🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: comp.sys.dec
Article #1857

Re: SUMMARY: VT1200

#1857
From: Kenneth Cleary
Date: Sun, 12 Aug 1990 13:59
39 lines
2970 bytes
After the hours of work lost due to VT1000 crashes, I'll be interested
to see how the VT1200 performs.

More interesting still, the VT1300 or VT2000, or whatever they call it.
It is supposed to be a stripped-down VAX running VAXeln.  This is not
a diskless workstation.  It is just one VAX chip dedicated to being an
X server.  They just better not try putting in the same chip as in the
VAX 2000 series.  Unless it has the 3800 series chip, it won't be much
better than the VT1000 (or the VS2000)  Even if it has that color
tube, a $6000-$7000 price tag won't be justified for an over-glorified
VS2000.  If they do use the 3800 chip, then that 3.8 VUPS will be worth
about an 11MIPS RISC chip. :-)
[Comparing a VS3100/38 with a DS2100: the VS is a bit slower for some DECW
actions, but it is satisfactorily competitive, even with running VMS.]
For this price range, you might arrange for a diskless DS2100, but then
all those MIPS are useless, when spending all your time paging.

Though I was only using unofficial benchmarking programs, like muncher and
maze, performance did not seem to matter whether I ran the CLIENT on a
VAX8550(6 VUPS) or a DECsystem5400(16 MIPS).  Performance seemed to depend
more on the display device.  VT1000 was slowest.  A local-paging VS3100/38
w/16MB was better, and roughly equivalent to a Tektronix XN7.  A local-
paging DS2100 was a little better still.  As soon as you iconify those
maze & muncher & plaid windows, watch CPU usage climb to 90%, since the
drawing bottleneck is removed :-)  Client & server were running on different
CPU's, and those cases with both on same, did not seem to make a big
difference.

Finally, the VT1000 was beta test, no question in my mind.  It certainly
has some advantages over other devices, if they can just make it what it
promises to be.  I've grown to love its features.  Now, just get rid of
those bugs!  Plan on figuring out BDF-to-SNF conversion or whatever you
need to do if you buy a non-DEC server, and need to make sure you have
all the MIT fonts (not to mention font-alias files).  A certain other
manufacturer shipped only a partial set of fonts, and I did not find out
until I tried running DECWindows apps that tried calling standard fonts.
The ones that did run looked like SH*T.  Others just crashed.  No such
problems with the VT1000, and it had no problem displaying non-DEC apps
running on both UNIX & VMS concurrently.  DEC has done good integration...
Technical Details
Message-ID: <3838@husc6.harvard.edu>
Path: arc1-ca.pugleaf.net!utzoo!attcan!utgpu!news-server.csri.toronto.edu!clyde.concordia.ca!uunet!snorkelwacker!husc6!husc9!cleary
References: <1990Aug10.135729.17631@wam.umd.edu>

Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!utzoo!attcan!utgpu!news-server.csri.toronto.edu!clyde.concordia.ca!uunet!snorkelwacker!husc6!husc9!cleary
From: cleary@husc9.harvard.edu (Kenneth Cleary)
Newsgroups: comp.sys.dec
Subject: Re: SUMMARY: VT1200
Message-ID: <3838@husc6.harvard.edu>
Date: 12 Aug 90 13:59:47 GMT
References: <1990Aug10.135729.17631@wam.umd.edu>
Sender: news@husc6.harvard.edu
Reply-To: cleary@husc9.UUCP (Kenneth Cleary)
Organization: Harvard Arts & Sciences Computer Services
Lines: 39