Their options were. 1. 64k slow dram + a single cheap static ram (which might have been free at one point as they had a surplus from the pet). 2. faster/more expensive 64k dram & vic 2 with double the clock speed (more heat/more failures etc), so that it could fetch the font/bitmap and the color on alternate cycles. 3. more expensive vic 2 that stores the color as well as the characters in a buffer, which would waste another 1/8th of the cpu (this is the approach that the ted machines used). You probably shouldn't question someones decisions if you don't understand the reason why they made them. "Tinkering with the address decoding logic" cannot do anything to get rid of the static ram. On 03/11/2021 21:37, Claudio Sánchez wrote: > > I think that was a solution of compromise so they could stop tinkering > with the address decoding logic. Why should it be a problem to have > color RAM mapped in DRAM in the first place? They had to include a > SRAM chip in *every* computer (later revisions would have it included > in that SHARP chip). >Received on 2021-11-04 10:00:02
Archive generated by hypermail 2.3.0.