[Coco] format problems

[Gene Heskett]

On Sunday 20 January 2019 11:26:12 Tormod Volden wrote:

> On Sun, Jan 20, 2019 at 3:00 PM Gene Heskett <gheskett at shentel.net> 
wrote:
> > gene at coyote:/opt/genewww/nitros9$ hg pull -u
> > pulling from http://hg.code.sf.net/p/nitros9/code
> > searching for changes
> > no changes found
> >
> > is that the correct address?
>
> Yes. If you run "hg tip" it should show a commit from me, timestamped
> Aug 29, although I only pushed it to the repo a week ago.
>
> Regards,
> Tormod

It sure does. and the subject being the wordpack cards which brings up a 
question: early in the original co80 code is a 16 byte data table, 
written to the cg chip in the wordpack, and I wondering if it uses the 
oem values yet, or if my kit of changes was used. They had the effect of 
giving it a 25 line display, about 1.5" larger overall and considerably 
more readable than the oem version which gave a 24 line postcard in the 
middle of a 13" monitor. Wasted over half of the monitors screen real 
estate, and burning into the screen in a year or so.

The oem version pretty much maintained the 15.75 kilohertz hscan rate, 
which the high voltage in a magnavoc pc80 amber screen monitor was 
designed to work at.  But with a 4x color subcarrier crystal, needed 
huge blanking timings to make it run that slow for hscan.

Knowing I could go faster than 15.75 without any danger of hurting the 
monitor, (you cannot go slower due the H.O.T. transformer core 
saturation which quickly breaks the mirror and lets all the smoke out of 
those expensive parts which now, 30 years later, are made out of pure 
unobtainium ) with the only side effect being a few percent loss of crt 
high voltage, which would increase the deflection sensitivty both ways, 
so I reduced the blanking time both before and after the line of text 
until the slightly dimmer display filled a much larger portion of the 
screen and was many times more readable than the oem postcard sized 
display in the middle of the screen, and since it had sufficient memory 
to contain the 25th line, adding that brought the vertical scan rate 
back to within a couple hertz of 60 cycles/sec while the hscan remained 
at about 18 kilohertz.  That, and a sheet of light steel between it and 
the 8cm515 color monitor brought the magnetic crosstalk wiggle between 
the two monitors down to a tolerable pixel or so.

I can supply a co80 with those changes. The higher frequency also reduced 
the heating in all those unreplaceable parts, which as a C.E.T., I also 
call a Good Thing.

Recommended by Grandpa Gene. But likely not compatible with all monitors. 
I've also hacked a db9 connector into mine so I could send the h & v 
syncs separately, getting rid of the windy image at the top.  We've 
likely got 25 guys here on this list that can do that.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>