The Official Dirac Live Thread: Start w/1,2,3 🙋‍♂️

[jayfabs]

Hey guys couple things I need help with. The first one is during the setup for Dirac I raise my mic till I have a -40db noise floor and the set my speakers to match around -20db except the sub I have at around -17db. When doing measurements I can see on the screen that Dirac does not capture the full signal on certain speakers and I will get a noise warning at the end. The house is dead quiet. Some times it takes me 10 times it get it to take the full measurement. Should I drop my mic gain and raise the master volume?

The second thing is even with a +3db house REW with show a +12 db house curve. I have tried this a couple time and I keep coming up with a huge house curve. This is in my living room with a huge opening. I’m guessing one side has a lower sub level so it’s compensating.

[marcl]

Jan 17, 2024 23:38:40 GMT -5 jayfabs said:

Hey guys couple things I need help with. The first one is during the setup for Dirac I raise my mic till I have a -40db noise floor and the set my speakers to match around -20db except the sub I have at around -17db. When doing measurements I can see on the screen that Dirac does not capture the full signal on certain speakers and I will get a noise warning at the end. The house is dead quiet. Some times it takes me 10 times it get it to take the full measurement. Should I drop my mic gain and raise the master volume?

The second thing is even with a +3db house REW with show a +12 db house curve. I have tried this a couple time and I keep coming up with a huge house curve. This is in my living room with a huge opening. I’m guessing one side has a lower sub level so it’s compensating.

A quick try with a couple things ...

1 - Leave the mic level at 100% and don't touch it.  Don't worry about noise floor for now.
2 - Start the level check for your sub(s) and raise the master gain until the sub level is around -18.
3 - Click the level check for each of the other speakers and let each play for 2-3 seconds ... you will see which speaker has the highest level and which is the lowest.
4 - Adjust all the speakers to match the lowest level speaker ... no need to click the level check, just pull down the sliders and you'll see the levels go down.

Try a calibration and let us know how it goes.  If levels are too high and you get a clipping error, lower the master gain a bit.  If levels are too low and you get an error, raise the master gain a little bit.  Don't touch the mic level.

Regarding REW ... when you say house curve, are you referring to all the channels being +12db in the bass, or just the LFE channel (Output 4 in REW)?  How many subs do you have?

[jayfabs]

Jan 18, 2024 8:47:10 GMT -5 marcl said:

Jan 17, 2024 23:38:40 GMT -5 jayfabs said:

Hey guys couple things I need help with. The first one is during the setup for Dirac I raise my mic till I have a -40db noise floor and the set my speakers to match around -20db except the sub I have at around -17db. When doing measurements I can see on the screen that Dirac does not capture the full signal on certain speakers and I will get a noise warning at the end. The house is dead quiet. Some times it takes me 10 times it get it to take the full measurement. Should I drop my mic gain and raise the master volume?

The second thing is even with a +3db house REW with show a +12 db house curve. I have tried this a couple time and I keep coming up with a huge house curve. This is in my living room with a huge opening. I’m guessing one side has a lower sub level so it’s compensating.

A quick try with a couple things ...

1 - Leave the mic level at 100% and don't touch it.  Don't worry about noise floor for now.
2 - Start the level check for your sub(s) and raise the master gain until the sub level is around -18.
3 - Click the level check for each of the other speakers and let each play for 2-3 seconds ... you will see which speaker has the highest level and which is the lowest.
4 - Adjust all the speakers to match the lowest level speaker ... no need to click the level check, just pull down the sliders and you'll see the levels go down.

Try a calibration and let us know how it goes.  If levels are too high and you get a clipping error, lower the master gain a bit.  If levels are too low and you get an error, raise the master gain a little bit.  Don't touch the mic level.

Regarding REW ... when you say house curve, are you referring to all the channels being +12db in the bass, or just the LFE channel (Output 4 in REW)?  How many subs do you have?

I had my mic gain at +2.7 yesterday. I will lower it to 100% and try again. For my subs I lower the sub trim level to -8db to make sure I don't clip my signal with large house curves and then I adjust my gain to bring it down more. I have all my speakers match -20db using the master slider. The center and subs are the only ones I have to pull down. 

After running dirac I'm back checking my house curve in REW. I'm just doing a sweep from 5hz to 1000hz with subs and mains. For some reason in REW it recognizes my XMC-2 but I can only get sound to play in Java so I just use that. I can turn my mains off and subs off independently so It works for me.  My subs are (2) stereo integrity HS-24's in sealed boxes being driven buy a FP-14K clone amp. I can see that with a 3db house curve in dirac that I have a 12db house curve in REW. Maybe I'm doing my sweeps wrong I don't know. 

[ttocs]

Jan 18, 2024 10:32:48 GMT -5 jayfabs said:

I had my mic gain at +2.7 yesterday.

Setting the Mic Gain higher than 100% uses DIGITAL GAIN and is better to be avoided.

[marcl]

Jan 18, 2024 10:32:48 GMT -5 jayfabs said:

Jan 18, 2024 8:47:10 GMT -5 marcl said:

A quick try with a couple things ...

1 - Leave the mic level at 100% and don't touch it.  Don't worry about noise floor for now.
2 - Start the level check for your sub(s) and raise the master gain until the sub level is around -18.
3 - Click the level check for each of the other speakers and let each play for 2-3 seconds ... you will see which speaker has the highest level and which is the lowest.
4 - Adjust all the speakers to match the lowest level speaker ... no need to click the level check, just pull down the sliders and you'll see the levels go down.

Try a calibration and let us know how it goes.  If levels are too high and you get a clipping error, lower the master gain a bit.  If levels are too low and you get an error, raise the master gain a little bit.  Don't touch the mic level.

Regarding REW ... when you say house curve, are you referring to all the channels being +12db in the bass, or just the LFE channel (Output 4 in REW)?  How many subs do you have?

I had my mic gain at +2.7 yesterday. I will lower it to 100% and try again. For my subs I lower the sub trim level to -8db to make sure I don't clip my signal with large house curves and then I adjust my gain to bring it down more. I have all my speakers match -20db using the master slider. The center and subs are the only ones I have to pull down. 

After running dirac I'm back checking my house curve in REW. I'm just doing a sweep from 5hz to 1000hz with subs and mains. For some reason in REW it recognizes my XMC-2 but I can only get sound to play in Java so I just use that. I can turn my mains off and subs off independently so It works for me.  My subs are (2) stereo integrity HS-24's in sealed boxes being driven buy a FP-14K clone amp. I can see that with a 3db house curve in dirac that I have a 12db house curve in REW. Maybe I'm doing my sweeps wrong I don't know. 

When you sweep Output 4 in REW the processor sees it as the LFE channel and adds +10db as compared to the other channels.  Since you have two subs, there's +3db additional gain for bass management with whatever speakers you have set to small.  is this what you're seeing?

[jayfabs]

Jan 18, 2024 11:06:01 GMT -5 marcl said:

Jan 18, 2024 10:32:48 GMT -5 jayfabs said:

I had my mic gain at +2.7 yesterday. I will lower it to 100% and try again. For my subs I lower the sub trim level to -8db to make sure I don't clip my signal with large house curves and then I adjust my gain to bring it down more. I have all my speakers match -20db using the master slider. The center and subs are the only ones I have to pull down. 

After running dirac I'm back checking my house curve in REW. I'm just doing a sweep from 5hz to 1000hz with subs and mains. For some reason in REW it recognizes my XMC-2 but I can only get sound to play in Java so I just use that. I can turn my mains off and subs off independently so It works for me.  My subs are (2) stereo integrity HS-24's in sealed boxes being driven buy a FP-14K clone amp. I can see that with a 3db house curve in dirac that I have a 12db house curve in REW. Maybe I'm doing my sweeps wrong I don't know. 

When you sweep Output 4 in REW the processor sees it as the LFE channel and adds +10db as compared to the other channels.  Since you have two subs, there's +3db additional gain for bass management with whatever speakers you have set to small.  is this what you're seeing?

I'm not using output 4. I'm using Java L+R. But I guess your math would add up. +10db for LFE +3db since all speakers are small and +3db for house curve. Its hard to tell how much house curve I really have. Sound wise everything is good though. I was just curious where the higher house curve came from. 

[jayfabs]

Jan 18, 2024 11:02:24 GMT -5 ttocs said:

Jan 18, 2024 10:32:48 GMT -5 jayfabs said:

I had my mic gain at +2.7 yesterday.

Setting the Mic Gain higher than 100% uses DIGITAL GAIN and is better to be avoided.

good to know. Thank you. 

[marcl]

Jan 18, 2024 11:12:43 GMT -5 jayfabs said:

Jan 18, 2024 11:06:01 GMT -5 marcl said:

When you sweep Output 4 in REW the processor sees it as the LFE channel and adds +10db as compared to the other channels.  Since you have two subs, there's +3db additional gain for bass management with whatever speakers you have set to small.  is this what you're seeing?

I'm not using output 4. I'm using Java L+R. But I guess your math would add up. +10db for LFE +3db since all speakers are small and +3db for house curve. Its hard to tell how much house curve I really have. Sound wise everything is good though. I was just curious where the higher house curve came from. 

Okay seems like everything is in the ballpark.   ASIO4ALL driver should let you measure 7.1  individually.

If you have trouble maybe this will help   www.dropbox.com/s/51jpnxet3bvew2k/REW%20101%20HTS%20Current%20Version.pdf?dl=1

BTW ... setting levels in Dirac is a challenge because from one system to another it varies considerably.   For example, I leave the sub slider all the way up and set the level to -18 ... then I set all the other speakers to -24 to -27.  When Dirac does the sweeps they play at about 75db ... all the speakers and the subs.   Another forum member with a totally different speaker/amp setup uses the same settings ... and his sweeps play at 95db.  Neither of us gets a clipping or noise floor warning.

[jayfabs]

Jan 18, 2024 11:19:02 GMT -5 marcl said:

Jan 18, 2024 11:12:43 GMT -5 jayfabs said:

I'm not using output 4. I'm using Java L+R. But I guess your math would add up. +10db for LFE +3db since all speakers are small and +3db for house curve. Its hard to tell how much house curve I really have. Sound wise everything is good though. I was just curious where the higher house curve came from. 

Okay seems like everything is in the ballpark.   ASIO4ALL driver should let you measure 7.1  individually.

If you have trouble maybe this will help   www.dropbox.com/s/51jpnxet3bvew2k/REW%20101%20HTS%20Current%20Version.pdf?dl=1

BTW ... setting levels in Dirac is a challenge because from one system to another it varies considerably.   For example, I leave the sub slider all the way up and set the level to -18 ... then I set all the other speakers to -24 to -27.  When Dirac does the sweeps they play at about 75db ... all the speakers and the subs.   Another forum member with a totally different speaker/amp setup uses the same settings ... and his sweeps play at 95db.  Neither of us gets a clipping or noise floor warning.

Originally I could not get any sound out of ASIO4ALL so I stuck with Java since I'm only messing with my subs. 

Does it really matter where you set you levels in Dirac for the individual speakers? I had my subs at -17db I believe and the rest were around 20db. 

[marcl]

Jan 18, 2024 11:31:55 GMT -5 jayfabs said:

Jan 18, 2024 11:19:02 GMT -5 marcl said:

Okay seems like everything is in the ballpark.   ASIO4ALL driver should let you measure 7.1  individually.

If you have trouble maybe this will help   www.dropbox.com/s/51jpnxet3bvew2k/REW%20101%20HTS%20Current%20Version.pdf?dl=1

BTW ... setting levels in Dirac is a challenge because from one system to another it varies considerably.   For example, I leave the sub slider all the way up and set the level to -18 ... then I set all the other speakers to -24 to -27.  When Dirac does the sweeps they play at about 75db ... all the speakers and the subs.   Another forum member with a totally different speaker/amp setup uses the same settings ... and his sweeps play at 95db.  Neither of us gets a clipping or noise floor warning.

Originally I could not get any sound out of ASIO4ALL so I stuck with Java since I'm only messing with my subs. 

Does it really matter where you set you levels in Dirac for the individual speakers? I had my subs at -17db I believe and the rest were around 20db. 

It doesn't matter as long as you are within the range and don't get errors.   Many of us have found it's best to set the subs at least 6-8db higher than the other speakers.  It has an effect on how Dirac sets the playback levels in the processor.

[ttocs]

Jan 18, 2024 11:31:55 GMT -5 jayfabs said:

Does it really matter where you set you levels in Dirac for the individual speakers?

Yes, it does. But this varies by system, so the only way to know for sure how your settings affect the outcome is to experiment and take notes.

After trying many variations of settings, this is what works for my system.
Set Mic Gain to 100%.
Play pink noise in every channel for 3 seconds, then I don't need to play the pink noise anymore.
Without playing noise, adjust the Master Output until my weakest channel is at about -24dB.
Without playing noise, adjust all other channels to match at -24dB except for Subwoofer which I set to -18dB. The 6dB difference in Volume Calibration makes the subwoofer channels play at the same level as the speaker channels in Measurement.

In this example the Left Surround is the weakest channel.

[jayfabs]

Jan 18, 2024 11:48:43 GMT -5 ttocs said:

Jan 18, 2024 11:31:55 GMT -5 jayfabs said:

Does it really matter where you set you levels in Dirac for the individual speakers?

Yes, it does. But this varies by system, so the only way to know for sure how your settings affect the outcome is to experiment and take notes.

After trying many variations of settings, this is what works for my system.
Set Mic Gain to 100%.
Play pink noise in every channel for 3 seconds, then I don't need to play the pink noise anymore.
Without playing noise, adjust the Master Output until my weakest channel is at about -24dB.
Without playing noise, adjust all other channels to match at -24dB except for Subwoofer which I set to -18dB. The 6dB difference in Volume Calibration makes the subwoofer channels play at the same level as the speaker channels in Measurement.

In this example the Left Surround is the weakest channel.

That all makes sense but from a lot of trial and error this last week it seems you need to lower the trim level on the sub channel and adjust your gain as necessary. This is especially important with a minidsp as you will clip it. I’ll still make the sub around -18db and everything else at -24db. Thanks

[jayfabs]

I messed around with everything you guys pointed out but my mic just randomly drops out during measurements. Even during volume calibration you can see the signal randomly drop out. Even in REW checking for spl with the tone generator you can see the mic dropping out. I have changed cables and Dirac was doing with both the Emotiva and umik-1 mics. I think it must be my computer. It’s only a couple years old. I have an older Mac that I was going to try and use but something is wrong and can’t even login. Looks like I need to take it in.

[ttocs]

Jan 18, 2024 23:05:43 GMT -5 jayfabs said:

I messed around with everything you guys pointed out but my mic just randomly drops out during measurements. Even during volume calibration you can see the signal randomly drop out. Even in REW checking for spl with the tone generator you can see the mic dropping out. I have changed cables and Dirac was doing with both the Emotiva and umik-1 mics. I think it must be my computer. It’s only a couple years old. I have an older Mac that I was going to try and use but something is wrong and can’t even login. Looks like I need to take it in.

Are you using an extension cable? Extension cables don't seem to work well, or at all. Also, cables that are too long don't work either.

[jayfabs]

Jan 18, 2024 23:13:21 GMT -5 ttocs said:

Jan 18, 2024 23:05:43 GMT -5 jayfabs said:

I messed around with everything you guys pointed out but my mic just randomly drops out during measurements. Even during volume calibration you can see the signal randomly drop out. Even in REW checking for spl with the tone generator you can see the mic dropping out. I have changed cables and Dirac was doing with both the Emotiva and umik-1 mics. I think it must be my computer. It’s only a couple years old. I have an older Mac that I was going to try and use but something is wrong and can’t even login. Looks like I need to take it in.

Are you using an extension cable? Extension cables don't seem to work well, or at all. Also, cables that are too long don't work either.

I’m using a 6’ cable and I tried the 3’ cable that came with the Umik. I got my Mac working but there seems to be an issues with the mic in Dirac but seems others might be having this issue also. I can’t get it register unless I pump in the gain to +20db

[ttocs]

Jan 19, 2024 0:18:21 GMT -5 jayfabs said:

Jan 18, 2024 23:13:21 GMT -5 ttocs said:

Are you using an extension cable? Extension cables don't seem to work well, or at all. Also, cables that are too long don't work either.

I’m using a 6’ cable and I tried the 3’ cable that came with the Umik. I got my Mac working but there seems to be an issues with the mic in Dirac but seems others might be having this issue also. I can’t get it register unless I pump in the gain to +20db

Those cable lengths should work ok.

I have some ideas, but let's start with the following.

Please share your settings in Volume Calibration, or a screenshot would be even better. What I'm looking for are: Master Output, Mic Gain, and just give me one channel's settings for the the actual Slider position which will be all the way up at 0.0dB or less than that like -2.5dB, and, the value shown for the pink noise level meter.

Be aware that if you have played the pink noise, and then change the Mic Gain, this will affect the levels in the channel sliders and the pink noise will need to be played again to correct or verify.

After I set the Mic to 100%, I adjust the Master Output to a low setting - then start the pink noise and adjust the Master Output to a comfortable level - then play the pink noise on each channel for 3-4 seconds, one right after the next, until all channels have played pink noise, then I stop the noise. This is when I adjust the Master Output for the final time and watch to see which channel is the weakest and set the Master Output for my desired level. Once this is done, I then pull down all the other sliders to match except the subwoofer which will be at a higher level by +6dB than the speaker levels. I don't adjust my settings based on room noise, like, so much higher than the noise floor, etc.

When Dirac is first opened and after selecting the processor, the input device(s) can be seen. If using a UMIK mic, what does it show as the GAIN next to the UMIK-1 mic name? Mine shows -18dB, which is the sensitivity setting of the mic. Also, this is where the calibration file is shown for the mic, as you can see in my example.


One last thing. Change your Speaker Size settings so your system has a minimum amount of speakers while you're testing these things. It save a lot of time and hassle. Right now, because I was checking something for someone else yesterday, my system is setup as 3.1, Left, Center, Right, and one Subwoofer channel. Having more channels just adds frustration and time.

[jayfabs]

Jan 19, 2024 1:16:42 GMT -5 ttocs said:

Jan 19, 2024 0:18:21 GMT -5 jayfabs said:

I’m using a 6’ cable and I tried the 3’ cable that came with the Umik. I got my Mac working but there seems to be an issues with the mic in Dirac but seems others might be having this issue also. I can’t get it register unless I pump in the gain to +20db

Those cable lengths should work ok.

I have some ideas, but let's start with the following.

Please share your settings in Volume Calibration, or a screenshot would be even better. What I'm looking for are: Master Output, Mic Gain, and just give me one channel's settings for the the actual Slider position which will be all the way up at 0.0dB or less than that like -2.5dB, and, the value shown for the pink noise level meter.

Be aware that if you have played the pink noise, and then change the Mic Gain, this will affect the levels in the channel sliders and the pink noise will need to be played again to correct or verify.

After I set the Mic to 100%, I adjust the Master Output to a low setting - then start the pink noise and adjust the Master Output to a comfortable level - then play the pink noise on each channel for 3-4 seconds, one right after the next, until all channels have played pink noise, then I stop the noise. This is when I adjust the Master Output for the final time and watch to see which channel is the weakest and set the Master Output for my desired level. Once this is done, I then pull down all the other sliders to match except the subwoofer which will be at a higher level by +6dB than the speaker levels. I don't adjust my settings based on room noise, like, so much higher than the noise floor, etc.

When Dirac is first opened and after selecting the processor, the input device(s) can be seen. If using a UMIK mic, what does it show as the GAIN next to the UMIK-1 mic name? Mine shows -18dB, which is the sensitivity setting of the mic. Also, this is where the calibration file is shown for the mic, as you can see in my example.
View Attachment

One last thing. Change your Speaker Size settings so your system has a minimum amount of speakers while you're testing these things. It save a lot of time and hassle. Right now, because I was checking something for someone else yesterday, my system is setup as 3.1, Left, Center, Right, and one Subwoofer channel. Having more channels just adds frustration and time.

I can try to get a screen shot later today. Everything is exactly as you say. My system is a 5.1 and I only have those speakers turned on in my XMC-2. I can get through 9 positions of measurements but it takes me an hour because the mic randomly does not record. I tried raising my master level and still it will just randomly stop recording. I uninstalled Dirac and reinstalled. no fix. I'm going to go back into windows tonight to makes sure my audio enhancement is not checked off in the properties for the umik-1. I do have the 90 degree calibration installed and it is -18db. 

[ttocs]

Jan 19, 2024 9:08:08 GMT -5 jayfabs said:

I can try to get a screen shot later today. Everything is exactly as you say. My system is a 5.1 and I only have those speakers turned on in my XMC-2. I can get through 9 positions of measurements but it takes me an hour because the mic randomly does not record. I tried raising my master level and still it will just randomly stop recording. I uninstalled Dirac and reinstalled. no fix. I'm going to go back into windows tonight to makes sure my audio enhancement is not checked off in the properties for the umik-1. I do have the 90 degree calibration installed and it is -18db.

Ok, good info, thanks.

When I am trying to figure things out I just do a 1 mic position Dirac run. Then it's rinse and repeat to go through any steps I might think of to solve the issue. This way, if successful, I can get through all the steps including loading the filter and then check the results with REW. So I highly suggest doing just the one mic position at the main listening position and see what happens.

It's the random part of it that makes me wonder what's going on here, especially since you're getting this issue with different operating systems. I've not heard of a mic being kicked out of operation before. I've seen people do a Mic Drop, but not in this context :) . I'm wondering if it actually is a problem in the mic cable, or maybe the mic itself?

If you have a SPL meter, or a SPL app on your cell phone, please measure the SPL when the Measurement sweeps in Dirac are playing. This would be good info to have. Or, do you need to plug your ears with your fingers when the sweeps play - meaning they are loud? or are they really easy to listen to when playing, or somewhere in between?

[marcl]

Jan 19, 2024 9:08:08 GMT -5 jayfabs said:

Jan 19, 2024 1:16:42 GMT -5 ttocs said:

Those cable lengths should work ok.

I have some ideas, but let's start with the following.

Please share your settings in Volume Calibration, or a screenshot would be even better. What I'm looking for are: Master Output, Mic Gain, and just give me one channel's settings for the the actual Slider position which will be all the way up at 0.0dB or less than that like -2.5dB, and, the value shown for the pink noise level meter.

Be aware that if you have played the pink noise, and then change the Mic Gain, this will affect the levels in the channel sliders and the pink noise will need to be played again to correct or verify.

After I set the Mic to 100%, I adjust the Master Output to a low setting - then start the pink noise and adjust the Master Output to a comfortable level - then play the pink noise on each channel for 3-4 seconds, one right after the next, until all channels have played pink noise, then I stop the noise. This is when I adjust the Master Output for the final time and watch to see which channel is the weakest and set the Master Output for my desired level. Once this is done, I then pull down all the other sliders to match except the subwoofer which will be at a higher level by +6dB than the speaker levels. I don't adjust my settings based on room noise, like, so much higher than the noise floor, etc.

When Dirac is first opened and after selecting the processor, the input device(s) can be seen. If using a UMIK mic, what does it show as the GAIN next to the UMIK-1 mic name? Mine shows -18dB, which is the sensitivity setting of the mic. Also, this is where the calibration file is shown for the mic, as you can see in my example.
View Attachment

One last thing. Change your Speaker Size settings so your system has a minimum amount of speakers while you're testing these things. It save a lot of time and hassle. Right now, because I was checking something for someone else yesterday, my system is setup as 3.1, Left, Center, Right, and one Subwoofer channel. Having more channels just adds frustration and time.

I can try to get a screen shot later today. Everything is exactly as you say. My system is a 5.1 and I only have those speakers turned on in my XMC-2. I can get through 9 positions of measurements but it takes me an hour because the mic randomly does not record. I tried raising my master level and still it will just randomly stop recording. I uninstalled Dirac and reinstalled. no fix. I'm going to go back into windows tonight to makes sure my audio enhancement is not checked off in the properties for the umik-1. I do have the 90 degree calibration installed and it is -18db. 

Keep going with ttocs guidance ..... but here's something else that might help if all else fails ...

"Users experiencing recurring imprecise measurements on Windows PCs now have the possibility to opt-in for using the microphones in Exclusive Mode"

www.dirac.com/live/faq/helpdesk/dirac-live-beta-releases/

This is a Beta version of Dirac.  I'm using it now so it's stable.  I haven't done the Exclusive Mode thing yet.   If you decide to try it, be sure to uninstall your current Dirac version before installing 3.9.1

[jayfabs]

Jan 19, 2024 9:24:54 GMT -5 ttocs said:

Jan 19, 2024 9:08:08 GMT -5 jayfabs said:

I can try to get a screen shot later today. Everything is exactly as you say. My system is a 5.1 and I only have those speakers turned on in my XMC-2. I can get through 9 positions of measurements but it takes me an hour because the mic randomly does not record. I tried raising my master level and still it will just randomly stop recording. I uninstalled Dirac and reinstalled. no fix. I'm going to go back into windows tonight to makes sure my audio enhancement is not checked off in the properties for the umik-1. I do have the 90 degree calibration installed and it is -18db.

Ok, good info, thanks.

When I am trying to figure things out I just do a 1 mic position Dirac run. Then it's rinse and repeat to go through any steps I might think of to solve the issue. This way, if successful, I can get through all the steps including loading the filter and then check the results with REW. So I highly suggest doing just the one mic position at the main listening position and see what happens.

It's the random part of it that makes me wonder what's going on here, especially since you're getting this issue with different operating systems. I've not heard of a mic being kicked out of operation before. I've seen people do a Mic Drop, but not in this context . I'm wondering if it actually is a problem in the mic cable, or maybe the mic itself?

If you have a SPL meter, or a SPL app on your cell phone, please measure the SPL when the Measurement sweeps in Dirac are playing. This would be good info to have. Or, do you need to plug your ears with your fingers when the sweeps play - meaning they are loud? or are they really easy to listen to when playing, or somewhere in between?

The problem is that sometimes its not till the third or 4th position that the mic will drop out. When it drops its only drops for several seconds and then comes back. Even during volume calibration you will see the mic randomly drop out. I have tried different cables with no success. I even tried and different master volumes from -40db to -30db which is loud and still random drop outs. I believe its a computer issue. The sound does not drop at anytime during the measurements the mic just stops recording. I was having the same issue with the emotiva mic a couple months ago when I first tried Dirac on this computer. That's why I bought the Umik-1. I tried my Mac yesterday but I have to raise the mic gain to +20db for it to register.  I found others on the minidsp forums having the same issue.  So that's an other issue.