NAD M33 Streaming Amplifier Review

[vanwalsum]

To suppress the impulse correction in DL I have generated an ideally measured flat curve, which can be changed as described here via a target curve:
Link

I have degraded DL to a Devialet Sweet Room and get better results.
My flat curve is the result of a bypass measurement (M33 pre-out -> mixing channel L+R to one channel as input for DL) that is used instead of a real measurement.
As a side effect you can also equalize your headphones

So the purpose of your project is that you can do a sort of EQ at your speakers. The pro is that you don't have the impulse correction on your speakers to take out the "life". The downside is perhaps that you still have the peeks and resonance in the bass area, if you have them standard like in my room. Thats why I want an amp with Dirac, to make the low end sound cleaner and better. The past days I'm switching my Dirac measurements on and off. With Dirac the sound is flatter, but more controlled in the lower regions. But with Dirac off I get harder and better drum kicks, more space around the speakers, which I prefer. After all I want the good low end with the life.... grr...

 

[pogo]

The downside is perhaps that you still have the peeks and resonance in the bass area, if you have them standard like in my room.

That's why I previously identified my problem areas with REW and created an opposite target curve in DIRAC LIVE and that works fine.

Interestingly, I recently had a second pair of my speakers optimised by a tuner, replacing the mid-bass drivers with 'faster' ones. Result: My room modes are no longer excited so strongly that I was even able to deactivate my DSP correction in the range of 50-100Hz and I can hear more 'micro details'!

 

[fcracer]

An FYI for those thinking of adding this beautiful subwoofer to your setup. I picked up a white KC62 to pair with an NAD M33, but sadly had to return it after seven days. There are a number of issues I experienced:

• The KC62 won’t wake from sleep using the M33’s sub-out unless I increase the volumes to high levels. This is partially due to the M33’s low voltage output (max 1.1V) and the KC62’s wake threshold being too high (even their own KEF products struggle to wake the sub).
• We all know about the inter sample clipping issue which rears it’s ugly head when using the subwoofer outputs. The temporary fix is to turn on tone controls and turn the volume up 10db to compensate.
• There was a hum or ruffling sound from the subwoofer that would come and go. Changing the ground lift switch would sometimes fix the issue, only for it to reappear again randomly.
• When using the M33’s pre-outs, there was an ENORMOUS pop from the subwoofer when it would come out of sleep. This would be a full excursion movement of the woofer cones. The first time it happened, I’m certain it shortened my life by a few years. I posted a video of it on the KEF owner’s forum on Facebook if you’d like to see the incredible excursion of the little woofers.

The KC62 is a marvel of mechanical engineering, but it feels like a beta product. I’ll definitely be looking at the V2 version of this subwoofer, when all the kinks are worked out.

As an aside, for those that dislike Dirac, I would think there has been some error in measurement or the filter creation process. The most recent version of Dirac has some bugs that result in the filter putting one channel out of phase. You can test this yourself if you’re getting funky outcomes from Dirac; just flip the wiring on one speaker and load Cheskey’s phase test single. You’ll know right away if you’re affected by the bug. While Dirac has confirmed this bug exists in 3.1.1, I’ve actually experienced this bug in previous versions as well, so it may be some interaction issue with the M33, or between Dirac and MacOS.

When Dirac is working correctly, it’s glorious, especially with the NAD curve which boosts the low end by 7-8db and gentles curves down at the high frequencies. I can never go back to listening to music without Dirac.

 

[unexperienced]

NAD никогда не говорил, что Дирак будет доступен на M32. У меня был один в течение многих лет - отличный усилитель, но я обменял его на еще лучший M33. Я бы раскритиковал NAD за то, что они сказали, что их слоты для модулей сделают M32 «перспективным». Единственным достойным модулем была карта BluOS - функция, включенная в стандарт M33.

Ранее вы предположили, что М33 не хватает «жизни». В этом я соглашусь с вами, если Дирак будет заниматься. К сожалению, как и во всех усилителях, которые включают DSP, дополнительная обработка сигнала, которую требует DSP, высасывает часть жизни из музыки. Есть способы избежать этого, но не если вы используете один полноразмерный усилитель для питания ваших динамиков. Даже версия Dirac Live, поставляемая с M33, отрицательно влияет на верхние частоты, хотя она регулирует только диапазон 500 Гц. Я просто отключаю его, так как предпочитаю больше волнения и «жизни», а не точности басов в любой день. К счастью, я могу достичь и того, и другого, так как мои динамики предлагают DSP в своих активных басовых усилителях.

You're wrong. When the NAD M32 was presented on the official website and in the news, and in the section of this product, the Dirac logo was displayed. It was written in the product history that Dirac will also be available in the future. Perhaps it was meant that it would be implemented through the Bluos module. But after a while, the Dirac icon disappeared from the product description page, and the text was edited. I am one of the first who bought the M32 in Europe. It's a pity that I didn't save the screen with that first page of the M32 description from the offsite.

 

[pogo]

Perhaps it was meant that it would be implemented through the Bluos module.

Here it is:
MDC2-Module

I don't know whether the M32 will be enabled in the future.

 

[fcracer]

As an aside, for those that dislike Dirac, I would think there has been some error in measurement or the filter creation process. The most recent version of Dirac has some bugs that result in the filter putting one channel out of phase. You can test this yourself if you’re getting funky outcomes from Dirac; just flip the wiring on one speaker and load Cheskey’s phase test single. You’ll know right away if you’re affected by the bug. While Dirac has confirmed this bug exists in 3.1.1, I’ve actually experienced this bug in previous versions as well, so it may be some interaction issue with the M33, or between Dirac and MacOS.

When Dirac is working correctly, it’s glorious, especially with the NAD curve which boosts the low end by 7-8db and gentles curves down at the high frequencies. I can never go back to listening to music without Dirac.

Dirac has released 3.1.2 and I can confirm it has fixed the phase issue introduced in 3.1.1. I will say it again, Dirac Live is the biggest single improvement that I've experienced in my music system.

MacOS: https://artifacts.connect.dirac.com/public/accord/release/macos/diraclive-v3.1.2-setup.zip

Windows: https://artifacts.connect.dirac.com/public/accord/release/win64/diraclive-v3.1.2-setup.exe

 

[unexperienced]

Here it is:
MDC2-Module

I don't know whether the M32 will be enabled in the future.

Wow!

 

[vanwalsum]

Dirac has released 3.1.2 and I can confirm it has fixed the phase issue introduced in 3.1.1. I will say it again, Dirac Live is the biggest single improvement that I've experienced in my music system.

MacOS: https://artifacts.connect.dirac.com/public/accord/release/macos/diraclive-v3.1.2-setup.zip

Windows: https://artifacts.connect.dirac.com/public/accord/release/win64/diraclive-v3.1.2-setup.exe

Okay, a bit of good news. I just saw the new version, updated, loaded my old project measured in 3.1.1 (I think that isn't a problem?). Then I made a new export with the NAD target curve on another slot. And wow, the difference is immense. I was so disappointed in my M33 with Dirac, just got it a few weeks and only tested it with 3.1.1. Now switching back and forward widens the soundstage, nicer kicks...

 

[fcracer]

Okay, a bit of good news. I just saw the new version, updated, loaded my old project measured in 3.1.1 (I think that isn't a problem?). Then I made a new export with the NAD target curve on another slot. And wow, the difference is immense. I was so disappointed in my M33 with Dirac, just got it a few weeks and only tested it with 3.1.1. Now switching back and forward widens the soundstage, nicer kicks...

I tried a few things last night:
1. Build new 3.1.2 filters for measures taken with 3.0.14 = works great
2. Build new 3.1.2 filters for measures taken with 3.1.1 = out of phase
3. Re-measure with 3.1.2 = works great

I‘d recommend checking your phase by running Chesky’s Phase ID track and ensuring the in and out of phase is correct. It’s very obvious using that track.

 

[vanwalsum]

I tried a few things last night:
1. Build new 3.1.2 filters for measures taken with 3.0.14 = works great
2. Build new 3.1.2 filters for measures taken with 3.1.1 = out of phase
3. Re-measure with 3.1.2 = works great

I‘d recommend checking your phase by running Chesky’s Phase ID track and ensuring the in and out of phase is correct. It’s very obvious using that track.

Because i received my M33 at the begin of september my only impressions were with the bug. I tested en redo the measurements in 3.1.2 and wow, nog i know why i invested in the M33. Stupid how software can ruin the sound!

 

[pogo]

Stupid how software can ruin the sound!

The DL algorithms have also become audibly better over time, but where is the end?
Can there be more

 

[Ulmer]

I purchased an M33 recently and ran DIRAC with the included mic using v3.1.1 of the software. Didn't seem awesome. Ordered the a UMIK-1 calibrated mic. I just recalibrated with the new mic and v3.1.2. Wow! So much better and a big leap forward versus having correction turned off. Better soundstage, less boomy, more detail. Loving it.

I assume the software bug was most of my problem.

 

[pogo]

I assume the software bug was most of my problem.

On which platform was DL running for both measurements (Win, Mac, IOS, Android)?

 

[Ulmer]

On which platform was DL running for both measurements (Win, Mac, IOS, Android)?

Original measurements were with the bundled mic, connected to the M33 USB port, and the Mac DIRAC app 3.1.1.

New measurements were with the UMIK-1 with Cross Spectrum Labs calibration file, mic connected to Mac directly + Mac DIRAC app 3.1.2.

 

[Stefan]

Two hours ago I did the first measurement with the UMIK-1 I received yesterday.

Setup:

NAD M33 on newest BluOS 3.14.26
MacBook Pro with latest MacOS Big Sur 11.6
UMIK-1 ver.2 attached to the MacBook Pro via USB-C
Dirac Live 3.1.2

Result:

Speakers are out of phase

Any advice?

 

[Stefan]

Now that I have changed polarity on my speakers (first only the left channel, then only the right channel) to check which channel is actually affected, it seems that the problem is more complicated than that. None of the two changes yielded a positive result regarding the overall polarity / soundstage...

 

[Ulmer]

Now that I have changed polarity on my speakers (first only the left channel, then only the right channel) to check which channel is actually affected, it seems that the problem is more complicated than that. None of the two changes yielded a positive result regarding the overall polarity / soundstage...

Is it possible that they're not actually out of phase? Maybe you just don't prefer the DIRAC adjusted sound? What speakers are you using?

Are you applying the adjustment to the full frequency range, or just under 500 Hz? I have the full license and am applying it to the full frequency range. I feel like soundstage is improved, but we're all different.

 

[Stefan]

Is it possible that they're not actually out of phase?

That is basically what I said in post #896

Maybe you just don't prefer the DIRAC adjusted sound?

No offense taken. Not that I was unhappy with what I already achieved with Dirac. I can switch back to filters that I have. Just wanted to give the UMIK-1 and new DL version a chance. But the result has nothing to to with preferation, the soundstage is completely off, even worse than polarity. Definitely an issue on the DL side.

What speakers are you using?

Wharfedale Elysian 4

 

[Ulmer]

That is basically what I said in post #896

No offense taken. Not that I was unhappy with what I already achieved with Dirac. I can switch back to filters that I have. Just wanted to give the UMIK-1 and new DL version a chance. But the result has nothing to to with preferation, the soundstage is completely off, even worse than polarity. Definitely an issue on the DL side.

Wharfedale Elysian 4

That's a bummer. Didn't know your history with DIRAC. If you have prior filters that are much better, that makes it easy to judge. My first experience was with the supposedly bad 3.1.1 version, so I have no prior frame of reference. I hope you get it sorted out!

 

[Stefan]

I hope you get it sorted out!

Thanks. Later on today I'm going to do a measurement with DL 3.1.2 and the NAD supplied mic to rule out that the UMIK-1 is the reason for this issue.