Beta Test: Multitone Loopback Analyzer software

[pkane]

The FR sweep is interesting

View attachment 224313

Once you get to about 1/2 the measured bandwidth, most of the harmonics fall outside that bandwidth. Normally, sawtooth has too many tones to measure IMD (maximum of 5 are currently supported by MT). As the result, IMD is only measured when you get close to the limit, when only a few of the sawtooth harmonics are included in the measurement bandwidth.

 

[Sokel]

Ok,I got to 192 since the harmonics are way up,ok.

 

[pkane]

Is it the nature of the signal not to work over 7K? (tested it)

Pay attention to your measurement range. If you have doubts, look at the spectrum plot to see what's included.

Sawtooth includes all the harmonics of the main signal up to the specified bandwidth. The second harmonic of 7k is at 14k, the third is 21k, which is outside your measurement range of 20k.

 

[Sokel]

If that's correct is a nightmare...

 

[pkane]

If that's correct is a nightmare...


View attachment 224314

You specified 90k as the highest multitone frequency. Your sampling rate is only 96k, so you've exceeded the allowed frequencies in the generated input signal by 42kHz. The maximum allowed by Nyquist is 48k, not 90k. Set High Multitone to 1/2 the sampling rate or less to stay within the bounds allowed by the sampling theorem.

 

[Sokel]

You specified 90k as the highest multitone frequency. Your sampling rate is only 96k, so you've exceeded the allowed frequencies in the generated input signal by 42kHz. The maximum allowed by Nyquist is 48k, not 90k. Set High Multitone to 1/2 the sampling rate or less to stay within the bounds allowed by the sampling theorem.

(bookmark)

 

[Grooved]

Wanted to ask your help to test MT output to a multi-channel device. This should work with WASAPI and possibly ASIO drivers, outputting to channels other than just left and right...

Hi @pkane , this is great news!
I only did a quick test with ASIO and it shows all inputs and outputs (22) on the multichannel interface I had at hand, but:
- outputs looked mapped like this: L = L Main, R = R Main, 1 = L Main, 2 = R Main, 3 = Analog1 4 = Analog2,... but once I selected 9 (should have been Analog7), a bug appeared:

Then tried 10 input, and MT crashed.

- regarding inputs: so I used one of the L to 8 outputs to send signal, OK, and the signal was coming back on the inputs, OK (saw signal on interface panel), but I didn't find any input in MT that was receiving it, I only had the noise back, so mapping was incorrect

 

[Rantapossu]

It would be perfect if there would be a unique drop down setting for all the tabs:

Spectrum Settings
Freq Response Settings
Waveform Settings
Level Sweep Settings
Freq Sweep Setting

On the "Freq Response Settings" and "Waveform Settings" could at least be the "Title" and "Font Size" options.


Could you also add the default "title text" for all the tabs like this (The are blank, except the Spectrum tab, where it's "Spectrum"):

Saves some time when you dont have to write all the text again if you add for example the sound card name...


I would also like if the bitrate settings on the main page windows were in the same order than on the settings page:

Not like this...

...but like this!

Trips me every time

 

[pkane]

Hi @pkane , this is great news!
I only did a quick test with ASIO and it shows all inputs and outputs (22) on the multichannel interface I had at hand, but:
- outputs looked mapped like this: L = L Main, R = R Main, 1 = L Main, 2 = R Main, 3 = Analog1 4 = Analog2,... but once I selected 9 (should have been Analog7), a bug appeared:
View attachment 224318

Then tried 10 input, and MT crashed.

- regarding inputs: so I used one of the L to 8 outputs to send signal, OK, and the signal was coming back on the inputs, OK (saw signal on interface panel), but I didn't find any input in MT that was receiving it, I only had the noise back, so mapping was incorrect

Thanks, @Grooved! As I said, input multichannel support isn't implemented yet, so no surprise any channel above 2 didn't work.

Output errors above channel 8 are a bit strange. I'm not sure if I can reproduce these here, but I'll try. What device and driver were you using to test with 22 channels?

 

[pkane]

It would be perfect if there would be a unique drop down setting for all the tabs:

Spectrum Settings
Freq Response Settings
Waveform Settings
Level Sweep Settings
Freq Sweep Setting

On the "Freq Response Settings" and "Waveform Settings" could at least be the "Title" and "Font Size" options.


Could you also add the default "title text" for all the tabs like this (The are blank, except the Spectrum tab, where it's "Spectrum"):
View attachment 224326
Saves some time when you dont have to write all the text again if you add for example the sound card name...


I would also like if the bitrate settings on the main page windows were in the same order than on the settings page:

Not like this...
View attachment 224327

...but like this!
View attachment 224329

Trips me every time

I plan to add the properties for all the plots, I just focused on the ones that had more custom/specific settings, first.
I'll fix the order, not sure how these came out differently, but I assume it has to do with the labels being different.

 

[Grooved]

Thanks, @Grooved! As I said, input multichannel support isn't implemented yet, so no surprise any channel above 2 didn't work.

Output errors above channel 8 are a bit strange. I'm not sure if I can reproduce these here, but I'll try. What device and driver were you using to test with 22 channels?

OK, I missed that input part
I was using the only multichannel interface I keep at home, an old MOTU but working good with ASIO in general. I will try with other audio interfaces soon and let you know

 

[Rantapossu]

As I said, input multichannel support isn't implemented yet, so no surprise any channel above 2 didn't work.

Any input channel above 2 shouldn't work, but how about the input channels 1 and 2? Should they be comparable to L and R inputs?

A single RCA-cable between left output and left input, the right channel disconnected:
Output 1 -> Input 1 works (It should, there is a cable)
Output 1 -> Input 2 works! (It shouldn't, there is no cable!)

A single RCA-cable between right output and right input, the left channel disconnected:
Output 2 -> Input 1 doesn't work (It shouldn't, there is no cable)
Output 2 -> Input 2 doesn't work! (It should, there is a cable!)

Everything works, if I use the L->L and R->R channels and not the new channel numbers. Is there some kind of routing bug on the input channel mapping? I'm using E-MU 0404 USB with ASIO.

All output channels 1-4 works as expected, 3 and 4 being SPDIF outputs.

 

[Blumlein 88]

Clipping is almost certainly intersample-overs. Check that the blue waveform on the Wave plot doesn't exceed the bounds from -1 to 1. If it doesn't, but the recording clips anyway, it's due to the intersample overs. This is possible even with the same in/out rates or synch'ed clocks, due to internal oversampling by the DAC, but even more likely when two independent clocks are used (DAC and ADC).

You broke it good in that one I'll check, but I'm pretty sure the sawtooth test signal is fine. The problem is with the horizontal (time) scale on the waveform plot -- it appears to scale everything to the recording sampling rate, even though the blue waveform is generated for the output sampling rate. Only an issue with the plot itself, and only when the input/output sampling rates differ. I'll fix it!

Yes, the generated waves stay within 1 and -1 so it isn't that. I get the same thing with square waves, but not triangle waves. It looks on the square waves like ringing in the filter is what is causing it to clip.

I knew the signals had to be fine and it is just a graphing issue. It may have been going on in recent versions, but I didn't check. In any case currently if the two devices are at different sample rates the graph of the recorded waveform scales in that ratio. Does it for sawtooth, square, triangle and sine wave. Also in the twin tone CCIF. I've not checked others, but it looks to be a global bug right now.

 

[pkane]

Yes, the generated waves stay within 1 and -1 so it isn't that. I get the same thing with square waves, but not triangle waves. It looks on the square waves like ringing in the filter is what is causing it to clip.

I knew the signals had to be fine and it is just a graphing issue. It may have been going on in recent versions, but I didn't check. In any case currently if the two devices are at different sample rates the graph of the recorded waveform scales in that ratio. Does it for sawtooth, square, triangle and sine wave. Also in the twin tone CCIF. I've not checked others, but it looks to be a global bug right now.

View attachment 224345

The issue is with the waveform plot when using two different sampling rates. Since blue plot is of the test waveform at the output sample rate, and white is the recorded signal at input rate, the scales for the two are not the same and need to be adjusted. I don't think this ever worked correctly since I added support for different rates (but works correctly if the same rates are used).

 

[Blumlein 88]

The issue is with the waveform plot when using two different sampling rates. Since blue plot is of the test waveform at the output sample rate, and white is the recorded signal at input rate, the scales for the two are not the same and need to be adjusted. I don't think this ever worked correctly since I added support for different rates (but works correctly if the same rates are used).

Yes, it doesn't effect anything measurement-wise. I just hadn't noticed it before now.

 

[pkane]

Any input channel above 2 shouldn't work, but how about the input channels 1 and 2? Should they be comparable to L and R inputs?

A single RCA-cable between left output and left input, the right channel disconnected:
Output 1 -> Input 1 works (It should, there is a cable)
Output 1 -> Input 2 works! (It shouldn't, there is no cable!)

A single RCA-cable between right output and right input, the left channel disconnected:
Output 2 -> Input 1 doesn't work (It shouldn't, there is no cable)
Output 2 -> Input 2 doesn't work! (It should, there is a cable!)

Everything works, if I use the L->L and R->R channels and not the new channel numbers. Is there some kind of routing bug on the input channel mapping? I'm using E-MU 0404 USB with ASIO.

All output channels 1-4 works as expected, 3 and 4 being SPDIF outputs.

Output channels 1 and 2 are exactly the same in Multitone as L and R -- this shouldn't make any difference if you pick L as output or 1.

Don't use anything other than L or R on the input selector, as that's not implemented yet.

 

[Rantapossu]

Output channels 1 and 2 are exactly the same in Multitone as L and R -- this shouldn't make any difference if you pick L as output or 1.

Don't use anything other than L or R on the input selector, as that's not implemented yet.

Yep, all input channels 1-4 maps as L and nothing maps as R. I just found it like 10 minutes before you replied

I'll stick with L and R. Thanks!

 

[Rantapossu]

Now I'm getting greedy, but would it be possible to add Import and Export functions?

When you select a certain measurement from the history windows and press "Export" -button, it would ask for a file name and a location and saves the measurement as a file. And if you press "Import" -button, it would ask for a location and a file name and imports previously saved measurement to be the last measurement as it would be the last measurement you did with a sound card.

It would be fun to have some contest (but not too serious) too about who has the best equipment and TD+N...

I would really like to get for example your previous measument to drool and to compare:

 

[Rantapossu]

If you want to add the Import/Export -function, it could save the key settings (FFT size, avarages, sample rates etc) to the ouput file and when you import it, MT would asks something like "Do you want to replace your (certain) settings (Y/N)?".

So when you want to reproduce your old or some else's measurement, it would be easy: Import -> Yes -> Record.

We could even have the standardized competition settings for the "TD+N king class" (Virtual / digital loopbackers and other cheaters will be disqualified...)

 

[Sokel]

If you want to add the Import/Export -function, it could save the key settings (FFT size, avarages, sample rates etc) to the ouput file and when you import it, MT would asks something like "Do you want to replace your (certain) settings (Y/N)?".

So when you want to reproduce your old or some else's measurement, it would be easy: Import -> Yes -> Record.

We could even have the standardized competition settings for the "TD+N king class" (Virtual / digital loopbackers and other cheaters will be disqualified...)

Let the others go for a single freq,we will go for the multitone King!
(Or one of our own ,that takes no prisoners )