I had three 9-tracks I'd picked up a few years ago. None worked
out of the box; they collected dust. Last weekend I knew I'd pass
by http://www.comco-inc.com/ in Bettendorf, Iowa, one of the few
9-track sales and service places I'd found. I didn't have much
advance warning, so I just brought the drives with me and left them
at his door step because the shop was closed. I left an M4 9914,
an Overland Data 5622, and an HP 88780.
Diagnosis is $495, deductable from repairs if I proceed. Ouch!
He says he still sells "a few" 9-tracks a year. His offer to me
was a reconditioned HP 88780 for $1795 including manual, cleaning pads
and a scratch tape.
He said "BIG IDEA... Here's a wonderful chance to corner the 9-track
business: I'll sell you 1000 lbs of parts for $1,000 (FOB Bettendorf).
I'll even throw in graphics, manuals, etc. At the very least, you
will be able to build several drives. I am not kidding."
The building he's in has a commercial real-estate "for sale" sign
out front. Maybe he was a renter and he needs to move.
- John
And yes that does render them vintage dag nabbit!
I'm not too proud to accept donations (I'll pay the media rate postage of course). PO could give you a hard time (ooh they contain advertising, can't claim media rate. Kindly explain that 22 year old advertising doesn't count). I could pay a buck or so per though (except the real oldies, I'm sure I could do better). Just no onesies or twosies. I'll go barkrupt trying to complete my set.?
Hi all,
finally I got my real Iron delivered. I nearly can't express how happy I am
about that.
Now the usual Inspection begins (PSUs etc...)
Could please one of you have a look at my "board-layout" in the backplane
and correct me if I am completely wrong?
>From top downwards I would insert the modules as follows:
M8189 - 11/23 -CPU
M8067 - MSV-11 - 256 KW RAM
M8043 - DLV11 - Quad Serial Card
M8061 - RLV12 - RL02 Controller
M7516 - DELQA - Ethernet Card
Goal is to bring this System up with VTSERVER for starters. I hope to add
one or two RL02s in Summer. A RX50 and a RD50/51 or 52 (don't remember at
the moment) are sitting on my shelf too and I'd like to put them in there.
But first I have to get a controller for that.
This all sounds like quite an amount of work and a lot of fun for me :)
Regards,
Wolfgang - who wanted to share his happiness
--
Wolfgang Eichberger - OE5EWL
Operating System Collector
Blog: 5ewl.blogspot.com
Homepage: www.eichberger.org
On 2012-01-25 19:00, Holm Tiffe<holm at freibergnet.de> wrote:
> allison wrote:
>
>> > On 01/25/2012 06:40 AM, Holm Tiffe wrote:
>>> > >Hi,
>>> > >
>>> > >I have A M8186 (KDF11-A Rev D) and an M8047-CA (MXV11-AC) currently in
>>> > >an H9275 Backplane.
>>> > >I've fiddeled around an wondering why I can't access the ROMs on the M8047
>>> > >at 173000. Finally found it, they get active ad 773000 in the ODT, since
>>> > >the M8047 is using the BBS7 Signal from the CPU to decode the ROM Address.
>>> > >My question is now regarding the Bootstrap setting, the Manual says that in
>>> > >ODT mode the BDAL16 and BDAL17 are enabled to get access to the higher
>>> > >memory contents. How is this handeled while bootstrapping? I can't set a
>>> > >bootstrap address of 773000 on the KDF11, it boots now in ODT Mode and
>>> > >displays 173000 after Reset, thati's the jumpered address.
>>> > >How is this (BBS7) handeled while bootstrapping? Is the BBS7 low while
>>> > >bootstrap at 173000 or is the M8047 incompatible with this processor?
>>> > >(I've not tested to bootstrap since I don't know what is in the ROMs and I
>>> > >don't have an RX02 nor TU58 to connect, want to put in a M7555 so I have to
>>> > >change the ROMs for sure). As far as I know this KDF11 should be able to
>>> > >handle 22 bits (420HD0 on the Handle, 5013326 D-P4 on the Board) is that
>>> > >the case?
>> >
>> > Most all KDf11s are Q22 make sure both of the hybrid chips are there the
>> > second
>> > one has the MMU for the 22 bit mapping!
>> >
>> > Also make sure the backplane is Q22.
>> >
>> > Lastly the M8047 make sure its jumpered correctly. If the boot in it
>> > fails due to lack of device I think the report is 173003?.
>> >
>> > ODT only knows 16bit octal. And BBS-7 is activated when the upper three
>> > address bits are active for addresses greater than 160000 which is IO
>> > and boot address space.
> Yes, but I mean the difference between 773000 vs 173000.
> I see the ROMs in ODT at 773000 but not at 173000!
> Anyway, I'll jumpering the CPU to bootstrap and will look what happens.
Here is how it actually works:
When the CPU is addressing anything in memory, it goes through the MMU.
Always. However, the MMU can act in three different modes.
16-bit
18-bit
22-bit
In 22-bit mode, all CPU accesses are virtual addresses, and they get
translated by the MMU into other 22-bit addresses. Pretty straight forward.
In 18-bit mode, all CPU accesses are also virtual addresses, and they
get translated by the MMU into 22-bit addresses. Only part of the PAR
registers are used. However, memory access to addresses in the high 4kW
are mapped to the high end of the 22-bit address range, even though you
only have 18-bit addresses.
In 16-bit mode, all CPU accesses are physical addresses. However, the
MMU recognize addresses in the high 4kW of the address space, and move
those accesses to the high 4kW of the 22-bit address space.
Next is the BBS-7 signal, which is active if the physical address then
is above 17760000. Anything in the I/O page should never decode the
A21-A13. BBS-7 will be the logical AND of A21-A13.
In fact, it might be that BBS-7 is shortcircuited inside the MMU, and
the MMU don't even drive A21-A13 when you are in 16-bit or 18-bit mode.
Anyway, the ROMs are *not* at 173000, if you talk about 22-bit or 18-bit
addresses. They are at 173000 if you talk about 16-bit addresses only.
In short, they are at address 13000 in the I/O page. Wherever your I/O
page is. If you enable 22-bit addressing, and put the I/O page at page
0, the ROM memory will appear at virtual address 13000, really.
I hope this makes things clear. Otherwise feel free to ask. It's really
not that complicated.
>>> > >Next question about RAMs: I do have some QBUS Memory boards with PMI
>>> > >connectors (M6708, NatSemi NS638, Chrislin CI-MV8 to be concrete) does
>>> > >anyone know if I can use one of them with the KDF11 (or at least the
>>> > >KDJ11?) I hve a KA630 too, but I need PDP11 RAM, the 16KW on the M8047
>>> > >is really a bit small....
>>> > >
>> > PMI can be used with KDJ11 if its the right PMI. KA630 is not PDP11 but it
>> > is microVAX.
> Yes, know this. But what's the right PMI for the KDJ11?
> I have soldered an additional Megabyte to the KDJ11 and it has now 1,5MB,
> which is enough tu run 2.11BSD, therefore I don't have tried to use one of
> the memory Boards additionally until now.
No. It is not compatible.
The PMI memory for VAXen does not work the same way as PMI memory for
PDP-11s (even those that do have PMI memory).
VAX PMI uses a flat cable between the CPU and the memory boards. PDP-11
PMI memory uses signals in the CD slots to communicate with PMI memory.
In addition, for VAXen, PMI is used to be able to address more than 4 MB
as well as speed, while on the PDP-11 PMI memory only have PMI for the
speed gains, as a PDP-11 cannot address more than 4 MB anyway.
Also, the PDP-11 PMI memories works just fine as any normal Qbus memory.
It only acts as PMI memory under some specific circumstances, else it's
just normal Qbus memory. VAX PMI memory does not work as normal Qbus
memory under any circumstances, as far as I know.
If you want the designation for the KDJ11 PMI memories, just search
through the field guide that is on the internet.
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
Hi,
I have A M8186 (KDF11-A Rev D) and an M8047-CA (MXV11-AC) currently in
an H9275 Backplane.
I've fiddeled around an wondering why I can't access the ROMs on the M8047
at 173000. Finally found it, they get active ad 773000 in the ODT, since
the M8047 is using the BBS7 Signal from the CPU to decode the ROM Address.
My question is now regarding the Bootstrap setting, the Manual says that in
ODT mode the BDAL16 and BDAL17 are enabled to get access to the higher
memory contents. How is this handeled while bootstrapping? I can't set a
bootstrap address of 773000 on the KDF11, it boots now in ODT Mode and
displays 173000 after Reset, thati's the jumpered address.
How is this (BBS7) handeled while bootstrapping? Is the BBS7 low while
bootstrap at 173000 or is the M8047 incompatible with this processor?
(I've not tested to bootstrap since I don't know what is in the ROMs and I
don't have an RX02 nor TU58 to connect, want to put in a M7555 so I have to
change the ROMs for sure). As far as I know this KDF11 should be able to
handle 22 bits (420HD0 on the Handle, 5013326 D-P4 on the Board) is that
the case?
Next question about RAMs: I do have some QBUS Memory boards with PMI
connectors (M6708, NatSemi NS638, Chrislin CI-MV8 to be concrete) does
anyone know if I can use one of them with the KDF11 (or at least the
KDJ11?) I hve a KA630 too, but I need PDP11 RAM, the 16KW on the M8047
is really a bit small....
Next question is about an M7656 Graphic board that I have laying around
here. Where was this used and are there any docs available?
Regards,
Holm
--
Technik Service u. Handel Tiffe, www.tsht.de, Holm Tiffe,
Freiberger Stra?e 42, 09600 Obersch?na, USt-Id: DE253710583
www.tsht.de, info at tsht.de, Fax +49 3731 74200, Mobil: 0172 8790 741
Hmm, have people on here ever found a *sensible* way of defining our
interests such that it's obvious we're *not* interested in that x86 clone
PC running Windows that you bought in 2002 and have just unearthed in your
garage?
I've been trying to put the word about a bit locally that I'm interested in
rescuing vintage machines, but boy is it ever a struggle - sifting through
offers of "very big, old computers" that turn out to just be PCs running a
decade-old version of MS Windows is getting a little tiring (that's when
the "very big, old computer" doesn't turn out to be just a CRT monitor, of
course, which happened to me twice last week)
Maybe someone's cracked this particular nut and got the wording down just
right - or maybe everyone just lives with it. Setting a cut-off date of
1990 (say) wouldn't work, I think, because a) it might cut out a few
interesting non-PC/non-Apple items, and b) a lot of people out there really
seem to have no idea how *new* the item that they have is.
Idle thoughts for an afternoon, anyway...
cheers
Jules
