REW vs SpectraLAB

[krunk]

Hi, I use these 2 software packages to get a good respons from my Adire Tumult subwoofer in my room. I also have the Behringer FBQ2496 and the Behringer EMC8000 mic.

To my optinion, the disadvantage of REW is that you can only make 1 measurement and use this one for creating all your filters.
I've noticed that the predicted respons is not exactly the same as the final real life respons.
I would like to correct peak by peak, in stead of all at once.

SpectraLAB is also a good way to measure your room. With this package and manually entering filters, I get a flatter respons than with REW.

A few questions:
1. With SpectraLAB I use pink noise, not a sweep that REW uses. Is there a difference?

2. Does REW uses a C-curve correction when measuring the room?
So is the target a C-corrected curve, or a flat curve?

3. In REW, can you start measuring at a higher frequency than 2 Hz? Let's say 10 or 15 Hz?

Regards,
Bart

[Wayne A. Pflughaupt]

What is SpectraLAB? I can’t find anything on it with Google. From the way you describe it, I assume it uses real-time analysis? The nice thing about REW is that you don’t have to endure the test signal running the whole time you are tweaking filters.

1. With SpectraLAB I use pink noise, not a sweep that REW uses. Is there a difference?

Pink noise is a more random (read unstable) signal. REW uses a sine-wave signal, I think (at least that's what it sounds like to me). So yes, they are different.

2. Does REW uses a C-curve correction when measuring the room?
So is the target a C-corrected curve, or a flat curve?

The program uses a compensation file to convert a C-weighted SPL meter’s mic output to flat.

3. In REW, can you start measuring at a higher frequency than 2 Hz? Let's say 10 or 15 Hz?

Actually, it starts at 0 Hz. As far as I can tell, you can’t change the lower limit, just the upper.

Regards,
Wayne

[Sonnie]

I think it's SpectraPlus... and if it is, it ain't nearly as functional and robust as REW.

I would suggest that if your measured response is not at least fairly close to the predicted response, then you may not be entering the filters correctly or your mic may not be stationary between measurements... or something else it going on.

[krunk]

This is some more info about SpectaLAB.
http://www.pmgrp.com/prod01.htm
It's one of the four versions that is available. It's also the most complete one and quite solid software.

Anyways,

First point
I understand that using pink noise can give interference at other frequencies, a sweep is just 1 frequency. So you may be right at that.

Second point.
I use a mic, not a SPL meter.
But I understand that in that case, no correction is applied. Through?

Third point.
Do you have the option (somewere in the software or registry) to change the start frequency? Otherwise, this may be something for the next version?

I would suggest that if your measured response is not at least fairly close to the predicted response, then you may not be entering the filters correctly or your mic may not be stationary between measurements... or something else it going on.

What I've done is moving the sub around in the room for a obtimal location. It seems that turning the sub 180 degrees, improves the respons enourmously. Now I have a flatter point to start from. I've noticed that the predicted respons is better now.

The only thing that's left is phase respons. This could be good info to get your subwoofer alligned with the main speakers. Applying filters also change the phase, so some more info on this could be handy.

Thanks for your answers so far.

Bart

[Wayne A. Pflughaupt]

krunk wrote:

This is some more info about SpectaLAB.
http://www.pmgrp.com/prod01.htm
It's one of the four versions that is available. It's also the most complete one and quite solid software.

I like REW’s price better.

First point
I understand that using pink noise can give interference at other frequencies, a sweep is just 1 frequency. So you may be right at that.

I forgot to mention, using pink noise with a stand-alone RTA, the readings jump around a lot, making precise EQ difficult. That’s why I prefer a sine-wave signal like REW has. However, this might mot be as much of a problem with a software-based RTA, as I expect it will let you take a snapshot.

Second point.
I use a mic, not a SPL meter.
But I understand that in that case, no correction is applied. Through?

You have to disable the C-weighting if you use something other than a SPL meter for your measuring. But yes, that’s essentially correct.

Third point.
Do you have the option (somewere in the software or registry) to change the start frequency?

As they say on those lawyer TV shows, “Asked and answered.” There’s no good reason I can think of to have the start frequency sweep of a sub adjustable. I suppose you have one? End frequency, sure, but start frequency?

The only thing that's left is phase respons. This could be good info to get your subwoofer alligned with the main speakers. Applying filters also change the phase, so some more info on this could be handy.

I think this is more of a problem with analog equalizers than digital ones like the BFD we typically use for subs, but I’m admittedly a bit fuzzy on that. JohnM would probably know. In any event, a filter will only change phase at and around the frequency that’s being adjusted, so trying to adjust your sub’s alignment based on that sounds like an exercise in extreme frustration to me. Phase as it relates to equalizer adjustments is one of those subjects where there seems to be more theories and speculations floating around than factual, scientific (read “reproducible”) information.

Regards,
Wayne

[krunk]

I forgot to mention, using pink noise with a stand-alone RTA, the readings jump around a lot, making precise EQ difficult. That’s why I prefer a sine-wave signal like REW has. However, this might mot be as much of a problem with a software-based RTA, as I expect it will let you take a snapshot.

That's the main difference. With SpectraLAB you make real time changes to your respons and you don't use a snapshot.
The advantage is that you don't have to work with a predicted respons.
But if the predicted respons is the same as the real respons, there is no issue.

As they say on those lawyer TV shows, “Asked and answered.” There’s no good reason I can think of to have the start frequency sweep of a sub adjustable. I suppose you have one? End frequency, sure, but start frequency?

Why would you start at 0 Hz? You can't change the respons, and how accurate is the measurement below 10 Hz? Most of the time the soundcard and mic is not liniair any more.
10 Hz is a better startingpoint if you ask me. (I understand that the software doens't provide this, that's why I asked if it was possible to change the value on a different level like the registry)
Also, a flexible starting point could be handy if you only want to measure the frequencies around the crossover frequency. You don't need a full sweep for that.

I think this is more of a problem with analog equalizers than digital ones like the BFD we typically use for subs, but I’m admittedly a bit fuzzy on that. JohnM would probably know. In any event, a filter will only change phase at and around the frequency that’s being adjusted, so trying to adjust your sub’s alignment based on that sounds like an exercise in extreme frustration to me. Phase as it relates to equalizer adjustments is one of those subjects where there seems to be more theories and speculations floating around than factual, scientific (read “reproducible”) information.

I thought that a digital EQ didn't have this issue. But in the review of the Behringer, I saw that the phase does shift. Individual I have a good respons for my fronts and sub, but together there is a small dip. I'll have a look at some other speakerdesign software that I've got, to see if I can get this point clear.

Thanks for you help so far. I'm just giving some feedback, please don't take this in a negative way.

[Sonnie]

Why would you start at 0 Hz? You can't change the respons, and how accurate is the measurement below 10 Hz? Most of the time the soundcard and mic is not liniair any more.
10 Hz is a better startingpoint if you ask me. (I understand that the software doens't provide this, that's why I asked if it was possible to change the value on a different level like the registry)
Also, a flexible starting point could be handy if you only want to measure the frequencies around the crossover frequency. You don't need a full sweep for that.

The measurement below 10Hz is as accurate as your mic is accurate... and there are mics accurate below 10Hz.

I'm not sure why this is such a big deal. If you only want to see from 10Hz on up, simply set your graph axis limits from 10Hz on up to however far you want. It's as if anything below 10Hz wasn't measured.

You are only talking about a few seconds of extra sweep... it's not hurting a thing...

[Wayne A. Pflughaupt]

Why would you start at 0 Hz? You can't change the respons, and how accurate is the measurement below 10 Hz? Most of the time the soundcard and mic is not liniair any more.

Which has anything to do with the capabilities of REW. As you note, the limiting factors, if any, will be external. The soundcard calibration – at least for my Soundblaster – makes it flat down to 10 Hz, and only 2 dB down at 3 Hz. As Sonnie said, the limitation ends up being the mic, or more likely other electronics, like the receiver itself, for instance. More and more people are custom-building high-performance subwoofers that are extending below 10 Hz, so the need for measuring capability below that point is real. Check this chart from Rodny Alvarez’s IB Makeover thread:



Personally I wish he had extended his chart below 10 Hz!

The sweep is already programmed to extend an octave above your cut-off point, just to make sure you get a complete reading. No reason not to start “too low” for the same reason!

Also, a flexible starting point could be handy if you only want to measure the frequencies around the crossover frequency. You don't need a full sweep for that.

A full 0-200 Hz sweep displays a response chart in less than 10 seconds. Are you really in that big of a hurry that you need it done faster?

Bottom line, there’s just no good reason to have an adjustable starting point for the sweep. None of the reasons I’m seeing here I would consider good reasons. (No offense I hope – you did ask for feedback! )

I thought that a digital EQ didn't have this issue. But in the review of the Behringer, I saw that the phase does shift. Individual I have a good respons for my fronts and sub, but together there is a small dip.

Do you have variable phase on your sub? Or a distance setting on your receiver? If not – fortunately a “small dip” typically is only a small problem.

Regards,
Wayne

[krunk]

This is my respons. Tumult in a closed box, no boost and no smothing.
Quite flat, only a dip around the 25Hz region.

Personally I wish he had extended his chart below 10 Hz!

Most subs don't reach 10Hz, this is only room gain. Probably his respons dropped also below this point.

A full 0-200 Hz sweep displays a response chart in less than 10 seconds. Are you really in that big of a hurry that you need it done faster?

How many and with wich length do you make a sweep?

Bottom line, there’s just no good reason to have an adjustable starting point for the sweep. None of the reasons I’m seeing here I would consider good reasons. (No offense I hope – you did ask for feedback! )

Point taken.

Do you have variable phase on your sub? Or a distance setting on your receiver? If not – fortunately a “small dip” typically is only a small problem.

Yesterday, I did measure the subwoofers phase. There are some phase shifts when applying filters, but they are not major and only small banded.

I can set the distance of the sub on my reveiver, but I don't have a phase shift on my sub to change to overall phase. Whatever I do, I always have a small dip at the crossover point. But I'll do some more testing for this.

Bart

[PeteD]

krunk wrote:

Most subs don't reach 10Hz, this is only room gain. Probably his respons dropped also below this point.

It is true that most sub don't reach 10 Hz, but many IB's do...

[Sonnie]

Yep... most IB and LLT subs will reach 10Hz.... some lower than that... and we have a lot of members with IB and LLT subs. Sure some of the response may be room gain, but the sub has to produce a signal for the room to have an effect on it.

[Wayne A. Pflughaupt]

Sonnie wrote:

Yep... most IB and LLT subs will reach 10Hz.... some lower than that... and we have a lot of members with IB and LLT subs. Sure some of the response may be room gain, but the sub has to produce a signal for the room to have an effect on it.

Yup. Room gain only boosts the level of a bass signal. It won’t produce something from nothing!

Regards,
Wayne A. Pflughaupt