REW w/software EQ?

[andreh]

Hi All-

I'm trying to figure out a way to analyze my room with corrections I've made in a Mac software application called Pro Tools, but I'm not sure if this is possible:

- For one, I haven't been able to get the Mac version of REW to work properly, and without that all is lost.
- Beyond that obstacle, Pro Tools doesn't let other audio applications have access to the Core Audio Driver while it's loaded, so I wouldn't be able to run REW at the same time anyway. This could be overcome by using a different audio application to design the EQ plugin curves, and then save them as a preset for later use in Pro Tools.
- Even if REW can run its tests while another audio application is running, the software plug-in's applied in those programs won't affect the signal that REW sends so the calibration won't have any affect on REW readings.

Unless I'm missing something, the only way for this to work would be for REW to support a software plugin format such as RTAS or VST, which would allow you to use a software plugin directly from the program, save the changes, and then recall them in your audio application of choice later on.

Thoughts?


Andre

[Wayne A. Pflughaupt]

Hi Andre,

I’m not sure if you’re going to find many people here who are familiar with Pro Tools, since few of us do any live recording. JohnM or Dr. Who might know something.

Regards,
Wayne

[andreh]

Wayne A. Pflughaupt wrote:

Hi Andre,

I’m not sure if you’re going to find many people here who are familiar with Pro Tools, since few of us do any live recording. JohnM or Dr. Who might know something.

Regards,
Wayne

Hi Wayne-

I thought that might be the case, but that I'd throw it out there anyway and see what all these great minds can come up with!

This would be a great solution to a problem that a lot of studio guys face; maybe that'll be incentive for the REW developer to add plug-in support?


Andre

[DrWho]

What OS are you running? Tiger? Panther?

And on a broader scope, what exactly are you trying to achieve?

[andreh]

DrWho wrote:

What OS are you running? Tiger? Panther?

And on a broader scope, what exactly are you trying to achieve?

Hi Mike-

I'm running OSX 10.4.8. My goal is to correct my room with software plug-ins instead of hardware. I can do this now by estimating what kind of EQ curve to apply based on REW analysis, but I can't measure the results since the room corrections aren't applied during the tests.


Thanks,
André

[Sonnie]

Couldn't you simply use another computer for REW? Borrow a PC laptop and do your REW measurements since it's something you'll only have to do in one session. This way it wouldn't interfere with your Mac and Pro Tools.

[andreh]

Sonnie-

I am actually using a Windows laptop for measurement currently; I tried my Mac with no luck.

The problem is that the EQ'd signal isn't being measured, because only signals that go through the application using the EQ plugin are affected.


Andre

[pinduro]

http://software.muzychenko.net/eng/vac.html

Hi Andre,

I'm a little in the same boat as I am trying to use a convolver to do some correction. The above link is for some s/ware that lets you "patch" the output of one program to another...early tests are very encouraging (although the demo is driving me nuts...it actually says "trial" every 5 secs!)

I was able to play measurement sweeps from REW though Audacity and then to the soundcard. Next step is to try REW though a track in Sonar with the SIR convolution plug-in loaded w/ impulses exported from REW treated and with DRC. Should keep me off the streets for a bit!

Good luck,

Mike

[andreh]

pinduro wrote:

http://software.muzychenko.net/eng/vac.html

Hi Andre,

I'm a little in the same boat as I am trying to use a convolver to do some correction. The above link is for some s/ware that lets you "patch" the output of one program to another...early tests are very encouraging (although the demo is driving me nuts...it actually says "trial" every 5 secs!)

I was able to play measurement sweeps from REW though Audacity and then to the soundcard. Next step is to try REW though a track in Sonar with the SIR convolution plug-in loaded w/ impulses exported from REW treated and with DRC. Should keep me off the streets for a bit!

Good luck,

Mike

Hi Mike-

This looks like a potential solution. It'll take several steps for me since my application is on a Mac, but fortunately the EQ plugin I want to use (Waves Linear) is crossplatform - so once I get the settings on the PC I should be able to copy them (manually if necessary) to my Mac with identical results.

One concern in reading this program's specs is that there are "1.2ms of latency per interrupt," which could throw off REW's frequency analysis. My recommendation would be to set the sweeps to the slowest speed possible to minimize this problem; apparently slower sweeps are more accurate anyway.

Thanks for the reference... let us know how your trials go!


André

[DrWho]

The software that you'll be using the EQ with....does it have the ability to play and record audio at the same time? If so, you can manually generate your own sweeps and then convolve the measured sweep into an impulse response that you can load into REW.

The only downside is the only convolving software I know about is PC based. I will do some asking around to see if anyone is doing this on a Mac. Or heck, if this is a one time thing you can send me your measured sweep and I'll send you back the impulse response.

But for what it's worth, I don't think EQ is the best way to go about addressing the acoustical issues in your control room.

[andreh]

DrWho wrote:

The software that you'll be using the EQ with....does it have the ability to play and record audio at the same time? If so, you can manually generate your own sweeps and then convolve the measured sweep into an impulse response that you can load into REW.

It does. Can you go into more detail about this process or provide a link?

DrWho wrote:

The only downside is the only convolving software I know about is PC based. I will do some asking around to see if anyone is doing this on a Mac. Or heck, if this is a one time thing you can send me your measured sweep and I'll send you back the impulse response.

Thanks for the offer! How would I measure the sweeps to include the software EQ?

DrWho wrote:

But for what it's worth, I don't think EQ is the best way to go about addressing the acoustical issues in your control room.

I agree...that's why I have almost 30 bass traps in my control room! I'm doing critical mixing though, and the dimensions of my room (12x16x8) create problems that I haven't been able to eliminate with trapping.


André

[DrWho]

Creating an impulse response manually:

First you need to generate a frequency sweep test file. There are various ways to do this, but the easiest would probably be just downloading something free off the internet. For what it's worth, I think you could even use music for the "test tone" as long as it hits every single frequency.

Load this sound into your program with the EQ enabled. Then just as you do when laying a new track, playback the sweep channel while recording the output of your speakers onto another channel. The waveform of the measured response should look similar to the frequency sweep, but will vary in amplitude and probably have some distortion making it look more jagged.

Then you dump the original test file and the recorded test file into a program that does the convolving. The program will output the impulse response and should be identical to the impulse response REW generates. (In fact, it's the same process).

You can then load this impulse response into REW and it will treat it just like any other REW measurement...basically letting you see the effect of the EQ on your system.

----------------------------

30 bass traps?!? Yikes. What frequencies are you having problems with? Do you have any before and after measurements?

When treating a room (especially a control room), you absolutely want to be looking at the ETC - not the frequency response. You should check out the book "Sound System Engineering" by Davis. I think you'll find the section on small room acoustics extremely helpful.

[andreh]

Thanks for all the great info, Mike. I'll give the convolution process a whirl; it doesn't sound too hard now that you've described it.

To be more specific about my room treatment, some of the traps are actually for mid and high frequency absorption, but because I have them spaced away from the walls they help with lower frequencies as well. They vary in thickness between 2-6 inches, and while most of them are 2'x4' I have several that are 2'x8'. Also, most of them are wrapped rockwool (pressed fiberglass insulation, aka OC703) but I have 4 cutom-made wood panel traps for frequencies lower than rockwool can deal with (50-120Hz).

Sorry, no before and after measurements...if only I'd heard of REW before I started the room!

Thanks for the book recommendation...I'll check it out for sure.


André

[DrWho]

Here's a link to the book:
http://www.soundsystemengineering.com/

The reason I asked about before and after measurements is I wonder if maybe your current treatment hasn't worsened the acoustical situation. It's cool to see someone going the extra mile

What kind of studio are you running?

[andreh]

DrWho wrote:

Here's a link to the book:
http://www.soundsystemengineering.com/

The reason I asked about before and after measurements is I wonder if maybe your current treatment hasn't worsened the acoustical situation. It's cool to see someone going the extra mile

What kind of studio are you running?

Cool! I'll check it out.

It's possible I've harmed my room with so much treatment, but very unlikely. For one thing, its dimensions are 8x12x16, and with an even divisor into all three dimensions it's got some pretty pronounced modes (though not as bad as they would be were all the dimensions to be divisible by one another).

Also, the variety of acoustic traps (and diffusors) I've put up cover a wide range of frequencies, so it's not like there are 30 traps sucking up alll the same range. And if the traps are attacking low frequencies, then you really can't have too many...in a studio envirunment, at least, the only [sometimes] desirable reflections are higher frequencies that can add life to a recording, not low frequencies that can cause modal ringing and mudiness or boominess.

You probably know this Mike, but there's a common misconception about bass traps "eating bass;" some people think if their treated room is bass-shy it may be because of too much bass trapping. However, bass traps don't remove any bass from what's coming directly out of the speakers, only the from bass frequencies being reflected off walls - and it's these reflected frequencies merging with the direct signal at a slightly later time that causes uneven dips and peaks in what we hear.

Now, it is possible to perceive "better" (meaning "preferable") bass reponse with less treatment if some of the peaks or dips are to one's liking, but if your goal is to have as flat a room as possible then the more trapping the better; EQ boosting the low end in a flat room is a better solution than accepting artificially [acoustically] boosted bass, since the latter approach yields a much smaller sweeter spot (and usually the opposite of a sweet spot in many other areas in the room).

Of course, all this trapping is unnecessary if your room is built to acoustically "sound" dimensions in the first place; that's why many pro studios don't have much room treatment up. Theatres are somewhat cubicle, but they're so big that only the lowest of low frequencies have long enough wavelengths to become "modal," and often very frequency-specific traps (that are usually not visible to the audience) are employed to correct these issues.

Nonetheless, I do wish I had measured the room before I treated it just to see what each trap is doing; I'd probably be better armed to address my remaining room issues if I had.


Andre

[DrWho]

Agreed....but not all reflections are "detrimental"

For instance, let's say you've got two reflections that arrive 180 degrees out of phase. If you effectively cancel one of these reflections, then you have removed the "damping" that occurs after the first half cycle, basically resulting in a longer resonance than you had before. I know this is an abstract concept, but I've certainly read about times when adding more actually worsens the issue. I don't mean to imply that this is certainly the case...just offering it up as a possibility.

With before and after measurements, we wouldn't need to conjecture about it - we could just interpret the data Nevertheless, it wouldn't hurt to look at your current data after the treatment.

Btw, is there any chance that you might be able to change the geometry of your room? Namely splaying the side walls and perhaps flush-mounting the mains if that's feasible. Here's a link to a forum dedicated to studio acoustics:
http://www.johnlsayers.com/phpBB2/index.php

[andreh]

DrWho wrote:

Agreed....but not all reflections are "detrimental"

For instance, let's say you've got two reflections that arrive 180 degrees out of phase. If you effectively cancel one of these reflections, then you have removed the "damping" that occurs after the first half cycle, basically resulting in a longer resonance than you had before. I know this is an abstract concept, but I've certainly read about times when adding more actually worsens the issue. I don't mean to imply that this is certainly the case...just offering it up as a possibility.

I see what you're saying, and I think it's both logical and quite possibly occurring in my space. However, wouldn't the phenomenon only create an equal (and therefore desirable) "balancing" effect in one very particular spot in the room? As soon as you moved the smallest amount in one or another direction the wavelengths would cease to be "equal" and would therefore cease to cancel one another out. Not particularly conducive to achieving a wide listening area in a home theatre, though ironically not the worst problem to have in a control room like mine where I'll be only listening from one spot anyway!

DrWho wrote:

With before and after measurements, we wouldn't need to conjecture about it - we could just interpret the data Nevertheless, it wouldn't hurt to look at your current data after the treatment.

I'll post it soon.

DrWho wrote:

Btw, is there any chance that you might be able to change the geometry of your room? Namely splaying the side walls and perhaps flush-mounting the mains if that's feasible. Here's a link to a forum dedicated to studio acoustics:
http://www.johnlsayers.com/phpBB2/index.php

I'm afraid contruction is out of the question (I'm renting), or I'd be all over improving the acoustics that way.

I've been to John Sayers' Forum; it is a great resource that I've learned a lot from. Thanks for the reference!


Andre