PDA

View Full Version : Project64 and other platforms


nogagplz
5th February 2009, 07:48 AM
I can appreciate the difficulty in porting an application made for Windows to another platform such as Linux, BSD or OS X, and can very well see that it will never happen.

However, last time I tried Project 64 in Wine (under Linux), the performance was nearly what you'd expect running it natively on Windows, save for some niggly things that ranged from plain annoying to show stopping (but not completely capable of stuffing the experience). This lead me to think that instead of having to worry about porting it to different OS, perhaps some time and effort can be dedicated to ensuring that it continues to run in Wine, be it in the form of opening a bug for Wine and submitting patches to fix the problem, keeping an up to date appdb (http://appdb.winehq.org) entry for people attempting to get it running, etc.

Obviously some plugins and components will not function correctly due to the way Wine is implemented, but at least it will enable terrific N64 support without having to move heaven and hell.

Cheers

squall_leonhart
5th February 2009, 10:20 AM
Wine is not ours, or the developers of this emulators problem.

Thomas
5th February 2009, 08:17 PM
Going by the earlier versions the emulator is self does not lend it self to portability. There are other emulators available that are more potable you may be more interested in. They may not have the same performance but that's a price you pay.

squall_leonhart
5th February 2009, 10:25 PM
Mupen64++ is fairly good, and is available on Linux

HatCat
5th February 2009, 10:59 PM
I don't understand "portability," but I don't necessarily code either. So someone teach me?

Why is this important if each operating system affects what an emulator is given? If you make things multi-OS aren't you just kinda neutralising everything...it's like Nb1-c3 in chess!! Nf3 doesn't create any permanent weaknesses either, but Nc3 essentially neutralizes the initial startup position while Nf3 loves it and cheers it on.

As far as calculable algorithm coding or something like that there's no Direct3D/OpenGL intervention hardly in something so straightforward, so multi-OS consideration still won't hurt.

Thomas
6th February 2009, 08:38 AM
There are core aspects to operating systems and CPUs that are relatively common. If you stick to these you can generally happily port things with relatively little work. Many programs can be a simple as compiling for what you want to run it on. Or some take it further such as Java are intended to run on whatever systems. Other programs are more dedicated and use specific features of a OS or its hardware. This can make it very tricky to port things but often gives greater performance. In the old day is was not uncommon to go as far as writing programs in assembly that tied you to a specific CPU. Also the more ports you have the more work there is to update everything.

Alpha Dog
6th February 2009, 09:14 AM
Project64 can be made portable. But looking at the 1.4 sources, it would be a pain because Windows specific calls are littered all over the place ;P . They are mostly for GUI though, with some used for multi-threading.

HatCat
6th February 2009, 10:51 PM
Anything of that subject would be better than neutralizing code abandoning each system's uniquity to someone else's time to do the porting for Author.

I can more see manual porting then since it's not like there are aspects of each operating system definitively encouraging emulation of particular systems. At the same time you reminded me of assembly variants based on hardware architecture, so consistency without contradiction can never be maintained to the extent of the hardware the software runs on.