3Do Source Avaliable?


1magus

Member
Joined
Jun 29, 2007
Messages
462
I came accross this and it appears the developer of FreeDO the 3DO emulator has released some part of his source code: 3DO Source
I have no idea how much is there but this is the comments of the developer:

"We've just released sources of main portion of FreeDO (the emulator core).
It is compilable under gcc. Go grab it from http://code.google.com/p/freedo/.
Example QT Gui frontend sources, PBUS plugin and XBUS plugin sources will be added soon.
Admin"

Please delete my other thread I seemed to of posted it on the wrong forum.

I was wondering can somebody check this out and see if this is enough to go on for 3DO emulation?
 
Wow, seriously?

Hot damn, pigs can fly.

Looks on the level. Someone should give a port a try, the modular library core style organization lends itself well to fast ports. But I can't say whether or not it'll be easy enough. It has a few areas that might be problems; take that with a grain of salt though, I just say it after giving it a cursory glance. If it ends up that way someone should run it through a profiler.

What's most interesting to me is I couldn't find a trace of HLE. Since 3DO is defined as a high level spec and not much of a hardware implementation per se, and since all of the leaked documentation was regarding this spec, I always assumed this was a very HLE'd emulator. The FourDO emulator started at some point was also looking to take this approach. But I don't see any suggestion of it here, which begs the question of where all of the hardware understanding came from. It'd be great if we had documentation, then new 3DO emulators could be done (not to say anything is necessarily wrong with this one).
 
Exophase said:
Wow, seriously?

Hot damn, pigs can fly.

Looks on the level. Someone should give a port a try, the modular library core style organization lends itself well to fast ports. But I can't say whether or not it'll be easy enough. It has a few areas that might be problems; take that with a grain of salt though, I just say it after giving it a cursory glance. If it ends up that way someone should run it through a profiler.

What's most interesting to me is I couldn't find a trace of HLE. Since 3DO is defined as a high level spec and not much of a hardware implementation per se, and since all of the leaked documentation was regarding this spec, I always assumed this was a very HLE'd emulator. The FourDO emulator started at some point was also looking to take this approach. But I don't see any suggestion of it here, which begs the question of where all of the hardware understanding came from. It'd be great if we had documentation, then new 3DO emulators could be done (not to say anything is necessarily wrong with this one).
I have no idea what you are saying but I can give it a shot, maybe he analysed how the 3DO worked with the unstable PCI version of it which might explain why it is missing something you are looking for? Btw was I right is this the source code or is it really missing things that would prevent it from compiling?
 
Last edited by a moderator:
Wonderful, fantastic and amazing news :)
I promise a LARGE donation for anyone who can get this to run Need for Speed at 80% speed on the Pandora :)
 
TylerAW said:
I have no idea what you are saying but I can give it a shot, maybe he analysed how the 3DO worked with the unstable PCI version of it which might explain why it is missing something you are looking for? Btw was I right is this the source code or is it really missing things that would prevent it from compiling?

I took a look and all this here appears to a plugin or library for use with a main application which doesnt exist in this repo (yet)
 
Last edited by a moderator:
Gah! Always a step away. Would be so sweet if 3DO source was fully available.

Edit: I went to freedo.org and took a look at the status:

We've just released sources of main portion of FreeDO (the emulator core).
It is compilable under gcc. Go grab it from http://code.google.com/p/freedo/.
Example QT Gui frontend sources, PBUS plugin and XBUS plugin sources will be added soon.

So is it the other way around? Plugins are available and the core isn't?

and then in the forums:

So XBUS and PBUS sources have been released, though that doesn't seem like the entire core, what about CPU, CEL, VDL, DSP, MEM, DMA, CDROM, Interrupts, Timers, Portfolio, Fence, CELDMA+VRAM,CDROM, DAC?

Not quite right. Actually everything is inside the libfreedo.Majority of HW is emulated in madam.cpp and clio.cpp : CEL+DMA+PBUS+timers+Matrix engine are in Madam.cpp, DSP,DMA,interrupts are in Clio.cpp
VDLP and SPORT are separate classes. XBUS/ISO (cdrom) are also separate. The library emulates entire Opera HW.


So then libfreedo, madam.cpp and clio.cpp are also provided right? All classes? I am just wanting to make sure that nobody has to suddenly code from scratch a function of freedo. :p

Yes, all they will need to do is to provide cd data, user input, draw bitmap and play sounds. Typical front end. Why don't you check the sources yourself. http://code.google.com/p/freedo.
 
Yeah, all that stuff they mentioned is there... it didn't seem apparent than anything crucial for a port was missing...
 
I wonder how well this would run on the Pandora?

I know the 3DO CPUs were pretty weak but it had loads of custom crap in it that is hell for emus. Kind of like SNES, weak system but is hell to emulate.

The 3DO had some very good games though for it's time.
 
would be nice to see it run on the pandora.Is the whole thing there then for a potential port?
 
wrath of khan said:
would be nice to see it run on the pandora.Is the whole thing there then for a potential port?

Love to see a port of this...one day.
 
Last edited by a moderator:
Back
Top