Audio Dropouts

[BarneyRubble]

Sound artefacts are now also reproduceable.
It happens in the beginning of a jingle but only after jingle has been stopped and restarted again.

[BarneyRubble]

Artefacts are only on button row 1 column 3, sample sent to you by email.

[Couin]

Hi,

About the small delay on changing device is possible, I got it too, but as well we should not often change device during shows, I put this problem as not critical. I will dig to see if I can get better results.

About controller soundcard blocked, I will check about Freeing audio devices procedure.

About device freq @48kHz, I tried of my PC, its soundboard is set to 48 kHz and I get no problem. One year ago, I asked on the BASS (the audio library) forum and I got as answer that even if we initialize the device @44100 Hz, it will take the device native format on Windows Vista and above. I will ask some confirmation.

I got your sample file, thanks. You get this only on the third jingle button of first row?
If yes, this is very strange because all buttons are made as a copy of the first at JP loading (the "Building main interface" step, on the progress bar), so you should either have the problem on any button, either not at all

I changed some code order in the Play function, can you try if you still get part of sample played at the beginning of jingle play, by replacing the jingle palette.exe (rename it to keep a backup) by these included in this zip file? Don't worry about file size difference, I did not compressed it.

[BarneyRubble]

Testing all buttons with new .exe, same thing.
Fragment sounds like the last part of the jingle where it was stopped.

[Couin]

This is strange, because I added (even if I never got this on any machine where I used JP) a Stop event before cueing at start posinio and applying jingle volume and play it.

Code: Select all

51                Call BASS_ChannelStop(Jing(btnum).Strm)
52                Call BASS_ChannelSetPosition(Jing(btnum).Strm, Jing(btnum).CueStart, BASS_POS_BYTE)
53                Call BASS_ChannelSlideAttribute(Jing(btnum).Strm, BASS_ATTRIB_VOL, Velocity / 127, 10)
54                Call BASS_ChannelPlay(Jing(btnum).Strm, BASSFALSE)

Do you get the same thing with this exe? (You have to place it in the JP folder so it requiere bass.dll)

Also, could you try running JP on another cleanest machine? I mean, without a lot of softwares around. Just a simple machine with its soundcard.

[Couin]

Hi,

I made some modifications of code, can you try with the exe included in this zip and see if you still have the part of audio played before the beginning?
Also, dropouts still there? I added a small buffer to the internal mixers.
Honesty, if this problem still present, I don't know where to search more after Perhaps some incomptibilities with your system, that looks, like you wrote, a little bit complicated (I have difficulties to imagine and understand your setup lol).

Second, can you try to reproduce this?

Trying straight to DJ controller sound card no problem.
After closing JP sound card driver was still blocked, had to reset Controller.

I applied some modification about freeing audio devices too.

Third (but less significative), code modifications made changing Audio Device no longer having delay (I has not it anymore, perhaps for you too).

[BarneyRubble]

I will hopefully get to test again by weekend.
This is a short video about my setup.
https://drive.google.com/file/d/1J3-86v ... drive_link

Here is a video how it is used.
https://drive.google.com/file/d/1_hVA_j ... drive_link

[Couin]

I was meaning about a schematic view (I already saw your FB live video ), I ask myself why a lot of softwares (and what about IP connexion, ASIO4ALL too) are needed for a live radio show, myself I use a simpler setup for mines:
2 SL1200
1 TTM57 SL
1 PC for audio broadcasting (with BUTT) for webradio + video broadcasting (with OBS) for video player popup on website chat
1 PC for JP (I also could use the first PC with two monitors, but I don't like multiple monitors, I always loose the mouse pointer AHAH)
1 PC for Serato (for tracks I do not play with real vinyl)
Don't care about small monitor to the left, it was for the previous PC dedicated to audio stream.

If you can try before this week-end, even for a short time, to see if my code modifications are good
Thanks !

[BarneyRubble]

I am DJ since 1978 and a professional in electronics and used to run a complete sound system since the early 90ies. I had my self made console
with 5 channel Dynacord Mixer, 3 1210s and built in monitor speakers, digital live sampler with the technics pitch control fader + bpm display,
effects rack with enhancer, subbass generator, echo, delay etc. + Amiga500 for my efx keybord in the console.
Additionally 2 amp racks and 24 speaker boxes and lots of vinyls and heavy dubplates with aluminium core and used to tour in Europe.



Over the years the heavy load did start give trouble to my spine bone and I startet do dip into digital audio.
My new goal was to emulate my old analog system the digital way. No more cables means much less work to set up and service
and most important no noise at all. The best part of it: Endless VST effect plugins!
I am using no more analog cables except mic cables for the 2*2 mic interface Behringer UMC202HD.

The heart of my emulation is Reaper which is connected to the DJ-Controller via Rearoute ASIO driver.
This opens creativity when it comes to sound design with VSTs and JS-efx.
My efx chain is emulating hardware like Aural Exciter, BBE enhancer, DBX 5BX dynamic expander, Richter Scale and some more.
There is also an Akai MPK mini connected to launch effect samples and to control STEM mute buttons in Traktor.

Now the problem starts. Most DJ controllers, especially my AA VMS4, have a crappy mic input.
It is not possible to add a second ASIO device to the same DAW. Even if you interconnect them via ASIO4ALL it will not work
because clock drift will create audio dropouts and other problems.
The solution is a next session of Reaper for the ASIO Mic interface. Turning down mic input gain on the controller and boost
output gain on the Mic interface hardware was my solution to work around my crappy mic input. Since the mic mix in my
DJ-Controller is analog there is only interface latency which can even set up with other buffer size than the first Reaper session.
This allows mic use with live artists and any kind of all purpose efx like delay, reverb, autotune, gate and a ducker.
The ducking control signal is processed here and forwarded via loopMIDI to the first Reaper session damping the volume there
with very low and not noticeable latency.

Now the elder DJ starting get tired of party and radio comes into the game. A Whatsapp phone line needs to get connected and
a radio jingle player too! Playing media from any internet source would be also a useful feature but no more digital inputs
available.
My solution is a 3rd session of Reaper to control all auxiliary inputs via ASIO4ALL driver and preferred amount of virtual audio cables.
Latency again depends on buffer size and can be set independent from other sessions.
Audio is routed via Reastream plugin to the 1st Reaper session which is causing additional latency.
This is no issue since none of these sources need to be synced with anything, practically I have no problem at all.
Mic is also routed via Reastream to the 3rd Reaper session to feed Whatsapp.
This session is also handling the ducker for the Whatsapp phone line and of course Jingle Palette Reloaded.
Again this Reaper session is also connected to all other sessions via loopMIDI.

All sessions must be set to the same sampling rate to avoid dropouts!

BUTT can be connected to the 1st Reaper session via Rearoute. The stream signal can now be processed differently
from the output and recording signal in Reaper.
I am not using BUTT any more but would run it on the streaming PC if I would need it again.

Part 2 of the radio system is the streaming PC which is quite simple.
Audio signal is routed from the 1st Reaper session on the Laptop to Prism Live Studio on the PC via NDI Tools
Reastream plugin did work too but did cause dropouts once and then.
Midi is received via rtpMIDI to switch cams. The switch signals are created by the first Reaper session on the Laptop
by a midi sequencer and the mic button on the DJ-Controller.
OBS delivers the scenes via NDI Tools to Prism for the green screen background.

I wish I had an easier solution, it took years development and testing on PA systems and takes about 100 plugins
to reach where it is now. Feedback about sound quality is appreciated

Some stylish schematics are here

FB_IMG_1749561450690.jpg (28.12 KiB) Viewed 106 times

Schematic Laptop.jpeg (1.06 MiB) Viewed 106 times

Schematic PC.jpeg (567.17 KiB) Viewed 106 times

[BarneyRubble]

Testing with BASSmini.exe, no sound artefacts!
Testing BASSMini.exe for dropouts: No issue!