Re: Reccomending a monitor?

[mjmahon@aol.com (Michael J. Mahon)]

David Stark replied (top-posting corrected):

>Michael J. Mahon wrote:
>> David Stark wrote:
>> 
>> 
>>>Many TVs today have higher resolutions than what was available when your 
>>>Apple ][ was in its hey day.  My Wega has a couple of high definition 
>>>modes.  I would bet that it would work with an 80-column card or other 
>>>graphics card that you might have (provided that the output is still 
>>>composite NTSC)
>> 
>> 
>> Although the comb filters of modern sets allow a somewnat greater
>> luminance bandwidth than the older sets, they do not allow enough
>> bandwidth to produce a good 80-column display.
>> 
>> All NTSC TV sets must roll off the video passband around 4MHz,
>> to avoid getting the sound subcarrier (at 4.5MHz) in the video.
>> 
>> Older sets rolled off before 3MHz, to keep the 3.58MHz color
>> subcarrier from being annoyingly visible.
>> 
>> A sharp 80-column display, with isolated dots about as bright
>> as horizontally adjacent dots, requires bandwidth in excess of
>> 10MHz.  (In fact, the dot frequency of 80-column pixels is 14MHz.)
>> 
>> 
>>>Might be fun to hook your old Apple up to a 32" screen and load up some 
>>>of those old arcade games.
>> 
>> 
>> They will loolk great, since they were designed to be viewed on
>> composite monitors or TVs.
>
>The problem that you are describing holds for all NTSC composit signals, 
>so it would also hold for color composit monitors.

Not _all_.  Some composite monitors sense the absence of the
color burst signal in a monochrome signal and switch off their
chroma filter, allowing a much greater luminance bandwidth.

The Applecolor Composite monitor is one of these.  It provides
good support for 80-column text, whose quality is limited only
by the pitch of its shadow mask.

>So what you are arguing is that an 80-column card could not produce a 
>composit signal (I agree with this).  It must be either be black and 
>white, or RGB (three seperate signals).  If it is black and white, then 
>I think the component inputs should work (just use the Y input, the Pb 
>nad Pr aren't being generated anyway).  On high end sets these are 
>designed to take high definition signals, and 1080 overscan is higher 
>resolution than an Apple 80-column card (560x192?)  But like I said, you 
>could probably get a cheaper monitor.  But I like the idea of hooking it 
>up to a wide screen :-)

Yes, the luminance input should support much higher bandwidth
for a monochrome signal.

>The other issue is if the 80 columns are interlaced or not.  The 
>standard composit output from the Apple is 30 frames (not 60) 
>interlaced.

No, the Apple II video is 60 fps, non-interlaced.

>  The 80 column card may assume a 60 frame/sec non-interlaced 
>monitor.  If it operates at 30 frames/sec interlaced then the minimum 
>bandwidth required should be around 5 MHz (assuming 560 pixel --- you 
>have twice as much time per scan line).  Of course higher bandwidths 
>will give a crisper image.  The point is you might have to find a TV 
>takes progressive inputs if you wanted to do this, and you would have to 
>make sure the frames rates match.

The vertical scan rate does not determine the bandwidth required--
it determines the number of horizontal lines.

The duration of a horizontal line is a fraction of the inverse of the
horizontal scan frequency, and the number of video transitions
across a line determines the horizontal resolution.

>It could also be the case that the component inputs are assuming a 
>different number of scan lines than the usual 525.  The card would have 
>to agree with this as well.

Don't worry--all 80-column cards produce frame rates which are
well within the lock range of a nominal 60Hz monitor.  And all that
I know of use non-interlaced video, since the display of characters
with single-line features creates annoying flicker if an interlaced
display is used without an unusually long-persistence phosphor.

-michael

Check out amazing quality sound for 8-bit Apples on my
Home page:  http://members.aol.com/MJMahon/