[Simh] RSTS/E 10.1-L and Paper tape

[Johnny Billquist]

On 2016-01-07 03:08, Mark Pizzolato wrote:
> On Wednesday, January 6, 2016 at 4:24 PM, Johnny Billquist wrote:
>> On 2016-01-07 01:00, Mark Pizzolato wrote:
>>> Meanwhile, the 'crude' way to exchange data on most simulators can
>>> actually be done with cut and paste in console or telnet sessions.
>>
>> Indeed. That also works. But it is really problematic when copying data into
>> the simulator, since this goes through the OS terminal driver, which have
>> limitations. Pasting in data often leads to data loss in my experience. But
>> copying out works fine.
>
> Actually, copying data into a simulator with paste has improved vastly recently.
> 10's of Kbytes of data pasted at once without data loss.  Give it a try.

Uh? Really. How recent do I need it to be then?
I'm on:
  PDP-11 simulator V4.0-0 Beta        git commit id: e9b312f2

Pasting the following:
---
.w

RSX-11M-PLUS BL88   (JOCKE ) 2044K PDP-11/94 at 02:41:45.0, 7-JAN-16
Users=5  Idle=95%  Task=KERT5   Up: 0 days, 08:05.  Ticks=6  Errseq=0.

user             task      uic     pri  where  state  size(K) ioc     par
J Billquist     EMAT2    [ Self  ]  50  _TT2  WaitFor  5/31          GEN
J Billquist     EMAT3    [ Self  ]  50  _TT3  WaitFor  5/31          GEN
J Billquist     WHOT4  * [ Self  ]  50  _TT4  Running  1/4           GEN
J Billquist     KERT5    [ Self  ]  50  _TT5  Running  11/72         GEN
J Billquist    (In CCL)  [ Self  ]      _TT1
---

Into another terminal gives:
---
.pip nl:=ti:
J Billquist     EMAT2    [ Self  ]  50  _TT2  WaitFor  5/31          GEN
J Billquist     EMAT3    [ Self  ]  50  _TT3  WaitFor  5/31          GEN
J Billquist     WHOT4  * [ Self  ]  50  _TT4  Running  1/4           GEN
J Billquist     KERT5
---

I doubt simh can do much more, except if I manage to get some delay 
involved, as it looks like the RSX terminal driver gets overflowed.

The end result is the same, though. You basically can't paste that much 
into simh.

>> But I still maintain that KERMIT is really the easiest solution to start with.
>
> Given that the user can get beyond the chicken and egg problem of getting
> Kermit on the simulated system to start with...

Yup, there is still that problem. The only "comfort" is that if you can 
just transfer one chunk of printable text into the host, then you're 
good. KERMIT have a bootstrap procedure that only needs transfer of two 
text files to get started...

	Johnny

-- 
Johnny Billquist                  || "I'm on a bus
                                   ||  on a psychedelic trip
email: bqt at softjar.se             ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol