Why so strange plots in Audacity

[Balle Clorin]

This is a common stepped sine sweep from Ultimate Analog test record form Analog productions, track A6, recorded with RIAA active
Why does the plot look normal in CoolEditPro2=Adobe audition but ro ragged and sloped in Audacity-Right side. What am I doing wrong, many people use Audacity but I find it to be quite useless, Any comments to help me onto the right track with Audacity?

In REW it looks like this

In Mutitone Analyzer

 

[Apesbrain]

Below the Audacity spectrogram plot there are some settings not shown in your screenshot. Try some different options.

 

[Balle Clorin]

Did not make any improvement

Pink noise looks more similar ...but still the stupid axes control

Not a very linear pink noise track from Ultimate Analog....above 1k is true ,below not

 

[danadam]

Audacity is very bad at scaling its spectrum plot. Making the window wider usually helps somewhat:

 

[Aero]

Audacity does a great job of showing the amplitude of unwanted resonance down to 20 Hz.
Here are two 20Hz-20kHz sweep signals used to determine the frequency response of a phono cartridge. CA-STR-1007 (1A/1B)
Left Channel/Right Channel.
As you can see, the amplitude is different, so something is wrong with the tonearm-cartridge setup.
Some kind of unwanted resonance suppression system would have to be used because the amplitude is very high.
The ideal situation would be such that there is no resonance in the range up to 20 Hz, but unfortunately this is impossible.

 

[Blumlein 88]

If you dumped the FFT into a txt file which is an option you might find it has similar results in Audacity. It is pretty terrible at its display. As someone said try and widen it. I think you'll find if you put the cursor on each peak it reads much higher than is displayed. It is likely the actual readings are even, but don't display that way. So the readings are there they just display very poorly. I don't know why this hasn't been fixed after so many years.

Now even with its faults there was a long time when no other free software did what Audacity would do. For purposes of testing Multitone and REW are much better. I still use Audacity for quick checks or some editing of music, but REW and Multitone for any serious analyzing of test tones or test sweeps.

 

[Balle Clorin]

Thanks, your comment confirms to me that Audacity is crap and useless as I thought. Totally irrelevant and misleading presentation of data.

 

[Aero]

Thanks, your comment confirms to me that Audacity is crap and useless as I thought. Totally irrelevant and misleading presentation of data.

Plot Spectrum - Audacity Manual

manual.audacityteam.org

 

[Balle Clorin]

Audacity does a great job of showing the amplitude of unwanted resonance down to 20 Hz.
Here are two 20Hz-20kHz sweep signals used to determine the frequency response of a phono cartridge. CA-STR-1007 (1A/1B)
Left Channel/Right Channel.
As you can see, the amplitude is different, so something is wrong with the tonearm-cartridge setup.
Some kind of unwanted resonance suppression system would have to be used because the amplitude is very high.
The ideal situation would be such that there is no resonance in the range up to 20 Hz, but unfortunately this is impossible.
View attachment 404361

View attachment 404362

Can you please share the file so I can plot it in another program? The Audacity presentation is very misleading , one problem is the slope due to the sweep nature, but that can possibly be corrected by a 10db/obtave eq/calibrationt (can that be done in Audacity?), ,but worse are the peaks that are not existing in reality...

THIS is the way the CA TRS-1007 plot should look like, notice the difference?

 

[Aero]

The same as before in Audacity, now in Adobe Audition

 

[Balle Clorin]

Yes Audacity does something that makes wobbles, here is s digital sweep compared. How de we get rid of the wobbles add also make the plot horisontal in Audacity?

 

[Aero]

File_1.wav

drive.google.com

File_2.wav

drive.google.com

As many results as programs.
Each one different

 

[Balle Clorin]

which cartridge and arm/tt?

 

[Aero]

which cartridge and arm/tt?

Unfortunately, I don't remember

 

[danadam]

How de we get rid of the wobbles ...

One way is to increase fft window overlap, but Audacity doesn't offer such option. Here's in REW, 10 seconds log sweep, 50% vs 75% overlap:

... add also make the plot horisontal in Audacity?

Use linear sweep instead of log sweep.

 

[Balle Clorin]

Thanks guys!

 

[Balle Clorin]

@Aero , here is your file

and audacity same file

 

[Blumlein 88]

As far as displaying if you'll use a lower size FFT Audacity will look better. Its problem is Pixel matching the data. That is also why the wider you make the window size the better it does.

 

[JP]

Can you please share the file so I can plot it in another program? The Audacity presentation is very misleading , one problem is the slope due to the sweep nature, but that can possibly be corrected by a 10db/obtave eq/calibrationt (can that be done in Audacity?), ,but worse are the peaks that are not existing in reality...

THIS is the way the CA TRS-1007 plot should look like, notice the difference?
View attachment 404411

Turn off your smoothing and see what happens.

 

[Balle Clorin]

you mean smoothin on my file?. her with and without smoothing, files made horisontalby 10db/decade adjustment

My file and Aero's file unsmoother smoothed