david@wubios.wustl.edu (David J. Camp) (11/26/90)
I just had an idea. Perhaps it is possible for a 34010 to emulate an 8514/A. Is there someone out there familiar with the hardware that can verify or deny this? Is there any reason to do so? I cannot think of a reason, other than to allow programs written for the 8514/A to run on systems with a 34010. This is an unlikely scenario, since the OS usually handles the graphics hardware these days. Thanks for any replies. -David- david@wubios.wustl.edu ^ Mr. David J. Camp david%wubios@wugate.wustl.edu < * > +1 314 382 0584 ...!uunet!wugate!wubios!david v "God loves material things."
c37189h@saha.hut.fi (Harri "Haba" Suomalainen) (11/27/90)
In article <1990Nov25.173919.4313@wubios.wustl.edu> david@wubios.wustl.edu (David J. Camp) writes: >I just had an idea. Perhaps it is possible for a 34010 to emulate an >8514/A. Is there someone out there familiar with the hardware that >can verify or deny this? Yap. It has been done by various producters. As an example I could mention Hercules Graphics station. It has 34010 on board. It emulates SVGA, 8514/A and TIGA. There may be hercules etc as well. Not quite sure of that... -hs
geoffs@gssc.UUCP (Geoff Shapiro) (11/29/90)
In article <1990Nov26.203552.18302@santra.uucp> c37189h@saha.hut.fi (Harri "Haba" Suomalainen) writes: >In article <1990Nov25.173919.4313@wubios.wustl.edu> david@wubios.wustl.edu (David J. Camp) writes: >>I just had an idea. Perhaps it is possible for a 34010 to emulate an >>8514/A. Is there someone out there familiar with the hardware that >>can verify or deny this? > >Yap. It has been done by various producters. As an example I could mention >Hercules Graphics station. It has 34010 on board. It emulates SVGA, 8514/A >and TIGA. There may be hercules etc as well. Not quite sure of that... > >-hs Well, if the truth be known, the 34010 cards DO NOT emulate an 8514/A at the hardware level. When people speak of 34010 cards emulating 8514/A they mean that there is a software emulation available for the 8514 AI which provides a compatible software i/f to the 34010 cards. Thus, a program which talked to an 8514/A card ONLY via the AI would most likely also talk to a 34010 card which had an equivalent AI written for it. Any programs that talked directly to the hardware of the 8514/A would not be able to run on a 34010 card without a massive amount of dedicated circuitry on the card to emulate the 8514/A hw i/f. As far as I know there are NO 34010 cards which do this. -- UUCP: tektronix!sequent!gssc!geoffs INTERNET: geoffs@gssc.gss.com PHONE: (503) 641-2200 USMAIL: Graphic Software Systems * GSS is a wholly owned subsidiary * 9590 S.W. Gemini Dr. * of Spectragraphics, Corp. * Beaverton, OR 97005 USA
c37189h@saha.hut.fi (Harri "Haba" Suomalainen) (11/29/90)
In article <6468@gssc.UUCP> geoffs@gssc.UUCP (Geoff Shapiro) writes: >Well, if the truth be known, the 34010 cards DO NOT emulate an 8514/A at the >hardware level. When people speak of 34010 cards emulating 8514/A they mean >that there is a software emulation available for the 8514 AI which provides >a compatible software i/f to the 34010 cards. You are absolutely right on that. >Any programs that talked directly to the hardware of the 8514/A would not be >able to run on a 34010 card without a massive amount of dedicated circuitry on >the card to emulate the 8514/A hw i/f. As far as I know there are NO 34010 >cards which do this. As IBM did not release documentation to the hardware at the start, most programs (all i've run in to) do not use 8514 on the hardware level. Even the original 8514 was used as the beginning (as fortunately still is) used via the device driver software. As there are many software compatible cards available by now, i think it would be extremely stupid to write a program using 8514 only on hw level. -hs
lance@motcsd.csd.mot.com (lance.norskog) (11/30/90)
IBM never documented the 8514/A register set (I think). They have, however, released their sources to the 8514/A-based X Window Server. Sooooo, you can read it and figure it out. Also, several manufacturers have reverse-engineered the chip spec, built their own chips, and publish chip books which contain the 8514/A register set. Any processor could emulate the 8514/A register set by capturing register accesses, holding the bus, and then feeding a value to the 80x86. It would be hokey, but actually fairly easy to implement, if the 8086 bus spec allows playing with timing in this fashion. Lance
geoffs@gssc.UUCP (Geoff Shapiro) (12/01/90)
In article <1990Nov29.132641.11209@santra.uucp> c37189h@saha.hut.fi (Harri "Haba" Suomalainen) writes: |As IBM did not release documentation to the hardware at the start, most |programs (all i've run in to) do not use 8514 on the hardware level. |Even the original 8514 was used as the beginning (as fortunately still is) |used via the device driver software. As there are many software compatible |cards available by now, i think it would be extremely stupid to write a |program using 8514 only on hw level. Unfortunately (or fortunately, depending on how you view it), the two most important sw programs on the market DO talk to the 8514/A interface directly. Namely, the Windows and Presentation Mgr. display drivers! Also, any 3rd party sw which needs to milk the adapter for all the speed it can deliver also talks directly to the hardware (for example, the 8514/A Autocad display drivers available from Panacea, Inc.). -- UUCP: tektronix!sequent!gssc!geoffs INTERNET: geoffs@gssc.gss.com PHONE: (503) 641-2200 USMAIL: Graphic Software Systems * GSS is a wholly owned subsidiary * 9590 S.W. Gemini Dr. * of Spectragraphics, Corp. * Beaverton, OR 97005 USA
robk@cpqhou.uucp (Rob Kiesler) (12/04/90)
in article <1990Nov29.132641.11209@santra.uucp>, c37189h@saha.hut.fi (Harri "Haba" Suomalainen) says: > > As IBM did not release documentation to the hardware at the start, most > programs (all i've run in to) do not use 8514 on the hardware level. > Even the original 8514 was used as the beginning (as fortunately still is) > used via the device driver software. As there are many software compatible > cards available by now, i think it would be extremely stupid to write a > program using 8514 only on hw level. > > -hs It would be stupid, except: 1) Apps which use AI will be slower than those going direct to the hardware, given that the app writer knows what he's doing. 2) AI is only available for Real Mode environments (Read: MS-DOS). AI is NOT available for protected mode environments! I doubt that it will ever be available in protected mode. At the very least, a massive rewrite would be in order, since an app talks to AI using a software interrupt. Rob Kiesler Compaq Computer "Is that a real poncho, or is that a Sears(TM) poncho?" - FZ