> Chris M wrote:
>
> >But Apple made a mistake with
> >the Mac by not supplying a distinct video ic,
> allowing
> >the 68k to do all the work, and therefore was
> lacking
> >in speed.
> >
>
> That's not quite true. Both the original Mac and
> the Lisa shared memory
> access with the video hardware. The video hardware
> was actually much
> simpler than what most computers used a dedicated
> display chip for. It
> was basically a nothing more than a shift register
> that walked memory
> and spat out video signals.
>
> Half the time the CPU had access to the memory bus,
> the other half the
> video system.
>
> Other contemporaries of the time may have used a
> dedicated IC to do the
> video, *BUT* in most cases, these also shared access
> to memory with the
> CPU. So it was no better. Infact, they were more
> complex because they
> were text mode (40x25 or 80x25) and needed a
> character generator ROM.
> The video IC would read a byte from main memory,
> then turn around an
> read the bitmaps for that character from a ROM and
> display that.
>
> I remember there were various tricks done to get
> various styles
> displayed too. For the Commodore line, there were
> several bitmaps (aka
> fonts today) that implemented primitive graphics.
> The high bit (128)
> was used to invert the bitmap, so the scheme to
> display the cursor was
> to use XOR 128 on and off every second to flash the
> character. There
> was a patent for this simple scheme. Other displays
> used another chunk
> of memory that mapped along with the text to
> implement attributes such
> as underline, flash, inverse, and another set for
> color.
>
> Things like the VIC20 and Commodore 64 had some
> dedicated hardware to do
> sprites and such, it's true, but for normal
> operations, it wasn't too
> much better what the Mac/Lisa had. There were of
> course vector systems
> out there, but these were mostly for games and
> worked in a totally
> different way than raster displays like om the Mac,
> Lisa, Commodore's,
> and PC's.
>
> Even so, they generally had to share the memory with
> the CPU, so there
> was a slowdown due to that. This can be exposed on
> the Commodore 128 by
> going into FAST mode which ran at 2Mhz instead of
> the usual 1Mhz. The
> 40column display would be shut off. (The 80 column
> one which ran off a
> chip similar to the CGA controller still worked.)
> Even the lowly
> TS/1000 had a fast mode that disabled the video
> because it too shared
> it's small memory with the video system.
>
>
> I don't recall whether you had to do special stuff
> to access IBM PC's
> video memory on the CGA cards, perhaps it was
> accessible in memory
> though the video ram as it lived on the ISA card,
> but I do recall it
> displaying snow if you directly wrote to the video
> memory and didn't use
> the INT21 routines in the BIOS. Lots of program
> wrote directly to the
> screen for speed, but had to do so in the vertical
> retrace. (The BIOS
> routines were very slow.)
>
>
> The Lisa ran at 5MHz even though the 68000 was an
> 8MHz cpu due to the
> video circuitry needing access to memory. I'm not
> sure how they fixed
> this for the original Mac. Perhaps faster RAM, or
> more likely the
> smaller screen real estate did the trick. In some
> ways, if you look at
> the Mac and the Lisa, the Lisa actually had
> something like 5 CPU's
> (68000, 6504, COPS, COPS in keyboard, and an
> optional AMD/TI FPU for the
> early I/O boards, and a Z8 in the Profile/Widget).
> The Mac had to
> rely entirely on the 68000.
>
> They could have added one more CPU just to do
> graphics, but, that would
> have added a lot more expense and complexity.
> Besides, in that sort of
> system, whenever the main CPU would need to transfer
> a big chunk of data
> to the graphics controller instead of just
> instructions that say, draw a
> line from this point to that point in this color,
> there would be a
> bottle neck there.
>
> Also, back then having a dedicated video processor
> didn't mean you could
> do graphics primitives with it. i.e. the chips did
> not have the silicon
> to draw lines, boxes, in "hires" bit mapped display
> modes. Rather the
> CPU had to do that work and there were various
> algorithms for it.
> QuickDraw just happened to be a better
> implementation that all of those. :-)
>
> I'm not sure many computers had video chips that
> could offload graphics
> work from the main CPU at that stage (i.e. hardware
> accelerated
> graphics), except maybe perhaps for the Amiga, but
> that came later on.
> Most were just good old fashioned frame buffers in
> bit mapped mode, and
> character generator based displays.
>
> --
> LisaList is sponsored by <http://lowendmac.com/>
> and...
>
> Shop buy.com and save.
> <http://lowendmac.com/ad/buy.com.html>
>
> Support Low End Mac
> <http://lowendmac.com/lists/support.html>
>
> LisaList info:
> <http://lowendmac.com/lists/lisa.html>
> --> AOL users, remove "mailto:"
> Send list messages to:
> <mailto:lisalist_at_email.domain.hidden>
> To unsubscribe, email:
> <mailto:lisalist-off_at_email.domain.hidden>
> For digest mode, email:
> <mailto:lisalist-digest_at_email.domain.hidden>
> Subscription questions:
> <mailto:low_end_mac_at_email.domain.hidden>
> Archive:
>
<http://www.mail-archive.com/lisalist%40mail.maclaunch.com/>
>
> iPod Accessories for Less
> at 1-800-iPOD.COM
> Fast Delivery, Low Price, Good Deal
> www.1800ipod.com
>
-- LisaList is sponsored by <http://lowendmac.com/> and... Shop buy.com and save. <http://lowendmac.com/ad/buy.com.html> Support Low End Mac <http://lowendmac.com/lists/support.html> LisaList info: <http://lowendmac.com/lists/lisa.html> --> AOL users, remove "mailto:" Send list messages to: <mailto:lisalist_at_email.domain.hidden> To unsubscribe, email: <mailto:lisalist-off_at_email.domain.hidden> For digest mode, email: <mailto:lisalist-digest_at_email.domain.hidden> Subscription questions: <mailto:low_end_mac_at_email.domain.hidden> Archive: <http://www.mail-archive.com/lisalist%40mail.maclaunch.com/> iPod Accessories for Less at 1-800-iPOD.COM Fast Delivery, Low Price, Good Deal www.1800ipod.comReceived on 2006-03-07 17:10:58
This archive was generated by hypermail 2.4.0 : 2020-01-13 12:15:20 EST