Beta Test: Multitone Loopback Analyzer software

[Rantapossu]

Actually it's the THD which is wavy, the noise floor is pretty flat:

 

[Rantapossu]

Almost 600 points:

 

[Rantapossu]

Could the X and Y values of the point (with the mouse movement over the curve) be printed as black when using the dark theme too?

It's a little hard to read on my not-so-good monitor when the text is white:

Much better with a black ink:

And could the width and height of the capture plot be changed independently of each other?

You could almost match the aspect ratio of the capture by moving the divider between the settings and graph portion to the left or right by trial and error (The red arrow), but it almost never matches exactly. After about 5 or 6 tries I managed to get only 1 pixel too high:

There could be a "Lock the ratio" -selection which locks the aspect ratio as same as the graph and when unlocked, both values could be independend of each other...

 

[Sokel]

Actually it's the THD which is wavy, the noise floor is pretty flat:

View attachment 261110

No,no,that's nice,what I mean waves is that:

 

[Rantapossu]

No,no,that's nice,what I mean waves is that:

View attachment 261159

Oh, it's same for me if I use the analog output too (Green with Direct Monitor on):

I think that is wavy because there is a slight time delay between the buffered normal signal and non-buffered Direct Monitor signal and they are mixed together, which causes a slight echo in the test signal.

It doesn't affect the optical output I use, only the headphone and main analog output.

 

[pkane]

New official version of Multitone, v1.0.71 is available. This incorporates all the changes in preview versions since 1.0.67, and adds a few new features and changes (highlighted in red):

Changes in 1.0.71​

• Add: support for phase specification in test signal format (for example, 1kHz@90)
• Add: ability to unlock aspect ratio when copying plot to clipboard or saving to file
• Add: measurement bandwidth controls (low/high) under Measurement settings to separate measurement from display bandwidths
• Change: better labels selection for linear frequency axis
• Change: add a significant digit to frequency axis labels
• Change: bandwidth selector changed to work consistently across all chart types
• Change: allow mixing of different sampling rates and FFT sizes from history on the same plot
• Change: multitone IMD calculation switched to using CCIF method. Previously the default was SMPTE
• Change: new property section created ‘Device’, with in/out device settings moved to it from Test Signal
• Change: number tracker display always uses black text on light blue background now, regardless of contrast setting
• Fix: certain test signals exceeding 0dBFS previously would result in a missing fundamental, now clipped to 0dBFS
• Fix: IMD, Noise, SNR calculations didn’t work correctly with variable amplitude multitones (e.g., FM signals)

 

[Sokel]

New official version of Multitone, v1.0.71 is available. This incorporates all the changes in preview versions since 1.0.67, and adds a few new features and changes (highlighted in red):

Changes in 1.0.71​

• Add: support for phase specification in test signal format (for example, 1kHz@90)
• Add: ability to unlock aspect ratio when copying plot to clipboard or saving to file
• Add: measurement bandwidth controls (low/high) under Measurement settings to separate measurement from display bandwidths
• Change: better labels selection for linear frequency axis
• Change: add a significant digit to frequency axis labels
• Change: bandwidth selector changed to work consistently across all chart types
• Change: allow mixing of different sampling rates and FFT sizes from history on the same plot
• Change: multitone IMD calculation switched to using CCIF method. Previously the default was SMPTE
• Change: new property section created ‘Device’, with in/out device settings moved to it from Test Signal
• Change: number tracker display always uses black text on light blue background now, regardless of contrast setting
• Fix: certain test signals exceeding 0dBFS previously would result in a missing fundamental, now clipped to 0dBFS
• Fix: IMD, Noise, SNR calculations didn’t work correctly with variable amplitude multitones (e.g., FM signals)

Square,triangle and sawtooth don't work.
That's what I tested so far.

 

[pkane]

Square,triangle and sawtooth don't work.
That's what I tested so far.

Looks like I broke this when I added phase angles. I'll fix ASAP.

 

[Blumlein 88]

Square,triangle and sawtooth don't work.
That's what I tested so far.

That did not take long.

 

[pkane]

Square,triangle and sawtooth don't work.
That's what I tested so far.

Should be fixed now. Give it a few minutes for the website download to update, then try again

 

[Sokel]

That did not take long.

I have an intuition for the least used and usually forgotten ones

 

[pkane]

I have an intuition for the least used and usually forgotten ones

Your intuition is amazing!

 

[Sokel]

Yep.

 

[Rantapossu]

Min Frequency seems to be stuck at zero...

 

[pkane]

Min Frequency seems to be stuck at zero...

View attachment 261828

Actually, it shouldn't even be there. It's now automatically set to 0, and can't be changed. Instead, you can control the portion of the display you want to see with the freq controls above the plot. I'll remove it (it's already gone in the Settings dialog).

 

[DonH56]

Min Frequency seems to be stuck at zero...

View attachment 261828

You want it to go negative?

 

[pkane]

You want it to go negative?

Hmm, I can do that...

 

[DonH56]

Hmm, I can do that...

Cool, I could really use a time machine...

 

[Rantapossu]

• Add: ability to unlock aspect ratio when copying plot to clipboard or saving to file

Thanks, this feature is super!

Unfortunately there's a small bug, MT forgots the state of the aspect ratio lock when you close and reopen the capture window and it starts always locked, so it calculates the aspect ratio wrong. You'll have to unlock the aspect ratio and insert the right values again to match the previous capture.

At the first time, when the values are correct:

After shutting and reopening the window:

Now I'm a little greedy, but there could even be a drop down menu for the most common sizes:

Paper: US Letter
Paper: US Legal
Paper: A3
Paper: A4
Screen: 3840x2160
Screen: 1920x1080
Screen: 1280x720
etc...

The paper sizes can be calculated easily, A4 is for example 3508 x 2480 pixels.
(Link: https://www.adobe.com/uk/creativecloud/design/discover/a4-format.html )

Of course the title, the axis texts and the distortion panel might be too small, because they doesn't scale to the larger resolutions, but for example by halving the values to 1754 x 1240 pixels makes everything to look ok if you are using the A4 size. If you keep the horizontal resolution as 1754 pixels, you can easily calculate the US vertical resolutions to match it.

Or you can use the following calculator and use 150 PPI:

US Paper Sizes to Pixels

pixelsconverter.com

The calculator says that the US Letter is 1648 x 1276 pixels and US Legal 2102 x 1276 pixels when using the 150 PPI (Same as A4 above).

 

[pkane]

Thanks, this feature is super!

Unfortunately there's a small bug, MT forgots the state of the aspect ratio lock when you close and reopen the capture window and it starts always locked, so it calculates the aspect ratio wrong. You'll have to unlock the aspect ratio and insert the right values again to match the previous capture.

At the first time, when the values are correct:

View attachment 261850

After shutting and reopening the window:

View attachment 261853


Now I'm a little greedy, but there could even be a drop down menu for the most common sizes:

Paper: US Letter
Paper: US Legal
Paper: A3
Paper: A4
Screen: 3840x2160
Screen: 1920x1080
Screen: 1280x720
etc...

The paper sizes can be calculated easily, A4 is for example 3508 x 2480 pixels.
(Link: https://www.adobe.com/uk/creativecloud/design/discover/a4-format.html )

Of course the title, the axis texts and the distortion panel might be too small, because they doesn't scale to the larger resolutions, but for example by halving the values to 1754 x 1240 pixels makes everything to look ok if you are using the A4 size. If you keep the horizontal resolution as 1754 pixels, you can easily calculate the US vertical resolutions to match it.

Or you can use the following calculator and use 150 PPI:

US Paper Sizes to Pixels

pixelsconverter.com

The calculator says that the US Letter is 1648 x 1276 pixels and US Legal 2102 x 1276 pixels when using the 150 PPI (Same as A4 above).

The lock by default was by design, although I can see that it might be useful to remember the last state.

Different standard paper sizes, on the other hand, doesn't seem all that useful to me, since printing a full page is likely not how these plots will be commonly used. The "locked" state allows you to preserve the aspect ratio of the plot as it is currently on the screen, and apply a scaling factor to it, if desired to scale it up or down. Anything beyond that seems to me to be a custom setting, better left up to the user.