doorofnight
Member
Same calibration clipping problems. I used this instruction also. I can only calibrate at low volumes.
-
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
- Thread starter amirm
- Start date
What version of dirac live are you using?
doorofnight
Member
Always the newest version. With a calibrated UMIK microphone AND with the NAD mic.
If the pipeline is working correctly, and if my db math is correct, here are some power figures for the M33, assuming 0db on the unit is max power:
0db = 200W
-3db = 100W
-6db = 50W
.
.
.
-19db = 1.6W
-22db = 0.8W
-25db = 0.4W
In my apartment, with the speakers 3m away (~9ft), with 90db sensitivity, the volume set at -25db will produce 83db at the listening position. That's a decent volume and seems reasonable. At 0db, the same setup would produce 110db at the listening position.
That could be the issue. Dirac Live website says to use version 3.4.4 with NAD. Anything newer can have problems. Dirac change log here: https://www.dirac.com/live/faq/helpdesk/software-changelog/
doorofnight
Member
Ok, I've done a new measurement with the DIRAC 3.4.4. version. Also I set my subwoofers at 50% volume instead of 40%. It was possible to make measurements at a slightly higher level than before (but by far not in between -12dB and -15dB as the manual states).
Now the NAD can play much louder than before. I do not even want to come close to the max.
Thank for the help, again. I hope this also helps the other poster.
Now the NAD can play much louder than before. I do not even want to come close to the max.
Thank for the help, again. I hope this also helps the other poster.
Glad to hear it was resolved for you. FWIW, when I did my Dirac measurements, I recall the system volume was something like -50db or -45db. Anything higher and the room modes would trigger the clipping warning. I have a room mode that adds 20db at 47Hz, so I can't blame Dirac for triggering a warning.
I did receive the M33 back from the service. I can only conclude that it is working as designed. It is just not powerful enough when using Dirac in my setup, I don't like that I can max the volume (down to -10dB with dirac on) when I listen to some classical music like this one for example: https://tidal.com/browse/track/12720229
I also do not feel like adding a M23 in bridge mode to get a bit of extra headroom. The mandatory 50 ms delay of Bluos also gives me headaches for the integration with my surround system managed by my Marantz. Marantz has a limitation of max 20 feet between farthest and closest speakers (max buffer side for channels), the result is that the Marantz is not able to apply enough delay to surrounds to match the delay of the front speakers attached to the M33.
When I create filters I did also find out that running my subwoofer louder than the main level match in Dirac gives less volume attenuation, I have almost no volume difference with dirac on and off now. The sub output was measured 6dB quieter than the main speakers preout in some test with the tester mentioning it as being inexplicable. I don't have the link to the article anymore, that might explain.
I really like the M33, but I'm starting to think that it may not be the right solution for my setup.
These are my targets curves, crossover at 70Hz because of terrible bass response of the speaker in the right corner.
Attachments
I'm not fully opening the curtains because I like the natural response of the main speakers better than with Dirac.
For the sub I'm not sure how to make a full range filter that follows the natural response.
I had to save the target curve suggested by dirac for the mains to reuse with the sub, so yes I've them saved.
Indeed you have suggested many times to add a few dB to the control points of the target curves to avoid loosing too much headroom, I can play with that, how much would you add looking at the curves?
3.4.4 as advised. Waiting for the latest version to be fully bug free...
Thanks, I will try.
I've elevated the target curves by a conservative 6 dB to start with, this is what they look like.
It does indeed bring more volume, I have maxed out the volume to 0dB on the music provided in my previous post and I can clearly hear the volume difference when I switch between the 2 filters.
So this workaround works to restore some of the lost gain, Dirac and NAD should take note I think.
Attachments
Your room curve should typically slope down in the higher frequencies. The curve you have now will sound too bright. If memory serves me right, I have around +6db in the low frequencies and then slope down about -1db per octave. You can find the ideal room frequency preference in Floyd Toole’s book.
The filter did give the following REW response at the listening position when using the curtains, 1 measure only, and I'm quite happy with the the result, indeed for some people it could be too bright in the highs. My philosophy is more to not correct anything in the mids and high and to trust the engineer of the speaker, Cabasse in this case, I think that the Cabasse Murano Alto doesn't have a flat response in anechoic chamber like the older designs did, the result is totally non fatiguing and I can listen for hours.
Unfortunately I can't use the curtains anymore to correct only the bass response if I want to get some of the lost gain with Dirac.
