Audio Measurement System

[whatatrip]

I got an email from PE advertising the EMM-6 mic, Behringer XENYX 802 analog mixer and True Audio TrueRTA Audio Spectrum Analyzer Software. I also have Joseph D'Appolito's book "Testing Loudspeakers" but haven't had time to read it yet. Would this equipment and some cables be all I need to get started testing loudspeakers?

[romanbednarek]

Re: Audio Measurement System

I got an email from PE advertising the EMM-6 mic, Behringer XENYX 802 analog mixer and True Audio TrueRTA Audio Spectrum Analyzer Software. I also have Joseph D'Appolito's book "Testing Loudspeakers" but haven't had time to read it yet. Would this equipment and some cables be all I need to get started testing loudspeakers?

The mic and the mixer are a good combo (very similar to the setup that I use) but I'm not sure if the software is the best for designing speakers (although I'm not completely familiar with the capabilities of TrueRTA). The reason that I feel that the software may fall short is that I don't think that it will allow you to take measurements that can isolate the driver's frequency response from the interactions of the room. You really need a software package that is capable of taking "gated" measurements which, while often limited to frequencies above around 300 Hz, will provide an accurate measurement of the true response of a speaker without including artifacts due to room reflections. I've been taking a break from the hobby for a while so I'm not real familiar with the latest software. I started out using Speaker Workshop which is free but can be frustrating when used to take measurements due to the calibration routine and interface jig required. I later got SoundEasy (v10) which was a lot better for taking measurements but had a user interface that wasn't very intuitive and costs a few hundred dollars (they have newer versions that may be more intuitive in terms of the user interface, but regardless of the version, SoundEasy always seems to be the most powerful software package for the money in terms of capabilities including a crossover emulator and many different measurement types available including non-linear distortion and more). Another software package that people seem to like is LSPCad which seems to have a much better user interface than SoundEasy but has fewer features (and I'm not sure on the current price for the basic version... the pro version had more features but was well beyond my budget at the time).

I'm sure that somebody who is still active in the hobby can give better recommendations on the best software, but unless I underestimate the capabilities of TrueRTA, I think that you'll need something that can take different types of measurements more suited for use in a crossover simulator (and by the way, I've always used Speaker Workshop for simulating crossovers regardless of the measurement software used).

[johnlmurphy]

Re: Audio Measurement System

The mic and the mixer are a good combo (very similar to the setup that I use) but I'm not sure if the software is the best for designing speakers (although I'm not completely familiar with the capabilities of TrueRTA). The reason that I feel that the software may fall short is that I don't think that it will allow you to take measurements that can isolate the driver's frequency response from the interactions of the room. You really need a software package that is capable of taking "gated" measurements which, while often limited to frequencies above around 300 Hz, will provide an accurate measurement of the true response of a speaker without including artifacts due to room reflections. <snip>

Hello PE Forum!

This is John, the author of TrueRTA. I'd like to expand a bit on Roman's comments above.

Roman is correct in saying that TrueRTA will not do quasi anechoic measurements. While quasi anechoic measurements are a nice supplement to in-room frequency response measurements during the design phase of a project the in-room response is ultimately what the listener will hear...and that is what TrueRTA measures.

Consider this. Most of you will be using loudspeakers to listen to CDs and DVD movies in a home listening environment. For the most accurate reproduction you will want your playback system to sound like the system used to mix the audio for that CD or DVD. For the bulk of the existing music and film on CD or DVD those products were typically mixed in studios that were voiced using an RTA measuring the net in-room response of the monitoring system. The typical target voicing for these systems would be the X-Curve as specified in the technical specifications: ANSI/SMPTE 202M and 222M. TrueRTA is used by many of the engineers who set up these audio and video mixing rooms. So if you use TrueRTA to voice your playback system you will be in very good company.

Do this test. Try to turn your home playback system up to the same loudness you hear in the theater when you go out to a movie. Does your wife scream "turn it down!" as your ears begin to hurt? Then most likely your playback system does not sound like the reference monitoring system on which that program material was mixed/mastered.

Those speaker builders who set as a target a "flat quasi-anechoic frequency response" as their final voicing will be ignoring the effects of the room. This is not how the audio professionals do it, nor is it what you actually hear in the room. So while quasi-anechoic measurements are an interesting supplement to in-room measurements while designing, in the end it is the in-room response (and measurement) that really matter when voicing a professional audio monitoring system. That same voicing is what I recommend for home playback systems. For more on voicing home theaters see my Application Note 2: "Measuring and Voicing a Home Theater Sound System" in the TrueRTA Help Topics available in the Free version.

The responses I have available on my own reference systems include: Flat, Small-Room X-Curve and X-Curve. X-Curve is probably used the most. Flat is used only for very select reference material.

I apologize for this commercial interruption but I felt my product needed to be placed in perspective.

Regards,

John

[whatatrip]

Re: Audio Measurement System

Hello PE Forum!

This is John, the author of TrueRTA. I'd like to expand a bit on Roman's comments above.


I apologize for this commercial interruption but I felt my product needed to be placed in perspective.

Regards,

John

Thank you for the response John. I appreciate the "commercial" but I do have a question. I understand that the Level 2 software being advertised has upgrades Level 3 and Level 4 available that cost twice as much as the original $40.00 software being offered at PE. Keep in mind I am a newbie so I don't yet understand the advantages of going to levels 3 and 4 but I also want to know what I might be getting myself into should I find I might want levels 3 and 4. What are the differences between levels 2, 3, and 4 and how much does each upgrade cost?

[dlr]

Re: Audio Measurement System

Do this test. Try to turn your home playback system up to the same loudness you hear in the theater when you go out to a movie. Does your wife scream "turn it down!" as your ears begin to hurt? Then most likely your playback system does not sound like the reference monitoring system on which that program material was mixed/mastered.

I'm not sure I'd want that. Personally I'm only concerned with stereo, not home theater, though many are and I don't know if studio monitor sound is what I ultimately desire. And when I'm in a movie, I really, really, really wish that I could say "Turn it down!!!" for the majority of movie theaters I've been in.

I'm also not going to have any idea if my system sounds similar to a studio-monitored recording for stereo based on movie theater sound. I'd have to have first hand experience in a studio to have any idea on that aspect.

dlr

[Jeff B.]

Re: Audio Measurement System

Hello PE Forum!

This is John, the author of TrueRTA. I'd like to expand a bit on Roman's comments above.

Roman is correct in saying that TrueRTA will not do quasi anechoic measurements. While quasi anechoic measurements are a nice supplement to in-room frequency response measurements during the design phase of a project the in-room response is ultimately what the listener will hear...and that is what TrueRTA measures.

Consider this. Most of you will be using loudspeakers to listen to CDs and DVD movies in a home listening environment. For the most accurate reproduction you will want your playback system to sound like the system used to mix the audio for that CD or DVD. For the bulk of the existing music and film on CD or DVD those products were typically mixed in studios that were voiced using an RTA measuring the net in-room response of the monitoring system. The typical target voicing for these systems would be the X-Curve as specified in the technical specifications: ANSI/SMPTE 202M and 222M. TrueRTA is used by many of the engineers who set up these audio and video mixing rooms. So if you use TrueRTA to voice your playback system you will be in very good company.

Do this test. Try to turn your home playback system up to the same loudness you hear in the theater when you go out to a movie. Does your wife scream "turn it down!" as your ears begin to hurt? Then most likely your playback system does not sound like the reference monitoring system on which that program material was mixed/mastered.

Those speaker builders who set as a target a "flat quasi-anechoic frequency response" as their final voicing will be ignoring the effects of the room. This is not how the audio professionals do it, nor is it what you actually hear in the room. So while quasi-anechoic measurements are an interesting supplement to in-room measurements while designing, in the end it is the in-room response (and measurement) that really matter when voicing a professional audio monitoring system. That same voicing is what I recommend for home playback systems. For more on voicing home theaters see my Application Note 2: "Measuring and Voicing a Home Theater Sound System" in the TrueRTA Help Topics available in the Free version.

The responses I have available on my own reference systems include: Flat, Small-Room X-Curve and X-Curve. X-Curve is probably used the most. Flat is used only for very select reference material.

I apologize for this commercial interruption but I felt my product needed to be placed in perspective.

Regards,

John

Hello John,

I just wanted to add that I strongly agree with your points. I have used quasi-anechoic responses to help get my crossover to a reasonable point in simulations. Then I use an RTA to meaure the set-up in my room with the speakers placed where I expect they will end up. At this point I do most of my tweaking by listening to music that I feel I am familiar with and checking the spectrum with the RTA. I have done this for years and it has become my standardized approach. It has worked out very well for me.

Jeff B.

[romanbednarek]

Re: Audio Measurement System

I must admit that a very recent experience has taught me some of the value of knowing how the room influences the sound of a pair of speakers.

I recently set up my Asterion two-way speakers (XT18WH/OW1) in my current listening room which is larger than the room that I designed them in and also has different bass reinforcement based on the listening position being located near more bass nulls than peaks. Anyway, these speakers sounded very thin in this room so I decide the mess around with the crossovers and balance them out a bit (as well as tackle a few issues that I had with them in the past). Although I used my ears as the "spectrum analyzer" in this case, I can understand where such a tool can be valuable in adjusting a speaker crossover response based on room interactions.

I still believe that a "reflection free" measurement is essential when it comes to simulating a crossover design in order to give you a foundation to work from. I think that measurements that include the room response should be used in conjunction with the crossover simulation. I always say that you can't have too many tools in your toolbox, especially when it comes to designing speakers and the more information that you have to go by, the less guessing you have to do. My dilemma (if the Asterion project could still be built, considering the XT18WH is no longer produced) is that my site includes projects that are intended for a wide audience with many different room configurations which would essentially require many different crossover versions for them to sound balanced in each environment. I always knew that designing crossovers was tough but with so many possible unknown variables it can seem nearly impossible at times (especially if you are very critical).

I think that the important thing is that the original poster is getting a better idea about some of the factors that influence crossover design and some of the tools that can help.

[Deward Hastings]

Re: Audio Measurement System

It’s a given that no amount of BSC is ever “correct”, since by its nature BSC or the lack thereof causes on-axis and power response to diverge, producing the inevitable “monkey coffin” signature. But BSC is necessary, to maintain the illusion of “flat” bass response.

For a DIYer the answer is to “tune” for the “best” sound in the intended listening room . . . for the loudspeaker designer it’s a problem, since you don’t know where the speaker is going to be used. One solution is to implement BSC as a functional block independent of the rest of the crossover so that it can be adjusted by the end user (Behringer does that with their 2030a monitors, for example). Another (fairly common in HT receivers) is to let the receiver do the BSC (as Yamaha does with their YPAO feature). It is not particularly difficult even with a passive crossover to make BSC adjustable (at least switchable) so the end user can move the speaker without a complete crossover re-design. Of course if you can be sure that the speaker will never be closer than 4 ft to any wall or reflecting surface in a minimum 300 sq. ft. room then the full 6 dB of BSC is probably safe (and it’s necessary if the speaker is to be “demonstrated” in a large room like at a DIY event).

More difficult is dealing with the other effects such as: too close to the walls, not the design height off the floor, or the simply unsatisfactory (but unfixable) listening room. Simulations may give some hint what to do about that, but in the end you have to measure, listen, measure, listen, measure, listen . . .

[WWWJD]

Re: Audio Measurement System

Thank you for the response John. I appreciate the "commercial" but I do have a question. I understand that the Level 2 software being advertised has upgrades Level 3 and Level 4 available that cost twice as much as the original $40.00 software being offered at PE. Keep in mind I am a newbie so I don't yet understand the advantages of going to levels 3 and 4 but I also want to know what I might be getting myself into should I find I might want levels 3 and 4. What are the differences between levels 2, 3, and 4 and how much does each upgrade cost?

The biggest difference between levels 1 through 4 is resolution; level 4 provides you with 1/24 octave resolution, which in my opinion, is the only way to fly. We purchased this license for work, and I've been chomping at the bit to get it home and tinker with the speakers.

If you want to see "everything" happening in the spectrum, resolution is priority. $100 ain't bad, of course, that's easy for me to say because it wasn't my $100.

[whatatrip]

Re: Audio Measurement System

The biggest difference between levels 1 through 4 is resolution; level 4 provides you with 1/24 octave resolution, which in my opinion, is the only way to fly. We purchased this license for work, and I've been chomping at the bit to get it home and tinker with the speakers.

If you want to see "everything" happening in the spectrum, resolution is priority. $100 ain't bad, of course, that's easy for me to say because it wasn't my $100.

I downloaded the trial 1 octave version. Everything went fine except system calibration and actually getting a pink sound out of the speakers. It was a matter of pushing the right buttons at the right time. I hooked up a Radio Shack SPL meter and got some results so I have a feel for what to expect.

You are right, higher resolution is a must so I ordered the level 4 version for $100.00.

[mgrabow]

Re: Audio Measurement System

I am going to order the level 4 in February as I agree the higher resolution is a must. I downloaded the fee version and it was very simple to use. Most of the other more expensive software available on the internet is too tricky to figure out on the trial version. I would hate to spend several hundred on something because it does more only to find out a simple frequency response measurement is like pulling teeth... I wish it were available on PE so I can use paypal...

[diy speaker guy]

Re: Audio Measurement System

Why not try Room Eq Wizard or HolmImpulse? You get all the resolution you can handle for free. They are also both fairly intuitive to use - I got both of them working (the main features, at least) without looking at the documentation.

[brianp]

Re: Audio Measurement System

Thank you, John, for the clarification. I too am curious about the applicability of True RTA measurements to crossover design, and have a few questions/comments.

It would seem that, as a starting point for crossover simulation, I would want the most accurate measurements of the responses of the individual drivers IN the actual cabinet, excluding all near boundary reflections EXCEPT floor bounce, which will always be present and basically the same in any room and hence should be included. If I measure these responses (on-axis, and various degrees off-axis) outdoors, on a nice quiet day, would True RTA (the highest rez version) give me accurate enough results to plug into a simulator like Jeff's PCD? Or are there better options at this stage of the design process?

Once I have modeled a crossover that yields flat summation of the drivers based on these measurements, and tested it (again outdoors) to confirm that the actual anechoic response matches the simulation, I could bring the speakers indoors, set them up in their intended positions, and measure the response again from the listening position. Whatever variance from the measured anechoic response is revealed by True RTA could then be ascribed to the room, and corrective measures (crossover mods, EQ, or room treatment) could be applied. Is this basically correct?

I think the reason most speaker designers target a flat anechoic response is that this provides a universal, albeit "one-size-fits-none" reference standard. Commercial designers have no idea how the end user will set the speakers up in what room, so they voice them in a way that will at least be replicable under lab conditions. It is left up to each end user to correct for room-based anomalies. Recent developments in digital EQ (Audyssey and others) make this a WHOLE lot easier than it used to be.

If I build a one-off pair of speakers, optimized for best response in a specific room, I may find that these speakers perform far from optimally when placed in a different room -- as may happen if, for instance, I move to another house. True RTA, under these circumstances, could be extremely useful in telling me how the response in the new room departs from that in the old, thus allowing me to undertake corrective strategies.

Hello PE Forum!

While quasi anechoic measurements are a nice supplement to in-room frequency response measurements during the design phase of a project the in-room response is ultimately what the listener will hear...and that is what TrueRTA measures.

Most of you will be using loudspeakers to listen to CDs and DVD movies in a home listening environment. For the most accurate reproduction you will want your playback system to sound like the system used to mix the audio for that CD or DVD. For the bulk of the existing music and film on CD or DVD those products were typically mixed in studios that were voiced using an RTA measuring the net in-room response of the monitoring system. The typical target voicing for these systems would be the X-Curve as specified in the technical specifications: ANSI/SMPTE 202M and 222M. TrueRTA is used by many of the engineers who set up these audio and video mixing rooms. So if you use TrueRTA to voice your playback system you will be in very good company.

Those speaker builders who set as a target a "flat quasi-anechoic frequency response" as their final voicing will be ignoring the effects of the room. This is not how the audio professionals do it, nor is it what you actually hear in the room. So while quasi-anechoic measurements are an interesting supplement to in-room measurements while designing, in the end it is the in-room response (and measurement) that really matter when voicing a professional audio monitoring system. That same voicing is what I recommend for home playback systems. For more on voicing home theaters see my Application Note 2: "Measuring and Voicing a Home Theater Sound System" in the TrueRTA Help Topics available in the Free version.

The responses I have available on my own reference systems include: Flat, Small-Room X-Curve and X-Curve. X-Curve is probably used the most. Flat is used only for very select reference material.

John

[mgrabow]

Re: Audio Measurement System

Why not try Room Eq Wizard or HolmImpulse? You get all the resolution you can handle for free. They are also both fairly intuitive to use - I got both of them working (the main features, at least) without looking at the documentation.

Looks nice but where is the pink noise generator? I searched the manual and the word pink appears no where...

[diy speaker guy]

Re: Audio Measurement System

I don't know. I've only ever done the frequency sweeps. I know ARTA has pink noise for sure though and that's free too. There's no shortage of free measurement software. I'm sure one of them will be right for you.