Subject: Re: MV2 speeds on various NetBSD VAX releases
To: Ken Wellsch <kwellsch@tampabay.rr.com>
From: NetBSD Bob <nbsdbob@weedcon1.cropsci.ncsu.edu>
List: port-vax
Date: 10/31/2000 10:45:06
> > Where can I get a load of the 1.0A or 1.1B for the VAX?
> > I have not seen that anywhere.  I would like to try those for
> > comparisons, too, and keep a set around for historical reasons.
> 
> I happened to have a snapshot release of 1.0A stashed on an old account
> archive of mine, and the 1.1B just happened to be where I'd left this
> particular uVAX-II when I stopped having time to do development work.
> 
> But I thought it worth testing them as more sample points seem better.
> 
> Maybe it would be better to solicit official release copies?  I just
> didn't have any around that I could find quickly.

OK, I only have 4 drives in the beast, so Ultrix runs on one, then
1.2, 1.3.2, 1.5B1 on the other three.

IF you would not mind passing along the 1.0A thing, I would appreciate
it, for the record.  I try to keep a spare set of anything VAX related
for play and historical reasons.  It is good for my learning.

.....


> > This is just about what I was getting, it would seem.  That is
> > about a 35 percent slowdown difference roughly between 1.3.2 and 1.5?
> 
> Why did you drop the 1.4.x release data?  I'm confused why you didn't
> instead compare 1.4.2 with 1.5_BETA?  The 1.4.2 release seemed to
> continue a trend of improved performance, and in fact demonstrated
> the best performance I encountered while testing.

I did not have room in HD's to run up the 1.4.2, so I opted for the
end of the previous line 1.3.2.  I may pull the TK50, and drop in one
more HD (the controller will support 6), and run up a 1.4.2 there.
If I can find room to mount more drives in, in half height size,
I may lash up 6 drives internally so I can run up one on each different
release, plus 4.3BSD and Ultrix.  I initially figured the 1.5B1 would
be better than the 1.4.2, so loaded it up instead of 1.4.2.  I did
load 1.4.1 initially, but it kept giving me halt core dumps so I took
it off, in favor of the 1.5B1.

The 1.2 suite seemed rather zippy, and did not have the halt core dumps
I get in the 1.3.2 suite or the 1.4.1 suite.  I did not find its kernel
though, and had to bring up a 1.3 kernel instead, then compile a 1.2
generic kernel.  Tonight I will strip it down and compile the stripped
kernel I am using for comparisons.

.....

> Well, a uVAX-II is a uVAX-II.  A 0.9 VUP machine.  Slow.  About all
> I can offer is to say from my past recollections that DEC always advised
> putting the DEQNA as close to the CPU as possible.

I have heard the ethernet and the tape should go as far up as possible.
Is there any particular preference to the ordering between those two
cards?  I would tend to give preference to the ethernet, but I do write
a lot of tapes, too.  I have noticed some sensitivity in the NetBSD
TK50 tape drivers that give rise to errors, although they seem to be
only timing errors, since the tape keeps writing, most of the time.
Ultrix, on the other hand, has no such problems.

> In your case, I'd not give a hoot about where the DHV lives as long
> as you place it properly to provide grant continuity, driving a printer
> is not exactly major stuff.  Presuming you're not planning to print
> 24 hours a day 8-)

Well, my thought was 1 port for a printer somewhere might be nice and
1 port for an emergency spare login terminal if I want to get dualheaded.
I figured a postscript dumper ought to be fine for troff/TeX work on
the old critter, if nothing else.  If I could get a reasonably steady
1200 baud out, that would run most old postscript printers of the early
apple/HP kind, relatively well, and in idiot mode with no handshake.

> While disk controllers, or at least the third-party kind, I think tend
> to have sufficient buffering capability to survive longer interrupt
> processing latencies.

OK, I have both the SCSI SQ706A (in the BA123 MVII) and an ESDI DQ696-15
(in the BA23 box MVII), and have always put them last on the bus.
Is there anything strange about either of those cards?  I have no info
or data or manuals on them, and the company does not seem to either.
I sent them an email a week or so back, and got nothing in reply.
If anyone in the group does have manuals for the SQ706A or the DQ696-15
cards, I would appreciate a xerox of the manuals.

My BA23 MVII has an early cpu and 2 meg ram board and an 8 meg ram board
third party.  Alas, it only has one 300mb esdi drive, and I need to go
dredging in my storage bins for a 700mb esdi drive to stick in it to
get some space.  IFF I get it loading satisfactorily, I may pull the
TK50 and add in a second esdi drive.  Are there any heating problems
in replacing a TK50 with a second esdi drive in a BA23 cabinet?  My
expectations are that they should be rather similar, and removing the
TK50 card will cool it down some, too.

> Evil cards like the DJV which can totally overwhelm a CPU with interrupts
> are best kept for entertainment sake 8-)  You'll want a PDP-11 CPU for
> one of those as it actually has a very fine reputation for handling
> the interrupts in a "timely fashion." 8-)

Well, one of my goals is to find a PDP-11 in the BA23 box, for the fun
of it, and see if I can get a 2.11BSD up.  Someday.....

Also, I want to find a 9 track reel tape for it (the table-top flat one
that I think is a TU81 or something like that) and and a Qbus interface
card so I can write some antique 9 track unix tapes.  Someday.....
 
Thanks, Ken,

Bob