Beta Test: Multitone Loopback Analyzer software

[Sokel]

According to the tooltip I am using a logarthmic one:
View attachment 508032

You got me intrigued and I gave it a try the way I wrote:

(scrap the beginning, I was lazy to change the lowest setting)
I use ASIO at Multitone and exclusive Java at REW though, maybe the shared output you use makes it so?

 

[morillon]

So REW isn't doing any internal smoothing? The difference is due to multitone's method of measurement being inherently "noisier"?
(I wasn't trying to measure distortion, just PEQ behaviour).

Between a lack of skills and...the language barrier, I won't be able to help you much ;-)

(MT has the enormous advantage of being able to use two different machines for input and output...which can be very important when used as a headphone measurement bench...
This aspect that MT shares with Virtins is a great strength...
)

 

[Sokel]

...and that's the original Multitone Log Chirp:

 

[IsaacOscar]

A minor bug that was annoying me before: you can't take a capture of a spectrogram, e.g.

This is what it gives me (which is very much not what I wanted):

Obviously I can just use Window's Snipping Tool to take a screenshot of the spectrogram instead, so it's not a big deal.
(However, as you can see above, my HDR monitor does weird things to screenshots...)

 

[pkane]

A minor bug that was annoying me before: you can't take a capture of a spectrogram, e.g.

View attachment 508634
This is what it gives me (which is very much not what I wanted):
View attachment 508635

Obviously I can just use Window's Snipping Tool to take a screenshot of the spectrogram instead, so it's not a big deal.
(However, as you can see above, my HDR monitor does weird things to screenshots...)

Right... I've seen that before but it's been a while. These are two separate plots, and MTA defaults to the scale plot, instead of the main one. I'll see what I can do.

 

[pkane]

Thanks, I'll give this a try. Strange that averages has an effect, as it's using a single average internally... or at least that's what I thought

Didn't have much time to test the new version yet, but just letting you know about progress:

1. Warm-up issue with Log-chirp is fixed, as is average setting
2. DSD log-chirp analysis works, but currently requires the signal generator to match the recording sample rate

 

[IsaacOscar]

Today when I loaded multitone, it didn't find any devices (despite me litterally listening to music through my headphones at the time).

Error message on opening multitone:

Pressing OK and trying to select a device (none are listed):

Then pressing ctrl-D to try and refresh the device lsit:

I ofcourse tried turning my computer off and on again, but that did not fix the issue.

Multitone used to work before; however I recently installed Equalizer APO, Peace and HeSuVi, but I haven't touched it yet, and didn't have Equalizer APO enabled on any devices.

EDIT: the issue has gone away now that I've installed Equalizer APO to a device, removing it from the device does not bring it back.

 

[Guy.Cohen]

Hi Paul, from time to time I am getting this comments, after I click OK, the software is turned of. I have 60GB RAM in this computer.
It happened when I ran 128K, AP32, DSD 512, base 44.1, 2AVG

 

[Guy.Cohen]

Hi Paul, two more questions:
1. I measured noise floor level for my cosmos ADC, no windowing, 128k, 4 avg, now AES, 1hz-20khz.
In REW about 122db in stereo mode and 126db in mono mode (4.5V sensitivity).
With MTA, about 126db and 129db, respectively. It seesm one computation is wrong.

2. In one of my computer that has a strange behaviour (I mentioned once) I got now a message that might help me to understand his problem, but I do not understand its meaning. Could you explain?

Thanks,
Guy

 

[pkane]

Hi Paul, two more questions:
1. I measured noise floor level for my cosmos ADC, no windowing, 128k, 4 avg, now AES, 1hz-20khz.
In REW about 122db in stereo mode and 126db in mono mode (4.5V sensitivity).
With MTA, about 126db and 129db, respectively. It seesm one computation is wrong.

2. In one of my computer that has a strange behaviour (I mentioned once) I got now a message that might help me to understand his problem, but I do not understand its meaning. Could you explain?

Thanks,
Guy

View attachment 510322

Hi Guy,

In REW settings, try to toggle the full scale setting and see if the noise levels match after this:

As to the ASIO error, it appears to come from the NAudio library I'm using to send/receive data from the audio device. It appears there's something not working with ASIO buffer sizes, inferring the problem from the two error messages, it appears some ASIO buffer was not allocated. Could it be an out of memory problem? Is this with a DSD test signal?

 

[Guy.Cohen]

Thanks for your answer.
In my REW the "Full scale sine rms is 0 DBFS" is checked (active). With this the noise floor is about 122.8 to 126.7 between stereo to mono mode. The same cosmos setup with MTA shows 126 to 129. If I unchecked the "Full scale sine ..." both MTA and REW show the same.
So, MTA is presenting the peak value? not RMS 0dbfs ?

 

[Guy.Cohen]

Hi Guy,

In REW settings, try to toggle the full scale setting and see if the noise levels match after this:

View attachment 510366

As to the ASIO error, it appears to come from the NAudio library I'm using to send/receive data from the audio device. It appears there's something not working with ASIO buffer sizes, inferring the problem from the two error messages, it appears some ASIO buffer was not allocated. Could it be an out of memory problem? Is this with a DSD test signal?

Also. yes the error occurred with native DSD test but not something terrible hard. The specific computer is AMD 7 with 60GB RAM.
By the way this computer, something really strange is happening, It is a new one with Win 11. At the beginning everything worked fine but I wanted to test COSMOS ADC with ASIO, hence I installed ComTrue driver from their website. Then I realized I can not change the volume slider (that the reason I noticed the too good noisefloor). I can change the input level only with COSMOS tweak application, not with Windows slider at the 'sound' control panel. So, I decided to roll back to the default windows driver and then the input level slider (of windows) is working but MTA does not want to work with the ADC (now only WASAPI of course). REW working with both drivers.
Now, without ComTrue driver MTA is always 'waiting...'. If I install back the ComTrue driver then is is working but only Tweak slider is active.

Maybe you have an idea?
I have several Windows computers, each one has some problems. I also have 3 MacBook (Late 2011. 2013, 2022) pro and all of them working excellent (no drivers of course).

 

[pkane]

So, MTA is presenting the peak value? not RMS 0dbfs ?

Correct.

 

[pkane]

Also. yes the error occurred with native DSD test but not something terrible hard. The specific computer is AMD 7 with 60GB RAM.
By the way this computer, something really strange is happening, It is a new one with Win 11. At the beginning everything worked fine but I wanted to test COSMOS ADC with ASIO, hence I installed ComTrue driver from their website. Then I realized I can not change the volume slider (that the reason I noticed the too good noisefloor). I can change the input level only with COSMOS tweak application, not with Windows slider at the 'sound' control panel. So, I decided to roll back to the default windows driver and then the input level slider (of windows) is working but MTA does not want to work with the ADC (now only WASAPI of course). REW working with both drivers.
Now, without ComTrue driver MTA is always 'waiting...'. If I install back the ComTrue driver then is is working but only Tweak slider is active.

Maybe you have an idea?
I have several Windows computers, each one has some problems. I also have 3 MacBook (Late 2011. 2013, 2022) pro and all of them working excellent (no drivers of course).

I've had problems with ComTrue drivers and Cosmos ADC before. There were a few different versions of the driver and it took some installing/uninstalling/reinstalling different versions to find one that actually worked well.

 

[Rantapossu]

Also. yes the error occurred with native DSD test but not something terrible hard. The specific computer is AMD 7 with 60GB RAM.
By the way this computer, something really strange is happening, It is a new one with Win 11. At the beginning everything worked fine but I wanted to test COSMOS ADC with ASIO, hence I installed ComTrue driver from their website. Then I realized I can not change the volume slider (that the reason I noticed the too good noisefloor). I can change the input level only with COSMOS tweak application, not with Windows slider at the 'sound' control panel. So, I decided to roll back to the default windows driver and then the input level slider (of windows) is working but MTA does not want to work with the ADC (now only WASAPI of course). REW working with both drivers.
Now, without ComTrue driver MTA is always 'waiting...'. If I install back the ComTrue driver then is is working but only Tweak slider is active.

Maybe you have an idea?
I have several Windows computers, each one has some problems. I also have 3 MacBook (Late 2011. 2013, 2022) pro and all of them working excellent (no drivers of course).

Are you changing the input sensitivity or the mono/stereo setting with the slider?

 

[Guy.Cohen]

Correct.

As I remember the small measurement window (when one can see a summary of the measurement) says RMS (in dbfs) and not peak. Maybe you will change it later or I don't understand something.

 

[Guy.Cohen]

I've had problems with ComTrue drivers and Cosmos ADC before. There were a few different versions of the driver and it took some installing/uninstalling/reinstalling different versions to find one that actually worked well.

Maybe Ivan had old driver version or we wait for a new one.
By the way, same cosmos, same driver doesn't make problems with another Windows 11 computer. Mysterious of Microsoft.

 

[Guy.Cohen]

Are you changing the input sensitivity or the mono/stereo setting with the slider?

To compare noise floor I put the same settings for both softwares.
The ADC was at 4.5V range, not populated XLRs and then one time measure integrated noise (in dbfs) with slider at mono or stereo mode.

I didn't try to test the noise floor of the ADC at all, that the reason I didn't short the inputs and put the 1.7V range. By the way, the COSMOS has so low input impedance that shorting or not or 1.7V or 4.5V yields more or less the same performance.

 

[pkane]

As I remember the small measurement window (when one can see a summary of the measurement) says RMS (in dbfs) and not peak. Maybe you will change it later or I don't understand something.

Different RMS. The setting in REW determines what is considered to be the 0dBFS level (RMS value or peak value of the signal).

For a sine wave, RMS value is 3dB lower than peak, which is the reason you saw a 3dB difference before you changed the setting.

 

[Guy.Cohen]

OK, of course it doesn't change proportion like THD, THD+N.
When considering DBV then we fill the full scale value in dialog. But then this full scale is the rms or peak?
I didn't check but if I put full scale of 1V in REW and MTA.
Then if I put 1Vrms sinewave, both softwares will show the same DBV values? I will check but maybe you know.

Another thing, the measurement of Volt, not DBV. The vertical axes is linear. Of course it is correct but it is not easy to see the low values.
For example in REW, even if of you choose V you see the vertical axes in logarithmic spacing.