At 15:12 18/09/98 +0200, you wrote:
>At 11:30 PM 9/16/98 +0000, you wrote:
>
>>-For most screens, there is a relationship between the exact VRAM
>>address, and the VRAM page/pixel-coordinates that is about what you
>>would expect. In some screen modes (I think Graphic 5 & 6 modes, that
>>is, SCREEN 6 & 7), this relationship is different, so the contents of
>>for instance 2 SCREEN 5-pages is not visible in one SCREEN 7-page,
>>but 're-arranged' somehow (does anyone know how exactly?). In SCREEN
>>8, this is 'back to normal'.
>
>I thought VRAM arrangements were different in SCREEN 7 & 8.
>
>The reason is in hardware. In SCREEN 7 & 8, more pixels per second are
>needed to generate the image. Because fast RAM is expensive, instead they
>used 2 slower RAM ICs as VRAM. In SCREEN 7 & 8, of every byte the
>odd-numbered bits come from one RAM IC, and the even-numbered bits come
>from the other RAM IC. In all other modes, all bits come from the same RAM
IC.
>
>This is also the reason why the V9938 databook says that SCREEN 7 & 8 are
>only available in 128K configurations. In theory, you could have SCREEN 7 &
>8 with just 64K VRAM, because one page takes 54K. But 64K VRAM means just
>one RAM IC, and that's why SCREEN 7 & 8 won't work on 64K VRAM.
>
>Note: I'm not sure I got every detail right, but this is the general idea.
The right way is to say that there are two banks of memory and each bank is
composed ot 2 chips of 64K by 4bits (so two chips gives you 64K 8bits)
Since the DRAM memory used for the bank was slow YAMAHA decided to switch
between two bank to get double speed of pixel rate (also called DOT SPEED
of something like that)
(your explanation was good for this)
Martial.
****
MSX Mailinglist. To unsubscribe, send an email to [EMAIL PROTECTED] and put
in the body (not subject) "unsubscribe msx [EMAIL PROTECTED]" (without the
quotes :-) Problems? contact [EMAIL PROTECTED] (www.stack.nl/~wiebe/mailinglist/)
****
