MidiLink 2.0 looking for testers. (no MIDI devices required)

https://github.com/MiSTer-devel/Main_MiSTer/wiki

Moderators: Mug UK, Zorro 2, Greenious, spiny, Sorgelig, Moderator Team

BBond007
Captain Atari
Captain Atari
Posts: 402
Joined: Wed Feb 28, 2018 3:23 am

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby BBond007 » Fri Jan 18, 2019 9:55 pm

Sorgelig wrote:
BBond007 wrote:Tomorrow I will start updating the WIKI reflecting the new features.

it's better to move Midilink.ini to /media/fat/linux folder. Config folder is usually cluttered with many files. Config folder is mostly for automatically created configs not requiring manual editing.


OK, will change it

NegSol
Captain Atari
Captain Atari
Posts: 326
Joined: Sat Dec 05, 2015 9:22 pm

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby NegSol » Fri Jan 18, 2019 10:50 pm

BBond007 wrote:
NegSol wrote:Ok - I tried to get midi running but failed. No "internal" midi sound. Guess I will have to wait until you update the wiki. It looks like everything is running but I can see in htop that multiple instances of fluidsynth and mt32d spawn. Is this behavior expected? Should it be just one?


Did you use the new ao486.rbf?

viewtopic.php?f=117&t=35031&start=100#p364323

If that does not work:

#cat/dev/MrAudio

and see if the write count is increasing.


No, used the old ao486 one. i will try again tomorrow. Thank you :)

BBond007
Captain Atari
Captain Atari
Posts: 402
Joined: Wed Feb 28, 2018 3:23 am

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby BBond007 » Fri Jan 18, 2019 11:07 pm

NegSol wrote:
No, used the old ao486 one. i will try again tomorrow. Thank you :)


I forgot to mention that the multiple instances of MUNT and FluidSynth is normal. I believe this is really multiple threads of a single instance.

Sorgelig
Ultimate Atarian
Ultimate Atarian
Posts: 5592
Joined: Mon Dec 14, 2015 10:51 am
Location: Russia/Taiwan

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Sorgelig » Sat Jan 19, 2019 9:07 am

released!

Since all binaries are inside Linux already, install script must be updated to skip copying the same files.
Actually script needs to download soundfonts only now.

BBond007
Captain Atari
Captain Atari
Posts: 402
Joined: Wed Feb 28, 2018 3:23 am

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby BBond007 » Sat Jan 19, 2019 11:08 am

Sorgelig wrote:released!

Since all binaries are inside Linux already, install script must be updated to skip copying the same files.
Actually script needs to download soundfonts only now.


EDIT:

Had to replace the script, unfortunately I forgot to include the creation of the directories.
You do not have the required permissions to view the files attached to this post.
Last edited by BBond007 on Mon Jan 21, 2019 8:18 am, edited 5 times in total.

NegSol
Captain Atari
Captain Atari
Posts: 326
Joined: Sat Dec 05, 2015 9:22 pm

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby NegSol » Sat Jan 19, 2019 11:53 am

Did a quick test - everthing works nominally :) - Also the external MIDI support still works like charm. :mrgreen:

NegSol
Captain Atari
Captain Atari
Posts: 326
Joined: Sat Dec 05, 2015 9:22 pm

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby NegSol » Sat Jan 19, 2019 1:10 pm

I did some more tests. Great stuff so far...
Would be nice to be able to switch gm soundfonts from within a core - so far I created shell scripts that can be executed from menu core and modify the MidiLink.ini.

I tested the following sf2

SC-55.sf2 (original set - 130 MB)
32MbGMStereo.sf2 (32 MB - Soundblaster set ??)
OmegaGMGS2.sf2 (272 MB - Biggest set I could load successfully)
SGM-V2.01.sf2 (242 MB)

Sound quality is great!

I failed to load
FatBoy-v0.784.sf2 (323 MB)
fluidsynth keeps reseting when running out of DDR Ram :) - I assume around 300 MB is the hard limit here for sf2

Anyway I am really happy with how it turned out. It is a dream come true. Thank you both Sorgelig and BBond007!!! :cheers:

Sorgelig
Ultimate Atarian
Ultimate Atarian
Posts: 5592
Joined: Mon Dec 14, 2015 10:51 am
Location: Russia/Taiwan

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Sorgelig » Sat Jan 19, 2019 1:58 pm

Linux has only 512MB of RAM. And of course some RAM is occupied by Linux to work.
Even 200MB soundfont sounds too much to me. I believe these huge soundfonts use very high sample rates. MiSTer audio uses fixed 48KHz sample rate - everything higher will be converted to 48KHz, so it is just a waste of RAM.
Don't go too crazy :)

NegSol
Captain Atari
Captain Atari
Posts: 326
Joined: Sat Dec 05, 2015 9:22 pm

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby NegSol » Sat Jan 19, 2019 3:31 pm

No worries :mrgreen: I expected limitations here. I just wanted to push it to the limit :)

Anyhow just as a background for those interested - it is not only about sampling frequency as far as I know. It is also about number and length of the samples. So more can give you a more accurate representation of the imitated instruments. So for example in a small soundfonts file a piano could use only one waveform simply pitched higher for each note. In a more accurate set of soundfonts the piano would have maybe 4 waveforms to have different expressions in different octave ranges. But it is all a bit magic. Nobody seriously need a waveform for each note of every instrument although I am told that these kind of "virtual instruments" do exist reaching sometimes multi GB file for the soundfonts.

Really glad MIDI is now a strong suit of MiSTer. Hope this will draw some more musically inclined people to the project.

Sorgelig
Ultimate Atarian
Ultimate Atarian
Posts: 5592
Joined: Mon Dec 14, 2015 10:51 am
Location: Russia/Taiwan

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Sorgelig » Sat Jan 19, 2019 8:01 pm

The primary intention is for games. Ok, may be midi listening.
No serious MIDI composing can be expected here.

Hewhoisred
Atari freak
Atari freak
Posts: 51
Joined: Sun Dec 09, 2018 6:53 pm

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Hewhoisred » Sun Jan 20, 2019 12:59 am

Just wanted to report that midilink is working for me on AO486.

Softmpu settings are:
SOFTMPU.EXE /SB:220 /IRQ:5 /MPU:330 /OUTPUT:COM1

Fluidsynth seems to play fine, however, I am definitely getting some tempo problems with MUNT on AO486.

I tested dos based midi player Megamid and the dos games Flashback and Ultima 6. All exhibited choppy tempo with MUNT.

Fluidsynth sounds nice, but "wrong" for these games since the instruments are from the SC-55 sound font.

For reference, here is what real Ultima 6 (or at least emulated on PC) sounds like:
https://youtu.be/bDyIEIwYaJU?t=103
listen to munt-ultima6 attached below in comparison.

Attached are sounds captured from my MiSTer.
You do not have the required permissions to view the files attached to this post.

BBond007
Captain Atari
Captain Atari
Posts: 402
Joined: Wed Feb 28, 2018 3:23 am

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby BBond007 » Sun Jan 20, 2019 1:45 am

Hewhoisred wrote:For reference, here is what real Ultima 6 (or at least emulated on PC) sounds like:
https://youtu.be/bDyIEIwYaJU?t=103


Just FYI: I know what a real Roland MT-32 sounds like. I recorded these videos a while back with a real Roland MT-32 and my MiSTer

Flashback --> http://y2u.be/8FfU6gpbF28
Ultima 6 --> http://y2u.be/DZs6KVA4zvU

The problem is that the ARM CPU is on the low end of the requirements for MUNT. I think it can be better with compiler and run-time optimizations.

[/media/fat/linux/MidiLink.INI]

MUNT_OPTIONS =

-m : Manual buffering mode (buffer does not grow)
-a : Automatic buffering mode (default)
-x msec : Maximum buffer size in milliseconds
-i msec : Minimum (initial) buffer size in milliseconds
-l mode : Analog emulation mode (0 - Digital, 1 - Coarse, 2 - Accurate, 3 - Oversampled 2x, default: 2)

You could also try a MT-32 SoundFont with FluidSynth, but that will only work for things that don't reprogram the MT-32's default soundbank.

You can also use the UDP option to send the raw MIDI data to MUNT running on your desktop PC.

Sorgelig
Ultimate Atarian
Ultimate Atarian
Posts: 5592
Joined: Mon Dec 14, 2015 10:51 am
Location: Russia/Taiwan

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Sorgelig » Sun Jan 20, 2019 2:32 am

may be MUNT compilation option can be tweaked?
What compiler options you've used?

User avatar
Paradroyd
Captain Atari
Captain Atari
Posts: 293
Joined: Tue Sep 10, 2013 10:50 pm
Contact:

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Paradroyd » Mon Jan 21, 2019 5:21 am

BBond007 wrote:
EDIT:

Had to replace the script, unfortunately I forgot to include the creation of the directories.


It seems like somewhere along the line over the past day or so, the Atari ST UART stuff stopped working properly. I can't get it to respond to AT commands anymore. (I realize you may already know this).
- Paradroyd
@paradroyd on Twitter, @paradroyd@mastodon.sdf.org on Mastodon

Sorgelig
Ultimate Atarian
Ultimate Atarian
Posts: 5592
Joined: Mon Dec 14, 2015 10:51 am
Location: Russia/Taiwan

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Sorgelig » Mon Jan 21, 2019 5:33 am

BBond007 wrote:Had to replace the script, unfortunately I forgot to include the creation of the directories.

why you reboot at the end of script? I don't see anything there requiring reboot

BBond007
Captain Atari
Captain Atari
Posts: 402
Joined: Wed Feb 28, 2018 3:23 am

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby BBond007 » Mon Jan 21, 2019 7:34 am

Sorgelig wrote:
BBond007 wrote:Had to replace the script, unfortunately I forgot to include the creation of the directories.

why you reboot at the end of script? I don't see anything there requiring reboot


Nothing nefarious, I just thought it would be good to reboot in case F-synth or MUNT were already loaded... and it was a leftover from the other script.

I removed it...

Sorgelig
Ultimate Atarian
Ultimate Atarian
Posts: 5592
Joined: Mon Dec 14, 2015 10:51 am
Location: Russia/Taiwan

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Sorgelig » Mon Jan 21, 2019 8:34 am

Back to MUNT. Which compiler options you've used?
I think HPS on MiSTer is somewhat similar to RPi 2 but Cortex A9. May be A9 options will help a little.

BBond007
Captain Atari
Captain Atari
Posts: 402
Joined: Wed Feb 28, 2018 3:23 am

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby BBond007 » Mon Jan 21, 2019 8:44 am

Sorgelig wrote:Back to MUNT. Which compiler options you've used?
I think HPS on MiSTer is somewhat similar to RPi 2 but Cortex A9. May be A9 options will help a little.


https://retropie.org.uk/forum/topic/125 ... on-rpi-3/2

this is suggested for RP2, should I try that with -mcpu=cortex-a9

export CCFLAGS="-Ofast -mcpu=cortex-a7 -mfpu=neon-vfpv4 -mfloat-abi=hard -ftree-vectorize -funsafe-math-optimizations"
export CXXFLAGS="-Ofast -mcpu=cortex-a7 -mfpu=neon-vfpv4 -mfloat-abi=hard -ftree-vectorize -funsafe-math-optimizations"

Sorgelig
Ultimate Atarian
Ultimate Atarian
Posts: 5592
Joined: Mon Dec 14, 2015 10:51 am
Location: Russia/Taiwan

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Sorgelig » Mon Jan 21, 2019 8:49 am

found this:
https://gist.github.com/fm4dd/c66321793 ... c81b0aa845

they have Cyclone V5 HPS options listed.

Sorgelig
Ultimate Atarian
Ultimate Atarian
Posts: 5592
Joined: Mon Dec 14, 2015 10:51 am
Location: Russia/Taiwan

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Sorgelig » Mon Jan 21, 2019 8:52 am

so keep RPi2 other options, and just replace these: -mcpu=cortex-a9 -mfpu=neon

BBond007
Captain Atari
Captain Atari
Posts: 402
Joined: Wed Feb 28, 2018 3:23 am

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby BBond007 » Mon Jan 21, 2019 9:21 am

Sorgelig wrote:so keep RPi2 other options, and just replace these: -mcpu=cortex-a9 -mfpu=neon


OK, thanks!

it still works anyway, don't know if its better yet
You do not have the required permissions to view the files attached to this post.

Sorgelig
Ultimate Atarian
Ultimate Atarian
Posts: 5592
Joined: Mon Dec 14, 2015 10:51 am
Location: Russia/Taiwan

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Sorgelig » Mon Jan 21, 2019 9:32 am

It would be good if Hewhoisred would test it since he's complained :)

Sorgelig
Ultimate Atarian
Ultimate Atarian
Posts: 5592
Joined: Mon Dec 14, 2015 10:51 am
Location: Russia/Taiwan

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Sorgelig » Mon Jan 21, 2019 10:30 am

MUNT is the only MT32 soft-synth?

Sorgelig
Ultimate Atarian
Ultimate Atarian
Posts: 5592
Joined: Mon Dec 14, 2015 10:51 am
Location: Russia/Taiwan

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Sorgelig » Mon Jan 21, 2019 11:13 am

It's unrelated to MIDI directly but still related: Currently uartmode scripts kills all getty processes including the one on USB serial console. It would be good to find the way to kill getty on UART only.

may be make a copy of getty with other name to use it only for UART..

Hewhoisred
Atari freak
Atari freak
Posts: 51
Joined: Sun Dec 09, 2018 6:53 pm

Re: MidiLink 2.0 looking for testers. (no MIDI devices required)

Postby Hewhoisred » Mon Jan 21, 2019 6:02 pm

Sorgelig wrote:It would be good if Hewhoisred would test it since he's complained :)


The thread subject is asking for testers. I tested and posted results as best I could. No complaints. Midilink and MiSTer are amazing.

I will be happy to test new builds and post results.


Return to “MiSTer”

Who is online

Users browsing this forum: auron06, BigMac740p and 7 guests