• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

NAD M33 Streaming Amplifier Review

srslee

Member
Joined
Jan 29, 2022
Messages
23
Likes
6
What's the best way to compare Dirac on vs. off given the volume difference? I've tried making a Dirac filter that barely corrects anything, and have been using that as the 'Off' comparison setting, but not sure if this is at all a viable way to approach things.
 

Trondi

Member
Joined
Sep 16, 2021
Messages
14
Likes
7
What's the best way to compare Dirac on vs. off given the volume difference? I've tried making a Dirac filter that barely corrects anything, and have been using that as the 'Off' comparison setting, but not sure if this is at all a viable way to approach things.
As I understand it there are three states that i use to compare DL (full frequency version*)

1. DL switched off, so you are hearing your system with all natural timing inaccuracies and all room induced frequency spectrum peaks and hollows added to the sound quality that your system is able to provide.
2. DL switched on and curtains drawn together, so you are getting timing correction across the full frequency spectrum but no attempts to compensate for peaks and hollows as mentioned above. This is a very interesting state that allows us to see how important impulse timing issues are with our current systems.***
3. DL switched on and.... well at this point we, the end users of DL, will have tried our best to create/use a good target curve (**see below) , so hopefully we will be getting the best timing and frequency response from our hifi/room system.

* the bass only DL lite system will not let us have control of the high frequency curtain, which means that with DL on we will always have high (>500Hz) frequencies bound/altered by the target curve in use at the time.
** having bought my m33 some time ago, i have recently home demoed several new speakers. Each of them produce very different DL shaped FR curves (even positioned in the exact same points in the room) and therefore needed very different target curve tweaking to produce the best final result to my taste.
*** only one speaker demoed out of 4 speakers showed that it already had good impulse timing naturally (Russell K 150se). All the others showed a comparatively large change when state 2. above was engaged. ( B&W805; Audiovector R3; Kudos Titan 606)

Other points of note were:
- the Russell K150se, in room FR curve, most closely correlated with the standard DL target curve. The other 3 speakers each had varying degrees of naturally raised levels at tweeter frequencies (making them sound dull with using an untweaked standard curve, as would have to happen with DL Lite).
- i redid DL measurements several times for each speaker and found that my final results improved with practice at using the software and microphone placements. Astonishing ;)
- at no time did i notice a volume change with DL on compared to off. There may have been some difference but i didn’t notice any.
- ‘my taste’ with DL target curve tweaking is to get as natural frequency response as I can from my M33 through into my ears at my listening position. I am trying to wean myself completely from room boom and hot tweeters, both of which I have been accustomed to for 40 years or more. It is getting easier by keeping DL on at all times, and its sounding great!
- I have ended up with the Audiovector R3s because they sound good in my room AND they have the greatest sensitivity, which with my M33 solved the problem of maxing out of volume headroom with some of my input sources.
 

srslee

Member
Joined
Jan 29, 2022
Messages
23
Likes
6
As I understand it there are three states that i use to compare DL (full frequency version*)

1. DL switched off, so you are hearing your system with all natural timing inaccuracies and all room induced frequency spectrum peaks and hollows added to the sound quality that your system is able to provide.
2. DL switched on and curtains drawn together, so you are getting timing correction across the full frequency spectrum but no attempts to compensate for peaks and hollows as mentioned above. This is a very interesting state that allows us to see how important impulse timing issues are with our current systems.***
3. DL switched on and.... well at this point we, the end users of DL, will have tried our best to create/use a good target curve (**see below) , so hopefully we will be getting the best timing and frequency response from our hifi/room system.

* the bass only DL lite system will not let us have control of the high frequency curtain, which means that with DL on we will always have high (>500Hz) frequencies bound/altered by the target curve in use at the time.
** having bought my m33 some time ago, i have recently home demoed several new speakers. Each of them produce very different DL shaped FR curves (even positioned in the exact same points in the room) and therefore needed very different target curve tweaking to produce the best final result to my taste.
*** only one speaker demoed out of 4 speakers showed that it already had good impulse timing naturally (Russell K 150se). All the others showed a comparatively large change when state 2. above was engaged. ( B&W805; Audiovector R3; Kudos Titan 606)

Other points of note were:
- the Russell K150se, in room FR curve, most closely correlated with the standard DL target curve. The other 3 speakers each had varying degrees of naturally raised levels at tweeter frequencies (making them sound dull with using an untweaked standard curve, as would have to happen with DL Lite).
- i redid DL measurements several times for each speaker and found that my final results improved with practice at using the software and microphone placements. Astonishing ;)
- at no time did i notice a volume change with DL on compared to off. There may have been some difference but i didn’t notice any.
- ‘my taste’ with DL target curve tweaking is to get as natural frequency response as I can from my M33 through into my ears at my listening position. I am trying to wean myself completely from room boom and hot tweeters, both of which I have been accustomed to for 40 years or more. It is getting easier by keeping DL on at all times, and its sounding great!
- I have ended up with the Audiovector R3s because they sound good in my room AND they have the greatest sensitivity, which with my M33 solved the problem of maxing out of volume headroom with some of my input sources.
Huh, that's interesting. I'm not using the M33 right now as I'm still waiting for delivery, but with the M10 v2 (with pre-outs to another amp), I can noticeably spot a difference in SPL (between DL off & curtains drawn together). I do have the full frequency version, but for now, prefer correcting somewhere around the transition frequency (have a few profiles with curtain at 250-350hz). Makes a good difference in the transience and full-bodiedness of the bass at the MLP.

Just to confirm this, what would be the best way to measure SPL between the different profiles? I've been using a Umik-1 & REW's SPL meter, but wondering if there's a particular tone I can play to volume match. Maybe 1k or 3k sine wave with a tone generator?
 

doorofnight

Member
Joined
Mar 31, 2019
Messages
79
Likes
36
Location
The Netherlands, Rotterdam
Maybe I'll buy the M33, but I've been reading about some distortion problems.. Have these been solved by now? Are there still issues with this machine? And what to think about the jitter on the coax digital input? I like to read the measurmenents, but it's sometimes difficult for a nitwit to understand the implications.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,752
Location
Alfred, NY
Maybe I'll buy the M33, but I've been reading about some distortion problems.. Have these been solved by now? Are there still issues with this machine? And what to think about the jitter on the coax digital input? I like to read the measurmenents, but it's sometimes difficult for a nitwit to understand the implications.
Neither @Kal Rubinson nor I experienced any distortion problems in our respective reviews. The coax input functioned perfectly, as did digital optical.

I badly wanted to keep it, but sadly, NAD made me send it back.
 

Trondi

Member
Joined
Sep 16, 2021
Messages
14
Likes
7
Just to confirm this, what would be the best way to measure SPL between the different profiles? I've been using a Umik-1 & REW's SPL meter, but wondering if there's a particular tone I can play to volume match. Maybe 1k or 3k sine wave with a tone generator?

The app Decibel Ultra is one that i have been using recently. It has many settings and meters to measure sound.

The LEQ readings probably matches your requirements most closely. - see below their blurb copied from their info section.

Any repeated piece of music i guess would give you an indication of SPL, but some fuller frequency white noise would probably give you more accurate and precise values.

Leq : Equivalent continuous sound pressure level. Displayed as LAeqt , LBeqt , LCeqt , LDeqt or LZeqt , depending on the frequency rating is the average during the measurement period (t ) measured values , corresponding to the sound energy occurred . Does not show you the exposure that you were exposed !
 

doorofnight

Member
Joined
Mar 31, 2019
Messages
79
Likes
36
Location
The Netherlands, Rotterdam
Here is my latest from NAD on the issue

Hello Rene,

Thank you for your reply.

This issue is a high priority on the engineering team's list to resolve but not a simple fix, hence it has taken longer than expected to resolve.

The distortion does not affect all tracks, just a few that were recorded HOT with inter-sample clipping.

The distortion can be resolved by setting "Tone Controls - ON", or if one does not have subwoofers connected, setting "Select Attached Speakers - No Subwoofer".

Again I apologize that we do not have a more immediate solution for you now, but we will let you know when a fix is released.

Kind Regards,

Bob Moran
NAD Support Specialist


Are we still waiting for this bug to resolve? Yesterday I heard distortion in a Billie Eilish song: Ocean Eyes, in the last 20 seconds. (Tone control OFF, subs ON, DIRAC engaged and DIRAC off).

I am not sure if tone control on only effects the volume or also the dynamics and maybe other things. I bought the NAD M33 only 2 days ago.
 

Stefan

Member
Joined
May 29, 2021
Messages
67
Likes
13
Are we still waiting for this bug to resolve? Yesterday I heard distortion in a Billie Eilish song: Ocean Eyes, in the last 20 seconds. (Tone control OFF, subs ON, DIRAC engaged and DIRAC off).

I am not sure if tone control on only effects the volume or also the dynamics and maybe other things. I bought the NAD M33 only 2 days ago.
Yes, still waiting on this. Since I've sold my SVS SB-16 Ultra and don't need subwoofer OUT anymore, this problem is solved for me at the moment.

But I think it is a general problem regarding the headroom in the DSP curcuit. My experience (and I have been experimenting on this for a long time now) is, that switching either DL ON (obviously) or SUB ON or Tone Controls ON alters not only volume but overall sound (subjective listening).

Earlier in this thread (some weeks ago) I factory reset the M33 and have not touched any of the above mentioned controls. This is what I listen to for some time now and for myself am sure that this is the best sound quality I can get with the M33. Still I think that the M33 as a power amp does a marvelous job.
 

pogo

Major Contributor
Joined
Sep 4, 2020
Messages
1,242
Likes
382
You can also bypass the sub out and connect your sw to the pre out. Settings such as crossover frequency, level, ... are then made directly on the sw.
 

doorofnight

Member
Joined
Mar 31, 2019
Messages
79
Likes
36
Location
The Netherlands, Rotterdam
Thank you for the feedback. I am not sure what to do. Maybe bring the NAD M3 back to the shop ask for a refund.
You can also bypass the sub out and connect your sw to the pre out. Settings such as crossover frequency, level, ... are then made directly on the sw.

And will DIRAC recognise the subs and process them in DIRAC?
 
Last edited:

pogo

Major Contributor
Joined
Sep 4, 2020
Messages
1,242
Likes
382
And will DIRAC recognise the subs and process them in DIRAC?
I have a passive stereo sub and for me it works necessarily without problems. The whole thing is then seen as two full-range loudspeakers.
 

vanwalsum

Member
Joined
Sep 9, 2021
Messages
31
Likes
13
Changed my room a bit, so time for a new Dirac calibration on the M33. Version of the Dirac Software from 3.1.2 to 3.2.2. After measuring I discovered a thin sound. When I looked at the old project and the new. The higher frequency's drop about 10db in the new situation. Just put my speakers a meter or 2 further down the wall and moved a sofa. One speaker now something more in a corner. But this -10db difference is not normal I think? Used the Umik 1 and the right and same calibration.

Schermafbeelding 2022-03-14 om 19.31.58.png



Schermafbeelding 2022-03-14 om 19.57.29.png


What could be the problem?



----------------------------------------
Correction, I gave the amp a reboot and measured again. Weird!

Schermafbeelding 2022-03-14 om 20.34.10.png
 
Last edited:

pogo

Major Contributor
Joined
Sep 4, 2020
Messages
1,242
Likes
382
With this shown DL target curve you lose at least 6dB volume at the end. Either drag the DL target curve graphically upwards or increase the values in the target curve file by approx. 6dB. You don't have to give away the 10dB that DL allows itself in the M33;)

Here are the differences in performance when dBs are given away: Link
 

Astoneroad

Addicted to Fun and Learning
Forum Donor
Joined
Feb 16, 2022
Messages
999
Likes
2,046
Location
a Cave in the desert
Maybe I'll buy the M33, but I've been reading about some distortion problems.. Have these been solved by now? Are there still issues with this machine? And what to think about the jitter on the coax digital input? I like to read the measurmenents, but it's sometimes difficult for a nitwit to understand the implications.
I owned it, after a couple of firmware updates that fixed some irritations, I really enjoyed it, paired with Raidho C 1.1 stand mounts. Alas, and this is a personal observation, I found myself digging deep and thinking about the music more than just enjoying it. It reminded me of when taking "Method" acting classes... it took me a couple of days after scenes to "purge" my brain of the character. I quit that class forever... and the NAD was sold to someone with an appropriate approach to what it offered. I enjoyed both experiences, but needed to move on... or go mad... lol. :eek:
 

Calou

Member
Joined
Feb 20, 2022
Messages
29
Likes
5
New auto target curve: Link

Hi. If you see post #1050 and #1052, would be very neat if you could also run the same dB test, along with posting your speaker sensitivity and distance from the speakers etc.

Well, the guy from the HiFi shop suggested that it may be due to both the volume wheel being more linear than usual, combined with the fact that the amp has very low distortion (so loud listening doesn't trigger that earache you get from cheaper amps). I don't know - max volume has been enough so far, even with Dirac on, but for very dynamic pieces, like classical, it's only just enough. With my old NAD C375BEE, I had the volume wheel at max 40% when listening to classical music (with peaks well over 100dBa). About the same volume as 80% (where 90% is max with Dirac) on the M33. Though, I never tried cranking the wheel all the way, since it's less powerful than the speakers and would blow some expensive nautilus tweeters. For other music genres, I usually stop at between 65-75%, since the loudness war has killed just about all dynamics in modern music.

In your case, with Dirac and tone controls enabled, the loss should be around -17dB in theory - that's definitely a whole lot.

Overall I am happy with the sound too, except when listening to rock and metal. The snare drums sounded magical on the C375BEE, for some reason. No doubt it colored the sound somehow, to my liking.
Follow up on volume issues, I was playing with Dirac live App and I've noticed a significant volume difference when using pink noise with Dirac app or desktop during the volume configuration before measurement and with Tidal Connect.

In short when using Dirac during the volume setup I get good pink noise volume at -60dB
When playing pink noise from a test cd with tidal connect I need to get the volume to -20 dB to get a similar result.

Here is a quick video to illustrate it


I don't think such a difference is normal, I can't explain it.
 

Odradek

Member
Joined
Nov 9, 2020
Messages
43
Likes
10
Hello!

I finally received my M33 last week and have been experimenting a bit with it but I am not sure I understand how to properly use dirac with subwoofers. Following the instructions I suppose that I should select the "2 subwoofers" option and the corssover frequency in the M33's options, and then run the dirac calibration, but I find it odd that when dirac takes the measurements it doesn't seem to care for the crossover frequency though it is 80 Hz for example, the main's sweep goes down to about 30, and the reverse happens for the subs.

And then when I get to the filter design stage, I see the subs curve go up to beyond 200 Hz and I wonder, should I do the crossover filters manually here? It doesn't seem right.

Afterwards when in operation with dirac activated, I can change the crossover frequency in the settings and dirac doesn't switch off, so I wonder what is it really doing.

I am sorry for the convoluted text, but hopefuly someone can clarify this a bit for me.

I have not received my Umik mic to use with Rew which I also hope will help me undestand what is happening.
 

pogo

Major Contributor
Joined
Sep 4, 2020
Messages
1,242
Likes
382
I have not received my Umik mic to use with Rew which I also hope will help me undestand what is happening.
You don't have to wait and you can also use the included NAD mic under REW. Calibration data is also available for it.
 

Odradek

Member
Joined
Nov 9, 2020
Messages
43
Likes
10
You don't have to wait and you can also use the included NAD mic under REW. Calibration data is also available for it.
Thank you I will try that!

Do you know if there needs to be any ajusting of the crossover frequency in dirac??
 

mattyjman

Member
Joined
Mar 30, 2022
Messages
7
Likes
2
Hey guys - got maybe a basic question for you on the M33. I picked one up not too long ago, paired with some Focal Aria 936 towers, and plugged into two SVS SB1000 subs. I've found with this setup, the sub volume was pretty low. I maxed the volume on the subs, and also had to play with the 3 PEQ bands on each sub to increase the level of output to be complimentary to the Focal towers. I read some pages back about people having issues with auto-turn on from the 1.1v output of the sub preouts. Is there some way to fix this? Or boost the line output some way? I find it odd that with 2 subs with a decent amount of power is unable to match the output volume for volume between the towers and subs.
Anyone else had any issues like this? Thoughts?

My only real solution I could come up with is get a higher power set of subs, but that seems like a poor solution.
 
Top Bottom