[Coco] LCD TV with CoCo 3

[Gene Heskett]

On Wednesday 10 February 2010, Mike Pepe wrote:
>> -----Original Message-----
>> If it has an rca jack, it probably will work as well as composite ever
>> did on a coco3, which means the 80 column screen modes aren't really
>> usable because the text is to smeared when its squeezed into NTSC
>> bandwidth molds by using composite.
>>
>> One should bear in mind too, that even if such a tv/monitor claims a
>> triple comb filter, that this feature isn't very helpful for the
>> composite output of the coco since it is _not_ color interlaced as a true
>> NTSC signal would be.
>>
>> >I will be looking for one that also has a VGA connector, so that
>> >someday I could use the RGB to VGA converter with it.
>>
>> You would be better off to plan on doing that right up front, the results
>> will be much better visually.  I have a 22" Samsung HDTV with a db15
>> input that lost its tuner, but the rest of it still works like brand new.
>>  My milling machines control interface is absolutely stunning on a
>> 1680x1050 screen, but it won't fit in the monitor cubby hole so I'm stuck
>> with a
>> 1280x1024 screen on a 15" crt.  Bummer, but I get used to it.
>
>One of these days I'm going to wrap my head around the problem of the
> analog RGB to component video conversion. Then you could use the CoCo3 on
> any HDTV, probably with pretty good results.
>
Mike, that will need to back up and capture the data before it actually gets 
to the gime.  Only that way can adequate bandwidth be maintained.

Maybe there are faster gimes out there than mine, but frankly its video 
output rise and fall times are entirely too slow by a factor of at least 10, 
taking a blue signal a minimum of 6 dot clocks to reach its final voltage, 
which isn't much for the blue in particular since they did the amplitude 
matrixing in the gime proper, with the peak voltage delta at its output being 
only .7 volts for green.  Peak blue output for my gime is hard put to be more 
than .1 volts p-p.  You simply cannot get it up out of the noise with only 
that minimal level to start with.

We would need to read the data from memory, using the same clock the gime 
uses, and then feed it through a triplet of 2 bit d/a convertors with less 
than a dot clocks (14 mhz) rise & fall times, do also a 3x2 matrix on the 
same data to recover the luminance, and then do an analog multiply of all 3 
channels to get the brightness back.  That would, IMO, be much better done in 
the digital domain though.

The gime was a rather tour-de-force for the coco at the time in '86, but time 
has proven, at least to me, that it was a hugely power compromised design 
from the gitgo.  My guess is that the power budget was dictated by the 
refusal to do anything to improve the coco power supply's output capability, 
which on the 3, is taxed to the max with what they did do.  Virtually 
anything that we do, will require more psu, preferably a switcher that 
doesn't waste 30 watts as heat.  Even TI, in the 99/4A, had 3 or 4 x the 
coco's power available, with only 10% of the heat lost.  Unforch, while that 
psu was available on the surplus market for peanuts back then, now we'll need 
to locate a newer one.

Has anyone figured out how we could develop a detector that could watch the 
gime's outputs, and strobe our circuitry to grab that byte wide bit of data 
the gime gets when it is going to make video out of it?  That is the starting 
point to getting us truly 'eye candy' quality video IMO.

Unforch, while I have the schematics, I don't have the waveforms & timings, 
hence that question above.

-- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)

It is the wise bird who builds his nest in a tree.