VS1053b problem with percussive sound

Designing hardware that use VLSI Solution's devices as the system controller for the entire design.
av500
User
Posts: 14
Joined: Tue 2016-06-14 18:00

Re: VS1053b problem with percussive sound

Post by av500 »

I can confirm the issue with Hand Clap (note 39), it happens here too. the clap is "lost" about 25% of the time, even when it's the only note/sound playing and with a long pause between triggers.

what I hear is that the clap sound varies with each trigger, I assume that is to make it more "natural" sounding. I have no idea if this variation makes it also lose it's sound totally from time to time, but I would prefer a fixed clap that appears 100% of the time.
User avatar
pasi
VLSI Staff
Posts: 2019
Joined: Thu 2010-07-15 16:04

Re: VS1053b problem with percussive sound

Post by pasi »

There are three decays in the handclap, the second and third happening randomly between 0.55s to 0.743s from the beginning of the sound.

It looks like if the random function puts them in the reverse order, the note release field is set earlier than in the other case, so the load check routine which cleans out notes could trigger, but handclap should still always produce at least the first and second decays.
Visit https://www.facebook.com/VLSISolution VLSI Solution on Facebook
av500
User
Posts: 14
Joined: Tue 2016-06-14 18:00

Re: VS1053b problem with percussive sound

Post by av500 »

it does not :( I will try to make a recording tomorrow.
User avatar
pasi
VLSI Staff
Posts: 2019
Joined: Thu 2010-07-15 16:04

Re: VS1053b problem with percussive sound

Post by pasi »

I'll try it myself and try to figure out what the interaction is.

Edit: I have confirmed the issue, but it isn't caused by the order of the random intervals. They have scaling factors so the times are always ascending. I can fix it in the rtmidi application by forcing a fixed seed for the handclap, but I don't yet know what the actual reason for this behavior is.
Visit https://www.facebook.com/VLSISolution VLSI Solution on Facebook
User avatar
pasi
VLSI Staff
Posts: 2019
Joined: Thu 2010-07-15 16:04

Re: VS1053b problem with percussive sound

Post by pasi »

The new version of the rtmidi app for vs1053b is available from http://www.vlsi.fi/en/support/software/ ... tions.html ( http://www.vlsi.fi/fileadmin/software/V ... midi12.zip ).

The Hand Clap has been fixed, let me know if there are any issues with the code.
Visit https://www.facebook.com/VLSISolution VLSI Solution on Facebook
av500
User
Posts: 14
Joined: Tue 2016-06-14 18:00

Re: VS1053b problem with percussive sound

Post by av500 »

Thanks Pasi,

will try to test tonight
av500
User
Posts: 14
Joined: Tue 2016-06-14 18:00

Re: VS1053b problem with percussive sound

Post by av500 »

Great, it works now! thanks for the fix
rolf
User
Posts: 2
Joined: Tue 2022-08-30 15:47

Re: VS1053b problem with percussive sound

Post by rolf »

I tried the patch but for me it still sounds like the cutoff frequency of the CLAPs are changing every time the clap was triggered and often the CLAPs are muted. I would wish a static CLAP or a less changing Filter but never ever a MUTE.
I trigger the VS1053b with old fashion MIDI at pin 26 (RX) of the chip.

pasi wrote: Thu 2016-06-23 12:10 The new version of the rtmidi app for vs1053b is available from http://www.vlsi.fi/en/support/software/ ... tions.html ( http://www.vlsi.fi/fileadmin/software/V ... midi12.zip ).

The Hand Clap has been fixed, let me know if there are any issues with the code.
User avatar
pasi
VLSI Staff
Posts: 2019
Joined: Thu 2010-07-15 16:04

Re: VS1053b problem with percussive sound

Post by pasi »

rolf wrote: Tue 2022-08-30 15:51I tried the patch but for me it still sounds like the cutoff frequency of the CLAPs are changing every time the clap was triggered and often the CLAPs are muted.
You're starting the RT MIDI mode by loading the code every time, right? What are you setting CLOCKF with?

The random variance in the length is just 0..33%. As far as I remember the problem in the ROM code is that the handclap can immediately think that the note-off time is reached and doesn't generate anything. The patched version uses a counter, so should always generate output.

How many notes you have playing at the same time? Could it be that there are too many and thus the handclap is the first to get axed?
Visit https://www.facebook.com/VLSISolution VLSI Solution on Facebook
rolf
User
Posts: 2
Joined: Tue 2022-08-30 15:47

Re: VS1053b problem with percussive sound

Post by rolf »

sorry, absolutely my fault,
I forgot VS_WriteRegister(0x0A, 0, 0x50); so the plugIn could not work at all.
Thank you for your help.
Post Reply