I was wondering whether anyone else using MiniDSP hardware like the SHD, SHD Studio etc are seeing this when they run a room calibration.
When this occurs Dirac will not correct for phase or delay. This problem started when I updated to Dirac 3.9.7 from 3.8.2 which worked fine the last time I used it a few months ago. I have always used my UMIK-2 with Dirac without issue. I subsequently reverted to 3.7.3 and continued to have the issue until I switched to my UMIK-1 mic and I was finally able to do a calibration which seems valid. (Not one of my best outcomes however)
The UMIK-2 works fine with REW, HolmImpulse and audiotester.de., but suddenly not with Dirac.
I previously owned an SHD before replacement with the Studio a couple of years ago, and have a lot of experience with doing Dirac calibrations. I never had any problems until a week or so ago when a hardware replacement (Danley DNA SC48) dictated the need for a recalibration after I completed configuring it. (I did a lot of measurements with the UMIK-2 during that process and had no problems with it.)
Dirac provides access to previous versions of Dirac as well as release notes for each version, but I have not found a way to contact anyone there about the problem.
More disturbingly there are reports of this problem on ASR going back to 2019 so it has been plaguing MiniDSP users for some time (also other licenssees). MiniDSP indicated that they were not really able to provide assistance in this matter, and to contact Dirac for assistance. (FWIW - maybe not much.)
Without functional Dirac the Studio is sort of pointless in a sense, although the digital pre-amp functionality is still needed.
And yes the room is not great, I've made a lot of improvements, but serious issues persist.
When this occurs Dirac will not correct for phase or delay. This problem started when I updated to Dirac 3.9.7 from 3.8.2 which worked fine the last time I used it a few months ago. I have always used my UMIK-2 with Dirac without issue. I subsequently reverted to 3.7.3 and continued to have the issue until I switched to my UMIK-1 mic and I was finally able to do a calibration which seems valid. (Not one of my best outcomes however)
The UMIK-2 works fine with REW, HolmImpulse and audiotester.de., but suddenly not with Dirac.
I previously owned an SHD before replacement with the Studio a couple of years ago, and have a lot of experience with doing Dirac calibrations. I never had any problems until a week or so ago when a hardware replacement (Danley DNA SC48) dictated the need for a recalibration after I completed configuring it. (I did a lot of measurements with the UMIK-2 during that process and had no problems with it.)
Dirac provides access to previous versions of Dirac as well as release notes for each version, but I have not found a way to contact anyone there about the problem.
More disturbingly there are reports of this problem on ASR going back to 2019 so it has been plaguing MiniDSP users for some time (also other licenssees). MiniDSP indicated that they were not really able to provide assistance in this matter, and to contact Dirac for assistance. (FWIW - maybe not much.)
Without functional Dirac the Studio is sort of pointless in a sense, although the digital pre-amp functionality is still needed.
And yes the room is not great, I've made a lot of improvements, but serious issues persist.
Last edited:
Yes there is a licensed (to the hardware) Dirac plugin in the SHD Studio which works fine. Unfortunately the calibration tools run on a PC and you do a bunch of measurements in specific patterns using either a MiniDSP UMIK-1 or UMIK-2 mic (required).
Once the measurements are done the software generates an averaged measurement to display and all of the variables they need for phase, amplitude and time correction at the listening position. There is some manipulation of first arrivals claimed as well.
It suggests an initial target curve which you can manipulate or overwrite with your own, once that is done you upload the file to one of the 4 memory slots and listen.
I have to admit that while I am an experienced user I really have no clue about how it actually works. They're not particularly forthcoming about how it works, and I doubt I would understand very much of it if they were.
The UMIK-2 mic runs at 96K and the UMIK-1 at 48K. I think something has changed in Windows, literally for the past 5 and half years I never had significant problems doing the room measurements required for calibration, the last few runs have been nightmarish. The problem actually seems worse now. Based on what I am reading elsewhere this problem has existed for some users for at least 5 months, and there are reports of this same error back to 2019.
Ironically last night I heard that this problem only affects people using windows PC for calibration measurements and processing; Mac users don't have this problem. Is it time for a mac book?
The license will not permit it, only MiniDSP UMIK mics will work with MiniDSP hardware. You launch calibration from the MiniDSP console (DC) and Dirac looks for the SHD, when found you select and it connects to the SHD. (Dirac compensates for the average latency of the microphone and I think also does a loop back to determine the USB latency between the SHD and the PC - internal delays are known and quite consistent I would guess.)
The default mics for REW are UMIK-1 and UMIK-2 these days.
Roon runs on a remote server running linux and I would not want to change that. I have inherited my dad's old laptop with W10 that has not been updated in at least a year, so may try that - making sure to disable automatic updates before connecting it to my WIFI. It is I guess about 8 years old, not sure even what MCU it uses. Might be way too slow.
The default mics for REW are UMIK-1 and UMIK-2 these days.
Roon runs on a remote server running linux and I would not want to change that. I have inherited my dad's old laptop with W10 that has not been updated in at least a year, so may try that - making sure to disable automatic updates before connecting it to my WIFI. It is I guess about 8 years old, not sure even what MCU it uses. Might be way too slow.
It is part of the license agreement I guess, I can understand the logic since the mic codec and USB interface should behave consistently over the population, and each mic has its own accessible calibration file. Removes potential compatibility and misconfiguration issues since it is all handled automatically.
