[Simh] Rainbow100

[Johnny Billquist]

On 2017-07-22 17:15, Johnny Billquist wrote:
> On 2017-07-21 18:08, Timothe Litt wrote:
>> The 240 got rid of the Intel uPs for both technical and political
>> reasons.  The one T11 had to handle everything, including scan
>> interrupts.  This probably explains why it seemed slow - but I didn't
>> have to dig into the 240 internals to the same extent as the VT100.
>
> Not really. The VT240 have both a T11 and an 8085. (Not to mention the
> 8051 in the keyboard.) I can't really explain why it was so slow, but
> back at the time, people were talking about the interaction between the
> 8085 and the T11, and a bad system design, as a reason.

I decided to read through the system manual a little on the VT240, out 
of curiosity.

The T11 is the central processing. All external connections go into the 
T11. The T11 is then also responsible for all graphic handling. However, 
all text handling is done by the 8085. So what you get is that serial 
data is received by the T11, who has to do some processing and parsing, 
to see if this might be graphics stuff (REGIS, 4014). If not, it will 
send it over a communications channel to the 8085, who will then parse 
it again, to process the text mode stuff.
Both the T11 and the 8085 have access to the video controller, but only 
one at a time can access it, so there is a handshake between them for 
playing with the video. And the 8085 only work on the video at the 
blanking period of the video generation.

I think the overhead of having the T11 first process, and then pass the 
data on to the 8085 (over a channel which speed I'm not sure of), in 
combination with a shared access to the video controller itself, 
probably is a reason for the (bad) speed of the VT240.

	Johnny

-- 
Johnny Billquist                  || "I'm on a bus
                                   ||  on a psychedelic trip
email: bqt at softjar.se             ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol