Home Theater Forum and Systems banner

First time REW Plot

Tags
plot time
18K views 39 replies 7 participants last post by  eyleron 
#1 ·
Hello everyone, I just made my first plot using REW. I am kind of thinking that something must be wrong with the way I am measuring, and maybe some of the more experienced members of this forum can help me out. I just wanted to get familiar with the software, before I started measuring my home theater. I have a set of M-Audio studio monitor's that are connected to my PC that I measured. This plot just looked a little to good to me, but I have always loved these speakers. Just to make sure that everything was working right, I cut the mid's on my soundcard EQ, and sure enough the plot showed exactly what I did. So anyway here is my first plot, could anyone tell me why this might look just a little too good to be true?
 

Attachments

See less See more
1
#2 ·

Welcome to the Forum, baker!

Well, I doubt they really get response down to 2 Hz! LOL For one thing, your graph scaling is really off re-set the vertical axis for 45-105 dB, and the horizontal for 15 Hz – 25 kHz. Also, it looks like you’ve engaged maximum smoothing. Try smoothing for 1/3-octave, which is acceptable for a full-range graph.

Regards,
Wayne
 
#4 ·
Yes, I did use a loopback connection as shown in the REW help file. I am a bit confused though as to what you are saying the plot is showing, could you elaborate a bit. Is the plot showing only the differential of the feedback loop?

You have written a really nice piece of software here John. I did a lot of lab work at Kansas State University about 7 years ago when I was studying electrical engineering. We were using Texas Instruments Labview a lot at the time. I remember writing C files for differential equations, been a while, but boy you had a lot of hard work in this. Great job, and the only way I can thank you is to learn how to use this software to its full potential.
 
#6 ·
I suspect John had it right, that you are using the cable looping your output back to your input so you are measuring your soundcard response with a mic calibration file loaded. What microphone were you using, and how was it connected?

Below the Filter graph, there are boxes you can check to display the the Mic/Meter Cal and the Soundcard Cal along with the measured data. If you include these, any relationship with the data will become clearer.

Bill
 
#7 ·
Yes, went back and read the instructions again. I think this is a little more acceptable. I'm basically just trying to get a good understanding of the software at this point using the M-Audio's, so if you see anything that looks odd let me know. I really like this software so far! I fixed the graphs, and attached the requested plots.
 

Attachments

#9 ·
If you're using any RS SPL, limit your measurements to below 3kHz or so. That mic, even with the cal file, cannot be trusted any higher.
Often people will look at another scan, just up to 200Hz with no smoothing to get a real clear picture of the bass region.
 
#10 ·
Sounds good. If I wanted to do a full frequency plot of the system, what mic/pre-amp pair should I be looking at? The other question that this brought up is how is Audyssey EQ making appropriate changes to the system frequency response with a $.50 mic? A little off the topic, but just curious.

Thanks
 
#17 ·
Well after a bit of thought I am interested in measuring frequency response over the whole listening spectrum, so I purchased a calibrated ECM8000 that was refurbished from Cross Spectrum. I am still a little bit confused with the choice of pre-amp USB interfaces though. I looked at the Tascam US-122MKII and the M-Audio Mobile Pre, but the reviews are a bit mixed. I would like to use the pre-amp in my DJ gear also, so my question is does the sampling rate of the DAC in the USB-interface effect the measurements that I will be taking with the ECM8000. I'm not sure what the sampling rate of REW is, but as long as the sampling rate of the USB-interface is greater than what REW is measuring I should be good right? Additionally, I am running Windows 7 if anyone has had experience with a specific USB-Interface and Windows 7.
 
#19 ·

I looked at the Tascam US-122MKII and the M-Audio Mobile Pre, but the reviews are a bit mixed.
It’s good that you’re researching before making the plunge for an USB interface. :T

The thing to keep in mind with complaints in the reviews is determining if the issue they’re having is relevant to you or not. For instance, most of the reviewers are using their interfaces for recording to a computer, which is drastically different from what you’re going to be using it for. If they complain that the connectors seem cheap, or that it doesn’t sound good, or it’s noisier than they wanted, that has no bearing on your application, which is REW.

Basically, the only thing you’re concerned with is whether or not the interface will “play nice” with your operating system. Those are the comments you want to pay attention to. If there are compatibility issues you’ll probably have a problem generating a calibration file, and that’s the thing that frustrates new REW users the most. If you find positive reviews where the users have the same operating system as you - there’s your candidate. :T

Regards,
Wayne
 
#20 ·
Great advice Wayne, and I will take all that I can get. I am going to go with either the Focusrite Saffire 6 USB Audio Interface as it has no bad reviews that I can find, and its OS friendly, or the MBOX2 as a friend has one for sale. I'll let you all know when I get them plugged in, I'm sure I'll have some questions.

Thanks again!!
 
#21 ·
So I got my ECM8000 mic today, and I'm a bit confused on what a calibrated microphone is. I see the frequency plots with and without smoothing, and also the off axis polar plot. So I guess my question is the following: Is the microphone measured, and then the included mini-disk will perform the necessary corrections for measurements in REW, or is the physical microphone calibrated by variable resistors in mic circuit? I'm guessing that the disk included is to offset the response of the mic. Any help would be greatly appreciated.
 
#23 ·
So I hooked up the calibrated ECM8000 mic and the Focusrite USB interface, no problems there. I am receiving errors from REW that the mic signal is to low. Its saying that it should be greater than -10 dB, and I am getting somewhere around -90 dB. I have the input gain on the channel I am using within the USB interface almost maxed out. I did load the calibration files that came with the mic, and they are in .FRD file format. REW didn't seem to have a problem when I loaded those. I know the microphone is working correctly, because when I adjust the mixer knob to balance the input and the playback, the mic is very audible. The only other information I can provide is that I did set the level on the speaker to 75 dB with a SPL meter. So what am I doing wrong here?
 
#24 ·
Either REW is set to the wrong channel (e.g. mic is on left but REW has been told to listen on Right) or it is not set to the input the mic is connected to (for -90dB it is probably on a line input rather than the input the mic is connected to). Also make sure that you set up your mixer so that none of the mic signal is fed back out to playback when you are making measurements.
 
#26 ·
Well I need some more advise. Here are the plots, and from what I have read here on the shack forum, these look about right. I'm just measuring my M-Audio studio monitors that I have at my PC. If everything looks good, I will wait on my cables from Monoprice and measure the home theater. Thanks for all your help everyone!
 

Attachments

#29 ·

When I am taking measurements of my home theater should I place the mic in the primary listening position with the mic pointed towards the ceiling (90 degree angle)?
The short answer for subwoofer measurements is, it doesn’t matter.

If you’re doing full-range measurements (your main-channel speakers), the short answer is, you only point the mic towards the ceiling if you’re using a 90-degree incident (vertical) calibration file.

Horizontal orientation with the mic on-axis to the sound source (0-degrees incident) has been the traditional method for full-range, free-field measurement, where the room was an open space relatively free of reflections. However, that’s mainly because most stand-alone RTAs (which until several years ago was about the only thing available) came with mics that were calibrated for on-axis measurements.

Some RTA manufacturers offered the option of 90-degree orientation for random incident measurements (aka diffuse field), where the sound arrives from all directions more or less simultaneously, with equal probability and level. In other words, an exceedingly reverberant environment. Random-incident measurements required a different capsule for the measurement mic (which the manufacturer made available), for reason of the specific calibration as well as a housing better suited for 90-degree orientation.

That just refers to the mic’s position during its calibration, however. Everything I’ve seen for actual “in the field” measurements says the standard protocol for horizontal (on-axis) measuring is 20 degrees, and 70 degrees for vertical (i.e., angled slightly forwards towards the sound source). This may have something to do with compensating for interference from the mic’s housing with the sound waves, I forget exactly why.

The ready availability these days of mics with 90-degree calibration certainly opens up more measurement options. Others have their opinions, and maybe they’ll weigh in, but mine is that you will generally get the best results with on-axis measurements. It should be a no brainer to figure out that the home theater environment, while certainly not totally free of reflections and reverberation, more closely resembles a free-field environment than a random-incident environment.

Vertical orientation may add more upper-frequency information from ceiling reflections than you’d get with on-axis, and as such will probably influence what the RTA displays. How much so will depend on your particular room – how “live” it is, the height of the ceiling in relation to the distance between the sound source (speaker) and measurement mic (i.e. inside or outside the “first reflection” zone), etc. Even though the ECM8000 is omnidirectional, its capsule is rather large for a measurement mic. As a result, its off-axis response (compared to on-axis) starts skewing as low as 2 kHz. So differences >2 kHz are what you might see with horizontal vs. vertical readings. (Smaller-capsule omni mics typically retain uniform 0 vs. 90-degree response at least an octave higher.)

Regards,
Wayne
 
#30 ·
Thanks Wayne!!

One thing that I eventually wanted to measure was how much room reflections were playing into my home theater (living room). I will be adding some acoustic treatment in the next few months, and I would like to get a measurement before and after to see how much the treatments help. I haven't dug in to deep on the forum material concerning waterfall plots with REW, but I believe they can be used to show reflections through the decay time. So, from what I gathered in measuring the sub the orientation of the mic shouldn't make much difference, assuming I use the correct calibration file. When measuring the mains, I should tilt the mic to approximately 20 degrees for on-axis measurements (towards the speaker I am measuring), and use the 0 degree incidence file included with the mic.

Thanks again for all your time!
 
#33 ·
So I would love to get some comments on these, as they are my first round of measurements in my home theater. These were taken with only one mic location, in the sweet spot. I would like to get a better understanding of some of the other plotting features of REW, so I will start with the phase. Should I post the phase of the mains and sub separate around 80 Hz, and match the phase for blending?

I can re-affirm what I had heard in regards to Audyssey EQ. It is chopping the highs quite a bit, which may be okay for movies. It seems like it is giving a little more of a "house curve" like I have read about here. But for music, no Audyssey for this guy. It sounds like someone through a towel over my speakers, no cymbals at all!! I guess its a personal preference, which I am more inclined to prefer a flat response. Also wanted wanted to add that I have a very meager sub-woofer right now. The sub is from a Onkyo HTIB I ran for about a year. I will be upgrading to a SVS PB-12 Plus in the next few months, so that should fill in the low end frequencies a little more.

Any advice would be most appreciated, and REW v.5 is a very nice piece of software. Thanks so much for giving us such a powerful tool and your dedication to constantly improve the software John.
 
#34 ·
... But for music, no Audyssey for this guy. It sounds like someone through a towel over my speakers, no cymbals at all!! ...
Does your receiver offer a choice of the Audyssey Reference curve or the Audyssey Flat curve? As you mentioned, the Audyssey Reference curve is designed for movies. Some people find the Audyssey Flat curve works better in their room for music, where others find it is too bright in their environment and the Audyssey Reference curve works best for both.

Bill
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top