??? 09/05/04 01:13 Read: times |
#76898 - RE: Defining the project & new ideas Responding to: ???'s previous message |
The capability proposed for this platform is perhaps a set of interfaces. If these interfaces where conceptualize one way, a "central processor" exercises these interfaces to configure the "unit".
If the interfaces where logical, one version of this platform could implement as surplus keyboard, with TV output, SRAM Cartridge store. A binary compatible but alternate solution might implement PC type I/O (keyboard, mouse, Mag/Silicon Disk, true color VGA). I agree completely. I was just talking to George Pallis via Instant Messenger and he suggested the idea of a 160x160 PDA-style version. I'm not against this although I'd personally prefer higher resolution even for the initial version of the unit. But my thought seems to parallel yours--that this could be more of a "standard" interface for dealing with external devices that, when put together and run under an agreed upon OS "API", achieves a certain level of standardization such that the programs one person writes could be run on another unit. The program running wouldn't have to care whether the output is a 160x160 LCD, a TV, a monitor, or whether it is driven by an 8052 video or a VGA chipset. The user program would load into the OS which would, in turn, send standardized commands to each of the components (video, I/O, audio, etc.). How each of those components are implemented isn't actually important. Everyone could implement them any way they want to support their preferred manner of implementation. That said, I do think we should establish a standard base unit which is assumed to be the "standard" version. That way the schematics could be made available and anyone that put together the unit based on the schematics would have an "up to spec" functioning 8052 computer capable of running our OS. Also i believe this thread began in the spirit "I think an 8052 computer ought to rival or exceed the capabilities of those machines."
Art pretty much exists for stand-alone Disk, Kybd, Mouse and Network solutions using '51 technology. If these implementations where "unified" by an accepted interface, platform scaling could occur. It must be said, however, that doing this you just might lose the "spirit" of the ATARI 800 whose features where physically, instead of logically specific. I've kind of regretted mentioning the Atari 400/800 ever since I posted the original message. As you correctly point out, my intent was to foster a discussion on the possibility of designing an 8052-based computer that would roughly rival the capabilities of the Atari 400/800. My intent was NOT to precisely duplicate or emulate the Ataris with an 8052. As you point out, there is existing art for virtually every individual component of this project. I propose that we build on the existing open source art and make that part of the system. The main challenge here is agreeing on a standard design for a "base system", agreeing upon communication protocols between the primary 8052 and the standard 8052-based components (audio, video, and I/O as has been suggested), and desinging and coding an operating system capable of booting the 8052 off of the agreed-upon storage device and subsequently loading, running, and unloading programs from storage on demand. The OS would also provide a standardized API that compliant programs would call--for example, an LCALL to 0100h might be the API for writing a line from X1,Y1 to X2,Y2 onto video, 0103h might be the API for clearing the screen, etc. etc. But I suspect a lot of the low-level work for implementing this already exists and doesn't have to be written from scratch. This is mostly a challenge of standarization of hardware, inter-8052 communication protocols, and the underlying OS itself. Regards, Craig Steiner |