5.19 Release Setup/Preferences Questions

MellowDios

Registered
Thread Starter
Joined
Sep 6, 2017
Messages
5
Location
Whitby (near Toronto), Canada
Wow! It's been ten years since I did the room correction on my previous studio, using Doug Plumb's RplusD and REW 3.29 - 4.11. REW has blossomed unbelievably. Kind of intimidating actually. Dauntless, nonetheless. I leap into the breach.

My room is 17'7" x 11'1" x 7'3". as close to Fibonacci as I could build it, considering space limitations.

For initial tests (no room treatment) and speaker placement, I have REW 5.19 running on a Macbook Pro (USB, single analog port for input and output depending on what's plugged in. I have an EMM-6 (with cal file from Dayton (0 degrees. no sensitivity data) plugged into a simple Blue Icicle USB inline preamp (volume control, quoted as 0dB b to -40dB in the pithy manual). The Macbook's stereo analog out (headphone jack, level controllable in the Sound panel on the Mac) runs through unbalanced lines to the inputs on my Bryston power amplifier, to my Mains. I've yet to install my single sub.

This setup seemed to function, although I had considerable difficulty getting a useable level in REW for the first while. Here are a couple of screen dumps showing what fields are available in Soundcard Preferences and the setup in Mic/Meter Preferences.

REW519_SoundcardPrefs1.jpg


REW519_MicMeterPrefs1.jpg


There doesn't seem to be any way to make a Soundcard cal file, unless I'm missing it. As you can see, I had to reduce the Sweep Level drastically to get a reading (similar to what is suggested in Help for using a USB mic). Even though I was able to calibrate the SPL meter successfully, initially I could not get a level high enough to perform a measurement. After fiddling with the levels a few times I did successfully record a measurement, but I'm unsure of the accuracy of the results.

I've attached two mdat files. One full range and one 0-200 Hz. I'd appreciate it very much if someone could tell me if these graphs look reasonable. Also, an explanation of how to interpret the phase data, or how to record it in a way that is more relevant, would be great too.

It never ceases to amaze me how music and audio can be so frustrating and so much fun at the same time. Thanks in advance.
 

Attachments

  • LR_0-22k-NoTreatment.mdat
    1.8 MB · Views: 7
  • LR_0-200-NoTreatment.mdat
    153.7 KB · Views: 5
Last edited:

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
The same file was attached twice. The measurement looks fine, though the decay time is very long. As the space isn't a stadium I guess it has pretty solid walls.

The phase data doesn't tell you much in isolation, and doesn't really tell you anything that helps with treatment decisions.

Any reason you didn't make a selection in the input and output boxes?

Souncard cal doesn't make sense when using a USB mic.
 

MellowDios

Registered
Thread Starter
Joined
Sep 6, 2017
Messages
5
Location
Whitby (near Toronto), Canada
Thank you very much for your response, John.

I replaced the duplicate file. :redgrin:

Yes, the room has pretty solid walls. Double 1/2" fire code drywall suspended on RISC clips and hat channel all round. Studs de-coupled using rubber mounts. Subfloor between joists above also covered in double drywall and sealed. All flanking paths above ceiling and behind walls sealed. Cork floor floated on double 1/2" pine subfloor which sits on 1" of 703 (concrete underneath). All seams sealed. Doors (one closet, one entrance) lead lined per Rod Gervais' Super Door design. Two walls (front and left side) are backed behind the studs by the poured slab exterior basement walls. The other two are backed by double 1/2" drywall, again all seams sealed.

Undoubtedly, more information than you require or want. But yes, reasonably solid walls.

Where are you looking to see the decay time? Did you generate a waterfall? What do you mean by "very long" exactly? Should I worry?

I don't do this often, so could really use some handholding. :help:

The phase data is important when adjusting speaker placement. Correct?

I actually did make a selection in the Input and Output boxes. The choices for Output are Default Output and Headphones, both of which work equally well. The choices for Input are Default Input and Microphone (which is the internal mic). So non-choices really, but there you have it.

Admittedly, I did not understand the relevance of the Timing Reference and Loopback Input selections to my setup. I simply chose the speaker options that seemed logical. Is a loopback between the USB input and Headphone out possible? Can I use it as a timing reference? Or should I use an acoustic timing reference? Or none at all?

Yes, I had read that no Soundcard cal is needed with a USB mic, however, technically this setup has a preamp with a mic input and volume control, so I wasn't sure. I guess the levels I got from Check Levels reveal that REW is seeing this setup as a USB mic. Got it. I think.

Your support is deeply appreciated, John. Looking forward to your further comments.
 
Last edited:

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
Where are you looking to see the decay time? Did you generate a waterfall? What do you mean by "very long" exactly? Should I worry?
I was looking at the impulse response, but the RT60 plot shows the decay times most easily. Right now it is over 3 seconds at worst and it would ideally be below 500 ms, so there's work to do in treatment to damp those room modes and reflections.

rt60.jpg

The phase data is important when adjusting speaker placement. Correct?
No, it can be helpful when trying to align a sub to a main speaker or aligning drive units within a speaker, but for treating the space you can ignore phase.

I actually did make a selection in the Input and Output boxes. The choices for Output are Default Output and Headphones, both of which work equally well. The choices for Input are Default Input and Microphone (which is the internal mic). So non-choices really, but there you have it.
Selecting the named input/output rather than the default gives REW a bit more info, mainly the volume setting, so best to make those selections if they are available.

Is a loopback between the USB input and Headphone out possible?
No

Or should I use an acoustic timing reference? Or none at all?
You could use a timing reference if/when you get around to using REW to time align speakers, but it isn't needed for assessing treatment effects.
 

MellowDios

Registered
Thread Starter
Joined
Sep 6, 2017
Messages
5
Location
Whitby (near Toronto), Canada
It's been a couple of weeks since my last post. I've been doing my due diligence vis-a-vis speaker placement, how to identify and treat SBIR, and the various standards of house curves in mixing rooms and whether or not to use one.

Regarding the house curve question, there's an area of REW that is confusing me, and sorry to say, the help file isn't really ... er, helping. Maybe I'm just thick in this instance.

In Preferences/Equalizer some things about the Target Defaults area aren't clear to me.

How does the software apply the settings in this area? Do they skew the measurement results, do they somehow affect the display on the equalizer page, or are they simply a reference to be used while making EQ settings?

Do I understand correctly that they are not to be included when a house curve is being used? They don't seem to appear on the EQ page like a house curve does.

Here are my speaker system specs:
Mains: Dynaudio BM6 Passive, LF extension 43 Hz
Subwoofer: Dynaudio BM12S, LF normal 27 Hz, extension 18 Hz

Am I correct in thinking the options Full Range, Bass Limited, Subwoofer None are to be changed depending on the speaker(s) or combinatio of speakers being measured? If using a house curve, do I choose "None"?

Also, how do I use the LF Rise setting? A typical setting might be 100 Hz and 20 Hz, correct? So if I want the LF curve to peak at +2dB, the slope should be .6 dB/octave?

Similarly, if I want HF to be -6 dB from 1000 Hz to 20k Hz, the slope should be 1.2 dB/octave?

And these slopes are to be set at 0.0 dB/octave if I'm using house curve?

My math is a little weak here. A brief explanation or a link to some background reading would be greatly appreciated.

Thanks as always for your help.
 
Last edited:

John Mulcahy

REW Author
Joined
Apr 3, 2017
Messages
7,212
In Preferences/Equalizer some things about the Target Defaults area aren't clear to me.

How does the software apply the settings in this area? Do they skew the measurement results, do they somehow affect the display on the equalizer page, or are they simply a reference to be used while making EQ settings?
The default values in the equalizer preferences are used as the initial Target Settings for new measurements. They affect the Target trace which is shown in the EQ window and used when REW adjusts filter settings to match the target. They have no effect on the measurement.

Do I understand correctly that they are not to be included when a house curve is being used? They don't seem to appear on the EQ page like a house curve does.
When you say 'they' do you mean the defaults or the Target Settings? The target settings can be used together with a house curve if desired and, like the house curve data, affect the Target trace.

Am I correct in thinking the options Full Range, Bass Limited, Subwoofer None are to be changed depending on the speaker(s) or combinatio of speakers being measured?
Yes, the choice allows a different selection of parameters to represent the corresponding bass management filters that affect the chosen speaker type.

If using a house curve, do I choose "None"?
Depends whether you are accounting for the effects of bass management crossover filters in your house curve in some way.

Also, how do I use the LF Rise setting? A typical setting might be 100 Hz and 20 H, correct? So if I want the LF curve to peak at +2dB, the slope should be .6 dB/octave?
The target rises at the selected rate below the LF rise start frequency and stops rising below the LF rise end frequency. To have a 2 dB rise over the span from 100 down to 20 Hz the slope would need to be 0.8 (giving 1.9 dB) or 0.9 (giving 2.1 dB), but the idea is to just look at the target trace and adjust to get the result you want rather than calculate a setting.

Similarly, if I want HF to be -6 dB from 1000 Hz to 20k Hz, the slope should be 1.2 dB/octave?
1.4 dB/octave gives -6 dB drop from 1k to 20k. Again, look at the Target trace to see what effect the settings are having.

And these slopes are to be set at 0.0 dB/octave if I'm using house curve?
Yes, if your house curve already includes whatever slopes you want.
 

MellowDios

Registered
Thread Starter
Joined
Sep 6, 2017
Messages
5
Location
Whitby (near Toronto), Canada
Thanks for your explanations and your patience, John.

I was expecting to see two different curves (target and house) and was actually seeing the combination of the two (I thought there was an anomaly in the software). All work and no play makes a boy thick in the head.

Now that I understand how the target and house curves work, independantly and together, I can see the changes in "Target" in the EQ panel as I make them. Excellent!

I use a DEQ2496 as my room EQ and a DCX2496 for bass management (a term which I'm not particularly fond of for reasons I hope are obvious). The DCX is a decent crossover (at least for my needs) and can communicate with the DEQ via AES/EBU XLR ports, nice and clean.

Of course, none of this helps me decide what kind of curve would enable me to hear reference mixes as they are intended to be heard, or allow all my mixes to translate well to other systems. So far, I haven't been able to find any kind of "universal curve" that pro audio studios subscribe to. Maybe I'm dreaming pie in the sky.

Thanks once again, for your kind support.
 
Top Bottom