[Cs-dev] Cross compiling C++ for various virtual machines
Date | 2013-11-04 14:01 |
From | Michael Gogins |
Subject | [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
I have been following Steven Yi's reports on cross-compiling Csound to JavaScript bytecode using Emscripten. I got curious and did some research, and quickly found that Adobe has a similar initiative for ActionScript:
CrossBridge has an online demo using Unreal Citadel that looks virtually identical to the Emscripten demo. Does anyone have any experience with this?
Newer Flash runtimes support pthreads and OpenMP. As you probably know, both JavaScript and ActionScript are forms, more or less, of ECMAScript. I quote the first part of their README.md below.
Regards, Mike Crossbridge is the open-source version of Adobe FlasCC (formerly the Alchemy project). It provides a complete C/C++ development environment for targetting the Adobe Flash Runtime.
For more information about this project, please visit http://adobe-flash.github.io/crossbridge/. To get the latest release version, please visit the http://sourceforge.net/projects/crossbridge/files/.
Features: GCC 4.2 compiler with Flash backend. GDB debugger that debugs your code while running in the Flash player. SWIG backend that auto generates interop code allowing you to build Flash libs (SWCs) that wrap C/C++ libraries.
pthread based multithreading (in Flash 11.5+) Development The master branch contains all of the code necessary to build crossbridge 1.0.1, this is the current stable version. The future branch contains some work in progress to upgrade the toolchain to LLVM 3.2 and switch to using the Clang compiler instead of GCC.
Both the mac and cygwin SDKs are built on mac, you will need XCode and some macports packages installed to be able to build (pkg-config, glib, and mercurial). Doing a "make" in the root directory should be sufficient to build just the mac sdk, doing "make; make win; make deliverables" should build both mac and windows and package them as dmg and zips
Future Futures branch contains the upgrading code. Comparing to master, it will mainly has Clang/LLVM 3.2 toolchain Freebsd 9.1 Environment
Libc++ with support of c++11 std. =========================== Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com |
Date | 2013-11-04 19:20 |
From | Steven Yi |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
I'm very wary about Flash. Adobe seems to be moving away from it now for a while, Apple actively dislikes it, and others seem to merely tolerate it. With NPAPI going away, I don't know what the future of Flash will be. I personally don't find it attractive to invest time into, but then again, it's been a long while since I did any Flash development and hence I have only followed it from the outside. Seems to me that putting time into it would eventually be a loss as I think Flash is going to disappear within a few years. On Mon, Nov 4, 2013 at 9:01 AM, Michael Gogins |
Date | 2013-11-04 19:27 |
From | Victor Lazzarini |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
We should also investigate NaCl at some point https://developers.google.com/native-client/overview On 4 Nov 2013, at 19:20, Steven Yi wrote: > I'm very wary about Flash. Adobe seems to be moving away from it now > for a while, Apple actively dislikes it, and others seem to merely > tolerate it. With NPAPI going away, I don't know what the future of > Flash will be. I personally don't find it attractive to invest time > into, but then again, it's been a long while since I did any Flash > development and hence I have only followed it from the outside. Seems > to me that putting time into it would eventually be a loss as I think > Flash is going to disappear within a few years. > > On Mon, Nov 4, 2013 at 9:01 AM, Michael Gogins |
Date | 2013-11-04 19:36 |
From | Victor Lazzarini |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
It does have callback-based audio, which is a good start https://developers.google.com/native-client/devguide/coding/audio On 4 Nov 2013, at 19:27, Victor Lazzarini wrote: > We should also investigate NaCl at some point > https://developers.google.com/native-client/overview > > On 4 Nov 2013, at 19:20, Steven Yi wrote: > >> I'm very wary about Flash. Adobe seems to be moving away from it now >> for a while, Apple actively dislikes it, and others seem to merely >> tolerate it. With NPAPI going away, I don't know what the future of >> Flash will be. I personally don't find it attractive to invest time >> into, but then again, it's been a long while since I did any Flash >> development and hence I have only followed it from the outside. Seems >> to me that putting time into it would eventually be a loss as I think >> Flash is going to disappear within a few years. >> >> On Mon, Nov 4, 2013 at 9:01 AM, Michael Gogins |
Date | 2013-11-04 19:39 |
From | jpff@cs.bath.ac.uk |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None |
Date | 2013-11-04 19:40 |
From | Michael Gogins |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
I agree with you about Flash, Steven. I just thought I'd mention it because it's another route to the same goal. NaCl seems to be on hiatus. Victor, thanks for the information about PNaCl. I will look into it a bit more.
What are your thoughts regarding Emscripten vs. PNaCl? They seem to be about similar results. What is motivating me here is the thought of something like Csound 6 for Android that runs on any browser, or at least on Chrome on any platform, and ties into JavaScript as well as into the browser's image, video, and above all audio facilities. It something a lot closer to a "universal" Csound application. And it would be easy to do user-defined widgets. Frankly, this could save us all a huge load of work.
Best, Mike Best, Mike =========================== Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com On Mon, Nov 4, 2013 at 2:27 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote: We should also investigate NaCl at some point |
Date | 2013-11-04 19:56 |
From | Victor Lazzarini |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
In a way this is already achievable via Javasound and JAWS. We did a proof-of-concept study a couple of years ago that indicated it could be done, with the java Csound wrapper with Csound linked statically in it. With natively compiled csound jnis, JAWS then delivers the correct JNI to each client OS. Victor On 4 Nov 2013, at 19:40, Michael Gogins wrote: > What is motivating me here is the thought of something like Csound 6 for Android that runs on any browser, or at least on Chrome on any platform, and ties into JavaScript as well as into the browser's image, video, and above all audio facilities. It something a lot closer to a "universal" Csound application. And it would be easy to do user-defined widgets. Frankly, this could save us all a huge load of work. Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie ------------------------------------------------------------------------------ Android is increasing in popularity, but the open development platform that developers love is also attractive to malware creators. Download this white paper to learn more about secure code signing practices that can help keep Android apps secure. http://pubads.g.doubleclick.net/gampad/clk?id=65839951&iu=/4140/ostg.clktrk _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
Date | 2013-11-04 20:53 |
From | Michael Gogins |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
Yes, but you still have one binary for each platform that way. This way, you have one binary for all platforms. Regards, On Nov 4, 2013 2:56 PM, "Victor Lazzarini" <Victor.Lazzarini@nuim.ie> wrote:
In a way this is already achievable via Javasound and JAWS. We did a proof-of-concept study a couple of years ago that indicated it could be done, with |
Date | 2013-11-05 00:04 |
From | Michael Gogins |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
This page: compares Emscripten to PNaCl. The same demos will run either in Emscripten, or in PNaCL. You have get the latest Google dev channel Chrome Canary to run PNaCl, at least on Windows.
This is very thought-provoking. You are watching two mature runtimes for bytecode compiled from C or C++. According to the developer, using the latest Firefox gets even faster performance out of Emscripten. I am going to try that now...
Regards, Mike =========================== Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com On Mon, Nov 4, 2013 at 3:53 PM, Michael Gogins <michael.gogins@gmail.com> wrote:
|
Date | 2013-11-05 10:16 |
From | Victor Lazzarini |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
One advantage of NaCl over pure Javascript is the audio implementation, which I reckon is faster (by the looks of it). Victor On 5 Nov 2013, at 00:04, Michael Gogins wrote: > This page: > > http://flohofwoe.blogspot.com/2013/08/emscripten-and-pnacl-build-systems.html > > compares Emscripten to PNaCl. > > The same demos will run either in Emscripten, or in PNaCL. > > You have get the latest Google dev channel Chrome Canary to run PNaCl, at least on Windows. > > This is very thought-provoking. You are watching two mature runtimes for bytecode compiled from C or C++. > > According to the developer, using the latest Firefox gets even faster performance out of Emscripten. I am going to try that now... > > Regards, > Mike > > > =========================== > Michael Gogins > Irreducible Productions > http://michaelgogins.tumblr.com > Michael dot Gogins at gmail dot com > > > On Mon, Nov 4, 2013 at 3:53 PM, Michael Gogins |
Date | 2013-11-05 13:06 |
From | Michael Gogins |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
There appear to be some differences with respect to threading. PNaCl looks better for threading. I agree with Edward that standards are very, very important. One baseline assumption of Csound is that it rests upon software engineering and computer protocol standards. I'd hate to gain a universal platform and lose parallel execution in Csound, however. Here's what appears to be an informed comparison between asm.js (Emscripten with optimization, essentially) and PNaCl: http://games.greggman.com/game/thoughts-on-asm-js-vs-pnacl/ Regards, =========================== One advantage of NaCl over pure Javascript is the audio implementation, which I reckon is faster (by the looks of it).
Victor On 5 Nov 2013, at 00:04, Michael Gogins wrote: > This page: > > http://flohofwoe.blogspot.com/2013/08/emscripten-and-pnacl-build-systems.html > > compares Emscripten to PNaCl. > > The same demos will run either in Emscripten, or in PNaCL. > > You have get the latest Google dev channel Chrome Canary to run PNaCl, at least on Windows. > > This is very thought-provoking. You are watching two mature runtimes for bytecode compiled from C or C++. > > According to the developer, using the latest Firefox gets even faster performance out of Emscripten. I am going to try that now... > > Regards, > Mike > > > =========================== > Michael Gogins > Irreducible Productions > http://michaelgogins.tumblr.com > Michael dot Gogins at gmail dot com > > > On Mon, Nov 4, 2013 at 3:53 PM, Michael Gogins <michael.gogins@gmail.com> wrote: > Yes, but you still have one binary for each platform that way. This way, you have one binary for all platforms. > > Regards, > Mike > > On Nov 4, 2013 2:56 PM, "Victor Lazzarini" <Victor.Lazzarini@nuim.ie> wrote: > In a way this is already achievable via Javasound and JAWS. We did a proof-of-concept study a couple of years ago that indicated it could be done, with > the java Csound wrapper with Csound linked statically in it. With natively compiled csound jnis, JAWS then delivers the correct JNI to each > client OS. > > Victor > On 4 Nov 2013, at 19:40, Michael Gogins wrote: > > > What is motivating me here is the thought of something like Csound 6 for Android that runs on any browser, or at least on Chrome on any platform, and ties into JavaScript as well as into the browser's image, video, and above all audio facilities. It something a lot closer to a "universal" Csound application. And it would be easy to do user-defined widgets. Frankly, this could save us all a huge load of work. > > Dr Victor Lazzarini > Senior Lecturer > Dept. of Music > NUI Maynooth Ireland > tel.: +353 1 708 3545 > Victor dot Lazzarini AT nuim dot ie > > > > > ------------------------------------------------------------------------------ > Android is increasing in popularity, but the open development platform that > developers love is also attractive to malware creators. Download this white > paper to learn more about secure code signing practices that can help keep > Android apps secure. > http://pubads.g.doubleclick.net/gampad/clk?id=65839951&iu=/4140/ostg.clktrk > _______________________________________________ > Csound-devel mailing list > Csound-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-devel > > ------------------------------------------------------------------------------ > November Webinars for C, C++, Fortran Developers > Accelerate application performance with scalable programming models. Explore > techniques for threading, error checking, porting, and tuning. Get the most > from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk_______________________________________________ > Csound-devel mailing list > Csound-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-devel Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie ------------------------------------------------------------------------------ November Webinars for C, C++, Fortran Developers Accelerate application performance with scalable programming models. Explore techniques for threading, error checking, porting, and tuning. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-devel |
Date | 2013-11-05 16:08 |
From | Edward Costello |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None None None |
Date | 2013-11-05 17:01 |
From | Victor.Lazzarini@nuim.ie |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None None None |
Date | 2013-11-05 17:25 |
From | Michael Gogins |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
There many Internet (and other software-related) standards that began as proprietary standards, then became accepted and standardized by the Internet community, whatever that is, which is the closest approximation to functioning cooperative socialism that the world has yet experienced. When PNaCl is enabled by default in the standard edition of Chrome that people download on all platforms, that is when we will see if PNaCl becomes a true standard. And as you say, if that happens, then building Csound for PNaCl will be a no-brainer. As, probably, the primary target.
BUT NOTE: In the meantime, with my current Chrome installed in the office, Version 30.0.1599.101 m, if I start it up with the --enable-pnacl flag, it will run the PNaCl demo code just fine. So in reality, this is already out there, just not turned on by default.
Regards, Mike =========================== Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com On Tue, Nov 5, 2013 at 11:08 AM, Edward Costello <edwardcostello@gmail.com> wrote:
|
Date | 2013-11-05 17:50 |
From | Michael Gogins |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
Sorry to keep going on about this, but it fascinates me and may turn out to be important... PNaCl on Android will lag PNaCl on ChromeOS, Windows, and OS X because the plugin API has to be fitted to a non-POSIX style IPC facility.
I'm afraid I see this as a serious obstacle. I wouldn't proceed with Csound if a PNaCl binary didn't run in ALL browsers. This is very definitely NOT a problem with Emscripten, I just now tested that Lua runs (as a cross-compiled Lua to JavaScript Web page) on Android. NOT LuaJIT, but Lua: http://kripken.github.io/lua.vm.js/repl.html
Regards, Mike =========================== Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com On Tue, Nov 5, 2013 at 12:25 PM, Michael Gogins <michael.gogins@gmail.com> wrote:
|
Date | 2013-11-05 18:31 |
From | Richard Dobson |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
All in all, this sounds like a very good candidate for a Google "summer of Code" project. Good for Csound, good for Google. I hope the emscripten work continues too, even with the threads limitation. Maybe HTML5 "worker" threads can be recruited for it? For schools, even a cut-down single-thread Csound in a browser would be a major resource, and available even on the single browser (IE, inevitably) which the majority of Schools have, um, standadized on. Richard Dobson On 05/11/2013 17:25, Michael Gogins wrote: > There many Internet (and other software-related) standards that began as > proprietary standards, then became accepted and standardized by the > Internet community, whatever that is, which is the closest approximation > to functioning cooperative socialism that the world has yet experienced. > > When PNaCl is enabled by default in the standard edition of Chrome that > people download on all platforms, that is when we will see if PNaCl > becomes a true standard. > > And as you say, if that happens, then building Csound for PNaCl will be > a no-brainer. As, probably, the primary target. > > BUT NOTE: In the meantime, with my current Chrome installed in the > office, Version 30.0.1599.101 m, if I start it up with the > --enable-pnacl flag, it will run the PNaCl demo code just fine. So in > reality, this is already out there, just not turned on by default. ------------------------------------------------------------------------------ November Webinars for C, C++, Fortran Developers Accelerate application performance with scalable programming models. Explore techniques for threading, error checking, porting, and tuning. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
Date | 2013-11-05 18:52 |
From | Michael Gogins |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
Doing both PNaCl and Emscripten is way too much for me... if I do anything it will be one, or the other. Regards, Mike =========================== Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com On Tue, Nov 5, 2013 at 1:31 PM, Richard Dobson <richarddobson@blueyonder.co.uk> wrote: All in all, this sounds like a very good candidate for a Google "summer |
Date | 2013-11-05 19:53 |
From | Richard Dobson |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
I don't understand. Freom the original thread ("Csound on the Web....") Edward and Steven are working on emscripten, so that is at the very least a team of two. So I see no expectation or assumption that you (or anyone) would be doing both. Both seem like eminently (and independently) teamable projects. I have only become aware of them through these threads, so am in no position to contribute, need to do a bit of learning first. But if there has to be a choice, I guess I would vote for Emscripten simply because of the (almost) browser-independence. It also seems further along. There is a very cool demo of Gnuplot running under it, among other things. Richard Dobson On 05/11/2013 18:52, Michael Gogins wrote: > Doing both PNaCl and Emscripten is way too much for me... if I do > anything it will be one, or the other. > > Regards, > Mike > > ------------------------------------------------------------------------------ November Webinars for C, C++, Fortran Developers Accelerate application performance with scalable programming models. Explore techniques for threading, error checking, porting, and tuning. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
Date | 2013-11-05 20:07 |
From | Steven Yi |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
I'd note that with the Csound build we're working on, the source of timing that is driving Csound would be the Web Audio node's own audio callback, at least for this version. Not having threads isn't great, but we can live without it for the short term. Long term, I spoke on IRC with a developer who was working on implementing pthreads for emscripten, but it sounded like it would be a while before there was something to try out. I don't know IE's status regarding web audio. It might be the odd one out in all of this. On Tue, Nov 5, 2013 at 1:31 PM, Richard Dobson |
Date | 2013-11-05 20:09 |
From | Steven Yi |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
I think if we add an access layer, similar to CsoundObj, we might be able to sufficiently hide the implementation, such that one could use one or the other. That would protect anyone using either moving forward. But first thing is to get this Emscripten thing working. :) On Tue, Nov 5, 2013 at 1:52 PM, Michael Gogins |
Date | 2013-11-05 20:17 |
From | Michael Gogins |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
I hope I didn't give the impression that I wanted to do these projects in place of, or alongside of, anything Steven does! To expand upon what I was TRYING to say: I think that computing MAY change in some very good ways soon. It would be wonderful if a single binary target could execute on all common platforms, even if at somewhat slower speed, even if only in a browser. Because these are the early days, I feel a need to stay abreast of developments and, if possible, to help guide Csound into choosing the most effective path forward. This is not an easy sort of choice, but making the right choice now can lead to big payoffs in the near future.
I wouldn't dream of taking on developing any new port of Csound on my own. Not because I couldn't do it, because I could, but because it would interfere with my real job, which is making music. However, as with the Android port of Csound 6, I do find good reasons to take part so that I can ensure that the features that I want and need for composing are present.
I don't think it is nearly clear which, or indeed whether any, of these platforms are truly suitable for Csound. Right now, this is all I am trying to figure out. Once the platform/toolchain is clarified, I will become involved, if possible, to the extent necessary to ensure that, if at ALL possible,
(1) Csound will run multi-threaded, as it does now on computers. This is critical, not for me personally, but for the future of Csound in an increasingly parallel world. We can't afford to forget this for a second.
(2) Csound will support score generation from the orchestra header, as it now does either with Csound code, or with the Lua opcodes. Indeed, I would prefer to implement the ability to run my existing Lua code (this ought to be possible). This is critical for me personal.
(3) Csound will support user-defined widgets, as now is done with Cabbage or CsoundQt. This is not critical for me personally, but is highly desirable. (4) Csound will support writing both to real-time audio devices, and to soundfiles on the local filesystem. This is critical for me personally.
(5) Csound will be able to operate animated 3-dimensional graphics via control channels or opcodes or JavaScript or something. Not critical for me, but I think it would be so cool. If these things would all happen without me, I'd be very happy, but I'd also be happy to push things along. Regards, Mike =========================== Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com On Tue, Nov 5, 2013 at 2:53 PM, Richard Dobson <richarddobson@blueyonder.co.uk> wrote: I don't understand. Freom the original thread ("Csound on the Web....") |
Date | 2013-11-06 15:56 |
From | Stephen Kyne |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
Hi all, Interesting discussion. I've looked at PNaCL before and tinkered with the SDK a bit. Some of the examples are very impressive. Something related to Csound might be audiotool. I wouldn't mind taking a look at a Csound port after the VS2013 feature is more stable/completed. Stephen Date: Tue, 5 Nov 2013 15:17:12 -0500 From: michael.gogins@gmail.com To: csound-devel@lists.sourceforge.net Subject: Re: [Cs-dev] Cross compiling C++ for various virtual machines I hope I didn't give the impression that I wanted to do these projects in place of, or alongside of, anything Steven does! To expand upon what I was TRYING to say: I think that computing MAY change in some very good ways soon. It would be wonderful if a single binary target could execute on all common platforms, even if at somewhat slower speed, even if only in a browser. Because these are the early days, I feel a need to stay abreast of developments and, if possible, to help guide Csound into choosing the most effective path forward. This is not an easy sort of choice, but making the right choice now can lead to big payoffs in the near future.
I wouldn't dream of taking on developing any new port of Csound on my own. Not because I couldn't do it, because I could, but because it would interfere with my real job, which is making music. However, as with the Android port of Csound 6, I do find good reasons to take part so that I can ensure that the features that I want and need for composing are present.
I don't think it is nearly clear which, or indeed whether any, of these platforms are truly suitable for Csound. Right now, this is all I am trying to figure out. Once the platform/toolchain is clarified, I will become involved, if possible, to the extent necessary to ensure that, if at ALL possible,
(1) Csound will run multi-threaded, as it does now on computers. This is critical, not for me personally, but for the future of Csound in an increasingly parallel world. We can't afford to forget this for a second.
(2) Csound will support score generation from the orchestra header, as it now does either with Csound code, or with the Lua opcodes. Indeed, I would prefer to implement the ability to run my existing Lua code (this ought to be possible). This is critical for me personal.
(3) Csound will support user-defined widgets, as now is done with Cabbage or CsoundQt. This is not critical for me personally, but is highly desirable. (4) Csound will support writing both to real-time audio devices, and to soundfiles on the local filesystem. This is critical for me personally.
(5) Csound will be able to operate animated 3-dimensional graphics via control channels or opcodes or JavaScript or something. Not critical for me, but I think it would be so cool. If these things would all happen without me, I'd be very happy, but I'd also be happy to push things along. Regards, Mike =========================== Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com On Tue, Nov 5, 2013 at 2:53 PM, Richard Dobson <richarddobson@blueyonder.co.uk> wrote: I don't understand. Freom the original thread ("Csound on the Web....") ------------------------------------------------------------------------------ November Webinars for C, C++, Fortran Developers Accelerate application performance with scalable programming models. Explore techniques for threading, error checking, porting, and tuning. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-devel |
Date | 2013-11-06 18:25 |
From | Richard Dobson |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
But is that really using PNaCl? It clearly requires Flash. Richard Dobson On 06/11/2013 15:56, Stephen Kyne wrote: > Hi all, > > Interesting discussion. I've looked at PNaCL before and tinkered with > the SDK a bit. > Some of the examples are very impressive. Something related to Csound > might be audiotool. > https://chrome.google.com/webstore/detail/audiotool/bkgoccjhfjgjedhkiefaclppgbmoobnk > ------------------------------------------------------------------------------ November Webinars for C, C++, Fortran Developers Accelerate application performance with scalable programming models. Explore techniques for threading, error checking, porting, and tuning. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
Date | 2013-11-06 19:57 |
From | Stephen Kyne |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
Hmm damn, you're right. I must have been thinking of something else.
> Date: Wed, 6 Nov 2013 18:25:24 +0000 > From: richarddobson@blueyonder.co.uk > To: csound-devel@lists.sourceforge.net > Subject: Re: [Cs-dev] Cross compiling C++ for various virtual machines > > But is that really using PNaCl? It clearly requires Flash. > > Richard Dobson > > > On 06/11/2013 15:56, Stephen Kyne wrote: > > Hi all, > > > > Interesting discussion. I've looked at PNaCL before and tinkered with > > the SDK a bit. > > Some of the examples are very impressive. Something related to Csound > > might be audiotool. > > https://chrome.google.com/webstore/detail/audiotool/bkgoccjhfjgjedhkiefaclppgbmoobnk > > > > > ------------------------------------------------------------------------------ > November Webinars for C, C++, Fortran Developers > Accelerate application performance with scalable programming models. Explore > techniques for threading, error checking, porting, and tuning. Get the most > from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk > _______________________________________________ > Csound-devel mailing list > Csound-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-devel |
Date | 2013-11-11 13:44 |
From | Stephen Kyne |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
http://www.leaningtech.com/duetto/ http://allievi.sssup.it/techblog/archives/852 Duetto seems like another alternative not too dissimilar from emscripten. Might be worth taking a look. Stephen From: stevek@outlook.ie To: csound-devel@lists.sourceforge.net Date: Wed, 6 Nov 2013 19:57:36 +0000 Subject: Re: [Cs-dev] Cross compiling C++ for various virtual machines Hmm damn, you're right. I must have been thinking of something else.
> Date: Wed, 6 Nov 2013 18:25:24 +0000 > From: richarddobson@blueyonder.co.uk > To: csound-devel@lists.sourceforge.net > Subject: Re: [Cs-dev] Cross compiling C++ for various virtual machines > > But is that really using PNaCl? It clearly requires Flash. > > Richard Dobson > > > On 06/11/2013 15:56, Stephen Kyne wrote: > > Hi all, > > > > Interesting discussion. I've looked at PNaCL before and tinkered with > > the SDK a bit. > > Some of the examples are very impressive. Something related to Csound > > might be audiotool. > > https://chrome.google.com/webstore/detail/audiotool/bkgoccjhfjgjedhkiefaclppgbmoobnk > > > > > ------------------------------------------------------------------------------ > November Webinars for C, C++, Fortran Developers > Accelerate application performance with scalable programming models. Explore > techniques for threading, error checking, porting, and tuning. Get the most > from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk > _______________________________________________ > Csound-devel mailing list > Csound-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-devel ------------------------------------------------------------------------------ November Webinars for C, C++, Fortran Developers Accelerate application performance with scalable programming models. Explore techniques for threading, error checking, porting, and tuning. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/csound-devel |
Date | 2013-11-11 14:44 |
From | Michael Gogins |
Subject | Re: [Cs-dev] Cross compiling C++ for various virtual machines |
Attachments | None None |
Thanks for the links, Duetto looks very interesting. The technology is different, the end result is the same. I think the technology might be better. I will investigate further. Regards,
Mike ----------------------------------------------------- Michael GoginsIrreducible Productions http://michaelgogins.tumblr.com Michael dot Gogins at gmail dot com On Mon, Nov 11, 2013 at 8:44 AM, Stephen Kyne <stevek@outlook.ie> wrote:
|