Dirac works by interpreting audio sweeps, which would be affected by an amps poor frequency response. So, in the process of mitigating the room effects, it is also correcting equalization deficiencies in the amp. Dirac does not know the difference. That’s not what it’s meant to do, but ultimately it ends up doing it anyway. If I’m wrong, am not understanding how Dirac interprets the sweeps and somehow can see through an amps inability to m/n a flat response between 20Hz to 20kHz in the course of doing so then correct me. Otherwise, Dirac is inadvertently correcting the amp’s frequency response at the same time it is adjusting eq for the the room. Not what it’s for, but that’s what will happen. By definition, room eq and an amps frequency response are mutually exclusive. In practice, variations in an amps frequency response will be corrected by the room eq’s algorithms along with the room’s deficiencies as it is dependent on the signal sweeps from an imperfect source.

All this learning stuff makes for an interesting journey, no? ; )

You are correct technically but the outcome is not as effective as what one might expect in "correcting amp issues". Part of the problem is that such a correction is listening position dependent and this can be a huge negative when correcting "amp issues" at higher frequencies.

Amp FR issues don't typically result in dips and peaks of the type room resonance creates to be smoothed over and for which the room-eq is designed for.

The problematic amps have FR that fall off too quickly on either side or both sides. This is not something you can really fix with room eq. The amount of boost required at low frequencies may be too high for the available power (or headroom) to extend the FR in the low side. Boosting high frequencies can create very bad artifacts and localized issues depending on listening position, especially if you use narrow filters.

Rarely, amps might exaggerate the highs (or even flat responses may seem overly bright to some ears). These can be "corrected" with wide shelf filters in some room eq systems to correspond to a target curve that one prefers. This is probably where the room-eq most affects "amp issues" if you want to see it that way.

Of course, room eq can do nothing about noise or distortion issues. Other than rolling off the highs as above to reduce any harshness from noise or distortion issues more evident at high frequencies. Those are not fixing the amp problem but putting a band-aid over.

90%+ of the audible benefits of room eq comes from knocking off room resonances in the low frequencies, which exposes much more of the detail in the higher frequencies. If there are issues with the higher frequencies, it just exposes them more. Most room-eq systems are more or less equally effective in this part. Even a REW based manual correction can yield most of these benefits.

Dirac also seems to do a good job with phase alignment and they have proprietary algorithms to detect and correct for room issues that involve multiple reflections. The latter is what sets it apart from the other room eq systems out there and possibly gives that small extra clarity which can be subtle. With my experiments, in HT usage, Dirac can make the center channel voices clearer and the movements between surround and front speakers a bit more seamless than others (presumably by removing some multi-reflection artifacts but I am not sure of the technology here).

In 2 channel music, the difference between Dirac and other room eq like ARC or Audyssey is far less convincing unless you have a very problematic room with a lot of reflections.

 

[voodooless]

It would like comparing SINAD 90 to 100 to 110 dB in a room with noise floor at > 25 dB over the full audio band.

Except for the last part, with a decent enough DAC (just pick one with a > 110 dB SINAD from here), this can be done: just add distortion products to a music file to these levels, and ABX test them. That would actually be a very interesting test to do.

 

[3dbinCanada]

I don't think you are. Dirac will work based off a target, and the frequency response of the amps will be part of what it measures and thus makes correction filters for.

Its semantics. Room anomalies will play a much bigger factor than the poor frequency response of the NAD by a country mile. Its exactly what Vasir posted. NAD didnt put in Dirac to correct amplifier frequency response. They put it in to fix room response. Those are two different things altogether. To say Dirac fixes amplifier response is still incorrect because NAD's response isnt near as bad in room anomalies.

 

[3dbinCanada]

Sure, it gets obliterated—it is masked, laid over, superimposed, whatever you want to call it. Your point is that if the music signal is at, say, -80dB, you are not going to hear it so it doesn’t matter if it is drowned out by noise. That’s true, it won’t matter to some people. But the noise still does obliterate the signal. Some people are interested in achieving the highest fidelity possible to the source, and they are the ones reaching for technical perfection. Other people just want something to sound good but aren’t interested in fidelity. That’s fine, too, but I imagine it’s not what most readers of ASR are after.

Yes in theory it gets buried but you will never hear it so the problem from a blind listening test is moot. This is where subjective bias comes into play. You read the results and are convinced the NAD will sound mediocre or bad. I wonder if had you heard the NAD before reading the review if that would change your point of view?

My issue with NAD in the last 15 years is QC issues, not their sound. In a DBT test, they all sound same well within their power envelope. Things begin to sound different once you begin pushing the power envelope.

 

[Billy Budapest]

Yes in theory it gets buried but you will never hear it so the problem from a blind listening test is moot. This is where subjective bias comes into play. You read the results and are convinced the NAD will sound mediocre or bad. I wonder if had you heard the NAD before reading the review if that would change your point of view?

My issue with NAD in the last 15 years is QC issues, not their sound. In a DBT test, they all sound same well within their power envelope. Things begin to sound different once you begin pushing the power envelope.

I am not convinced it will sound mediocre or bad, although its noise performance could be audible at loud listening levels. Rather, my point is that it does not offer state of the art technical performance. I expect more from NAD, especially because this is NAD’s flagship AV receiver and is sold at a not-insignificant price point.

 

[Dmitri]

You are correct technically but the outcome is not as effective as what one might expect in "correcting amp issues". Part of the problem is that such a correction is listening position dependent and this can be a huge negative when correcting "amp issues" at higher frequencies.

Amp FR issues don't typically result in dips and peaks of the type room resonance creates to be smoothed over and for which the room-eq is designed for.

The problematic amps have FR that fall off too quickly on either side or both sides. This is not something you can really fix with room eq. The amount of boost required at low frequencies may be too high for the available power (or headroom) to extend the FR in the low side. Boosting high frequencies can create very bad artifacts and localized issues depending on listening position, especially if you use narrow filters.

Rarely, amps might exaggerate the highs (or even flat responses may seem overly bright to some ears). These can be "corrected" with wide shelf filters in some room eq systems to correspond to a target curve that one prefers. This is probably where the room-eq most affects "amp issues" if you want to see it that way.

Of course, room eq can do nothing about noise or distortion issues. Other than rolling off the highs as above to reduce any harshness from noise or distortion issues more evident at high frequencies. Those are not fixing the amp problem but putting a band-aid over.

90%+ of the audible benefits of room eq comes from knocking off room resonances in the low frequencies, which exposes much more of the detail in the higher frequencies. If there are issues with the higher frequencies, it just exposes them more. Most room-eq systems are more or less equally effective in this part. Even a REW based manual correction can yield most of these benefits.

Dirac also seems to do a good job with phase alignment and they have proprietary algorithms to detect and correct for room issues that involve multiple reflections. The latter is what sets it apart from the other room eq systems out there and possibly gives that small extra clarity which can be subtle. With my experiments, in HT usage, Dirac can make the center channel voices clearer and the movements between surround and front speakers a bit more seamless than others (presumably by removing some multi-reflection artifacts but I am not sure of the technology here).

In 2 channel music, the difference between Dirac and other room eq like ARC or Audyssey is far less convincing unless you have a very problematic room with a lot of reflections.

I get that. It will still be affected by the amplifier FR in how it interprets the sweeps, but will have little affect relative the room anomalies which are far greater in comparison. So, in the case of NAD’s analog input FR drop off 1.7 db between 5 and 20 kHz, does Dirac accommodate for that or not? And if it does, would that correction create your stated undesirable artifacts? I don’t get that...it’s a gradual drop off in FR, that should be corrected by Dirac room eq....knowing full well that’s not what Dirac is designed for nor it’s purpose.

 

[T3RIAD]

Why would I expect a $/€ 3000,- device to just sound "good"? The architecture of the NAD is a lot simpler than the Denon/Marantz. The lack of an analog path means they can rip out a lot of crap that actually degrades SINAD. Even with that it's 6dB off from Denon for the pre-amp section.

And yes, Dirac is indeed a major feature, and might result in an arguably better sounding system than a comparable Denon with Audyssey. It's still no excuse. Especially if you look at the implementation of the amp, they should have done better. It's not just the single -80 number. The original Hypex spec is not that far off at 5W, but below that, it actually looks much better on the Hypex spec sheet. It's clear that some corners were cut, also visible by the lower quality output coils used vs the ones on the Hypex OEM version. All so that they could make a bit more money. Als long as people keep finding this acceptable and no good alternatives are available, these products will be made and sold.

I see what you're saying, it's just that I see two opinions often expressed on this forum that seem to be contradictory:

1. Pretty much all DACs and amps sound the same, because distortion under -60 dB is nearly impossible to hear.

2. DACs or amps with better numbers are better products.

Why should AVR companies spend money making numbers go up that don't affect the sound? How is that any different from the $10,000 amps that have silver wiring and high-end capacitors and audiophile fuses in them? Those products are also technically better, but they get ridiculed here for good reason, IMO.

I don't think there's anything wrong with the review, and I was in fact going to send something to Amir to review soon. But these numbers aren't the only thing people look for in a receiver. I think the main thing most people want is an amp that's acceptably transparent, plus a bunch of surround sound codecs and DSP functions.

The existence of this site is a wonderful thing for audio, but I think people need to decide what the measurements are actually saying. Are they saying some products are better than others? Or are they saying most products sound the same? You can't have both.

 

[Vasr]

I get that. It will still be affected by the amplifier FR in how it interprets the sweeps, but will have little affect relative the room anomalies which are far greater in comparison. So, in the case of NAD’s analog input FR drop off 1.7 db between 5 and 20 kHz, does Dirac accommodate for that or not? And if it does, would that correction create your stated undesirable artifacts? I don’t get that...it’s a gradual drop off in FR, that should be corrected by Dirac room eq....knowing full well that’s not what Dirac is designed for nor it’s purpose.

I can't say it will never do so without knowing the technical details of the system but in my experience across multiple systems, I have never encountered corrections to raise the frequency response across a band in lower frequencies. There may be ways to force them to do so in manual tweaking but I have never tried to.

Note that the room-eq systems do not know if the fall-off is due to the amp or the speaker or some room mode. It is far more common to encounter such fall-offs due to speaker limitations than either of the other two in which case eq is not very effective in extending the lower FR of speakers and pumping large amounts of power to do so may be useless if not harmful.

Despite what the target curve is (or even what is shown as theoretical corrected curve which is pure fiction in some systems), natural slopes are left as is but they will smooth out deviations as they determine it. Just to illustrate from a known system, in REW, it will generate filters only when the response curve crosses the target curve successively in opposite directions in a local area. This detects peaks and troughs but will not consider fall-offs.

They may tame higher frequencies by cutting with shelf filters (ARC seems to do this in full range correction) to get closer to a target curve slope. But most corrections here will create at best locally optimal (in seating position) but likely artifacts everywhere else.

 

[Vasr]

The existence of this site is a wonderful thing for audio, but I think people need to decide what the measurements are actually saying. Are they saying some products are better than others? Or are they saying most products sound the same? You can't have both.

It is neither contradictory nor confusing if you understand that the reviews/discussion here almost always refer to "engineered better" not "sounds better". The latter is something that is explicitly avoided as it is a subjective experience and subject to biases and expectations.

The "engineered better" is with respect to an objective goal to achieve. Even if devices don't reach that top level of perfection, their relative measurements between comparable units still correlate with engineering effort objectively.

The only exception might be when a unit is so broken as to likely suffer from audible effects. But those are rare.

There may be any number of reasons why a unit is not engineered better - economics, incompetence, arrogance, laziness, pragmatism. But unless we were an insider we would never know and so cannot assume it was a conscious choice vs sloppiness. Which raises the question, if it was sloppy, what else is sloppily designed within the unit that may not be detected in hearing or testing immediately but may lead to problems down the road.... ability to withstand heat or shocks or surges, power cycles or any number of things. So, some people may avoid such units partly to not encourage such sloppiness and partly to feel safer with their investments with a better engineered unit. That is a personal choice, not a recommendation to others.

Personally, I would not depend on the measurements alone but read up as much as I can about it from other users, reviews, technical data, brand reputations, etc., all with a pinch of salt as they are prone to distortions from misinformation and false perceptions.

Contemporary NAD has been an engineering disappoint to me from the equipment I have owned before and what they stood for. That may not be relevant to others.

 

[yourmando]

I have already posted this earlier. It is difficult to use an external room correction box with an AVR if you want the sources to be connected to the AVR (for all the codecs to decode).

There is no way to get digital out from the AVR after the decoding to do room correctional externally with a digital input. If you take the pre-amp outs (and assuming you are ok with the ADC-DAC roundtrip in the external box and its DAC quality), you cannot attach it back to the AVR and have it go to the amps because you will need to have selected the original source to play to send its content out of the pre-outs and most AVRs doen't have the concept of a processor loop.

The only way to do this is by somehow connecting the source to the external box (after likely jumping through hoops for connector adapters, dongles, etc). They don't take HDMI and they certainly don't have all the codecs.

Cannot generalize from being able to do 2-ch PCM stereo for this application. That isn't the primary use of an AVR.

DSP Room correction in the PC should be possible if you have pre outs though. The latest Denon's have pre outs, for example.

pre outs -> something like the Motu 16a -> amps or active monitors

Then you have Dirac, Audiolense, Accourate, and many free options as well. For someone who really wants to customize.

@HighImpactAV made an excellent comment re Dirac and the use of IIR filters for the bass.

It would indeed be interesting to see a multi-channel head to head comparison of Audiolense vs Dirac.

If one really wanted to tweak and customize, I guess one could also use moderately priced Denon with pre outs or the HTP-1 to just to handle the HDMI switching and decoding.

Then use an AD/DA device such as the Motu 16a w/ 16 analog in & outs, and do DSP in the PC using Audiolense or Accurate or Dirac to compare, or to solve specific problems.

 

[Newman]

It is neither contradictory nor confusing if you understand that the reviews/discussion here almost always refer to "engineered better" not "sounds better". The latter is something that is explicitly avoided as it is a subjective experience and subject to biases and expectations.

Speaking as an engineer, I think I need to make you acquainted with the term "over engineered". You seem to be confusing "engineered better" with some pointless pursuit of ultimate perfection that is irrelevant to the intended consumer's experiences.

The "engineered better" is with respect to an objective goal to achieve. Even if devices don't reach that top level of perfection, their relative measurements between comparable units still correlate with engineering effort objectively.

See, you have done it again. The objective goal needs to be appropriate, not ultimate, for excellence in the engineering to emerge. Otherwise you just get pointless perfectionism, which is the opposite of excellence in engineering. It is sloppy, self-indulgent engineering.

...There may be any number of reasons why a unit is not engineered better - economics, incompetence, arrogance, laziness, pragmatism. But unless we were an insider we would never know and so cannot assume it was a conscious choice vs sloppiness. Which raises the question, if it was sloppy, what else is sloppily designed within the unit that may not be detected in hearing or testing immediately but may lead to problems down the road.... ability to withstand heat or shocks or surges, power cycles or any number of things. So, some people may avoid such units partly to not encourage such sloppiness and partly to feel safer with their investments with a better engineered unit. That is a personal choice, not a recommendation to others.

Personally, I would not depend on the measurements alone but read up as much as I can about it from other users, reviews, technical data, brand reputations, etc., all with a pinch of salt as they are prone to distortions from misinformation and false perceptions.

Contemporary NAD has been an engineering disappoint to me from the equipment I have owned before and what they stood for. That may not be relevant to others.

I was tempted to go one by one through the other points that you have made that are so biased by misperception in the paragraphs above, but in the end I actually decided that the tone was frankly insulting, and I don't feel inclined to go into any more detail.

In closing, let me just say that excellent engineering in an amplifier for example, with respect to distortion alone and assuming all of the parameters have been met, will be the cheapest design that has no errors that are audible. Any further reduction in errors is a waste and therefore poor engineering.

The idea that these receivers could all have 120 DB of SINAD, and for cheap, has not actually been proven. We haven't actually found one that achieves it, so we should not assume that it is easy. Also, nobody has yet pulled the DAC stage out of one of these receivers and tested it to still have the low performance that has been documented, nor has anyone replaced the DAC stage in one of these receivers with a cheap $100 DAC that performs so well, and still measured it to have such good performance in a receiver.

The main problem that I'm having with these units in relation to their performance, is price. They are probably over engineered, not under. But even then, purchase price is almost certainly out of the engineers' hands, and there is every possibility that they are being given insanely low budgets to develop the technical subunits. So stop blaming the engineers, and in particular keep the targeted insults to a minimum, please.

 

[3dbinCanada]

I am not convinced it will sound mediocre or bad, although its noise performance could be audible at loud listening levels. Rather, my point is that it does not offer state of the art technical performance. I expect more from NAD, especially because this is NAD’s flagship AV receiver and is sold at a not-insignificant price point.

Agreed. NAD has been dropping the ball for the last 15+ years.

 

[Anterantz]

thanks amir for the review it was the next av that i would have liked to try but having an 8500 i am not attracted to anything except dirac! I will continue in the shade enjoying this magnificent denon .. greetings

 

[Vasr]

DSP Room correction in the PC should be possible if you have pre outs though. The latest Denon's have pre outs, for example.

pre outs -> something like the Motu 16a -> amps or active monitors

Then you have Dirac, Audiolense, Accourate, and many free options as well. For someone who really wants to customize.

The problem is that you introduce an ADC-DAC cycle in that path (Denon pre-outs are analog) and you cannot use the amps in that Denon but you need another set of amps for all the channels or active speakers for all. If this is done for HT, the latency of that path relative to whoever is doing the video decoding and rendering may get quite a bit out of sync. So feasible but not very practical.

 

[T3RIAD]

Note that the room-eq systems do not know if the fall-off is due to the amp or the speaker or some room mode.

Amp vs. speaker is correct, but Dirac requires you to take nine different measurements in different locations, precisely so that it can distinguish between amp/speaker and room effects.

Question: the frequency response issue is with analog input only, correct? It wouldn't affect digital or HDMI inputs?