• 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!

Dirac Live Exclamation Mark issue

As you can see here, the UMIK-1 produces self-noise up to almost 60dB(Z): Link
DL calibration should be carried out at a minimum of 85dB to avoid masking effects. An UMIK-2 is better here.
 
I don't get why this is not an issue on REW.
What about reaching +10dB on the volume selection slide of Dirac? May it cause clipping?
 
What difference does it make except some closer tolerances for the Umik-2 ? Both mikes are specified by Dirac (and Minidsp) inluding sensitivity for their behavior without any preferences.
Most sellers just use it because they make more money selling it, when used as an example, but it doesn't make any relevant differences for this purpose.
Dirac expressetly recommends the Umik-1 for its measurements and even the less specified and in the box included mike does not make such a large difference, that it is worthwile to consider a "real change". The umik-1 is different as it has a conversion from analog to USB "on board", which simplifies hook up for any Standard-PC.

Most rooms and theire acoustics cause much larger deviations and usually nobody claims (with proof) anything negative about it... :)
While both UMIK-1 and 2 are recomended that does not mean they actually work correctly, especially the UMIK-1. My experience and that of others https://www.avnirvana.com/threads/acoustic-phase-measurements.13300/post-99941 is that for FR they are fine but they are not accurate enough for timing measurements. My guess is that is why DIRAC does fine with FR but not so well with timing.
 
I use a umik-2 and have tried 40dB difference but still get the error.
 
The thing is, if noise floor is the issue then the error should be about noise floor, not “imprecise measurement” with a million different possible causes.
 
Ok dumb question. How do I uninstall the current version?
Sorry I’m on PC. I uninstalled it in the same manner I’d do any other program, ie it wasn’t an option inside Dirac, its windows driven. Settings>Apps….uninstall
Also fyi I’m using a HTx.
 
While both UMIK-1 and 2 are recomended that does not mean they actually work correctly, especially the UMIK-1. My experience and that of others https://www.avnirvana.com/threads/acoustic-phase-measurements.13300/post-99941 is that for FR they are fine but they are not accurate enough for timing measurements. My guess is that is why DIRAC does fine with FR but not so well with timing.
Does it mean UMIK-1 is imprescise?
In my case It’s pretty obvious that I’m testing at very low volume to get accurate results, but it would be annoying that even at correct volume I wouldn’t get precission.
What’t the solution 200$ Umik-2? Any other recommended mic?
 
Does it mean UMIK-1 is imprescise?
In my case It’s pretty obvious that I’m testing at very low volume to get accurate results, but it would be annoying that even at correct volume I wouldn’t get precission.
What’t the solution 200$ Umik-2? Any other recommended mic?
For FR the UMIK-1 is precise enough, some here question the noise floor though. The big issue with the UMIK-1 is that is relies on USB for it's clock so it is not possible to get consistent timing measurements. If you look at the link you can see how the timing changes every run with the UMIK-1 where as the UMIK-2 with a built it clock it does not. Trying to use the UMIK-1 for "speaker design" where you are trying to time align drivers or trying to flatten phase is problematic since you can't get reliable and repeatable results. I can't confirm it but if REW can not get consistent timing results with a UMIK-1 I don't see how DIRAC can either. Timing is only used for the "phase" part of the correction and their are endless arguments about how audible phase is. I think it is safe to say FR is much more audible than phase but as DIRAC moves beyond just FR into timing and phase correction (for Sub integration and more) it seems like they are running into trouble (i.e. "!" issues and weird GD) which makes sense as if you can't get reliable measurements how can you make reliable corrections? It seems like the UMIK-2 is the safer bet with less noise more headroom and repeatable timing measurements.
 
I have the same measurement errors with the current versions and I'm back to 3.4.4

With 3.4.4 I have the same problems with phantom center on NAD M33. Deactivating delay/gain brings some improvement, but the phantom center is still not really there.

I've been hoping for a solution for months, Dirac currently cannot be used for me from Version 3.4.4 to 3.9.7

I have the same problems again with 3.10.1 and 3.10.3, I have measurement errors on the NAD M33.

Since both Dirac versions already worked with NAD M33, it can only be due to the NAD updates in the meantime.
 
Thanks for the feedback. :)

I tested the microphone on the NAD and on the laptop. I selected Line In and Phono as analog inputs, always the same error.
 
Dirac gives informations about disabling the noise filter on Windows and reenable it afterwards in his latest changelogs for Dirac Live.
This noise filter might seem to be (at least for some) the cause of those stated timing problems (exclamation marks).
Dirac Live seems to have problems running its later editions on a MAC, as stated in their accompaning informations.
 
The connection was fine and the microphone is fine, it worked during the last calibration too, the same error just occurred in the system again, whether on the Dirac side or the NAD side, I have no idea, I'm guessing NAD.
 
Back
Top Bottom