HP3000-L Archives

March 1995, Week 1

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jeff Kell <[log in to unmask]>
Reply To:
Jeff Kell <[log in to unmask]>
Date:
Sat, 4 Mar 1995 17:41:02 EST
Content-Type:
text/plain
Parts/Attachments:
text/plain (27 lines)
Just noticed a mistake in my original post...
 
>> There are THREE "modes" of terminal I/O on MPE/iX:
>>    * Block mode [...]
>>    * Line mode [...]
>>    * Character mode [...]
>>
>> In the latter case, yes, you can significantly impact an MPE/iX system.
>> In the center case, the DTC pre-processing has eliminated much of the host
>> overhead traditionally associated with block mode.
                                          ^^^^^-> should be character/line
And a brief addendum:
 
On the "ancient" 3000 systems there was a CPU interrupt per character in
character or line mode (and maybe even block) in the old Asynch Terminal
Controllers (ATCs).  The later classics had ADCCs and ATPs which offloaded
some overhead, but not that much.  The DTC and associated AFCP protocol
removed almost all of that overhead (unless you're doing single-byte reads
as noted - character mode).
 
Inbound telnet to a TAC in a DTC does the same thing - removing the character
overhead when possible (in block and line modes).  Host-based inbound telnet
will be "interesting" as it will return character-based overhead back to the
host.  Almost as if we've gone full-circle.
 
[\] Jeff Kell, [log in to unmask]

ATOM RSS1 RSS2