At what point did you stop REW measuring and buy treatments?

[hilde45]

Thanks to everyone for reading and responding to my questions so far. I'm still learning about REW and deeply appreciate how much it's done to help me grapple with the numerical and invisible factors at work in my listening space.

My question here is, perhaps, a naive one, but I need to ask it because I feel stuck. Stuck measuring, over and over, finding a million variables to check upon, etc. I need to move to a next step or just stop the process.

QUESTION: At what point did you realize that your room setup -- listening position, speaker location, and room orientation (long wall, short wall, etc.) were as good as they were going to get and it was time to invest in treatments?

To a large degree, a lot of my measuring was me just learning how to use the equipment, create stable experimental conditions, and eliminate unnecessary variables. But now I'm getting regular and repeatable scans and so I think I'm not tainting my own data (pats self on back, ironically). I think I've found the curves which have pushed nulls as far up the spectrum as possible.

Is it time to just take my best measurements which also sound good and present them to a treatment company, such as GIK, RealTraps, et al.? How do you know when your room is as good as it's going to get?

I'm attaching an image of what seems like my best data -- SPL, Waterfall, Spectrogram.
I'm also attaching what wound up on my short list for listening -- in other words, the scans which (to my eye) seemed best and worth examining via critical listening.
If anyone takes a look at the .mdat file I'm interested in what you see.

Thank you!

 

[Sonnie]

I don't really know that you are looking for treatments, unless you have some serious reflections you are dealing with. You probably won't notice much in the way of a response change if you add acoustic panels, although reflections can be tamed in many cases.

You will notice a greater improvement with speaker placement and Audiolense or Dirac Live... which will really smooth out your response. I think you may be able to work with Audiolense easier without having to add more equipment... depending on your source. Treatments won't fix the more serious issues you have.

 

[hilde45]

Thanks, Sonnie. Good advice. When you say the "more serious issues" I have, did you see something in my data or are you suggesting a general principle which would probably apply to my data, whatever it is?

 

[Sonnie]

Those large peaks and dips are pretty serious... and difficult to treat with acoustic panels.

 

[hilde45]

I see. I need to think about this some more. Thank you for your reply.

 

[DanDan]

On a quick look, your RT60 is not bad at all for an untreated room. Most of your measurements show a dip somewhere between 100-200. Have you identified what is causing that? I suspect Floor Bounce or Front Wall BIR.

 

[hilde45]

Hi DanDan -- thanks for chiming in. I'm racking my brain about that dip. I've moved LP and Spks to many many positions in the room, and while that dip moves a bit between 100-200, and sometimes a bit above but typically not, I cannot affect it. I DO have very low ceilings.

I will say that my setup sounds very nice, but I know with dips like these I am probably missing notes in the music.

 

[DanDan]

You are welcome. Can you tell me more about your speakers? Type? Subs? Xo frequency? I would fire up the RTA zoomed to 30-250, Pink Noise. Move the speakers towards or away from the Front Wall and observe the change in frequency of the dip. Also try moving the mic towards a spot on the floor half way to the speakers. Again let's see how the frequency changes.

 

[hilde45]

They are Salk SS 6M. I don't know the crossover. I've kept my sub off. I've not tried the RTA process yet. New thing to try for me, I'll see if I can manage it. Thank you!!!

 

[DanDan]

You may have to play around with the RTA, or the Spectrum, settings in order to get a useful visual. But it is a much faster way of working vs sweeps. Lovely looking speakers. Note LF is reversible, symmetrical. You could place a single speaker at your normal listening position, and move the mic towards the wall and towards the floor spot..... .

 

[hilde45]

Good ideas! I will need to play around. I just tried the RTA and it gave me such a fast readout I couldn't really figure out what I was looking at as I moved the mic. Maybe there's a video somewhere. I will also consult the REW manual again. I'm clearly playing with toys I don't fully understand. ;-)

 

[hilde45]

Sonnie,

You said, "You will notice a greater improvement with speaker placement and Audiolense or Dirac Live... which will really smooth out your response."

I have heard that DSP can help with peaks, but not nulls. Since I have a couple big problems with nulls, do you still think DSP is a good path forward? Or is the fact I quoted erroneous in some way?

 

[Sonnie]

Sonnie,

You said, "You will notice a greater improvement with speaker placement and Audiolense or Dirac Live... which will really smooth out your response."

I have heard that DSP can help with peaks, but not nulls. Since I have a couple big problems with nulls, do you still think DSP is a good path forward? Or is the fact I quoted erroneous in some way?

It might not deal with super massive nulls down in the ultra-low frequencies, but most of the time you can deal with those by moving the subs around, placing them in the corners, doubling or tripling subs, etc. As for the mid and upper frequencies, the dips are not so much of an issue. I have peaks and dips in my response, but no issues smoothing it out, and have zero issues with clipping or distortion... and I have my sub bass boosted a bit as well, although the SVS ultras can handle it with ease. Not everyone likes this much bass... but I do... and this is strictly for movies. My music response is not quite as boosted in the lowend.

 

[hilde45]

Thanks. The null in the data I posted is centered at 140 and 216, so I wasn't sure whether the comment I heard applied to those nulls or not. It sounds like you're saying that DSP could address them, but they also might not be that significant.

 

[Sonnie]

DSP won't do drastic adjustments in most cases from what I have experienced... but if you look at your post DSP response with say 1/6th octave smoothing, which is probably a finer resolution that what we can hear in most cases, it will look very smooth. The unprocessed dips will likely be very narrow if you view the response unsmoothed post DSP.

 

[hilde45]

Thanks, Sonnie! I just clicked on your Cedar Creek Cinema link and watched the video. What popped into my mind while I watched it was that scene in the old comedy Wayne's World where they're bowing and scraping, yelling, "We're not worthy! We're not worthy!" What a wonderful room and setup! Kudos!

 

[Sonnie]

lol... thanks for the comp!

I need to update that video. Maybe that can get done this weekend. It's not drastic, but I have new amps, new rack, new recliner, removed some recliners... added a couple more subs.

 

[Todd Anderson]

lol... thanks for the comp!

I need to update that video. Maybe that can get done this weekend. It's not drastic, but I have new amps, new rack, new recliner, removed some recliners... added a couple more subs.

That room is definitely something to be proud of! Killer!

 

[billrobbo]

Out of curiosity, what are your room dimensions?

 

[Red Tape Music]

To come back to your original question, isn’t this an iterative process? Measure - Treat - measure again etc etc?

 

[MediumRare]

Did you say you’ve kept your sub off? If so, turn it on! Nothing matters until you have your subs and mains set up properly. Having a sub in a location different from the mains can also even out the nulls and peaks significantly, as long as the crossover and phase are adjusted properly.

 

[w_sizer]

Good ideas! I will need to play around. I just tried the RTA and it gave me such a fast readout I couldn't really figure out what I was looking at as I moved the mic. Maybe there's a video somewhere. I will also consult the REW manual again. I'm clearly playing with toys I don't fully understand. ;-)

You may want to check out Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms, by Dr. Floyd Toole (if you haven’t already). It has been very helpful to me, and I think every audiophile should have it in his library. Dr. Toole addresses a variety of strategies for dealing with room issues, including EQ, acoustic treatments, and adjusting furniture and speaker position. I’ve been involved in the same process as you for almost a year now, and I’m finding that a combination of the above strategies is showing real promise. I’ll know for sure after my GIK soffit traps for the corners arrive in about a month.

 

[linearphase]

You are welcome. Can you tell me more about your speakers? Type? Subs? Xo frequency? I would fire up the RTA zoomed to 30-250, Pink Noise. Move the speakers towards or away from the Front Wall and observe the change in frequency of the dip. Also try moving the mic towards a spot on the floor half way to the speakers. Again let's see how the frequency changes.

Agreed that more data is needed. In addition to DanDan's request:

Dimensions of room
Position of speakers relative to all boundaries
LP " "
Room Construction

Optionally Any installed room treatment- where and what.

From the files you sent the mid bass dips appear to be non minimum phase and while you can try to EQ you will likely produce worse sound. They are likely caused by floor and /or ceiling bounce. Given the fact there is very strong reflection at 2.04 mS and you said your ceiling is low I would suspect this is a ceiling reflection. If your ceiling is 6.5' that is then almost assured. To correct the above measurements you can not really treat the floor much but you can the ceiling. Though if the ceiling is really that low it might be problem.

The 40 Hz dips are likely caused by the front wall and room modes combining. You can not absorb low frequencies below 100Hz very easily unless you use (semi) tuned absorbers like limp mass membranes. These are expensive if commercially bought but if you like DIY then they can be built into the house structure as they often have to be big, like tens of sq ft.
However, if you are using sub these don't need to matter, just place the sub closer to the front wall AFTER you have optimized the main only for smoothest frequency response above crossover and the best imaging.

RT60 appears to be good for HT , if that is your application, but arguably bit too dead for 2 channel. More low frequency absorption may be helpful but again this is tricky stuff and requires experimentation.

Usually you can produce satisfactory results with judicious EQ but to get the best available sound room treatment is often required. Unfortunately, most treatment is done without the aid of guided iterative measurements so it usually sub optimal.

When doing any measurements i prefer to find the problems using Psychoacoustic smoothing. To solve the problems 1/6 octave smoothing is usually the best.

When do you stop? When you are satisfied with the sound and/ or tired of measurement. If you enjoy measurement and the improved results and knowledge it gives you then carry on!

 

[hilde45]

Thank you, Linear phase. I'll upload a picture/diagram in case it helps. CAVEAT: one picture/diagram is of a closed rectangle. My room, as the another picture I will post should indicate, is not a closed rectangle, but has an entrance hall on one side and a hallway leading off the other way.

No treatment yet.
Construction: two walls (a long and a short wall along the right side) are brick, two others are drywall. The ceiling is drywall.

I'm pretty happy with the sound, overall, and am getting tired of measuring, but I've had help in the process from someone who's done this before, so I don't feel like I'm flailing.

 

[billrobbo]

Plugging those dimensions into Bob Gold's room calculator shows it isn't too bad. http://www.bobgolds.com/Mode/RoomModes.htm