REW Graphs of my Room

[ErVikingo]

I spent some time this weekend with REW (thanks to a member for his help in getting me to understand REW!); moved bass towers, changed toe on Mid/High towers, changed servo settings (see footnote) and crossover pots (on Mid/High), tried different seating positions. I then asked REW to model filters/EQ and graphed the predicted. Makes me wonder if I should try adding a MiniDSP to see how it feels to my ears.

I adjusted settings to get closer to a flatter line at 75dB and found the bass presence was too much for my taste so went back to my "by ear" settings for low frequencies. Optimizing crossover settings for mid/high frequencies left to right was a great gain! I must check caps (and driver) on the mid of the right channel where i can't get closer than 2dB from left channel.

I'm a total newb to REW and have for the past 4 decades relied on my ears to set up my listening spaces.

Thoughts?

Footnote:

Speakers are Infinity IRS Beta;
- Separate 4x12 woofer towers with Threshold Amps; signal passes through a Servo Circuit and Crossover
- Mid/High towers with LEMIM, EMIM, EMIT and SEMIT drivers direct feed from PreAmp to Manley tube amps

Room (28' L x 14'W x 11' H) has traps on all corners, abfussers on side walls and ceiling (hanging 12" from ceiling). Diffusers on rear wall

EDIT to re-insert graphs

 

[staticV3]

Your way of attaching images does not work. The hyperlinks don't open correctly.

 

[ErVikingo]

Your way of attaching images does not work. The hyperlinks don't open correctly.

Strange I can see them on my screen. Let me try again. Thanks for the feedback!

 

[staticV3]

Strange I can see them on my screen. Let me try again. Thanks for the feedback!

Now they're there!

 

[dualazmak]

Sorry, this post is not intending to specifically respond to your point, but quite general considerations and recommendations.

As a general thought and as a general discussion, measured "flat" SPL would not always give the best subjective hearing sensation/satisfaction for you.

Many other factors are involved for your best/better listening enjoyment, such as room acoustics (reflection, standing wave, negative-positive overlap of reflection and standing wave, etc.), time alignment among all the SP drivers, transient behavior matching/mismatching between SP drivers (especially sub-woofer to your woofer).

It would be also critically important, at least as I have leaned through my long and intensive audio tuning journey, you (we) should not rely on single measurement platform/software; we need to have at least two (or three) independent measurement platforms/tools, and it would be better to separate signal input/excitation procedures and the room air-sound recording/analysis process, hopefully using two independent PCs.

REW is really great tool/platform for audio measurement and tuning; I too often use REW, for example objective room acoustic observation by REW's wavelet 3D color representation which is powerful in analyzing total delay features/characteristics as well as standing wave observation.

I really would like to have, however, my own "another" primitive but reliable accurate fully-validated measurement and tuning methods/tools especially for precision time alignment (somewhat identical to phase matching) all over the SP drivers measured and tuned at my listening position by actual room air-sound recording/analysis.

And in such another method/tool, any black-box type (for you) analysis/tuning should be fully eliminated; the method should be well understandable for yourself, and you need to have several "validation/proof" procedures for reliability accuracy reproducibility (and limitation) of the method.

Just for example, for msec-level precision time alignment between sub-woofer and woofer, the intentionally forced 15 msec delay of sub-woofer sound against woofer by DSP should be accurately and repoducibly "measured as 15 msec delay" by the method hopefully in 0.1 msec precision.

Exactly the same for 0.1 msec-level time alignment between woofer and midrange; the intentionally forced 1.5 msec delay of woofer sound against midrange should be accurately and repoducibly "measured as 1.5 msec delay" by the method hopefully in 0.01 msec precision.

In this regard, if you would be interested, please refer to my summary post #520 here on my project thread. My recent post here (#240 there) would be also your reference.

This attitude/policy would be (should be) also applied for the method of SPL measurement; the rapid sine sweep and psychoacoustic smoothing not always give you accurate/enough SPL information of room air-sound at your listening position. Sometimes it has too much smoothing and/or too mach statistical deviations/fluctuations. This is why I again would like to have my another primitive but accurate reliable reproducible almost-not-statistically-deviating/fluctuating method for SPL measurement by analyzing "very rich amount of recorded room air sound data" at my listening position.

In this regard, my summary post #404 here would be of your interest and reference.

Finally, just only for your reference, my present best tuned SPL at listening position with my DSP-based multichannel multi-SP-driver multi-amplifier stereo audio setup in my room acoustic environment is like this (which does not visualize the established 0.1-msec-precision time alignment all over the SP drivers, though);

 

[ErVikingo]

Sorry, this post is not intending to specifically respond to your point, but quite general considerations and recommendations.

As a general thought and as a general discussion, measured "flat" SPL would not always give best subjective hearing sensation/satisfaction for you.


In this regard, if you would be interested, please refer to my summary post #520 here on my project thread. My recent post here (#240 there) would be also your reference.

This attitude/policy would be (should be) also applied for the method of SPL measurement; the rapid sine sweep and psychoacoustic smoothing not always give you accurate/enough SPL information of room air-sound at your listening position. Sometimes it has too much smoothing and/or too mach statistical deviations/fluctuations. This is why I again would like to have my another primitive but accurate reliable reproducible almost-not-statistically-deviating/fluctuating measurement method for SPL by analyzing "very rich amount of recorded room air sound data" at my listening position.

In this regard, my summary post #404 here would be of your interest and reference.

Your post is most appreciated! and your thought process follows what my ears are telling me. I will re-read your post and also access the links to your other posts.

THANKS!

 

[dualazmak]

You would please find the Hyperlink Index for my project thread here and here.

If you would be seriously interested in the precision tone burst signals I prepared for my "time alignment measurement/tuning" and "transient behavior assessment/tuning", please simply PM me writing your wish.

I assume my "audio sampler/reference playlist" summarized here and here, would be also of your interest and reference.

 

[ErVikingo]

Busy afternoon playing with REW. No DSP no EQ at all.