Preferred Frequency Response of Your Room

Greetings. I was able to do some extensive acoustic analysis over the weekend. Using the UMIK-1 microphone and REW software I was able to make adjustments to speaker placement, set crossover on the sub and add just a touch of shelving EQ.

Final result is a frequency response of the room that is +/- 10dB across the sound spectrum.

I'm interested in other's preferences when it comes to setting up a room. Do you prefer to bump the lower frequencies up? Do you like a soft roll-off on the high end? Do you like to bump the high frequencies up?

Just curious.

 

Sounds like you like things flat and even.

Personally, I have not analyzed my room. It's a big "T" shaped basement with carpet and drapes on both sides.

I don't currently use an EQ but have found, through trial and error, a wonderful cartridge>>turntable>>phono pre combination that is also fairly flat in presentation.

There is plenty of sparkle, pronounced upper mids, low mid warmth and detailed bottom end.

So I guess I like a flat, dynamic frequency response too.

The only thing I mess with is the volume level of my 10" front firing sub, so as not to disturb the family upstairs. Usually going from 12 o'clock to 1 o'clock on records that need it.

 

I was kind of waiting to see where this would go. I don't think there are many members that have measured their rooms to see what they like. I measured my room years ago and set it where I like. I like it where it sounds flat to me which brings up an interesting point. Sounding flat is relative to your volume level. I listen at around 85db peaks, so that may tell you something about how my room is set. Photobucket ate my old pictures from REW.

 

My listening room is only 12' X 10', so not much room to experiment with various placements, but I don't need any hardware or software telling me what sounds best...I have my ears for that.

 

I'm too afraid to measure my room. Ignorance is bliss.

 

That's about where I am at, but I'm saving for the right room treatment. I have a natural room node at 45 Hz, and that sends a ripple right up into the mid-bass. But yeah, I can get it all within a 10 db range. My mic pics up to about 15 Khz where there is a roll-off. I don't pay a whole lot of attention to the high end because it's beyond my measurement equipment and killing the bass nodes is really what I'm after.

 

I agree that sounding flat is relative to the volume level. As you know, increasing loudness flattens out the response on the low and high end. That said, when taking acoustic measurements, loudness doesn't matter other than it can be too low to get an accurate measurement.

 

I to have a nasty room node at 45Hz.

 

I should also note that my T30 decay time is below 0.5 sec which I was happy to see.

 

I have "treated" some of the corners and behind the stereo rack with large pillows.
And keeping the doors open keeps things from getting boomy on the outskirts of the room.

I have a small recording setup on an opposing wall and I use a graphic EQ between the mixer and the amp.
I've dropped the the 400hz and the 2.5khz just a skoch to keep playback from sounding muddy.
I do mix louder than I listen to records. But I have much bigger speakers in my stereo system and that makes the most difference.

The key for me is to get my recordings to sound good on the stereo system!

 

I would be interested to know what smoothing was used in your REW frequency graphs.

I had to change my room around so took the opportunity to take some measurements as I added room treatment. The room is 420cm x 386cm x 240cm high.

This is a frequency response from REW from 30-500Hz with no smoothing (really both are with no smoothing):



Green is before any room treatment; purple is with lots of bass traps added.

Here are the two waterfall plots for the same measurements.





Some of the 40Hz 'ringing' is external noise from traffic. Obviously there is still a major issue at 50-70Hz which I am looking to solve - it is not significantly affected by any speaker/ear positioning. The main points to note are the generally smoother frequency response and reduction in resonance times.

In the past, as well as room treatment I've used EQ to finish off aiming for a flat response but at the moment I like what I'm hearing with this current arrangement and no EQ. Here is a full frequency response after treatment at 1/24 smoothing:



The strange thing is it naturally has a so called 'BBC dip'.

 

Looks like you still have a huge bass suck out even after room treatement -- huge in terms of width -- like the full octave from 40 Hz to 80 Hz -- and in terms of depth, like - 15 dBs. Did you try moving the speakers and listening positions around? Bass nulls like that can be hard to deal with, most room treatments don't really help much below 80 Hz, and that's really where big room problems often are, thought it looks like you were able to smooth some of the peakiness in the upper part of that. Still have that hump in the 150-200 Hz half octave just below middle C, and a lot of peakiness in the 200-600 Hz range, then that big suck out from 2kHz to 4kHz -- I'm guessing that's in the vicinity of a crossover to a tweeter and maybe partially related to listening position relative to tweeter axis, but just a guess.

What kinds of treatment did you use? You should be able to deal with the 200-600 Hz peakiness relatively easily I would think. And again, did you play around with moving the speakers and listening position at all?

 

What's the easiest way to post images to this board?

 

Besides the low frequency suck out, I don't think the other one around 2kHz to 4kHz is all that bad. It's said the human error can only differentiate loudness starting at 3dB. I want to say there is about a 5-8 dB drop there from the average of the whole range.

That dip in the low frequencies is significant for sure.

 

I shoot the room and try to get it reasonably flat, no huge suckouts or peaks, which is not that easy to do unless have a cost no object budget and total control of parameter choices....so I live with some peakiness that I'd prefer not to live with. In my set up there's also some HF roll off that frankly bugs me and I've tried EQ my way around it, which works OK but was never fully satisfying to me. Just try to get the set up close to flat and live with the imperfections.

 

What I've found is that the best frequency response is not always best for imaging and soundstage. All you can do is know what compromises you are living with.

 

For me, in general I end up with a slightly tilted response, higher in the low end, lower in the upper end, but the tilt is moderate/mild. Basically a flat line though, as much as I can achieve it without eq. Also prefer a small increase towards the upper end from ~12K up.

+/- within 20-20Khz, I'd have to run it again, not sure where I'm at but likely around the OP's +/-10 dB or a little less. My listening room is a 16 by 24 garage, and speaker placement is governed not by sound but by where I can best fit it. It's far from ideal and worst than most for sound room optimization I'm certain. Hope to change that and some other things in the future.

Mic and RTA is a tool to assist with, not to govern what you will hear or prefer to hear.

 

I know that 3 dB threshold is a common rule of thumb. But I wouldn't take it as gospel depending on the source, the frequencies in question, and the listener's acuity and sensitivity, etc., one can definitely hear differences of less than 3 dB. In audio production people often make eq and loudness adjustments of fewer than 3 dB, and these can certainly make differences, like I said, depending. Trying to get things flatter than that in a home listening environment with a real-world budget and home furnishing needs, etc., is mostly not worth worrying about because it ain't gonna happen. But if there's some octave or octave plus wide kind of consistent gap in the response, like in that 2k-6k area in that chart, that's something I might look for the cause of because it's not just a lot of upper frequency reflection peakiness that you're gonna have in a home listening room, but a broad enough trough that it has to have some particular cause that may be able to be remedied. Like I said, my first guess might be something crossover and tweeter axis related, but I dunno.

 

Room Analysis by Ryan Spivey, on Flickr

Mine is actually better than I originally posted. Perhaps +/- 5-8 dB across the spectrum except for that nasty stuff around 45Hz. I do have a wicked room resonance around 45Hz and I also think that is about the frequency my Elac B6's ports are tuned to (if that's the correct terminology).

This is 1/24th scaling.

 

I've never had the opportunity.

 

if one has really +-10db from 20hz to 20khz, I think this is respectable.
I shoot for +-5 db from 20hz to 20khz and have achieved it with very careful speaker placement and lots of room treatment.
your suckout between 40hz and 60hz is room related and must be dealt with, not with room treatment but with proper speaker and listening position placement

 

Here is mine- sounds excellent. Feel free to copy it

 

With the more ear sensitive frequencies- 2Khz to 5Khz- differences are easier to detect than other frequency ranges and 1/2 to 1 db hearing sensitivity is common. This region can make or break the sound of a system.

 

^ how's your smoothing set?

 

1/6 octave