Beta Test: Room Correction for Android

[harkpabst]

so has it been determined yet that it's even worth purchasing and using a calibration mic like the imm6, or is it just as effective to use your phone with the case off.. eg. Samsung s23?

It's not so much a matter of calibration but a matter of accuracy. I agree with @jmudrick that external microphones pretty much always will be more accurate.

The term "calibration" is used a little ambiguously in this discussion. There are different aspects to it that are somewhat related: First, calibration can ensure SPL readings are correct in absolute numbers. We don't really need that for room correction, but it doesn't hurt either. Second, calibration can correct for frequency dependant inaccuracies of the microphone, ideally for each individual sample. Having that would be a good thing, but the better the mic, the smaller the error, that would need correction.

Personally I would always go with a USB mic. Not only don't you need an external ADC dongle. You also get the correct absolute SPL "for free" without the need for a calibration file. Even if this is not required for RC it allows more flexible use.

 

[bernard23]

The point is that an external mic like the Dayton will have a much flatter FR inherently than probably any Android phone. All the calibration file does is prove how accurate it is. For the purposes we use it, then the adjustments it might make are likely to be insignificant, just as Amir pointed out in his article. A cal file for the inbuilt S23 mic would actually be way more useful because it's likely that they exhibit far less consistent behaviour.

 

[canard]

The point is that an external mic like the Dayton will have a much flatter FR inherently than probably any Android phone. All the calibration file does is prove how accurate it is. For the purposes we use it, then the adjustments it might make are likely to be insignificant, just as Amir pointed out in his article. A cal file for the inbuilt S23 mic would actually be way more useful because it's likely that they exhibit far less consistent behaviour.

I'll have to search here..I had put here the calibration file of a non-USB Dayton 6... it's just a text file that opens without any worries..will give you an idea of the proportions ( for only microphone, not microphone+adc).... if it can be found, or other owners to re-post one....

 

[rcon7]

I'll have to search here..I had put here the calibration file of a non-USB Dayton 6... it's just a text file that opens without any worries..will give you an idea of the proportions ( for only microphone, not microphone+adc).... if it can be found, or other owners to re-post one....

here you go in txt form. 3.5mm imm6 nonusb calibration txt file.

*1000Hz-39.1
20.00-2.2
20.55-1.9
21.11-1.5
21.69-1.2
22.29-0.9
22.90-0.6
23.53-0.4
24.18-0.2
24.84-0.1
25.52-0.1
26.220.0
26.940.0
27.68-0.1
28.44-0.1
29.22-0.1
30.03-0.1
30.85-0.2
31.70-0.2
32.57-0.2
33.46-0.2
34.38-0.2
35.33-0.2
36.30-0.2
37.29-0.2
38.32-0.1
39.37-0.1
40.450.0
41.560.0
42.700.1
43.870.2
45.080.2
46.320.3
47.590.3
48.900.4
50.240.4
51.620.4
53.030.4
54.490.4
55.990.4
57.520.4
59.100.4
60.730.4
62.390.4
64.110.4
65.870.5
67.680.5
69.540.5
71.440.5
73.410.5
75.420.5
77.490.5
79.620.5
81.810.5
84.050.5
86.360.5
88.730.5
91.170.4
93.670.4
96.250.4
98.890.4
101.600.4
104.390.4
107.260.4
110.210.4
113.230.4
116.340.3
119.540.3
122.820.3
126.190.3
129.660.2
133.220.2
136.870.2
140.630.2
144.500.2
148.460.3
152.540.3
156.730.3
161.030.3
165.450.2
170.000.2
174.670.2
179.460.2
184.390.3
189.450.3
194.650.3
200.000.3
205.490.3
211.130.3
216.930.3
222.890.4
229.010.4
235.300.4
241.760.5
248.400.5
255.220.6
262.230.6
269.430.7
276.830.7
284.430.7
292.240.7
300.260.7
308.510.7
316.980.6
325.680.6
334.630.5
343.810.4
353.260.3
362.960.2
372.920.1
383.160.1
393.680.0
404.490.0
415.600.0
427.01-0.1
438.74-0.1
450.79-0.1
463.170.0
475.880.1
488.950.3
502.380.4
516.170.0
530.35-0.1
544.91-0.1
559.870.1
575.250.2
591.040.2
607.270.2
623.950.1
641.080.2
658.680.2
676.770.3
695.350.4
714.450.4
734.070.4
754.220.4
774.940.4
796.210.3
818.080.3
840.540.2
863.620.2
887.340.2
911.700.2
936.740.2
962.460.2
988.890.1
1016.040.0
1043.940.0
1072.61-0.1
1102.06-0.2
1132.33-0.2
1163.42-0.2
1195.37-0.1
1228.19-0.1
1261.91-0.2
1296.57-0.3
1332.17-0.3
1368.75-0.4
1406.34-0.4
1444.95-0.4
1484.63-0.4
1525.40-0.3
1567.28-0.3
1610.32-0.3
1654.54-0.3
1699.97-0.2
1746.650.0
1794.610.2
1843.890.5
1894.530.7
1946.550.8
2000.000.9
2054.920.9
2111.350.8
2169.320.6
2228.890.5
2290.100.4
2352.980.4
2417.590.5
2483.980.7
2552.190.7
2622.270.6
2694.270.5
2768.260.4
2844.270.3
2922.370.4
3002.620.4
3085.070.4
3169.790.3
3256.830.2
3346.260.1
3438.14-0.1
3532.55-0.5
3629.56-0.8
3729.22-0.8
3831.62-0.3
3936.840.5
4044.941.2
4156.011.5
4270.141.4
4387.391.0
4507.870.3
4631.65-0.5
4758.83-1.0
4889.51-1.6
5023.77-1.8
5161.72-1.7
5303.46-1.1
5449.09-0.3
5598.720.3
5752.460.6
5910.420.4
6072.720.0
6239.47-0.2
6410.800.0
6586.840.6
6767.711.1
6953.551.4
7144.491.4
7340.671.4
7542.241.5
7749.351.9
7962.142.4
8180.782.6
8405.422.6
8636.232.0
8873.371.1
9117.030.2
9367.380.3
9624.611.1
9888.891.7
10160.441.6
10439.440.8
10726.10-0.1
11020.63-0.1
11323.250.8
11634.181.0
11953.650.3
12281.890.0
12619.150.4
12965.660.7
13321.690.3
13687.50-0.1
14063.35-0.2
14449.520.0
14846.30-0.5
15253.97-1.1
15672.84-1.1
16103.21-0.7
16545.39-1.4
16999.72-1.7
17466.52-1.4
17946.14-1.7
18438.94-1.3
18945.26-0.6
19465.490.0
20000.003.2

 

[gremlin]

here you go in txt form. 3.5mm imm6 nonusb calibration txt file.

*1000Hz-39.1
20.00-2.2
20.55-1.9
21.11-1.5
21.69-1.2
22.29-0.9
22.90-0.6
23.53-0.4
24.18-0.2
24.84-0.1
25.52-0.1
26.220.0
26.940.0
27.68-0.1
28.44-0.1
29.22-0.1
30.03-0.1
30.85-0.2
31.70-0.2
32.57-0.2
33.46-0.2
34.38-0.2
35.33-0.2
36.30-0.2
37.29-0.2
38.32-0.1
39.37-0.1
40.450.0
41.560.0
42.700.1
43.870.2
45.080.2
46.320.3
47.590.3
48.900.4
50.240.4
51.620.4
53.030.4
54.490.4
55.990.4
57.520.4
59.100.4
60.730.4
62.390.4
64.110.4
65.870.5
67.680.5
69.540.5
71.440.5
73.410.5
75.420.5
77.490.5
79.620.5
81.810.5
84.050.5
86.360.5
88.730.5
91.170.4
93.670.4
96.250.4
98.890.4
101.600.4
104.390.4
107.260.4
110.210.4
113.230.4
116.340.3
119.540.3
122.820.3
126.190.3
129.660.2
133.220.2
136.870.2
140.630.2
144.500.2
148.460.3
152.540.3
156.730.3
161.030.3
165.450.2
170.000.2
174.670.2
179.460.2
184.390.3
189.450.3
194.650.3
200.000.3
205.490.3
211.130.3
216.930.3
222.890.4
229.010.4
235.300.4
241.760.5
248.400.5
255.220.6
262.230.6
269.430.7
276.830.7
284.430.7
292.240.7
300.260.7
308.510.7
316.980.6
325.680.6
334.630.5
343.810.4
353.260.3
362.960.2
372.920.1
383.160.1
393.680.0
404.490.0
415.600.0
427.01-0.1
438.74-0.1
450.79-0.1
463.170.0
475.880.1
488.950.3
502.380.4
516.170.0
530.35-0.1
544.91-0.1
559.870.1
575.250.2
591.040.2
607.270.2
623.950.1
641.080.2
658.680.2
676.770.3
695.350.4
714.450.4
734.070.4
754.220.4
774.940.4
796.210.3
818.080.3
840.540.2
863.620.2
887.340.2
911.700.2
936.740.2
962.460.2
988.890.1
1016.040.0
1043.940.0
1072.61-0.1
1102.06-0.2
1132.33-0.2
1163.42-0.2
1195.37-0.1
1228.19-0.1
1261.91-0.2
1296.57-0.3
1332.17-0.3
1368.75-0.4
1406.34-0.4
1444.95-0.4
1484.63-0.4
1525.40-0.3
1567.28-0.3
1610.32-0.3
1654.54-0.3
1699.97-0.2
1746.650.0
1794.610.2
1843.890.5
1894.530.7
1946.550.8
2000.000.9
2054.920.9
2111.350.8
2169.320.6
2228.890.5
2290.100.4
2352.980.4
2417.590.5
2483.980.7
2552.190.7
2622.270.6
2694.270.5
2768.260.4
2844.270.3
2922.370.4
3002.620.4
3085.070.4
3169.790.3
3256.830.2
3346.260.1
3438.14-0.1
3532.55-0.5
3629.56-0.8
3729.22-0.8
3831.62-0.3
3936.840.5
4044.941.2
4156.011.5
4270.141.4
4387.391.0
4507.870.3
4631.65-0.5
4758.83-1.0
4889.51-1.6
5023.77-1.8
5161.72-1.7
5303.46-1.1
5449.09-0.3
5598.720.3
5752.460.6
5910.420.4
6072.720.0
6239.47-0.2
6410.800.0
6586.840.6
6767.711.1
6953.551.4
7144.491.4
7340.671.4
7542.241.5
7749.351.9
7962.142.4
8180.782.6
8405.422.6
8636.232.0
8873.371.1
9117.030.2
9367.380.3
9624.611.1
9888.891.7
10160.441.6
10439.440.8
10726.10-0.1
11020.63-0.1
11323.250.8
11634.181.0
11953.650.3
12281.890.0
12619.150.4
12965.660.7
13321.690.3
13687.50-0.1
14063.35-0.2
14449.520.0
14846.30-0.5
15253.97-1.1
15672.84-1.1
16103.21-0.7
16545.39-1.4
16999.72-1.7
17466.52-1.4
17946.14-1.7
18438.94-1.3
18945.26-0.6
19465.490.0
20000.003.2

An interesting read.

 

[rcon7]

An interesting read.

 

[Smartplug]

Which curve most user use?

 

[Jozhix]

here you go in txt form. 3.5mm imm6 nonusb calibration txt file.

I have imm6c usb mic, and calibration file is almost flat from 40-10k Hz, with -4dB at the ends (20Hz and 20kHz).
Same with SoundID Reference mic, it is flat from 20-10kHz.
So missing calibration file will most likely not an issue. Positioning your mic just few inches away will affect your measurements more.

 

[canard]

I have imm6c usb mic, and calibration file is almost flat from 40-10k Hz, with -4dB at the ends (20Hz and 20kHz).
Same with SoundID Reference mic, it is flat from 20-10kHz.
So missing calibration file will most likely not an issue. Positioning your mic just few inches away will affect your measurements more.

if the deviations linked to the measurement conditions can of course be essential...the deviations observed here are not negligible...really...it is not class 1-2....must really be taken into consideration...

(flat mic... it's like +-0.5db 20-20k ;-) )

It would be really ""stupid"" in the case of these very modest micro solutions not to use the calibration.... They especially need it...
;-)

 

[Jozhix]

Here we are measuring stereo, not single individual speaker. What good 0 degree calibration file will do, if you point mic at 30 degree angle from speakers anyway? Or this is negligible?
Anyway, I agree that importing calibration file should be implemented, RC is such a powerfull tool.
I still prefer to do multiple measurements in REW, avereage and create EQ filters from there.

 

[canard]

Here we are measuring stereo, not single individual speaker. What good 0 degree calibration file will do, if you point mic at 30 degree angle from speakers anyway? Or this is negligible?
Anyway, I agree that importing calibration file should be implemented, RC is such a powerfull tool.
I still prefer to do multiple measurements in REW, avereage and create EQ filters from there.

from 100 to 10k or even 20-10k...the differences observed at 0 degrees will most certainly be found at 30 degrees... ;-)
invite dayton to propose correction a at 0-30 and 90 degres ;-)
ps: in general often only public 0 and 90dg...

 

[Jozhix]

Here are my calibration files. SoundID mic has also profiles for 30 and 90 degrees.
I would say for room correction (default range 40Hz - 4kHz) they can be used as is, no matter how you point them.

 

[canard]

the gap between your microphone soundid at 0 and 30 is rather small...
(and for the moment the calibration of the normal 6 from Dayton pointed here is in proportions unrelated to what you are presenting... I don't really understand...)
;-)

 

[harkpabst]

the gap between your microphone soundid at 0 and 30 is rather small...
(and for the moment the calibration of the normal 6 from Dayton pointed here is in proportions unrelated to what you are presenting... I don't really understand...)
;-)

I don't understand why you don't understand.

@Jozhix simply demonstrated that the calibration of his Dayton Audio iMM-6C (the USB type) is essential flat +/- 0.5 dB from 40 Hz to 4300 Hz, if we don't care about the absolute level.

Even if you insist on absolute level correctness (which isn't necessary here) it is still flat within ~+0.2/-1dB in the same frequency range.

It's simply good enough, even without the calibration applied, if you stay with in WiiM's default range of correction.

 

[canard]

I don't understand why you don't understand.

@Jozhix simply demonstrated that the calibration of his Dayton Audio iMM-6C (the USB type) is essential flat +/- 0.5 dB from 40 Hz to 4300 Hz, if we don't care about the absolute level.

Even if you insist on absolute level correctness (which isn't necessary here) it is still flat within ~+0.2/-1dB in the same frequency range.

It's simply good enough, even without the calibration applied, if you stay with in WiiM's default range of correction.

in detail it seems despite everything a little more than a +-0.5 if we observe in detail the normal 6 presented...and consider the smoothing adapted in rew... but hey, ok..if the idea is to consider just a 40 -4k... ... be careful the case of the usb does not consider the case of wandering or not on the side of the "adc" if not usb... seen like that, even if personally it is strange for me to consider a measurement microphone on such a narrow band..the dayton usb is rather usable without correction...even if necessarily a bit of a shame...

 

[wiimamp user]

Can I do the measurements with the mic of my smartphone?
It will be good enough I assume.
As a noob on this subject you totally have no Idea what you are doing.
All those graphics..they look nice but I have no clue at all what they mean...
I have some knowledge about audio but all the graphics and what it means no idea.
How to set up a peq or how to select the right amount of Q no idea at all .
Can a noob play with this if he doesn't know what he's doing?

Am I the only one?

 

[Jozhix]

Yes, you can. On Android phones results will vary. And better stay within default range of correction.

 

[wiimamp user]

Yes, you can. On Android phones results will vary. And better stay within default range of correction.

I will thanks.
Gonna try it soon.

 

[harkpabst]

the dayton usb is rather usable without correction...

My take away from your post.

Forget about smoothing. This is not an REW measurement. It's a display of the calibration file, which is very smooth.

 

[bernard23]

Yes, you can. On Android phones results will vary. And better stay within default range of correction.

Personally I would set the max correction to 6dB using the phone mic. That's for the Galaxy S23 Ultra