Spectrum analyzer?


Can someone suggest some options for acquiring some form of real time frequency analysis to help with speaker/room interactions?

I don't know if it makes sense to buy or rent hardware, go the laptop/software route, etc.

Thanks in advance.
madfloyd
Depends on your needs. In general, RTA is not the best tool for speaker/room interactions. The starting place for RTA is TrueRTA.

However, the best combination of capability and price is RoomEQ Wizard which is a freeware program. You will need a mic, a preamp/soundcard and a PC, though.

There are other options if you have specific requirements.

Kal
The spectrum analyser in the Behringer DEQ2496 will do the job for about $350 (including the microphone). 1/3 octave.
What's more, it can fix your problems once you have figured out what they are.
Post removed 
1/10 octave or better can be achieved by both TrueRTA and by REW.

The reason that RTA is not useful is that it has little memory while the major room influence is in the delayed reflections and their interactions. For that, you need to see "waterfall" plots or similar that show how the sounds at different frequencies decay in the room. Generally, one wants uniform decay of all frequencies.

For more info, I recommend Toole's new book "Sound Reproduction - Loudspeakers and Rooms." I am about half way through it now.

Kal
Post removed 
If your equalization requirement at some frequency is a "cut" rather than a "boost" the "feedback destroyer" function of the DEQ2496 can do 1/10 octave resolution.

Actually the DEQ2496 can simultaneously implement several sharp (1/10 octave) notches, several broad (multi-octave parametric equalizer) curves, and the 1/3 octave (31 band) graphic equalizer. It does not chew gum.
Kr4...The Behringer DEQ2496 autoequalization process gives the user a choice of FAST, MID, and SLOW response. Doesn't this answer your point about delayed reflections?

I use the FAST so as to quickly find the appropriate curve, and then switch to SLOW for the final tweak.
I have no experience with the Behringer. Does it show you a time-decay response?

Kal
I would be able to match the filter configuration to it and then check the results.

Kal
IMHO, the most pernicious effect of the room acoustics is the extended decay at certain frequencies. These frequencies are not necessarily those that show up as instantaneous magnitude peaks nor, if they do, is simply reducing the magnitude going to reduce the decay time appropriately. Ideally, one would like to have a room with a uniform decay time (analogous to RT60) for all frequencies but this is more difficult with small rooms (at home) than it is with large ones (concert halls).

Kal
Kr4...Thanks for your answers. Is there any device that can equalize delay at all frequencies? Anyway, I question your statement that delay should be uniform across frequency. In a highly reverberent hall, where delay is easy to hear, the tonal quality varies greatly as the sound dies away. In fact this variation is an essential characteristic of such sound. It wouldn't be realistic if it didn't happen that way.
There are many devices that attempt this with varying degrees of success. These include the usual suspects such as Audyssey, TacT, ARC and a slew of PEQs when coupled with RoomEQ Wizard. The one I like best is the Meridian MRC which measures the broadband decay above the Schroder frequency (~200-300Hz in most domestic rooms) and then calculates filters for band below that frequency. What is nice is the ability to adjust the decay target, to modify/delete/create filters and the ability to graph the results.

You are right about concert halls since there are effects there which are different from those in small rooms (short latency reflections, distance effects on frequency, etc.). Nonetheless, there is a definable RT-60 across most of the audible frequencies well into the bass in concert halls but small listening rooms have their low frequencies dominated by room modes below the Schroder frequency.

In addition, if you want to reproduce the concert hall acoustics when you play a recording of them, you do not want to superimpose the acoustical characteristics of the small room on them.

BTW, there's a great new book on all this from Floyd Toole.

Kal
I use and recommend SmaartLive (software on a PC) which is the industry standard for pro live sound work. In addition to RTA functions, it has real-time transfer function measurement . . . which is a much faster, more stable, and more accurate display mode than an RTA. Its FFT parameters include a FPPO (fixed-point-per-octave) setting, which greatly reduces the need to change FFT parameters to properly resolve the frequency domain of interest.

Phase and time-domain (impulse response) data are additional functions, and it can do so both with test signals or program material (music). It even has a good speaker-impedance testing function.
Hi, I think you are asking good questions and and you are headed in a good direction. There are probably a number of considerations but I think you can save some money and take a first pass with some PC-based software and a microphone. I've spent many years with hifi and several years ago after building my system to the max I got serious about figuring out room acoustics; the process convinced me that no matter how good your equipment is the room will be a huge part of the equation. If you start with some software and a microphone - and if you happen to get some equalizers so you can make the process interactive - you will get a very valuable sense for the challenge and importantance of room acoustics. After this first highly informative and relatively inexpensive pass at analyzing and correcting acoustics you might then decide to go on to room treatments or even a full room design - but I think you can learn a bunch with the PC software and a microphone; and while equalizers might not be the best or right long term solution, the ability to treat the signal (to perhaps a 1/3 octave level) and then see the results graphically on your computer and hear the results with your ears will add significantly to your knowledge and skill - which will help you make further decisions on how to best invest your time and money.

Check out these two older threads on audioasylum:

http://db.audioasylum.com/cgi/m.mpl?forum=speakers&n=177308&highlight=trueaudio&r=&session=

http://db.audioasylum.com/cgi/m.mpl?forum=speakers&n=178395&highlight=Yada&r=&session=

the spectrum analyzer is useless is you cannot alter the anomalies of your room.
In other words, the analyzer will show you the peaks, nulls in your room but thats it.
I bought a Goldline RTA few years back and while it showed me the peaks and nulls in my room, that all it did.
It tried using traps and panels, while it did improve it did not really cure my peak in my room which was around 60-80 hz.
I was about to sell the analyzer since it stayed most of the time in the box.
It was not until I bought a Accuphase digital voicing equalizer that i was able to exploit the use of the RTA.
Given that I had a peak of 63 hz in my room at +8 db, which the RTA showed, I then use the Accuphase bring up the 63 hz frequency and with a stroke of a pen, pull it down to about -7 db. Problem solved and everything begin to make sense.
In short, RTA is useless unless you have some sort of equalizer to alter your frequencies or room acoustics. But room acoustics is a bit tricky as in my experience, it did something else but not really cure the peaks in my room. If it did, just a small bit compared to the Digital room correction.

Another option as many guys said here is to use the Tact or Lyndorf room correction. THese works very well.
"the spectrum analyzer is useless is you cannot alter the anomalies of your room. In other words, the analyzer will show you the peaks, nulls in your room but thats it."

The RTA can only show you the frequency response in your room. It cannot identify variations in FR as peaks and nulls due to room modes since it cannot show you the room response with respect to time. Corrections based on RTA, only, are chancy.

Kal
Kr4...You say "Corrections based on RTA, only, are chancy."

Perhaps, but the $350 I bet on my first DEQ2496 has paid off better than any other expendature on my audio system.

As the saying goes "Don't let the search for perfection prevent the very good".
I said "chancy" intentionally; it does not mean that they never work. Besides, flattening the bass response, by itself, is advantageous.

Many people have had excellent results with the DEQ2496 using REW as the tool for setting up the filters. I do not know what tools for measurement are built into the DEQ2496, itself.

Kal
pc/laptop, nice sound card, like echo layla firewire + any condenser mic arround $100usd.-$200usd.
rode, mxl, adk, audio-technica, etc...

+ any software, from free Mda vst plugins to generate noise, to izotope ozone RTA to analize noise, to cubase/nuendo or sonar scope, to RME scope meter, etc..

but the best RTA its your ears,
make everything as silent as possible , then clap real hard
if you listen something other than your hands clapping, thats what you need to fix.

i dificult room could need +-24dB of EQ in some bands,
a good treated room needs less.
the sound of every eq its a bit diferent.

square rooms are dificult.
rectangular are better.
etc...
Juanpablocuervo wrote:
but the best RTA its your ears,
make everything as silent as possible , then clap real hard
if you listen something other than your hands clapping, thats what you need to fix.
Actually true since no RTA will give you that information about ringing and decay that your ears detect. Of course, REW will do it graphically and tell you how to filter it.

i dificult room could need +-24dB of EQ in some bands,
a good treated room needs less.
How true.

Kal