What programming language is used for programming ...
- KVRAF
- 6478 posts since 16 Dec, 2002
C, C++ and several forms of assembly languages.
http://en.wikipedia.org/wiki/C_(programming_language)
http://en.wikipedia.org/wiki/C++
http://en.wikipedia.org/wiki/Assembly_language
http://en.wikipedia.org/wiki/C_(programming_language)
http://en.wikipedia.org/wiki/C++
http://en.wikipedia.org/wiki/Assembly_language
-
- KVRian
- Topic Starter
- 514 posts since 2 May, 2004
What is the "production line"? Can an C++ file be used directly to burn a ROM in a chip?Or somebody has to rewrite everything in assembler?Kingston wrote:C, C++ and several forms of assembly languages.
http://en.wikipedia.org/wiki/C_(programming_language)
http://en.wikipedia.org/wiki/C++
http://en.wikipedia.org/wiki/Assembly_language
-
- KVRAF
- 3404 posts since 15 Sep, 2002
C compilers meant for embedded systems have linkers that can deal with generating code that can go to ROM.maki wrote:What is the "production line"? Can an C++ file be used directly to burn a ROM in a chip?Or somebody has to rewrite everything in assembler?Kingston wrote:C, C++ and several forms of assembly languages.
http://en.wikipedia.org/wiki/C_(programming_language)
http://en.wikipedia.org/wiki/C++
http://en.wikipedia.org/wiki/Assembly_language
-
- KVRist
- 214 posts since 29 Dec, 2006
GCC will do it, it's used in that capacity a lot. You just have to write your own startup file in assembly to intialise the stack etc. and create your own linker scripts.mistertoast wrote:C compilers meant for embedded systems have linkers that can deal with generating code that can go to ROM.
- KVRAF
- 6478 posts since 16 Dec, 2002
you basically code as you would on native platforms, but with lots more limitations to take into account with memory usage, bitdepths and datatypes. Then you compile the code into whatever format those sharc or motorola chips (most common in audio DSP) and their BIOSes expect, and flash them in. This part is usually quite transparent to the coder as the dev platforms for these tasks are very advanced these days. (I've used some sharc IDE and it was much like any other IDE)
These days there are a few synths and digital mixers that run (at least partly) on linux and coding for them is very much like it is on native platforms. You end up with "softsynths" and "plugins" for whatever proprietary formats the OS expects.
These days there are a few synths and digital mixers that run (at least partly) on linux and coding for them is very much like it is on native platforms. You end up with "softsynths" and "plugins" for whatever proprietary formats the OS expects.
-
- KVRAF
- 3404 posts since 15 Sep, 2002
Yeah. I keep meaning to write a game for MAME or for a Sega Genesis emulator. That's pretty much how you do it. I think there are some example startups around.vonRed wrote:GCC will do it, it's used in that capacity a lot. You just have to write your own startup file in assembly to intialise the stack etc. and create your own linker scripts.mistertoast wrote:C compilers meant for embedded systems have linkers that can deal with generating code that can go to ROM.
-
- KVRAF
- 2029 posts since 21 Jul, 2004
Isn't it ever from the ground up? chip design --> it's own assembly language --> it's own integrated memory. Like in high end reverbs arn't the chips designed to do that job very efficiently. I thought that was why it's so hard to reproduce the quality of hardware reverbs in computer software.
Do not lick the fablanky
-
- KVRAF
- 3404 posts since 15 Sep, 2002
I can't imagine it ever happens that way nowadays.funkadil wrote:Isn't it ever from the ground up? chip design --> it's own assembly language --> it's own integrated memory. Like in high end reverbs arn't the chips designed to do that job very efficiently. I thought that was why it's so hard to reproduce the quality of hardware reverbs in computer software.
-
- KVRian
- 1270 posts since 9 Sep, 2005 from Oulu, Finland
It probably wouldn't be too hard nowadays. IMHO, it just happens to be so that the companies owning the best algorithms for reverb (like Lexicon and TC) don't want them flying around freely as Intel/PPC/etc-code, as that would make it a lot easier to reverse engineer how the sound processing is made up. And of course, there is the piracy issue...Not really in the hardware makers' interests...funkadil wrote:Isn't it ever from the ground up? chip design --> it's own assembly language --> it's own integrated memory. Like in high end reverbs arn't the chips designed to do that job very efficiently. I thought that was why it's so hard to reproduce the quality of hardware reverbs in computer software.
-X
- KVRAF
- 6478 posts since 16 Dec, 2002
most commonly,funkadil wrote:What sorts of chips do DSP cards use?
http://www.google.fi/search?hl=fi&clien ... =Hae&meta=
http://www.google.fi/search?hl=fi&clien ... =Hae&meta=
there's a few other major DSP chip manufacturers (analog devices for one), and indeed occasionally it happens that a company designs their own audio DSP chip from ground up:
http://www.fairlightau.com/default_content.html (for example)
or sometimes a company scores a very powerful processor originally designed for other tasks than audio. The UAD-1 is a rather famous case of this as their card is powered by a specialised video accelerator chip.
and sometime in *very* near future we'll be able to completely unlock all those nvidia and AMD/ATI turbo GPUs for audio and general DSP usage. It's already happening actually, but the current solutions are very rough on the edges.
- Beware the Quoth
- 35517 posts since 4 Sep, 2001 from R'lyeh Oceanic Amusement Park and Funfair
Sometimes; there are a few companies around there with proprietary hardware; however I suspect a lot of that is hardcoded for very specific functions, and, I'd suspect, used in conjunction with something more generic.funkadil wrote:Isn't it ever from the ground up? chip design --> it's own assembly language --> it's own integrated memory.
Expect a lot of that stuff to be programmed in assembler, if its programmable; even if the company were able to port something like the GCC backend to their own hardware, a generic compiler wouldnt be able to optimise code as well as a talented human, and you'd probably need that human and a couple of his clones to even start working on a decent implementation of those optimisations for a compiler.
At least the common operations involved, yes; almost certainly on a heavily-parallelised basis, ie lots of the same operation all happening at once...Like in high end reverbs arn't the chips designed to do that job very efficiently.
Not really. I think its more to do with the fact that these companies have teams with hundreds of man-years of experience, a codebase to match, and are working on refinements of algorithms most programmers will never get to hear about.I thought that was why it's so hard to reproduce the quality of hardware reverbs in computer software.
An idiot on Set Theory:
"In some cases there is an object called red that contains everything that is red. In much the same way a pot is a plate."
"In some cases there is an object called red that contains everything that is red. In much the same way a pot is a plate."