Tools for room measurement

[terryj]

I use the deqx and it's software, and in addition use REW. Already had the mic (from the deqx) so all that was needed for REW was a mic pre, can pick them up for fifty bucks or so.

Re using a cal file, I guess it depends on what you want. When for example I use the mic to correct with the deqx, I don't even bother with loading the cal file. After all the deqx makes an inverse of the measurement, so the mic inaccuracy is automatically included in that.

Dunno about the Tact software, but the deqx software is a bit 'clunky' when I want to make quick measurements for adjusting levels and whatnot (gotta go thru wizrds and stuff) so I just whip the mic out of the deqx and throw it into the mixer then use REW. Best of both worlds.

I really do recommend REW.

This is one area that lot of audiophiles simply don't know what they are missing. I think that is changing tho. How long a go was it that this sort of stuff was hideously expensive and beyond the reach of mere mortals like myself??? Probably not even ten years ago. And for a computer idiot like me, to be able to 'master' REW certainly says something.

It is not the stone age anymore, and once you get a real grip on what the room is doing to your system, well for me at least, things like cables and all the other minutiae *most* audiophiles concentrate on become insignificant.

 

[audioguy]

For you TacT users, I suggest you get a better mic.
The LinearX M31 mic is plug-compatible with the TacT (9V phantom power) and comes with a TacT calibration file.

Very interesting. A couple of questions:

Do you have any measurement comparisons of the TacT mic versus this one that you could post?

You said this mic was plug and play so it plugs into the mic input on the back of the TacT (no additional preamp required), load the mic file and you are good to go?

Have you been able to detect any significant audible differences when using this mic versus the TacT?

 

[Kal Rubinson]

Re using a cal file, I guess it depends on what you want. When for example I use the mic to correct with the deqx, I don't even bother with loading the cal file. After all the deqx makes an inverse of the measurement, so the mic inaccuracy is automatically included in that.

I do not follow that logic. If what you say is so, then the compensation will be included in the filters even without the mic in the loop.

 

[RUR]

You said this mic was plug and play so it plugs into the mic input on the back of the TacT (no additional preamp required), load the mic file and you are good to go?

I'm not Dan but, yes, that's exactly how it works. I can't compare with the stock mic, since I've only used the LinearX.

 

[Bruce B]

We use an Earthworks M50 mic going into our Digital Audio Denmark AX24 for mic pre and A-D conversion. For software we use either SMAART or Spectrafoo.


Regards,

 

[audioguy]

I've tried to join that group for 3 months and have never been responded to. It says it is not moderated but when you sign up, it says I will have to wait to be approved!!! HELP!!

 

[steve williams]

Like REG's forum, membership is moderated.

I recently signed off as co-moderated of the forum.

Try to join again and indicate that you own TacT gear.

Unless you've had some type of run in with Anthony (Aberdeen/MauiMods), you should get approved.

I signed up under general interest and was approved the same day.

 

[LesAuber]

I'm currently playing around with a Behringer ECM8000 thru a little Rolls MP-13 preamp using TrueRTA running into a windoze laptop. The 3rd octive version of TrueRTA comes off a bit course but the results looked much like I'd expected by ear. Also use a Radio Shack sound level meter for speaker level matching etc. No idea where this fits in with some of the pro gear mentioned here though I know Steve has me beat hands down for portability.

Sounds like I need to try out REW and RPlusD at least.

 

[LesAuber]

Anyone have thoughts on CARA software? www.cara.de

 

[Kal Rubinson]

Anyone have thoughts on CARA software? www.cara.de

It is good for what it does but it does not measure anything. It only predicts results based on the information you program into it.

 

[terryj]

I do not follow that logic. If what you say is so, then the compensation will be included in the filters even without the mic in the loop.

That compensation will always be in the loop no matter what we do???

Let's say my mic is +5db at 10k. So I get a cal file which corrects it by adding -5 db at 10k, then make my measurements and corrections with the cal'd mic. (Let's also say the speaker is +2db at 10k).The speaker is corrected by the software (by adding-2b at 10k) to be flat.

Now if I don't use a cal file, the speaker will 'measure' +7db at 10 k (+5 from the mic and +2 from the speaker), the software will correct that by adding -7 db at 10k (-5 for the mic and -2 for the speaker). In other words the mic has automatically been corrected anyway.

I can do that either by adding a cal file, or just by creating the inverse of the measurement.

When we make a cal file, of the mic alone, ultimately it is simply and inverse of itself anyway no??

IF that logic is true...and I think it is...then I can make the further assumption that by correcting the mic by creating the inverse I will have 'infinitely' more points corrected than by (what I presume is something like notepad file) a mic cal file (???)

The only time I'd need to load the cal file is when I wanted to show you my speaker response, or my in room response. Then of course I don't want +5 db at 10 k to show.

Unless I have it completely arse backwards of course!!!



Hmm, I am just about to embark on a round of measurements. I will try and see if I can chase this down and post results. You have got me wondering.

I'll have to try and find a spectrum of the FR where there is something in the cal file operating (have no idea where my mic has been corrected), measure and correct without a cal file, then load and measure with the cal file, compare the results and see if they do indeed at the end of the day measure the same.

Will let you know.

I'm currently playing around with a Behringer ECM8000 thru a little Rolls MP-13 preamp using TrueRTA running into a windoze laptop. The 3rd octive version of TrueRTA comes off a bit course but the results looked much like I'd expected by ear. Also use a Radio Shack sound level meter for speaker level matching etc. No idea where this fits in with some of the pro gear mentioned here though I know Steve has me beat hands down for portability.

Sounds like I need to try out REW and RPlusD at least.

Try REW definitely. The other good thing about it is the forum, any questions then just get in and ask, very helpful.

 

[LesAuber]

Thanks Kal, I understand the no measurements. Predicting results without hefting heavy bits around can be rather handy.

Will do on the REW, Terry. Hard to argue with the price if nothing else. Thanks.

 

[terryj]

Ok, real quick measurement.

All done in room (I'm too lazy to lug them outside, or indeed to put them in the middle of the room or on a stand! Plus, I made them out of concrete so not that keen to do any of that...yet!)

Measurement and calibration done WITHOUT a cal file for the mic.

Then I measured the result without a cal file, and with a cal file loaded, and overlayed them. Just left the mic in position, real quick just one sweep.

There is a slight discrepancy up around 8 k that can be seen, whether or not that is within room measurement tolerance or not I don't know, ie an actual difference or measurement variability given the circumstances??

Still, pretty darn close. (???)

But NOT the same it must be admitted.

EDIT don't worry under 200, that is where I am high passing them. That is where the big difference is, prob a result of changing conditions in the room, which makes me 'less worried' about the slight discrepancy up high, could be just different conditions between measurements

 

[Kal Rubinson]

Let's say my mic is +5db at 10k. So I get a cal file which corrects it by adding -5 db at 10k, then make my measurements and corrections with the cal'd mic. (Let's also say the speaker is +2db at 10k).The speaker is corrected by the software (by adding-2b at 10k) to be flat.

Correct. Only the -2dB is added and necessary.

Now if I don't use a cal file, the speaker will 'measure' +7db at 10 k (+5 from the mic and +2 from the speaker), the software will correct that by adding -7 db at 10k (-5 for the mic and -2 for the speaker). In other words the mic has automatically been corrected anyway.

Nope. You need to add only -2dB for correction of room/speaker. Adding -7dB is wrong. The -5dB is a correction for the measurement, not a correction for the system.

Besides, how can -2dB correction and a -7dB correction both be correct?

 

[Kal Rubinson]

Still, pretty darn close. (???)

But NOT the same it must be admitted.

Exactly. Most corrections are small with a good mic.

 

[terryj]

Ok, thought experiment number three hundred and fifty nine for me....

Let's assume the *perfect* speaker..we know it does not exist, but nonetheless we finally have access to one...it measures +0.0000/-0.0000 db from 1Hz to 100,000Hz.

If we measure and correct with the *perfect* microphone, nothing will be added or subtracted from the speakers response.

Now, let's measure and correct with a non perfect microphone, then what will be superimposed upon the speakers response will be an inverse of the imperfect microphone. Which, I assume if we have the perfect speaker we really would not want eh?

If that imperfect mic had a *perfect* cal file, then what would be added or subtracted from the speakers response would, again, be nothing.

Moral?? Use as perfect a cal file as you can! Preferably one with a good phase response too one would assume.

(psst, that terry j fellow, I have seen him around on some other audio forums,...he's an idiot so my advice would be basically, ignore anything he has to say!!)

Cool, thanks Cal..oops, Kal haha.

 

[JonFo]

I use several measurement setups:

For general room acoustics and system validation:

R+D (AKA ETF) from Acoustisoft for measurement and analysis. Best impulse response display.

REW - additional measurement, and much better-looking waterfall plots.

Coupled with M-Audio Firewire 410 interface box and a set of DBX, Naint and Audyssey calibrated Pro mics + mic stands.

PC / Preamp Audyssey Pro

Audyssey 3.x software on Laptop + Audyssey Pro kit (mic, mic preamp, stand).

iPhone

on a 3G (as noted before, best internal mic) a copy of StudioSixDigital AudioTools. Ordering their mic interface in a few weeks, so I have a highly portable measurement system for visits to friends.



I try and balance the subjective impressions have of system performance with objective measurements and absolutely depend on these measurements to fine-tune the placement of treatments and gear.
The hard part is the years of learning about acoustics and psycho-acoustics to actually derive appropriate conclusions from the measurements.

IMHO any serious audiophile needs a decent measurement system. Much in the same way any decent cabinet maker needs a good tape measure.

 

[amirm]

Just a quick follow up on microphones. One of the guys in my company bought 4 Behringer mics with calibration files and found them to vary a lot! So much so that he returned them all and bought an expensive one that is supposed to be much better calibrated. Don't have the details of his testing other than his observation after the fact.

Has anyone else done consistency tests like this to verify how well these things are calibrated?

 

[kareface]

Just a quick follow up on microphones. One of the guys in my company bought 4 Behringer mics with calibration files and found them to vary a lot! So much so that he returned them all and bought an expensive one that is supposed to be much better calibrated. Don't have the details of his testing other than his observation after the fact.

Has anyone else done consistency tests like this to verify how well these things are calibrated?

When you say with calibration files, are you talking from the manufacturer, or from a lab? What range was he testing? Once you get too low, there will always be a large degree of variability. How did he eliminate external factors?

 

[DonH50]

I have found cheaper "measurement" mics vary 6 dB or more across the audio band. I know some manufacturers use the same capsule as various other mics in their lines and they just run the curves, include the plots, and call it a measurement mic. Earthworks and B&K measurement mics I have used tend to fall within a much tighter range, <3 dB and often <1 dB across the band, over a range of samples. Of course, even the cheap Earthworks mic I have lists for ~$600, out of sight for many audiophiles. You can also buy calibrated pairs that are tweaked at the factory to be matched to within 1 dB or less ($$$).

Still kicking myself for selling my B&K that had phase as well as amplitude curves...

HTH - Don