Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: foo_midi (Read 83037 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: foo_midi

Reply #225 –
Okay then. What exactly are you implying? That Microsoft's 2019 Community compiler is infected by design?

If you can produce builds that aren't detected at all, using the same toolkit and build environment that I use.

Fresh Windows 10 1909 x64 English ISO, downloaded from Microsoft. (size: 5,421,459,456; sha256: 01bf1eb643f7e50d0438f4f74fb91468d35cde2c82b07abc1390d47fc6a356be)

Installed into a fresh Linux Qemu/KVM virtual machine, using Gnome Boxes.

Switched to Virt-Manager for more flexibility post-install.

Install desktop updates to current status.

Install MSVC 2019 Community, and Clang-CL for some things now.

Install Git for Windows from its official site.

Git clone the entire bitbucket.org/losnoco/foobar2000 tree, recursively.

Build the component from the solution at <repository root>/plugins/foobar/foobar2000/plugins.sln

Re: foo_midi

Reply #226 –
Quote from: kode54 on
Fine. I'm pulling all binaries for my components. You fuckers can compile your own goddamn shit.
I'm done with this shit.
kode54, please don't let it get to you... :'(
and please don't de-list your components again... it saddens me when u do that...
I rely on your updates to keep Foobar2000 game music plugins working at their best and value your work...


I don't trust AV as far as I can throw them lol..

Kode54 all the componets I use were written by you and I have always considered u a trusted source.. ;^_^;

Re: foo_midi

Reply #227 –
kode54, v2.3.1 doesn't show Yamaha VSTi to me

Re: foo_midi

Reply #228 –
Quote from: kode54 on
Okay then. What exactly are you implying? That Microsoft's 2019 Community compiler is infected by design?

I imply nothing. I state the fact that two files in this component trigger alarms from my AV and apparently with a number of other AVs. I'm not the only one to notice this. That is all.

Re: foo_midi

Reply #229 –
I'd recommend sending detected files for false positive analysis to your antivirus vendor. That way you wouldn't stress the developer needlessly and you'd quickly hear from the vendor that the files are indeed clean. The false positive flagging won't stop unless paying customers complain loudly.

Re: foo_midi

Reply #230 –
GDI you guys! This is why we cannot have nice things.

Re: foo_midi

Reply #231 –
I'm putting them back, soon enough.

There's a new foo_midi, 2.3.2. The VSTi and Secret Sauce components require the MSVC 2015 runtimes, both 32 and 64 bit.

Re: foo_midi

Reply #233 –
Quote from: kode54 on
I'm putting them back, soon enough.

There's a new foo_midi, 2.3.2. The VSTi and Secret Sauce components require the MSVC 2015 runtimes, both 32 and 64 bit.

It works well and doesn't trigger any local or on-line AVs that I have access to. Thank you.

For the benefit of us many incredibly thick fucksticks (also known as every-day users, the target audience for Foobar2000 and its components,) did you find what was triggering all those AV engines? Or did our false-positive reports just catch up with their virus definition updates?

Re: foo_midi

Reply #234 –
Sorry, I need to clean up my language. I must apologize for calling you thick like that, I was really mad and in a bad place.

Basically, I had to do the following:

This step may not matter: Create new VM in place of old, install MSVC 2015 Update 3, final.

This step may matter: After getting everything compiled again, submitting my package to VirusTotal myself, and rescanning a few times to make sure it processed everything. It caught a few false positives, including in Microsoft's scanner.

This step matters: I submitted my software, packaged in an encrypted ZIP with the password "infected", as per instructions, to Microsoft, and kindly asked them to verify my assertion that it wasn't infected. Within about 8 hours, they verified that it was clean.

This step also matters: Once the above was completed, I rescanned the original upload at VirusTotal, which now comes out 100% clean. Note that three of the executables still register individually as "Malicious" or something with Cylance, even though the top level package does not.

It seems whenever I make updates to any of the executable subcomponents of foo_midi, or even the main component, I will need to complete the day long process of submitting it to Microsoft for whitelisting, every time. Oh well. Maybe I can convince someone to extend an EV certificate my way...

Re: foo_midi

Reply #235 –
Quote from: kode54 on
Oh well. Maybe I can convince someone to extend an EV certificate my way...
I would donate again when I have some $ to spare...
bills are killing me most of the time so its hard when I don't have much food most of the time...
sorry Kode54

Re: foo_midi

Reply #236 –
Nah, I didn't mean donations. More like another assist from the Mobile team. I'm already using their help to sign my macOS software. It was only an idea floated by Peter, in case it would help him with his false positive issues as well. Only considered since the last time he had a certificate, not EV, it sometimes resulted in worse experience than without a certificate.

Re: foo_midi

Reply #237 –
ooh ok..

Re: foo_midi

Reply #238 –
Corrected "with a certificate" to "without", oops. Yeah, there was a time when a "cheap" certificate was actually worse than no certificate.

Re: foo_midi

Reply #239 –
I am a old user of foo_midi when only can select between Emu de MIDI and FluydSinth (with Foobar2000 1.3.3).
I recently update to last Foobar version and plugins. And I see many changes to configure that plugin.

I am a ignorant about MIDI world and only want play/convert some general midi files with the best quality possible.
I suppose than BASSMIDI is like FluydSinth to use soundfonts and the best option to choice, but there are many, and I don't know the parameters to configure it.

I read this thread entirely and I don't see many info about it.
There are some extensive documentation about foo_midi configuration?
Or you can be so polite to answer my questions here?

Re: foo_midi

Reply #240 –
There is no need to change most of the defaults with BASSMIDI. You just need to load a SoundFont, or a list of SoundFonts with a priority order by the order they're listed.
 

Re: foo_midi

Reply #241 –
Thanks to answer. Only a few questions:

- Resampling: Is better 16pt Sinc interpolation?

- Cached: X MB / 2.62 GB (the 2.62 GB is a limit?, the Soundfont is 4 GB, is a problem?)
Also the Soundfont claim have 24 bits samples but when record to flac (or wav auto) is only 16 bits.

- The MUNT section, what is? Click to set what?

I suposse the rest of default settings are the recommended.

Re: foo_midi

Reply #242 –
Quote from: tebasuna51 on
Thanks to answer. Only a few questions:

- Resampling: Is better 16pt Sinc interpolation?
It's really only a placebo effect, but you're welcome to use it if you think it will improve the sound quality.

Quote from: tebasuna51 on
- Cached: X MB / 2.62 GB (the 2.62 GB is a limit?, the Soundfont is 4 GB, is a problem?)
Also the Soundfont claim have 24 bits samples but when record to flac (or wav auto) is only 16 bits.
BASSMIDI only supports 16 bit SoundFonts anyway, and even if that were not the case, you wouldn't be able to exceed about 3GB anyway, since the whole thing has to fit into a 32 bit process. I'll consider making an external BASSMIDI 64 bit process, maybe, if there's really a demand for hitting the limit of memory there.

(Also, I think I've tried this StGiga soundfont, and I wasn't really that impressed.)

Quote from: tebasuna51 on
- The MUNT section, what is? Click to set what?
MT-32 emulator, for MT-32 MIDI files. Supply your own MT-32 or CM-32L control and PCM ROM set, and it can play those files. Don't, and it won't.

Re: foo_midi

Reply #243 –
Quote from: kode54 on
BASSMIDI only supports 16 bit SoundFonts anyway, and even if that were not the case, you wouldn't be able to exceed about 3GB anyway, since the whole thing has to fit into a 32 bit process. I'll consider making an external BASSMIDI 64 bit process, maybe, if there's really a demand for hitting the limit of memory there.

Don't worry, is enough quality for me. I tested many free soundfonts an this one is the better for me, and seems work fine.

Thanks for all, your plugin and your answers.

Re: foo_midi

Reply #244 –
Hi, I have a couple of questions/requests regarding foo_midi.

1. I use Yamaha S-YXG50 as my default plug-in, though in many cases I'd prefer to be using Roland SCVA. So far, I have only been invoking SCVA for the purposes of rendering select midis to audio, and then switching back to S-YXG50 for general usage, because of the impracticality imposed by SCVA's exceptionally long initialisation time. Is there any possible workaround for this problem, or is continued case-by-case rendering my best option? I had also been using SCVA with mudlord's VST MIDI driver and using a separate MIDI player so that it isn't initialised on playback of every song, that is until Microsoft decided to horribly break MIDI devices in a recent Windows update...

2. This one's slightly out of left-field - is there any possibility of Ensoniq AudioPCI Wavesets (eapci2m.ecw / eapci4m.ecw / eapci8m.ecw) becoming supported in foo_midi/BASSMIDI? They were also used by some later SoundBlaster cards like Live! and Audigy, where the MIDI device named Creative SW Synth utilised them.
Failing that, are there any other apps I could use to play midis with these soundbanks? There was an ECW to SoundFont conversion utility (ecw2sf2.exe) released around a decade ago, I don't know if it is still online anywhere, though I still have it locally and could share it if it is of interest. Unfortunately it does not work very well, otherwise I'd just use the resulting SoundFonts in BASSMIDI and call it a day. These soundbanks were how I first experienced General MIDI in the late 90s, so I suppose I have a particular fondness for them.

Any pointers would be most welcome, thanks.

Re: foo_midi

Reply #245 –
For Sound Canvas VA, you might want to try the Secret Sauce option, which uses the Sound Canvas engine directly and as a result runs much faster than the official plugin. Just set the Secret Sauce path in the advanced settings to the location of the VSTi.

Re: foo_midi

Reply #246 –
Thanks for the reply Slender.

That sounds intriguing, though I'm not having any joy so far. I've copied the directory into the Secret Sauce path and closed/reopened the Preferences window, but I don't see anything about Secret Sauce appearing anywhere in the MIDI Player configuration; I'm assuming it would appear as a new output plug-in?

I read kode54's post earlier in the thread that Secret Sauce only works with whitelisted versions of SCVA, which may possibly be the issue as I'm using v1.00. Is there a list of versions that are known to work?

Re: foo_midi

Reply #247 –
Yes, it would appear as an output plugin. Make sure there are no quotes around the path as well. If it isn't recognized, you might want to post the MD5 hash of your copy of SCCore.dll.

Re: foo_midi

Reply #248 –
Ahh, success. I decided to update to v1.16 (S) (Build 1160) for some other reasons, and now this works. Gosh, it is significantly faster and actually usable. Thanks for the tip.

In case this is still useful information, I generated an MD5 of my previous version:
Code: [Select]
FD2FFC7229042738718278239D4B0EC4

A friend was also having the same problem with a different version, v1.11 (S) - Build 1110. Curiously it was said to be v1.00 when loaded in foobar2000, which could mean mine wasn't v1.00 either, but anyway:
Code: [Select]
DBD9A30C168EFEF577D40A28D9ADF37D

Re: foo_midi

Reply #249 –
Enjoy :). As a sidenote, you could actually just copy SCCore.dll somewhere in case the hash changes in the future. From what I can tell, as far as emulation is concerned, the project is basically complete, and any future changes seem to be focusing more on UI improvements and fixes. SCCore.dll is completely independent of the VSTi as well. You don't need to keep around any data files to use it, and in fact you don't even need Sound Canvas VA itself installed, either, just the Visual C++ 2015 redistributables.