[Coco] Higher graphics mode (256 color mode?)

[James Ross]

> From: flexser at fiu.edu
> Subject: Re: [Coco] Higher graphics mode (256 color mode?)
>
> Do these sorts of weird undocumented graphics effects tend to work the same
> with an emulator? Just curious.

The only way I could see that happening is if the emulator was emulating at the electrical transistor / logic gates level.  In which case you would have to start off w/ an exact schematic of the digital gates that make up the integrated circuit you would be emulating.
 
Perhaps there is software that does that for engineers --- however, most emulators work at a much higher abstract level interpreting op-codes and hardware register command codes.  Any behavior, documented or not, has to be clearly understood by the emulator writer who writes an implementation in software code to imitate/emulate precisely the exact same effects / features on the host computer running the emulator.  

>
> Art
>
> On Wed, Jul 15, 2015 at 5:43 PM, William Astle <lost at l-w.ca> wrote:
>
>> I wonder if this isn't something similar to the "64 column VDG" screen I
>> detailed some time back, which works by setting the coco2 text screen
>> and fiddling with the GIME registers.
>>
>> If my findings from then hold for the "VDG" graphics modes, then setting
>> a particular bit in one of the GIME registers (I can't remember which
>> one off the top of my head) should double the horizontal resolution.
>> That would give a 512 pixel wide "pmode 4" screen. It could also involve
>> a mismatch between the "VDG" graphics mode and the "SAM" graphics mode
>> settings which only makes it look like 4 areas by virtue of running the
>> data fetch counter at the wrong rate so it fetches the same data
>> multiple times to make up a full pixel row.
>>
>> The only thing I doubt is that it is actually 1024 pixels across, based
>> simply on the characteristics of the NTSC signal. If someone could
>> demonstrate the mode, it would be interesting. I'd try some experiments
>> but my coco3 is in a storage locker on the other side of the city and I
>> have nowhere to set it up.
>>
>> On 2015-07-15 15:25, Arthur Flexser wrote:
>>> P.S I didn't mean for my last message to seem negative. This could be
>> an
>>> interesting contribution, if in fact the mode is truly 1024 pixels wide,
>>> even if there is no way to uncouple the 4 areas. Maybe could be used for
>>> special effects in a game in situations where it doesn't matter if each
>>> quarter of the screen is identical to the others, and assuming that the
>>> 1024 resolution is appreciably sharper than 512 pixels on any type of
>>> monitor usable with a CoCo.
>>>
>>> Art
>>>
>>> On Wed, Jul 15, 2015 at 5:08 PM, Arthur Flexser <flexser at fiu.edu> wrote:
>>>
>>>> A couple of questions occur to me:
>>>>
>>>> 1. You say the graphics are repeated 4 times across the screen. Could
>>>> there be any way in this mode to change one of the 4 areas without
>>>> similarly affecting the other three? (I tend to doubt it.)
>>>>
>>>> 2. Are we sure that this is really 1024 pixel resolution? It could be,
>>>> for example, that the 4 areas are in fact only reproducing every 2nd or
>>>> every 4th horizontal pixel of the original screen. Indeed, maybe the 4
>>>> areas are not entirely identical, but 1/4 resolution and each offset by
>> 1
>>>> pixel horizontally from the adjacent area to yield the same number of
>>>> pixels as were there in the original screen.
>>>>
>>>> 3. Why exactly might this have anything to do with a 256-color mode?
>>>>
>>>> Art
>>>>
>>>> On Wed, Jul 15, 2015 at 4:49 PM, Paul Wittkoski via Coco <
>>>> coco at maltedmedia.com> wrote:
>>>>
>>>>> Hi everyone,
>>>>>
>>>>> At one time, I found a higher graphics mode that is not documented with
>>>>> the
>>>>> CoCo 3. It was 1024 pixels across and the pictures in pmode graphics
>> could
>>>>> be seen 4 times across the screen. I am unable to pull the exact method
>>>>> from
>>>>> my computer since my 5.25" floppy drive blew a chip some time back.
>> But,
>>>>> someone may be able to find it again from what I remember what I did.
>> It
>>>>> may
>>>>> have been odd thinking at the time but I wondered if pmode and hscreen
>>>>> could
>>>>> be used at the same time. Following the machine code for pmode and
>> hscreen
>>>>> in the unravelled series I found a specific poke to turn on pmode and
>>>>> another one for hscreen. When I used them together on a cm-8 monitor it
>>>>> produce a screen with 1024 pixel resolution across. I am not sure if
>> this
>>>>> would also lead to the 256 color mode if someone messed around with it.
>>>>> Interesting enough, I have not ever heard anyone else describe what I
>> have
>>>>> done. Maybe that's how Tandy hid the 256 color mode so well; right
>> under
>>>>> our
>>>>> noses with pmode. I hope this information is useful to someone in some
>> way
>>>>> after all of these years.
>>>>>