Strange results from ‘test convolution’

Post Reply
Delta-mike
Posts: 5
Joined: 21 Aug 2021 03:11

Strange results from ‘test convolution’

Post by Delta-mike » 23 Oct 2021 05:33

I have followed all the steps in order, but when I test convolution the frequency response has not changed.

I have Mitch’s book, and followed the both of the online guides. I have tried many (20) different excess phase and other changes and it’s always the same result.

My target curve is fine. I really can’t understand what I’m doing wrong.

Very grateful for any help! Thanks, Dion
Delta-mike
Posts: 5
Joined: 21 Aug 2021 03:11

Re: Strange results from ‘test convolution’

Post by Delta-mike » 23 Oct 2021 06:37

Measurement:
Attachments
Measurement.jpg
Measurement.jpg (635.56 KiB) Viewed 1537 times
Delta-mike
Posts: 5
Joined: 21 Aug 2021 03:11

Re: Strange results from ‘test convolution’

Post by Delta-mike » 23 Oct 2021 06:39

Here are my screen grabs
Attachments
Inversion.jpg
Inversion.jpg (768.7 KiB) Viewed 1537 times
Filter generation.jpg
Filter generation.jpg (793.98 KiB) Viewed 1537 times
Target.jpg
Target.jpg (653.24 KiB) Viewed 1537 times
Delta-mike
Posts: 5
Joined: 21 Aug 2021 03:11

Re: Strange results from ‘test convolution’

Post by Delta-mike » 23 Oct 2021 06:40

And my result after testing the convolution and applying smoothing
Delta-mike
Posts: 5
Joined: 21 Aug 2021 03:11

Re: Strange results from ‘test convolution’

Post by Delta-mike » 23 Oct 2021 06:41

After correction
Attachments
After correction.jpg
After correction.jpg (634.1 KiB) Viewed 1537 times
Ecwl
Posts: 48
Joined: 19 Feb 2021 01:05

Re: Strange results from ‘test convolution’

Post by Ecwl » 23 Oct 2021 13:27

I am a bit confused. Usually when you run Macro 5, the test convolution result is called Test1 & Test 2. But I see that in your display, they’re called Pulse96Lmp and Pulse96Rmp but those are usually the files you get after prefilter+original measurement+smoothing.
Moreover, if you’re to save the test convolution result (corrected frequency response), and you want to apply psychoacoustic smoothing to it, you have to empty curves 3-6 whereas your screen capture leaves the curves 3-6 intact from Macro 5.
I wonder if you just loaded the wrong files to be displayed.

As a total aside, I always think 6/6 and pre-ringing 2 as way too aggressive. I’m surprised you don’t have group delay above 100Hz (or maybe you do and was going to work on it after you figured the test convolution part out).
User avatar
UliBru
Posts: 445
Joined: 19 Oct 2019 13:58

Re: Strange results from ‘test convolution’

Post by UliBru » 23 Oct 2021 15:38

The pictures show a working samplerate of 48000 but Pulses with samplerate 96000. This does not make sense.
And yes, the pictures do not show the test convolution result. Usually the result is displayed unsmoothed. If you like to apply macro1 on the test pulses you have to switch the workspace to subfolder \Testconvolution and run macro1 once again
Stay well tuned
Uli

Moderator
Acourate system: JRiver/Roon -> AcourateConvolver -> miniDSP U-DIO8 -> TacT M/S2150 amps -> DIY horn speakers
Post Reply