Beta Test: Multitone Loopback Analyzer software

[Hee]

I tried again at 1.0.7.
dac : e50 trs out, 0.0db (max vol)
adc : cosmos 3.5v input, 2ch mode (my cosmos is 1ch 128.7dba, 2ch 125.7dba snr)
generator : multitone analyzer -1db (ver : 1.0.7)

The output was multitone analyzer, and REW only executed records.
Both were set to 256k, dbr, and dbc.

 

[pkane]

I tried again at 1.0.7.
dac : e50 trs out, 0.0db (max vol)
adc : cosmos 3.5v input, 2ch mode (my cosmos is 1ch 128.7dba, 2ch 125.7dba snr)
generator : multitone analyzer -1db (ver : 1.0.7)

The output was multitone analyzer, and REW only executed records.
Both were set to 256k, dbr, and dbc.View attachment 168714View attachment 168715

Hold on for a day or two. I'm making some changes to make TD+N computation more consistent by eliminating some of the more resolving/steeper FFT Windows, such as Kaiser. These windows have the extra resolution in the frequency domain, but significantly reduce resolution in time domain especially for very short recordings, affecting computations. The removal of these windows allowed me to simplify the logic quite a bit, so I'm hopeful this will help produce consistent results that are much closer to those expected.

 

[trl]

Please don't forget to add the feature of allowing external generator, at least for few sines like 100 Hz, 1 kHz or what you think it would be best. This way we can compare REW, ARTA, Multitone from the same external audio source and all figures should match perfectly. Many thanks in advance!

 

[Kessito]

Awesome piece of software, thanks!
A great addition would of course be an ability to make a level vs imd measurement like Amir does in his reviews (something REW sadly cant do )

 

[pkane]

Awesome piece of software, thanks!
A great addition would of course be an ability to make a level vs imd measurement like Amir does in his reviews (something REW sadly cant do )

Just what I was thinking yesterday

 

[pkane]

Based on feedback (@Hee , @trl ) here's a redone version 1.0.8 of Multitone.

Changes in 1.0.8​

• Changed: removed FFT windows that don’t work well with short time-domain recordings (Kaiser, Taylor, Gauss, etc.) Default window is now Blackman Harris 7
• Changed: simplified TD+N processing to produce more consistent results at different FFT sizes
• Added: support for using external WAV files as the test signal
• Added: setting to automatically resample external test WAV files to current output rate
• Changed: rearranged UI to accommodate longer test tone name/description

This is a significant update/rewrite, so please test well and report any issues!

To select an external WAV file as the test signal, click on Test Signal near top left, or pick "WAV File..." from the signal drop-down. Ideally, the test file should be the same sampling rate as the input/output of Multitone. If it's not, it will be resampled internally to the correct rate. Test file should contain at least as many samples as twice the FFT-size. More is OK, less may result in some artifacts in measurements.

Multitone will detect as many individual frequency tones as it can find and will process the recorded loopback by applying notch filters to remove these. Here's an example of a square wave generated in REW used for measurement using Multitone. Note that all the original test signal harmonics are notched out (white trace), just like they would be if the signal was generated by Multitone:

 

[pkane]

Posted with no explanation...

 

[Kessito]

you have just climbed 6 steps on the awesome stairs, level 3 accomplished !

 

[pkane]

Version 1.0.9 is now available.

The main new attraction is the IMD and TD+N vs. level sweep measurement. To activate it:

1. Select the desired test signal and frequency from/to for measurements. SMPTE 60Hz/7kHz 4:1 is normally used for this, but any 1, 2, 3, 4, or 5 tone signal should work (only TD+N will be measured with 1 tone or more than 5)

2. Check the sweep settings:

3. Select Sweep measurement from the drop-down next to the red record/measure button:

4. Click the red button and wait for the result:

 

[Hee]

Thank you for the good program.
I tried imd sweep.

Even if the filter of e50 is changed, the hook of imd is the same.

 

[Hee]

This is a screen of -30db and -27db with hooks.

Pkane, I have a favor to ask of you.
My computer is old, so 1280 resolution is the best.
It's hard to enjoy because the screen is cut off and the settings are not visible.
I'm sorry my computer was shit. ^^

 

[pkane]

This is a screen of -30db and -27db with hooks.

View attachment 169513

Pkane, I have a favor to ask of you.
My computer is old, so 1280 resolution is the best.
It's hard to enjoy because the screen is cut off and the settings are not visible.
I'm sorry my computer was shit. ^^

I'll try to size it a bit smaller, although on my screen it seems to be less than 1200 pixels horizontally. Of course, you can also resize the window yourself to make it a little smaller

 

[Blumlein 88]

New version works fine for me. This for a Topping D10 balanced and Babyface Pro FS.

 

[Grooved]

New version works fine for me. This for a Topping D10 balanced and Babyface Pro FS.

View attachment 169572



View attachment 169574

View attachment 169575


View attachment 169576

It looks to be working for me too. Did the same tests with Tone2 Pro and Cosmos ADC.
I did it with 131072 FFT size, but I think that for MT32 (AP) Amir is using 262144 and 8 averages. I don't know what he's using for the 1k Sine.
@amirm can you tell us if you average it and how many time please ?

 

[Blumlein 88]

It looks to be working for me too. Did the same tests with Tone2 Pro and Cosmos ADC.
I did it with 131072 FFT size, but I think that for MT32 (AP) Amir is using 262144 and 8 averages. I don't know what he's using for the 1k Sine.
@amirm can you tell us if you average it and how many time please ?

View attachment 169717

View attachment 169718

View attachment 169719

View attachment 169720

That Cosmos ADC looks to be working really good. I think my measurements are bottlenecked a bit by the BF ADC.

 

[Grooved]

That Cosmos ADC looks to be working really good. I think my measurements are bottlenecked a bit by the BF ADC.

It's certainly the case, you should get better results because the D10b has the same DAC chip than the Tone2 Pro, are both balanced, and has almost the same measurement on ASR reviews.
The D10b is even a little bit better, but I think Amir did not update the Tone2 Pro after the firmware update fixed the small problem on the first one :

Topping D10 Balanced Review (USB DAC)

This is a review and detailed measurements of the new Topping D10 Balanced. It was sent to me by the company and costs US $139. The D10 Balanced as the name implies, comes in my favorite USB DAC package with the same cute orange display: Back panel is where the difference is: Balanced...

www.audiosciencereview.com

Khadas Tone2 Pro Review (DAC & Amp)

This is a review and detailed measurements of the Khadas Tone2 Pro portable USB DAC and balanced headphone amplifier. It was sent to me by the company for testing. As configured, it sells for $199 from the company website direct. From the pictures, the unit didn't look all that nice but in...

www.audiosciencereview.com

That said, did you tried to set the Gain on OdB in Multitone Analyzer ? Is your D10b producing a lot more distortion ? this what I get, and maybe @pkane can confirm something on it : if that's the case, does't it mean that no audio should be sent to the DAC reaching OdB ?

 

[Blumlein 88]

It's certainly the case, you should get better results because the D10b has the same DAC chip than the Tone2 Pro, are both balanced, and has almost the same measurement on ASR reviews.
The D10b is even a little bit better, but I think Amir did not update the Tone2 Pro after the firmware update fixed the small problem on the first one :

Topping D10 Balanced Review (USB DAC)

This is a review and detailed measurements of the new Topping D10 Balanced. It was sent to me by the company and costs US $139. The D10 Balanced as the name implies, comes in my favorite USB DAC package with the same cute orange display: Back panel is where the difference is: Balanced...

www.audiosciencereview.com

Khadas Tone2 Pro Review (DAC & Amp)

This is a review and detailed measurements of the Khadas Tone2 Pro portable USB DAC and balanced headphone amplifier. It was sent to me by the company for testing. As configured, it sells for $199 from the company website direct. From the pictures, the unit didn't look all that nice but in...

www.audiosciencereview.com

That said, did you tried to set the Gain on OdB in Multitone Analyzer ? Is your D10b producing a lot more distortion ? this what I get, and maybe @pkane can confirm something on it : if that's the case, does't it mean that no audio should be sent to the DAC reaching OdB ?

Let me give that a try. Earlier versions at 0 db did give more distortion. I currently use - 1 db from the D10b. The output level won't get to 0 dbFS in the ADC unless I use just a little gain.

 

[Grooved]

Let me give that a try. Earlier versions at 0 db did give more distortion. I currently use - 1 db from the D10b. The output level won't get to 0 dbFS in the ADC unless I use just a little gain.

I got the impression that it was not the ADC producing more distortion, but the DAC itself, because even if I lower the ADC input, it still produce more distortion.

 

[MakeMineVinyl]

I just downloaded and ran it - there is no display area, just the upper controls. I'm using Windoze 11.

 

[Grooved]

@pkane
Just tried the same thing than above but setting the Tone2 Pro via ASIO, and it doesn't work now with 1.0.9 version