Does anyone know if there is any Asio bindings or direct D available that allows for IO?

On Thursday, 2 June 2016 at 04:02:36 UTC, Pie? wrote:
> Does anyone know if there is any Asio bindings or direct D available that allows for IO?

Check out vibe.d: https://vibed.org/ - it includes a fairly complete implementation of asynchronous I/O, among other things.

On Thursday, 2 June 2016 at 04:52:50 UTC, Mithun Hunsur wrote:
> On Thursday, 2 June 2016 at 04:02:36 UTC, Pie? wrote:
>> Does anyone know if there is any Asio bindings or direct D available that allows for IO?
>
> Check out vibe.d: https://vibed.org/ - it includes a fairly complete implementation of asynchronous I/O, among other things.

Oh, lol, I should have mentioned I meant for audio! ;)

On Thursday, 2 June 2016 at 06:28:51 UTC, Pie? wrote:
> On Thursday, 2 June 2016 at 04:52:50 UTC, Mithun Hunsur wrote:
>> On Thursday, 2 June 2016 at 04:02:36 UTC, Pie? wrote:
>>> Does anyone know if there is any Asio bindings or direct D available that allows for IO?
>>
>> Check out vibe.d: https://vibed.org/ - it includes a fairly complete implementation of asynchronous I/O, among other things.
>
> Oh, lol, I should have mentioned I meant for audio! ;)

It doesn't seem to exist but using bindings for FMOD you should be able to access ASIO as an audio driver.

On Thursday, 2 June 2016 at 11:15:59 UTC, Guillaume Piolat wrote:
> On Thursday, 2 June 2016 at 06:28:51 UTC, Pie? wrote:
>> On Thursday, 2 June 2016 at 04:52:50 UTC, Mithun Hunsur wrote:
>>> On Thursday, 2 June 2016 at 04:02:36 UTC, Pie? wrote:
>>>> Does anyone know if there is any Asio bindings or direct D available that allows for IO?
>>>
>>> Check out vibe.d: https://vibed.org/ - it includes a fairly complete implementation of asynchronous I/O, among other things.
>>
>> Oh, lol, I should have mentioned I meant for audio! ;)
>
> It doesn't seem to exist but using bindings for FMOD you should be able to access ASIO as an audio driver.

If FMOD is that commercial sound lib then I'm not interested. I guess I'll have to try and write some type of asio lib when I get around to it. Hopefully it is not too difficult.

I do have (Steinberg) ASIO binding in D.

The problem is I couldn't release the bindings. I've asked Steinberg if it was OK to release D bindings and they were strongly against it unfortunately (and this was over 3 years ago..).

Any kind of direct use of ASIO requires their approval first.. meaning you had to register on their website.

I would recommend using third party libs that abstract the underlying engine, like PortAudio  or RtAudio (the later of which I'm going to release a port of soon!).

I had a binding to PortAudio but the devs of that library insisted on only supporting interleaved audio, RtAudio supports both interleaved and non-interleaved audio, and the library is easy to port.



On 6/2/16, Pie? via Digitalmars-d-learn <digitalmars-d-learn@puremagic.com> wrote:
> On Thursday, 2 June 2016 at 11:15:59 UTC, Guillaume Piolat wrote:
>> On Thursday, 2 June 2016 at 06:28:51 UTC, Pie? wrote:
>>> On Thursday, 2 June 2016 at 04:52:50 UTC, Mithun Hunsur wrote:
>>>> On Thursday, 2 June 2016 at 04:02:36 UTC, Pie? wrote:
>>>>> Does anyone know if there is any Asio bindings or direct D available that allows for IO?
>>>>
>>>> Check out vibe.d: https://vibed.org/ - it includes a fairly complete implementation of asynchronous I/O, among other things.
>>>
>>> Oh, lol, I should have mentioned I meant for audio! ;)
>>
>> It doesn't seem to exist but using bindings for FMOD you should be able to access ASIO as an audio driver.
>
> If FMOD is that commercial sound lib then I'm not interested. I guess I'll have to try and write some type of asio lib when I get around to it. Hopefully it is not too difficult.
>

On Wednesday, 8 June 2016 at 23:19:13 UTC, Andrej Mitrovic wrote:
> I do have (Steinberg) ASIO binding in D.
>
> The problem is I couldn't release the bindings. I've asked Steinberg if it was OK to release D bindings and they were strongly against it unfortunately (and this was over 3 years ago..).
>
> Any kind of direct use of ASIO requires their approval first.. meaning you had to register on their website.
>
> I would recommend using third party libs that abstract the underlying engine, like PortAudio  or RtAudio (the later of which I'm going to release a port of soon!).
>
> I had a binding to PortAudio but the devs of that library insisted on only supporting interleaved audio, RtAudio supports both interleaved and non-interleaved audio, and the library is easy to port.
>

Why would bindings have any issues with licensing? People release VST source code all the time. Sure they will be against it because they are muddlefudgers! They can't officially endorse it without having to dealing with the user end to some degree and Steinberg is known for that kind of behavior(just look at all the hoops one has to jump through to get asio in the first place).

Of course, I can't convince you but and I'll probably have to re-create your work, but hosting something like that on git shouldn't cause any problems. At most, SB will send you a cease and desist type of letter. In which case you take it down. Think of mono, it essentially duplicated .net and MS hasn't done a thing about it(they can't do much but flex their big muscles, in which case they didn't).

I would appreciate it though if you thought about it again, it would save me a bunch of work!

If the problem is that you have included SB source code, then that can easily be remedied by removing and and placing an abstraction in it's place where others can plug in the source when they d/l it from SB.

I haven't got into writing any audio stuff yet but when I look in to it more I'll check out the options. I don't need anything overly complex but do need low latency IO.

On 6/9/16, Joerg Joergonson via Digitalmars-d-learn <digitalmars-d-learn@puremagic.com> wrote:
> Why would bindings have any issues with licensing?

Just to show that I'm not full of shit, here's the e-mail chain:

On 6/3/11, Andrej Mitrovic <andrej.mitrovich@gmail.com> wrote:
> Hello,
>
> I'm currently porting the Steinberg VST and ASIO SDKs for use with the D programming language. I would like to know if there are any special licensing requirements before I can publish the source code to a public repository.
>
> Specifically, do I need to print out the licensing agreement PDF document, sign it, and send it via snail-mail to Steinberg? I'm currently unemployed, so there shouldn't be any issues from my part with regards to signing any licensing agreements, if that's required.
>
> The bindings themselves won't distribute any C/C++ header or implementation files from the SDK, as the D language can't use C header files directly.
>
>
> Kind regards,
> Andrej Mitrovic
>



On 6/7/11, Yvan Grabit <y.grabit@steinberg.de> wrote:
> Hi Andrej
>
> i will speak with our internal Product planning people about your
> concern...
> I will keep you inform about our decision as soon as we have a decision...
>
> Cheers
>
> Yvan
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> Yvan Grabit                              mailto:y.grabit@steinberg.de
> Technical Manager - Technology Group     Phone: +49-40-21035125
> Steinberg Media Technologies GmbH        Fax  : +49-40-21035300
> Neuer Hoeltigbaum 22-32, D-22143 Hamburg/Germany
>                                          http://www.steinberg.net
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++



On 6/23/11, Andrej Mitrovic <andrej.mitrovich@gmail.com> wrote:
> Hello Yvan,
>
> I apologize if this e-mail is unnecessary and a decision is yet to be made. If you are too busy should I try to contact one of Steinberg's official representatives?
>
> I am willing to provide the source code for examination to Steinberg if this could help  verify that I'm not breaking any licensing requirements.
>
> Have a nice day,
> Andrej Mitrovic
>


On 6/23/11, Yvan Grabit <y.grabit@steinberg.de> wrote:
> Hi
>
> we have examine your request, and decide due to our current VST/ASIO
> license that you are not allowed to deploy/distribute a copy of the
> VST/ASIO SDK (partially or full).
> We will investigate a way to change our licensing for research or
> educational use.
>
> In your concrete case, could it be not possible to ask the people you give
> your source using the VST SDK (or ASIO)  to download the VST-SDK (or ASIO)
> (and agree the agreement) and provide a compiler macro (or something
> different) which will convert the VST /ASIO interface to something D
> compatible ?
>
> By the way how can we guaranty that the VST plugin or ASIO component developed in D could be load by a host developed in C ?
>
> Cheers
>
> Yvan
>
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> Yvan Grabit                              mailto:y.grabit@steinberg.de
> Technical Manager - Technology Group     Phone: +49-40-21035125
> Steinberg Media Technologies GmbH        Fax  : +49-40-21035300
> Neuer Hoeltigbaum 22-32, D-22143 Hamburg/Germany
>                                          http://www.steinberg.net
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++



On 6/23/11, Andrej Mitrovic <andrej.mitrovich@gmail.com> wrote:
> On 6/23/11, Yvan Grabit <y.grabit@steinberg.de> wrote:
>> In your concrete case, could it be not possible to ask the people you
>> give
>> your source using the VST SDK (or ASIO)  to download the VST-SDK (or
>> ASIO)
>> (and agree the agreement) and provide a compiler macro (or something
>> different) which will convert the VST /ASIO interface to something D
>> compatible ?
>
> Yes, I believe I can do that. For ASIO, I have the following:
>
> - A "Loader", which is a source file made from scratch which searches for ASIO entries in the Windows Registry and provides a class which can load and initialize any found ASIO drivers. Through this class user code calls functions like "canSampleRate", "outputReady", etc.
>
> The Loader only requires the type definitions found in the asiosdk\common\asio.h header file. This header file can be automatically converted to D via an existing tool, and the Loader would then import the newly converted header file and use the type definitions.
>
> The Loader requires the OpenAsio DLL that can be found here: http://www.martinfay.com/openasio.htm
>
> The reason OpenAsio is required is because the ASIO COM class has the "thiscall" calling convention which is not directly supported by the D language. It might be possible to use inline assembler to simulate the "thiscall" calling convention, but this would require some work on my part to get it working.
>
> No other ASIO SDK file is required.
>
> From my current understanding, you would require me that *no* ASIO SDK files are to be distributed with the library. Therefore I should provide only the Loader source file, and provide the links to the ASIO SDK Developer homepage where the users can agree to the license terms before they can download the SDK, after which they can extract the asio.h header file which is then converted to D. I would also provide a link to the OpenAsio DLL download page which is required for the 'thiscall' workaround.
>
> The library would not be able to compile without the header file, and therefore users would not be able to use the ASIO technology without agreeing to the license terms found on the Steinberg website.
>
> The VST binding (v2.4) is still in development and not ready yet. I
> have two C header files converted to D and a D host application
> example that can load VST DLL's and inspect their properties. The two
> needed header files are:
> aeffect.h
> aeffectx.h
>
> The D Host sample application can load a VST plugin and inspect its properties, process audio, and instantiate the GUI of the plugin (if available).
>
>> By the way how can we guaranty that the VST plugin or ASIO component developed in D could be load by a host developed in C ?
>
> D is a C ABI compatible language, meaning that it can directly call into C code and be called from C code without any need for wrappers or technologies like P/Invoke. All that is required is that the parameter types are typed the same, and that the calling conventions match. Calling conventions are specified in the D code by using "extern(C)"-style syntax, and this can be used to either export a D function to be used by C code, or import C functions to be used by D code.
>
> A D ASIO device driver could be written, however I don't think a manufacturer would ever consider using D for writing ASIO drivers, as it is a fairly young language and lacks the performance benefit of using pure C.
>
> A D ASIO host can directly use ASIO driver buffers, with no need for conversions between types. This is exactly why I have started the D ASIO bindings projects, as other solutions (such as PortAudio) use interleaved buffers and generally introduce additional overhead where it is not required.
>
> A D VST plugin can be created and loaded by a C host. A D VST Plugin can communicate with a C host and provide callback functions, which the C host can then directly call.
>
> A C VST plugin can also be loaded by a D host.
>
> I will provide example code for each of the above case, except writing an actual ASIO device driver in D, which I would discourage manufacturers from doing as there wouldn't be much benefit over using C directly.
>
> Kind regards,
> Andrej Mitrovic
>

--------

Unfortunately this is the last I've heard from them at the time..

On Friday, 10 June 2016 at 20:52:46 UTC, Andrej Mitrovic wrote:
> On 6/9/16, Joerg Joergonson via Digitalmars-d-learn <digitalmars-d-learn@puremagic.com> wrote:
>> [...]
>
> Just to show that I'm not full of shit, here's the e-mail chain:
>
> On 6/3/11, Andrej Mitrovic <andrej.mitrovich@gmail.com> wrote:
>>[...]
>
>
>
> On 6/7/11, Yvan Grabit <y.grabit@steinberg.de> wrote:
>> [...]
>
>
>
> On 6/23/11, Andrej Mitrovic <andrej.mitrovich@gmail.com> wrote:
>>[...]
>
>
> On 6/23/11, Yvan Grabit <y.grabit@steinberg.de> wrote:
>> [...]
>
>
>
> On 6/23/11, Andrej Mitrovic <andrej.mitrovich@gmail.com> wrote:
>>[...]
>
> --------
>
> Unfortunately this is the last I've heard from them at the time..

Well, I definitely didn't think you were full of shit! But I see no negative statements against doing what you did. They simply said that you can't include their code directly... they want people who download your code to have to download the sdk from Steinberg.

You own your code, no one else, not even Steinberg... but you don't own or have the right, as they mention, to publish their code with yours.

Creating bindings is not a license infraction.

>> [...]


Are you using any of their source code from the vst sdk? If you hand re-write any of their source code, it is yours.

You say that the only two files from the sdk are aeffect.h and aeffectx.h, right?

Is it these:

https://github.com/falkTX/dssi-vst/blob/master/vestige/aeffectx.h

https://sourceforge.net/u/geyan123d/freepiano/ci/54f876d52c6f49925495f7ed880bd2434bda0504/tree/src/vst/aeffect.h?format=raw

http://www.dith.it/listing/vst_stuff/vstsdk2.4/doc/html/aeffect_8h.html

https://source.openmpt.org/svn/openmpt/branches/devBranch_1_17_03/include/AEffect.h

etc...

If so, as you can see, people already do what your wanting to do and their is probably a reason they stopped responding, and it's because they can't do much as long as you don't include anything from them.

Of course, don't take my word... maybe ask on stack overflow or something? If you are simply worried about being sued then that problem is relatively easily fixed! Just ask Walter or Andrei to fund your defense ;)

Don't let your hard work go to waste! ;) I think you should be more afraid of all the questions I'll be asking you on how to use it rather than being sued by Steinberg! ;)

On Wednesday, 8 June 2016 at 23:19:13 UTC, Andrej Mitrovic wrote:
> I do have (Steinberg) ASIO binding in D.
>
> The problem is I couldn't release the bindings. I've asked Steinberg if it was OK to release D bindings and they were strongly against it unfortunately (and this was over 3 years ago..).
>
> Any kind of direct use of ASIO requires their approval first.. meaning you had to register on their website.
>
> I would recommend using third party libs that abstract the underlying engine, like PortAudio  or RtAudio (the later of which I'm going to release a port of soon!).
>
> I had a binding to PortAudio but the devs of that library insisted on only supporting interleaved audio, RtAudio supports both interleaved and non-interleaved audio, and the library is easy to port.
>

Any news on this?