Path: utzoo!attcan!uunet!husc6!ncar!hao!hull From: hull@hao.ucar.edu (Howard Hull) Newsgroups: comp.sys.amiga Subject: Re: Memory and the A2024 Hedley monitor. Summary: Too much work Keywords: Solder to silicon? Do us a favor! Message-ID: <540@ncar.ucar.edu> Date: 7 Aug 88 01:02:02 GMT References: <3627@louie.udel.EDU> Sender: news@ncar.ucar.edu Reply-To: hull@hao.UCAR.EDU (Howard Hull) Organization: High Altitude Observatory/NCAR, Boulder CO Lines: 90 Yeah, come to think of it, I did see your first request. I was thinking about how to answer it, but then there was one too many questions for the amount of time I had, plus the fact that I only wanted to address the last one. But now maybe I'll try both of them. >First. Can you have a screen (say workbench (and thusly CLI's etc)) >that is 1024 by 800, and then have other screens behind it >that are 640 by 200 or 640 by 400, and use screen to front >and back gadgets to flip between these things? A screen is a screen is a screen. The screen flipping key will flip the screens ok unless the Amiga software people put in some tests and locks to prevent it. Why would they do that, you no doubt ask? Probably because the 640 by 200 screens would be poorly proportioned and would look like crap, and they don't want to show people crap. They are trying to sell Amiga computers, not give them a poor reputation. The effect would be very similar to the effect you get when you show a 320 by 200 window on a 320 by 400 screen - squashed fonts, flicker, circles turned into ellipses, font strokes thinned to where you can't see them, etc. In one detail I am not sure what they'll do, that is, whether or not the Amiga programmers would put in something to expand the 640 to the width of the 1024 screen (it is 1024 wide (conventional w x h description), right?) or just run pixels into the blitter starting at the left side and going until there were no more pixels left in memory for that line, with screen background "color" filling in the rest of the line. I think this latter one is what they'd have to do, and that's why I think it would come out like crap. You will note that the only reason they can get away with creating 320 x 200 screens pulled down in front of 640 x 400 screens (and they have to go to interlace on both sreens, even so) is that the pixel count is _exactly_ twice for the interlaced hi-res screen - and that's something the electronics can handle in hardware with a divide-by-two binary counter. They don't do so well with dividing by 1.6 [though it would be a trivial matter to design a divide by 1.6 hardware counter for the task of synchronizing the Amiga blitter, but non-trivial to get it installed in 1E6 already existing computers in peoples' offices, homes, and closets. :-) ] >and thus have one mongo big screen, and then have other >applications running in the 640 by 200 screens??? As pointed out by others on the net, you have to have a lot of blitter address space - but it would be possible to do this. > >Second questions: >Have any of the genuiuses with the RFANG's (REALLY Fat Agnus) >figured out a hack to get 1 meg of chip >ram into the Amiga 1000????? >PLEASE PLEASE ? There are probably not very many hardware types that have RFANGs. Even those that do have them are busy with other stuff. When Amiga went from the regular Agnus to the fat Agnus, they did essentially the same thing Intel/IBM did when they went from the 8088 to the 80186. They took some of the surrounding MSI (Medium Scale Integration, i.e. MC7400) logic and integrated it onto the LSI (Large Scale Integration, i.e. MC68881) die. That means that the wires you need to solder to don't exist anymore, but rather are now little tiny silicon etchings on the die inside Fat Agnus. Now they could have exactly transported the outside logic to inside so that the new FA pins would exactly correspond to existing PC traces on some of the logic somewhat removed from the FA, but remember that designers like to change things and are always interested in improving the stuff they did before, taking advantage of certain economies in the new packaging. The result in the case of the Amiga is not only no correspondence between the new Fat Agnus pin names and functions and old A1000 printed circuit traces, but also no place where they particularly agree with points in the remotely located circuits, either. Of course this implies that they changed the detailed nature (not necessarily the function, though) of the surrounding logic packages when they replaced the A1000 logic with the A2000 logic. The equivalent of this also happened in the case of the IBM PC in going from the 8088 8-bit bus (and also the 8086 XT 16-bit bus) to the 80186 IBM PC AT. Things look similar in that there are rectangular cards with resistors, capacitors, and integrated circuit packages all over the place, but of course, no non-smoke-and-fire interconnection between the innards of these cards can take place on any grand scale, I can assure you! >I'm a soldering ace, and i'm not afraid to cut, hack and slice my machine. >give me accurate instructions.... and I'm in happy heaven. So how good are you at soldering to the inside of a glass passivated silicon die? Or if you don't want to do that, how good are you at laying out a new motherboard for your A1000 that incorporates all of A2000 support logic changes? And if you are good at this last one, how good are you at finding a sharp lawyer for when CA sues you for "look and feel"? >Jonathan P. Crone Howard Hull hull@hao.ucar.edu ### And it was said that when God and the devil were discussing the devil's negligence in keeping up his part of the fence line between heaven and hell, and it was mentioned that God had threatened to go to court, the devil said "Oh yeah? and where are YOU going to get a lawyer?!!!" ###