• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Setting Levels - Some Experiments


Jun 19, 2020
I posted a comparison between Auddssey's internal Mic setting of levels vs my UMIK-1 with the calibration file loaded in the Audyssey review thread. For me, on two seperate Marantz AVR's with 2 different mics tested on each the levels differ a lot. It's frustrating that the Marantz/Denon give us cheap mics with no corrections. Although I'm not sure how accurate the corrections are for the UMIK but I could hear problems with the default mic.

I've always had issues with not hearing my height channels with Atmos. Turns out with most content there isn't much going on up there anyway. Audioholics had a German installer on who did some experiments with disconnecting all the bed channels and just listening to height for movies. Hardly anything going on. Still I like upmixing and want to make sure my levels are bang on now.

We use omnidirectional mics to set levels with the mics pointed at 90 degrees to the bed channels. But they are pointed straight at the ceiling channels. Not sure if Audssey even attempts to take this into account, but with the shortcomings of the default mic it has bigger issues.

So I did a test of my UMIK with a 90 degree calibration file loaded and took an SPL log with it pointed directly up and then at 90 degree from the height channels (so 90 degrees from ceiling). Turns out there is a difference, it's small but noticeable 0.5db.

The initial plot is the mic pointed up at the ceiling. I then make a noise, which is the spike, and then place the mic 90 degree from the ceiling. It's 0.5db. Small difference, but worth running your heights 0.5db hotter if you can't specifically measure? Attached the raw plots too.

Edit: doesn't seem to be so simple as having an offset. In measuring all my positions and height channels they vary without a simple pattern. But I always need to turn it up by 0.5-1.5db but in one of my positions (1.5m away from center) I had to turn them down. I guess it depends on your room reflections.

Upon doing a listening test of Dolby leaf I immediately notice more of the leaf going on higher up now. That's a clip I've watched dozens of times now, so know it well.

Edit2: The above was all in my smaller office theater that's a 5.1.4 setup with bookshelf's pointing down from the walls on heavily angled brackets. On my main theater - with proper ceiling speakers there was very little difference. That room has more treatments in the form of heavy velvet curtains to help with PJ contrast. So I guess it all depends on your room.

Top Rear Right -0db.png

Top Front Right -0db.png


Last edited:
Top Bottom