What flavor of Linux do you recommend for 64 bit coding?

[MJaoune]

I would recommend you to use WXWidgets to create your next projects, since it is compatible with all Operating Systems, and it also have better functions, and the best documentation, but YMMV.

Best,
MJaoune

[Eccentro]

Wine: I'll learn what I can from it.  Ultimately, the more code that is under my direct creative control, the better anyway.

WXWidgets:  I'm checking it out.  It looks pretty handy.

Is there something like DirectX for Linux that I should work with in to order to make my apps cooperate with other apps that use audio buffer IO and plug-in capabilities?  Oh, maybe there's a VST for Linux.  I'll look for that now.

[Bryant Keller]

OpenGL/OpenAL is probably about the closest thing to DX on Linux. If you're going to be using OOP heavy wrappers like wxWidgets, you might also want to look into the Ogre framework which provides a platform independent wrapper to OpenGL/OpenAL/DirectX/etc.

[Eccentro]

Sweet.  My search for this kind of info had me banging up against broken links and misdirection for the last hour.  I knew about OpenGL, but not about OpenAL.

I've got some ideas for standards that might find a home in the Linux audio niche, but I should see what's already happening to avoid stepping on toes.

Over on the Windows side, I stripped my main app forms down to the minimum and created my own light weight consolidated control system for them.  It uses a lot less memory and greatly reduces the number of objects and messages that need to be managed.  It's actually pretty sweet.  Perhaps I can continue in that mode with Ogre - or maybe I should work directly with OpenGL.  I'll look at the pros and cons.

Thanks for the tips.

[Eccentro]

OpenAL looks very similar to one of the many projects that I have half done.  I probably shouldn't  describe it in a public forum.  There are sufficient differences to make mine worth finishing.  My dev history is documented, so there's no intellectual property issue.   

[MJaoune]

I use OpenGL on both, Windows, and Linux, so OpenGL is not for linux only, it is for all Operating Systems.

Best,
MJaoune

Edit: Inappropriate content removed by forum moderation.

[Eccentro]

Yep.  I bought a really big book a few years back.  I think it was called the OpenGL Bible.  I've intended to get into it.  I've always liked the idea of OpenGL.

I see in some docs, though, that, in order to draw a window, code still needs to access the API.  It then paints on it using OpenGL functions.  I want to incorporate my consolidated control system into my windows.

Anyway, I'm reading this:  http://www.aboutdebian.com/linux.htm

It makes me think that maybe I should gather up all of my audio ideas and bring them together to form a new shell, dedicated to the tasks that my applications facilitate.  The shell could support other people's software.  I would document the shell's systems, standards and libraries in an SDK.

What do you think of that idea?  It would be an audio production shell for Linux.  It could be expanded for video production and gaming.  Then, in the future, when I'm producing video games and cognitive models, I could target them to run under this shell.

Is this a good direction to go in?  Why or why not? 

[MJaoune]

Yep.  I bought a really big book a few years back.  I think it was called the OpenGL Bible.  I've intended to get into it.  I've always liked the idea of OpenGL.

I see in some docs, though, that, in order to draw a window, code still needs to access the API.  It then paints on it using OpenGL functions.  I want to incorporate my consolidated control system into my windows.

You can draw a window using WXWidgets, and you can control input using WXwidgets or SFML.

It makes me think that maybe I should gather up all of my audio ideas and bring them together to form a new shell, dedicated to the tasks that my applications facilitate.  The shell could support other people's software.  I would document the shell's systems, standards and libraries in an SDK.

Is this a good direction to go in?  Why or why not? 

I don't know, you can ask Cyrill about that, he is experienced in Linux, but I think you need an extra hand(s) for this idea. If you want a reference, you can always take a look at Linux's source codes.

Best,
MJaoune

[Eccentro]

Thanks.  I'm in brainstorming mode and am receptive to input.

BTW:  How does one make a living at this?  It would be nice to buy a better computer and a few cases of Ramen.   

[MJaoune]

Thanks.  I'm in brainstorming mode and am receptive to input.

BTW:  How does one make a living at this?  It would be nice to buy a better computer and a few cases of Ramen.   

Is your computer that bad? Can you give me your system Specs? And ofcourse, ramen is always important

Best,
MJaoune

[Eccentro]

No.  It's not that bad.  I just like the idea of using those 64bit functions that are probably disabled on this one.  You know, the batch instructions targeted at dsp.  I think I'll find that they don't work.  I could be wrong though.

I'm like Tim Taylor.  I always want more power.   

It's actually good to have limits, though.  I don't want to write code that nobody else can run.

I would love to turn this into a gig that produces a modest income.  I need to get a bit of sleep. cu

[MJaoune]

No.  It's not that bad.  I just like the idea of using those 64bit functions that are probably disabled on this one.  You know, the batch instructions targeted at dsp.  I think I'll find that they don't work.  I could be wrong though.

I'm like Tim Taylor.  I always want more power.   

It's actually good to have limits, though.  I don't want to write code that nobody else can run.

I would love to turn this into a gig that produces a modest income.  I need to get a bit of sleep. cu

You can write a code for many systems on your own single computer, and ask someone to test it, like if you don't have a x86(32) computer you can ask someone that has this type of a computer to test it for you.

Good Night ,
MJaoune

[Eccentro]

I've decided on a first program to write with NASM.

Back in 1995, I came up with a method to validate the accuracy of a given sine function with absolute precision out to n bits.  The test works perfectly, but I haven't yet been able to figure out how to modify it to generate perfect sine values, from scratch, in a timely manner.  I do hope to nail that one soon.   

I'll recreate the validation algorithm as a function that takes q (quadrant 0-3) and n (bits) as arguments and returns a true or false result to indicate whether or not the sine function is accurate out to n bits for the quadrant being tested.

If I build that shell we talked about, then I'll include this code in there somewhere.   

[MJaoune]

Good luck And if you need any help or support, NASM Forum is always ready

Best,
Mahmoud

[Eccentro]

Thanks.  I'm sure that I will and that any such help will be acknowledged with a special thanks in the shell "credits."

I've been familiarizing myself with gcc on Slackware and tonight, I'll install NASM from the command line.     <-- grasshopper

BTW:  Get the champagne ready.  I think I just solved the old sine problem in way that makes it a very useful algorithm.  <--  I'm soooo stoked right now.