Home Theater Forum and Systems banner

1 - 5 of 5 Posts

·
Registered
Joined
·
3 Posts
Discussion Starter #1
Hi,

Quick summary: I am getting a delay in my impulse response when I export it to use in an IR loader.

I currently do not have an ecm8000 but do have a c414 that I am using in omni. I calibrated the sound card and used the calibration numbers for the c414 TLII found in these forums for the mic calibration. I do not have a timing reference set up accoustically or through a loopback. I have t=0 at IR peak checked so it seems to be setting the major peak at 0 seconds but still when I export it there is a delay because the peak is showing up in my IR loader at 2 seconds. I tried focusing the IR window on the left to exactly where the peak is which got rid of the delay as much as I can tell but it seems like I am not getting an accurate representation of the room if I am moving that around and also, the lower end frequencies drastically change when I do it.

So I have a few questions:

1) What is causing the delay?

2) What are the steps I would take to obtain a proper room reading (aside from grabbing a measurement mic which I am working on)?

3) Is an SPL meter necessary?

4) How would I go about making a correction impulse response through the EQ window? I have tried but maybe because of the delay it isn't working properly. Just shows up as a couple bumps up to 0.1 seconds in the IR loader.

5) What information is in the filters impulse response wav file? Is it just EQ? Or is there the delay/reverb/distortion information?

6) On that note, what information is in the impulse response that is exported as a wav? Obviously there is delay data it would seem, based on my troubles haha

7) Lastly, how would I use the software to obtain the differences in eq response of mics or even distortion, if possible? Do I set the one mic as my calibration mic by making a file so that any measurement taken will give me the difference between the two? Or do I just take 2 measurements of mics with no mic calibration file? I used it for my mic pres by running signal through them and picking up measurements and then using the arithmetic function of A/B to get the difference. I then went into the EQ window to get the eq measurements through the filters to figure out the difference but these filters are giving me the eq numbers to counter the difference and not the eq values themselves.

Any help you can provide is very much appreciated. Thanks in advance.
 

·
REW Author
Joined
·
6,711 Posts
1) The impulse response contains information before and after the peak, if the export started at the peak significant information would have been removed. REW generally keeps 1 second of date before the peak. The export is the response that you see in REW, its first sample is the earliest sample you see in the REW response (the one at the earliest (negative) time. If you export a windowed response it will start at the left edge of the left window. Remember that the WAV file has no concept of time, it is just a sequence of samples at the sample rate. Time aligning the response needs to be handled by whatever you are using to process it.

2) The ones listed in the help!

3) No

4) If you are exporting the filters IR there won't be a great deal to see in it to the eye, make sure you normalise the response and use at least 24-bit export to avoid early quantisation of the data.

5) It is the impulse response of the filters, nothing else

6) The impulse response of the measurement

7) If you want the opposite of the filters that would flatten the response difference just invert the gains
 

·
Registered
Joined
·
3 Posts
Discussion Starter #3
So a timing reference loop back is not necessary? The issue is with the starting point I set in the IR loader?

What process would I use to compare mics? Do I make a cal file of the one mic to use as the reference and then just take a measurement of the other mic to get the difference of the two (of course in the exact same position to rule out placements and the room as contributing factors)?

How do I invert the gains of the filters? Or do you mean just manually invert them when I put the values into an eq?
 

·
REW Author
Joined
·
6,711 Posts
No, a timing reference is not needed. Accurately comparing mics is difficult but placing them close to the source can help to reduce room influence. Trace arithmetic can be used to compare the responses as you mentioned in the first post, depending on which you pick as the numerator you can get an inverted comparison. Inverting gains of filters just means manually changing the sign of the gain.
 

·
Registered
Joined
·
3 Posts
Discussion Starter #5
No, a timing reference is not needed. Accurately comparing mics is difficult but placing them close to the source can help to reduce room influence. Trace arithmetic can be used to compare the responses as you mentioned in the first post, depending on which you pick as the numerator you can get an inverted comparison. Inverting gains of filters just means manually changing the sign of the gain.
I appreciate your help! Thanks, John.
 
1 - 5 of 5 Posts
Top