> > The VAXstation 4000/90 is a 42 VUP system and the VAX 4000/400 is a
> > 16 VUP system, so you aren't doing too bad.
> The VS4k9 is around 24 VUPs. It compiles quite exact twice as fast as my
> 4000/60 that is rated at 12 VUPs. I have seen many confusing ratings for
> the VS4k90. It is a nice desktop machine. Especially as mine has 128MB
> RAM. But it is a joke to that VAX7650. Damn. I would sell my grandma
> for that machine. ;-)
I have a couple of VAX 7730 systems at work. They're pretty nice, and
they each have 2GB of RAM.
--
Eric Dittman
dittman(a)dittman.net
Check out the DEC Ethusiasts Club at http://www.dittman.net/
> > > The VAXstation 4000/90 is a 42 VUP system and the VAX 4000/400 is a
> > > 16 VUP system, so you aren't doing too bad. I seem to recall the
> > > VAX 4000/700A is about 40 VUPs and the /705A is about 45-50 VUPs.
> >
> > The VAX CPU Summary lists these numbers:
> >
> > VS4000/90 32.8 VUPS
> > 4000/700a 40 VUPS
> > 4000/705a 45 VUPS
>
> The numbers I have are from DSN, and the 42 VUP rating for the 4000/90 feels
> right to me.
The "VAX CPU Summary" numbers above are their SPECint92 numbers, NOT their
VUP rating. I'm glad to finally see VUP ratings for these systems. Eric, I
don't suppose you've also got any kind of full list from DSN? If not do you
happen to have info on the VS4000/90A and /96?
Zane
> > > > Actually that rack contains two VAX 4000/700As.
> > > Ahhhh! I get jealous reading that. The 4k VAXen are quite nice and the
> > > 700A has around 33VUPs. My best VAXen are a 4k400 and a VS4k90.
> >
> > The VAXstation 4000/90 is a 42 VUP system and the VAX 4000/400 is a
> > 16 VUP system, so you aren't doing too bad. I seem to recall the
> > VAX 4000/700A is about 40 VUPs and the /705A is about 45-50 VUPs.
>
> The VAX CPU Summary lists these numbers:
>
> VS4000/90 32.8 VUPS
> 4000/700a 40 VUPS
> 4000/705a 45 VUPS
The numbers I have are from DSN, and the 42 VUP rating for the 4000/90 feels
right to me.
> > I would rather go with a QBUS SCSI or HSDxx controller. Once I find
> > one I"ll get rid of my DSSI disks.
>
> Get rid of 'em in my direction, please. 8-)
Okay, when I do you've got first shot.
--
Eric Dittman
dittman(a)dittman.net
Check out the DEC Ethusiasts Club at http://www.dittman.net/
On September 19, Eric Dittman wrote:
> > > Actually that rack contains two VAX 4000/700As.
> > Ahhhh! I get jealous reading that. The 4k VAXen are quite nice and the
> > 700A has around 33VUPs. My best VAXen are a 4k400 and a VS4k90.
>
> The VAXstation 4000/90 is a 42 VUP system and the VAX 4000/400 is a
> 16 VUP system, so you aren't doing too bad. I seem to recall the
> VAX 4000/700A is about 40 VUPs and the /705A is about 45-50 VUPs.
The VAX CPU Summary lists these numbers:
VS4000/90 32.8 VUPS
4000/700a 40 VUPS
4000/705a 45 VUPS
> I would rather go with a QBUS SCSI or HSDxx controller. Once I find
> one I"ll get rid of my DSSI disks.
Get rid of 'em in my direction, please. 8-)
-Dave
--
Dave McGuire
Laurel, MD
On September 19, jkunz(a)unixag-kl.fh-kl.de wrote:
> > The 8700 and the 8600 are very different machines. The 8700 uses
> > VAXBI, and is mounted in a chasiss similar in size to that of the
> > 8600. I don't think there's any XMI in there.
> >
> > Incidentally, the 8700 can be turned into an 8800 by plugging a
> > second CPU card set into the machine.
> Hmmm. I got confused by:
> http://www.de.netbsd.org/Documentation/Hardware/Machines/DEC/vax/sections.h…
>
> There is the 8600 (aka 11/790) listed as a SBI machine.
> There is no reference to the 8700.
> The 8800 is listed as:
> Basically an 8550 processor in a bigger box with space for additional
> processors to make it into an 8820 or 8840
> The 8550 is listed:
> XMI processor/memory backplane with VAXBI I/O
>
> According to http://anacin.nsc.vcu.edu/~jim/mvax/vax-perf.html the 8k
> VAXen, except the 8600, are pure VAXBI, no XMI. XMI came with the 6k
> VAXen.
The only 8K machines that I have personal experience with are the
8200, 8250, and 8350. They're definitely all pure BI machines...the
CPU and memory are BI cards. The 8700 has BI busses for I/O but the
CPU (made up of multiple large boards) is definitely *not* BI.
> So I assume the "base" machine is the 8550. In a biger chasiss it is
> named 8700
I've never seen the innards of an 8550, but I can tell you that the
8700 processor is several very large non-XMI boards...maybe six boards
in all, about twice the size of XMI boards, that use ZIF connectors
similar to those used in XMI and BI.
The VAX CPU Summary lists both the 8550 and 8700 as 6 VUP machines.
> and with a second CPU the 8700 is called 8800?
Yes, I can confirm this.
-Dave
--
Dave McGuire
Laurel, MD
To all those that wanted to see the pictures of Brian Hechinger and I
unloading a truckload of IBM S/390 and DEC VAX equipment (you know who you
are), the pictures are up on
http://www.users.cloud9.net/~sridhar/mainframe/
Peace... Sridhar
> On September 19, jkunz(a)unixag-kl.fh-kl.de wrote:
> > BTW a stupid question: What is a 8700? Is it SBI based like the
> > 11/78[05] and 86[05]0, or is it XMI / VAXBI like the 85x0 or 6k VAXen?
> > I asume that a 8700 is a 8600 in a biger enclosure?
>
> The 8700 and the 8600 are very different machines. The 8700 uses
> VAXBI, and is mounted in a chasiss similar in size to that of the
> 8600. I don't think there's any XMI in there.
>
> Incidentally, the 8700 can be turned into an 8800 by plugging a
> second CPU card set into the machine.
The 8700/8800 systems were changed in designation about half-way through
their production. The 8700 became the 8810 and the 8800 became the 8820.
There was also another set of 88x0 systems that were different internally
in that they could be upgraded to more CPUs. All to confuse the customer,
I guess.
Does the 8700 have the VAX Console?
--
Eric Dittman
dittman(a)dittman.net
Check out the DEC Ethusiasts Club at http://www.dittman.net/
> > I missed the 4000 due to the shadow and the BA215 as I didn't scroll all
> > the way down. More nice toys.
> Very nice toys. I have a MV4k200 in a BA215. A very handy QBus VAX and
> it is nice to work on the BA215. No finger breaking and flor crawling
> like on the flor stand BA23.
Even worse is a rack-mount dual BA23. I used to have to pull all the
cards out and reseat them on one system. The quad-width cards had locks,
but the dual-width didn't, and I couldn't easily get to the dual-width
cards to properly reseat them with the quad-width cards in place. I
really like the BA213 and BA440. I have a 4000/500 in a BA440 at home.
> > As someone that cares for an 8820 at work,
> A 8820 still in production use?!
> I know from ragge, the NetBSD/VAX port-master, that there is a 8800 in
> north Sweden runing. AFAIK it is the machine behind ftp.luth.se, as the
> folowing picture implies:
> http://www.ludd.luth.se/gallery/ludd1-mII/ftp.luth.se_mark_IV-4.jpg
Yes, indeed. All because of licensing costs. The 8820 is their
slowest system and all it does is run that one piece of software.
Never mind the environmental and service costs are more than a
new system and license.
> > I really think the 8700 is nuts, but fun.
> Nuts are fun. ;-)
Yeah, people think I'm a nut, and my computer collection is not
nearly as large as it should be.
--
Eric Dittman
dittman(a)dittman.net
Check out the DEC Ethusiasts Club at http://www.dittman.net/
I'm rebuilding an RM03 head stack (I have two crashed drives, making one
good one from the two), and I got a problem here...
Drive one has all of the heads bad except for platter 2, head 1, and
surface 3, head two. (That is, I have one good UP head, and one good DOWN
head.)
Drive two has all of the heads good, except for the heads for platter 1.
That is, I have both the UP and DOWN heads for the topmost platter dead.
The problem is, the good UP head from the first drive is the servo head,
apparently - The pigtails for this head come out on the opposite side from
the others, and it has a double-ended connection. Can this head be used
as a normal data head anyway? The pinout looks the same, except that it
has two endings, both of which look alike...
Can someone look at the RM03 prints and find this out for me? This is my
only good UP head, so if I roach it, I'm done for.
PS: Ripping the drive apart and getting the head stack out to play with
was somewhat easier than it looked like; All I had to do was get the big
magnet out of the back, have someone hold the logic cage while I undid the
drive head pigtails, undo the track 0 microswitch, disconnect the big coil
at the end of the head assembly, and pull the whole mess out the back of
the drive.
Oh, and I'll have to take pictures of the crashed heads later; They're
really impressive! The original top head for the second drive is
COMPLETELY GONE. I have both the head and the platter it crashed onto
stored upstairs to take pictures of as soon as the camera works again.
