I would say I'm still in the conceptual stage. I've examined the Is and the Os in the documentation and I pretty much know what I want to build, with a few choices still remaining (e.g. SRAM vs. DDR2). But I'm stymied by the details and by the fact that when I started this I was unemployed and had time, although my motivation was low due to the associated emotional challenges. Now I am wonderfully employed, and very happy, but have very little spare time. I've read through "Designing Cards and Drivers for the Macintosh" (IIRC) but I need to go through in more detail and draw a diagram of the ROM structure. But it seems to me that some of what I need to know is in Inside Macintosh and extracting those tidbits is difficult (for me). Also, some of the information I want to examine is contained in Pascal coding examples, and I don't know Pascal. It will be easy enough to pick up, but it's another step back from the actual project. Of course, all of the Apple documentation assumes that the audience is already an expert at building the component they want to build and only needs the Apple specifics, and I've never designed a video card. I haven't found any kind of book about it anywhere. I picked up whatshisnames book about the video typewriter and the follow up book, but it doesn't seem quite applicable to this application. So I'm in one of those situations where I need bits of information in order to truly understand other bits of information. But I don't know in what order I really need the bits in order to make them comprehensible. Worse, I don't actually know which piece will be in which resource. So I need to find all the pieces of information, fail to understand them as I find them, but identify them, then after they've all been found, reorder them and go back, this time in logical sequence and put it all together. Maybe if I was smarter, one pass would be enough... I'm somewhere late in the finding all the pieces without understanding them yet stage, paused just before doing a quick learn of Pascal so I can examine/understand the examples in Inside Macintosh. But I still lack a resource on the finer points of video card design. I understand the basic fucntion, but there are always details which clever people have worked out, which a latecomer like me can greatly benefit from. For example, the various schemes to avoid drawing the top half of the screen and then using changed data for the bottom half of the screen. I'm familiar with the simple expedient of having two screen buffers. But I understand that there are more elegant solutions in use, otherwise video cards would need twice as much VRAM. Is it really all about tracking updates as they come in and only allowing ones which are earlier in the buffer than the current draw point, and delaying the others until the drawing point passes? If so, one would either hold up the host system, or need an additional buffer for the delayed changes. Yes? No? As far as cramming a lot more into the PDS slot with smaller components... Yes, the thing holding folks back is the time and skill to code, not the available technology. Taken to its logical conclusion, we would take a moderate sized FPGA, emulate the 68030 and FPU on board, put a fast memory interface on the FPGA, add a USB2 chipset and fast ethernet I/O, plus an ATA interface and video outputs, then just use the SE/30 board as an I/O peripheral and home to the Apple ROM, with the built-in 9" screen as one display and a connector for a large color display if desired. Alternatively, use a Coldfire MCU, patch the unimplemented 68030 instructions with the exception handler (there's a library at Freescale for this) and then copy the basic design of the Daystar CPU upgrades which already worked out how to put a faster 68030 into the PDS slot. With a little more work patching the firmware, one could use a Coldfire with built in USB and ethernet, but that's a lot more software work. One also loses the FPU functionality, and if your code calls many of the unimplemented instructions it could be pretty slow. But all those projects are dauntingly large. For now, I'd like a video card. Later, an ATA interface. Maybe USB after that, but frankly, without a more skilled programmer's help, I don't think I will be able/willing to write USB drivers. Eventually, it could all go on one card. Jeff Walther