[pdp7-unix] Asymmetries and other faults in the Graphic 2 characters

Lars Brinkhoff lars at nocrew.org
Thu Oct 31 15:48:40 AEST 2019

Diomidis Spinellis wrote:
> Thank you; this is better, but still not perfect.  I left an example
> on the GitHub issue https://github.com/dspinellis/cas2svg/issues/1

Yes, I see you are right.  It's more obvious with your rendition than
the Graphic II simulation.

All I can say is, we don't know for sure exactly how the hardware
worked, or what the input to cas means, or the output from cas.  Some of
them are based on guesses.

The current version of the simulator makes one-pixel dots.  Maybe (as we
know is the case with the Type 340), dots were larger than this so that
small gaps would dissapear.

>> I haven't seen anything to indicate a bitmap.
> The character images in https://github.com/simh/simh/issues/754.
> They appear pixelated to me, rather than vector drawn.

The characters are draw with a hardware scaling factor of two.  Because
of this, there is a small gap between each pixel.  This is also a bit of
a guess, but the theory is that the character generator uses the
hardware "incremental mode", and that is works similarly to the Type

> "The method of picture formation on the CRT display is point plotting
> on a 1,024 X [,024 raster, i.e., lines and characters are formed from
> closely spaced points. "

Our understanding of the Type 340 is that it does plot discrete points
on a raster, and it seems the Graphic II works much the same.  Note that
even lines are draw this way using digital logic, rather than a smooth

More information about the pdp7-unix mailing list