TOPIC:

For speaker measurements, use the normal (on axis) calibration file and point the mic at the speaker. You will/may also need to apply some gating. This app note may help:

www.minidsp.com/applications/acoustic-me...speaker-measurements

now we are more confused than before!

In general:

0° or 360° means pointing to the Top and 90° horizontal.
Like this

please is there one who knows it 100%?

The normal file is for the mike pointing at stuff
The 90 degree file is for the mike pointing up to the ceiling

If you are measuring speakers: normal file and point the mike at the speaker Left or Right on axis about 1 meter away.
If you are measuring your room: 90 degree file and point the mike to the ceiling
If you are measuring your surround system: be lazy and use the 90 degree file and point the mike to the ceiling. This will save you from moving it around to point to each individual speaker

The mike is omnidirectional until around 2 KHz. You can point it any where you like and use any file up till 2KHz.

You look like you are measuring speakers to flatten response or optimise an active crossover or whatever - just use the normal file and use the guide as John suggested.

The calibration file which gets downloaded by default when you enter the mic serial number is the pointed at ceiling version. Use that to find out which is which.

kathampy wrote:

The calibration file which gets downloaded by default when you enter the mic serial number is the pointed at ceiling version. Use that to find out which is which.

Hello, I think perhaps some further clarification is needed here.

1. It used to be that there was one calibration file. That is the on-axis file, which compensates the microphone response when pointed directly at the sound source. It has a name like "7004423.txt".

2. Now, there are two files downloaded from the UMIK-1 page. One is the same as above. The other is the 90-degree file, which has a name like "7004423_90deg.txt." The 90 degree file is the "pointed at ceiling version," which (as the UMIK-1 page says) is "for use with the nanoAVR DL" i.e. when measuring a multichannel system. The 90 degree file is auto-generated from the on-axis file.

www.minidsp.com/products/acoustic-measurement/umik-1

Thank you John,

that makes it 100% clear!

I just recalibrated with Dirac using the 90 degrees file and the treble is clearly improved! I'm able to hear high pitched "screeching" type noises which were not audible before. When I first ordered the Umik-1 there was no 90 degree file and I was using the ordinary file for Dirac. I presumed that was the one meant to be used for Dirac since even the manual mentioned pointing the mic at the ceiling.

I've generated graphs showing the two files.
plot.ly/~Locutus/17
plot.ly/~Locutus/15

Hi again, you'll find the graphs easier to read if you use a log scale horizontally.

Another way to do it is to import them into REW, then it's easy to overlay them.

Sorry for bringing this out of the grave, but shouldn't it be better to make couple measurements (5-6) at the listening position and than 5-6 around the room, but not at the extremes (near the wall or corners), average it all and than use those results to apply it to miniDSP?

Hi, it depends on what you want to do. If not using Dirac Live, I have found the best approach is to take the measurements around the listening area to understand how the room "behaves". (I'm assuming subwoofer frequency range.) Then you will know whether you should EQ to an average or whether it's adequate to EQ to a single point and make minor corrections later (much quicker).

PS if looking at a large listening area and willing to go multiple subs, perhaps take a look at multi-sub optimizer (maybe not at first but you can work up to it) - andyc.diy-audio-engineering.org/mso/html/

john.reekie wrote:

PS if looking at a large listening area and willing to go multiple subs, perhaps take a look at multi-sub optimizer (maybe not at first but you can work up to it) - andyc.diy-audio-engineering.org/mso/html/

This sounds rather interesting. Basically, I have two subwoofers (Rythmik F8) which I use to complement Harbeth P3s.
Main speakers (P3s) are connected directly to the amp, but are HPF (fixed) 24db/octave @100Hz at line level.
F8s are connected to the line level out of the amp with miniDSP in between.

I am actually, quite happy with my current setup and EQ I am using, but always wondered if there's a method which would make it better than I do it right now.