8251 troubles

[Allison]

>
>Subject: RE: 8251 troubles
>   From: "Chuck Guzis" <cclist at sydex.com>
>   Date: Mon, 28 Apr 2008 11:22:58 -0700
>     To: cctalk at classiccmp.org
>
>> Date: Sun, 27 Apr 2008 15:41:49 -0400
>> From: Allison 
> 
>> I have both in quantity.  Generally it's the quirks of the 
>> "improved part"that are annoying but for simple Async serial IO
>> they are identical enough.
>
>It could be that simple polled-mode async I/O works well enough, but 
>when I was trying to get a bisync package working with the 8251, I 
>wanted to chew my own arm off.  Among the issues (other than the 
>occasional device "hang") that I had was the lack of double-buffering 
>and the indeterminate control logic (it was possible to disable the 
>transmitter in the middle of a character).  The 8251A was *much* 
>better in that respect.
>

For async its ok bisync, you gotta be crazy.  While the A helped 
bisync it broke the CTS/ CTS to inactive during a TX meant the 
character would be resent!  The Fix was for both A and non A
is a gate to interlock the external CTS/ with TXE.

>In any case, Dwight has the NEC part which renders the topic moot.

Not really as it had the exact same bugs.

>
>I found at the time that the Signetics 2651 was a far better chip for 
>most utility purposes than the 8251, having the benefit of an 
>internal baud-rate generator for async operation. 

It was better.

Allison


>Cheers,
>Chuck