Michael Sokolov wrote:
> 
>    Jacob Suter <jsuter@intrastar.net> wrote:
> > Hell, on a 14.4k telnet its all the
> > same if its a PPro, a K6, or a VS3100.
> 
>    This is absolutely right! This is the whole point of professional
> retrocomputing.

I gotta get 1.3 DL'ed and running... I just seriously dislike the SCSI
controller in these m30's

> > > still true that most people on port-vax consider VAXen a retrocomputing
> > > diversion. I have seen lots of statements on port-vax like the
> > > following:
> >
> > Most, yes.
> 
>    OK, so you agree that most people here do hobbyist retrocomputing. Then
> how can their OS possibly be fit for professional retrocomputing?

Most people here are into hobbyist retrocomputing, but whats the real
difference between it and professional retrocomputing.  Do you think JKH
or Linus had any plan that their software would EVER be used
professionally?  Its just over time things improve (due to many people
working on it together for the greater good of the OS) to the point
where its usable in a business environment.

Commercial OS's overall seem to be of lower quality due to just this. 
They don't want/encourage others to improve their software.
 
> > >    - "I wonder how long will it take to, say, compile Lynx on
> > > NetBSD/vax? Two weeks?"
> >
> > 8 hours on a netbooted 3100m30...  took 10 on a 386DX/40 with its own
> > IDE drive, but half the ram.
> 
>    Something is definitely wrong here. I use a 80386DX-25 with 8 MB of RAM
> running DOS (means no swap), and although I don't have time to test it
> right now, there is no way it could take that long. How much RAM did your
> 80386 have? Was it running UNIX or DOS? How was it swapping? How much RAM
> did the VAX have? Was it swapping over the net?

My 386DX/40 box:

AMD 386DX/40
ISA only mobo w/ 128k cache (15ns)
cheap multi-io card (IDE, FDD, two 8250 serial, broken printer port)
Wang 512k VGA card
340mb Maxtor IDE from mid 1993
3.5" floppy (blah)
16 bit NE2000
FreeBSD 2.1.0 GENERIC kernel

100% self sustained, no NFS.  Had 3-4 shell sessions on it running
various things (BitchX irc, vi, jove, etc) and a small ircd.

My VS3100 test box:

VS3100 KA42-B (m30)
all NFS (see below)
16mb ram (one 8mb board, one 4mb board + 4mb onboard)
NetBSD 1.2G
A couple of very light shell sessions
swap on NFS, no apparent swapping occured
 
Boot server:

AMD 5x86/133
20mb ram
1gb WD EIDE, 2gb Maxtor IDE
PCI NE2000 card
FreeBSD 2.2.5 Personalized kernel
under light load, very light swapping.

> > Lusers that want to crash your box will NOT give a DAMN what you're
> > running.  You can hack/crash any OS on any platform if you know your
> > shit.  NetBSD/VAX is security by obscurity in reality.
> 
>    I disagree. NetBSD/vax is very unstable (especially on BabyVAXen), and
> therefore probably easy to crash, even unintentionally. You do something
> wrong, and here it goes. Up in flames.

Well of course.  Hell, Win95&NT is no better and look at how many people
run it (and pay for it, ugh!).  Gimmie my linux box and boink.c for a
few minutes....  :)

NetBSD/VAX has not been terribly stable (though I haven't tried
1.3-release yet) on my vs'es, but its getting there.  1.3-alpha was MUCH
better than 1.2G...

Jacob Suter
Intrastellar Internet Service