Re: LZ: Zterm & Modems

[Prev][Next][Index][Thread]

Re: LZ: Zterm & Modems

Mel Tsai wrote:
> 
> On Sun, 06 Oct 1996 11:32:45 -0500, you wrote:
> >Jarkko Vääräniemi wrote:
> >>
> >> I've found simple modem chip from National Semiconductor, that could
> >> perhaps be hooked straight to link port. I have ordered free sample
> >> and will have a look at it if it could be used. Though it's only
> >> 300bps, but that'll be just fine for irc/mud/telnet etc..
> >> One problem with this could be timing, but perhaps the timer used
> >> with busy indicator could be used with this, or there is some other
> >> some solution. ;-)
> >> This chip is "MM74HC943", put that into search field in
> >> www.national.com and you should find it.
> >> -------
> 
> I saw that chip a few months ago, and it looks pretty good.  However,
> the reason I didn't look further is because the vast majority of BBS's
> and dial up access accounts won't even let you connect at 1200 any
> more, much less 300.  Therefore it would only be useful as a novelty
> item or a simple data transferr method to a remote computer.  Note
> that the datasheet doesn't give you specs on the RJ-11 connection, and
> I doubt you'd want a handset coupled connection.  You might have to do
> some fishing to find out how to connect this thing properly.
> 
> >even though i am not doubting your way, the graphlink-ADAPTER-external
> >modem way of connection has already been proven to work. Also the g-A-em
> >way goes at 9600 baud which is the fastest baud rate that the graphlink
> >can handle.
> >PCoMP
> >Visit Maximus Productions home page at: http://www.ntsource.com/~pcomp
> >
> 
> Actually, it is theoretically possible to transferr at much higher
> rates than 9600.  With an efficient algorythm you can probably get up
> to 14.4 speed.  If you figure 10 bits per byte, and 1.4k bytes per
> second, that leaves a cycle time of about 70 uS.  At "raw" speeds the
> TI-85 can probably get up to about 1 uS per cycle, and with real
> software control I estimate cycle times of 25-50 us, meaning that we
> can probably go even faster than 14.4, maybe 28.8 on turboed calcs.
> This is all speculation, though, as the program would be so complex
> that real send/receive tasks may not be able to keep up, but in
> "burst" mode it should work.
> 
> -Mel


What this would be good for though, is if you have a friend at the other
end of town and you want to transfer a program to him...That would be
completely awsome to be able to do that!
<pre>
-- 
 _
(_
|_)teve
|eterson


mailto:stevep@comso.com
http://www.geocities.com/SiliconValley/Park/2636
</pre>
References: